Finley MI 1974 Studies in Ancient Society

RoutledgeRevivals

Studies in Ancient Society

Originally publishedin 1978, this volume comprisesarticles previously publishedin the historical journal, Past and Present,ranging over nearly a thousandyears of Graeco-Romanhistory. The essaysfocus primarily on the RomanEmpire, reflecting the increase,in British scholarshipof the post-war years, of explanatory, 'structuralist' studies of this period in Roman history. The topics treated include Athenian politics, the Romanconquestof the east,violence in the later Roman Republic, the secondSophistic, and persecutionsof the early Christians.The authors have all producedoriginal studies,a number of which have generated significant researchby other ancient historians.

'The papers vary from the interesting to the very good indeed and are all well argued.What they have in commonis a concern with theoretical and practical systems,political, social and economic and with relatedattitudesof mind.' - Ruth Padel, The Times EducationalSupplement 'As ProfessorFinley points out in his introduction, the authors are concerned not only with the explanation of change and movement,but with a perspectivewhich pays close attention to the complex and, to many, esoteric social structureof the classical world. It is the last point which will make this collection appealto many more than ancient historians.'- Economist

Studies in Ancient Society

Edited by M. I. Finley

(~ ;1 ...j\) Routledge ~

Iil \ .

~~,V~"

Taylor&FrancisGroup

First publishedin 1974 by Routledge& Kegan Paul Ltd This edition first publishedin 2011 by Routledge 2 Park Square,Milton Park, Abingdon, Oxon, OX14 4RN Simultaneouslypublishedin the USA and Canada by Routledge 711 Third Avenue, New York, NY 100l7

Routledgeis an imprint of the Taylor & Francis Group, an informa business © 1974 Pastand PresentSociety All rights reserved. No part of this book may be reprinted or reproducedor utilised in any form or by any electronic, mechanical, or other means, now known or hereafterinvented, including photocopyingand recording, or in any information storage or retrieval system, without permission in writing from the publishers. Publisher'sNote The publisherhas gone to great lengthsto ensurethe quality of this reprint but points out that some imperfectionsin the original copies may be apparent. Disclaimer The publisher has made every effort to trace copyright holders and welcomes correspondence from those they have beenunable to contact. A Library of Congressrecord exists under ISBN: 0710077815

ISBN 13: 978-0-415-69480-3(hbk) ISBN 13: 978-0-203-14548-7(ebk)

STUDIES IN ANCIENT SOCIETY Pastand PresentSeries GENERAL EDITOR: TREVOR ASTON

CONTRIBUTORS E. L. Bowie John Briscoe P. A. Brunt G. E. M. de Ste Croix M.1. Finley W. H. C. Frend

PeterGarnsey Keith Hopkins A. H. M. Jones A. N. Sherwin-White E. A. Thompson P. R. C. Weaver

FORTHCOMING The IntellectualRevolutionof the SeventeenthCentury editedby Charles Webster

STUDIES IN ANCIENT SOCIETY Past and PresentSeries GENERAL EDITOR: TREVOR ASTON

Editedby M. I. FINLEY

ROUTLEDGE AND KEGAN PAUL London, Henley and Boston

First publishedin 1974 by Routledge& Kegan Paul Ltd 39 Store Street London. welE 7DD, BroadwayHouse, NewtownRoad Henley-on-Thames axon. RG9 lEN and 9 Park Street, Boston, Mass. 02108, U.S.A. First publishedas a paperback1978 Printed in Great Britain by Alden & Mowbray Ltd at the Alden Press, Oxford

© Past and PresentSociety1974 No part of this book may be reproducedin any form without permissionfrom the publisher, exceptfor the quotation of brief passagesin criticism ISBN 0 ISBN 0

(c) 5 (p)

7100 7781 5 7100 8901

CONTENTS

pageix

INTRODUCTION

M. I. Finley, JesusCollege, Cambridge I.

1

ATHENIAN DEMAGOGUES

M. I. Finley, JesusCollege, Cambridge II.

ARISTOTLE AND ECONOMIC ANALYSIS

26

M. I. Finley, JesusCollege, Cambridge III.

ROME AND THE CLASS STRUGGLE IN THE GREEK STATES 200-146 B.C.

JohnBriscoe,University of Manchester IV.

THE ROMAN MOB

P. A. Brunt, BrasenoseCollege, Oxford V.

74

ELITE MOBILITY IN THE ROMAN EMPIRE

Keith Hopkins, Brunei University VI.

SOCIAL MOBILITY IN THE EARLY ROMAN EMPIRE: THE EVIDENCE OF THE IMPERIAL FREEDMEN AND SLAVES

121

P. R. C. Weaver,University of Tasmania VII.

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

141

PeterGarnsey,University of Cambridge VIII.

GREEKS AND THEIR PAST IN THE SECOND SOPHISTIC

E. L. Bowie, CorpusChristi College, Oxford V

166

CONTENTS

IX.

WHY WERE THE EARLY CHRISTIANS PERSECUTED?

210

G. E. M. de Ste Croix, New College, Oxford X.

WHY WERE THE EARLY CHRISTIANS PERSECUTED?AN AMENDMENT

25 0

A. N. Sherwin-White,SI John'sCollege, Oxford XI.

WHY WERE THE EARLY CHRISTIANS PERSECUTED?A REJOINDER

25 6

G. E. M. de Ste Croix, New College, Oxford XII.

THE FAILURE OF THE PERSECUTIONS IN THE ROMAN EMPIRE

W. H. C. Frend, University of Glasgow XIII.

263

288

THE ROMAN COLONATE

A. H. M. Jones,University of Cambridge XIV.

PEASANT REVOLTS IN LATE ROMAN GAUL AND SPAIN

E. A. Thompson,University of Nottingham INDEX

30 4 321

vi

ABBREVIATIONS

AmericanHistorical Review AmericanJournal of Philology CambridgeAncientHistory Corpus InscriptionumLatinarum ClassicalPhilology ClassicalQuarterlY CR ClassicalReview EHR EconomicHistory Review FGH Fragmenteder griechischenHistoriker GRBS Greek, Romanand ByzantineStudies HTR Harvard TheologicalReview InscriptionesGraecae IG IGR InscriptionesGraecaead res Romanaspertinentes ILS InscriptionesLatinae Selectae JEH Journal of EcclesiasticalHistory JHS Journal of Hellenic Studies JRS Journal of RomanStudies Journal of TheologicalStudies JTS PAPS Proceedingsof the AmericanPhilosophicalSociety PBA Proceedingsof the British Academy PCPS Proceedingsof the CambridgePhilological Society RE Realenryclopadieder classischenAltertumswissenschaft REA Revuedes EtudesAnciennes RH RevueHistorique RSI Rivista Storica Italiana SEG Supplementum EpigraphicumGraecum AHR AJP CAH CIL CP CQ

vii

ABBREVIATIONS

SIG3 SyllogeInscriptionum Graecarum TAPA Transactionsoj the AmericanPhilological Association TAPS Transactionsoj the AmericanPhilosophicalSociety

viii

INTRODUCTION

M. I. Finley

The first volume of essaysfrom Past and Present, published in 1965, had a central theme, as its title indicates, Crisis in Europe I560-J660.In the Introduction,the Masterof BaHiol, whoseassociation with the journal goesback to its beginning,wrote that the contributorsto Crisis in Europe 'would, I imagine, agreeon few points except that it is the duty of the historian to explain, not merely to record'. And he stressedthat changeand movement were what especiallyrequiredexplanation,as was alreadysaid in the statementof aims publishedin the first issue. The presentvolumehasno single themecomparableto Crisis in Europe,andrangesover nearlya thousandyearsof Graeco-Roman history. Nevertheless,every contributor would not only accept Mr Hill's dictum but would also, I believe, !accept the label 'structuralist'. By that I mean none of the different, and often incompatible, schools which have adoptedthat identification in recent decades,but something simpler and more elementary, namely,the view that neitherinstitutionsnor their transformations (pastor present)can be understoodexceptin their role within the social structureof their day, in the network of interrelationships that makeup any complexsociety-andit is sometimesnecessary to insist that these societies were complex. Few of the present contributors actually use the word 'structure',but what I have called structuralismis integral to their analyses,whether in the field of politics or law or in the study of social classesand social mobility or in the history of ideasand values. Structuralismis not to be confusedwith 'vulgar sociology' or ix

INTRODUCTION

'vulgar Marxism'. It is enough to note Mr Briscoe's carefully nuancedaccountof how Romanpreferencefor oligarchiesin the newly conquered territories was never allowed to override immediate political-imperial considerations;Mr de Ste Croix's conclusions that the 'main motives' behind the governmental persecutionsof the Christians were 'in the long run essentially religious'; or ProfessorThompson'sinsistencethat the revolts of peasantsand slavesin Gaul and Spain,extendingover severalcenturies, 'releasedno new productiveforces' and 'included no new modeof socialexistence'amongtheir ideas.Structuralismthrows up questionsto be answered,often new questionsabout old and evenwell-worn topics;it doesnot providetheanswers.Most of the politics, articlesarein fact devotedto familiar subjects-Athenian the Romanconquestof theeast,violencein thelaterRomanRepublic, the SecondSophisticand the persecutions.Yet without exception the authorshaveproducedfresh and original analysesand explanations,and someof thearticleshavealready,in a very few years, generatedimportant researchby other ancienthistorians. Becausethereis no centraltheme,it seemedbestto arrangethe essaysin a roughly, thoughnot pedantically,chronologicalorder, eventhough this left me in the embarrassingposition of opening the volumewith two articlesof my own. The heavyconcentration on the RomanEmpirebecomesimmediatelyapparent.In part this is accidental,but not entirely so: it also reflects the way British scholarshipin ancient history has developed in the post-war years,in the sensethat morework has beendone of an explanatory, structuralist nature on the Roman Empire than on either Greeceor the earlier period in Romanhistory. It is also striking that the studiesof the Empire constantlylook aheadto the end of the ancientworld, as much as, and perhapsmore than, they look backwards in time for complicated evolutionary explanations. (I say 'British scholarship'because,despitethe welcomePast and Presentoffers to historians elsewhere,ancient historiansabroad havefailed to respond,preferringto publish in classicaljournals.) Following the precedentset in Crisis in Europe, authors were askedto retain their original text, making only suchcorrectionsas were indispensable,and addingbrief bibliographicalsupplements when desirable. Dr Dorothy Crawford of Girton College, Cambridge,was kind enoughto revisethe notes for the chapter 'The RomanColonate'by the late ProfessorA. H. M. Jones. x

I

ATHENIAN DEMAGOGUES*l

M. 1. Finley

When the news of their defeat in Sicily in 413 B.C. reachedthe Athenians,they receivedit with disbelief. Then camethe realization of the full scale of the disaster, and the people, writes Thucydides,'were indignantwith the oratorswho had joined in promoting the expedition,as if they [the people] had not themselvesdecreedit [in assembly]'.2To this GeorgeGrote madethe following rejoinder:3 From theselatter words, it would seemthat Thucydides consideredthe Athenians,after having adoptedthe expeditionby their votes, to have debarredthemselvesfrom the right of complainingof thosespeakerswho had stood forward prominentlyto advisethe step. I do not at all concur in his opinion. The adviserof any importantmeasurealways makeshimself morally responsiblefor its justice, usefulness, and practicability; and he very properly incurs disgrace, more or less accordingto the case,if it turns out to present resultstotally contrary to thosewhich he had predicted. Thesetwo opposingquotationsraiseall the fundamentalprob*From no. 21 (1962). This is a revisedtext of a paperto the Hellenic Society in London on 25 March 1961, of which a shortenedversion was broadcaston the Third Programmeof the BBC and publishedin the Listener of 5 and I2 October 1961. I am grateful to ProfessorsA. Andrewesand A. H. M. Jones,MessrsP. A. Brunt and M. J. Cowling for advice and criticism. 2 Thuc., 8.1.1. 3 A History of Greece,new edn (London, 1862), v, p. 317 n. 3. I

I

M.

r.

FINLEY

lemsinherentin the Atheniandemocracy,the problemsof policymaking and leadership,of decisions and the responsibility for them. Unfortunately Thucydides tells us very little about the orators who successfullyurged on the Assembly the decisionto mount the great invasion of Sicily. In fact, he tells us nothing concreteaboutthe meeting,otherthan that the peopleweregiven misinformationby a delegationfrom the Sicilian city of Segesta and by their own envoysjust returnedfrom Sicily, and that most of those who voted were so ignorant of the. relevant facts that they did not evenknow the sizeof the island or of its population. Five days later a second Assembly was held to authorize the necessaryarmament.The generalNicias took the opportunity to seeka reversalof the whole programme.He was opposedby a numberof speakers,Athenianand Sicilian, neithernamedby the historian nor describedin any way, and by Alcibiades, who is given a speechwhich throws much light on Thucydideshimself and on his judgmentof Alcibiades,but scarcelyany on the issues, whetherthe immediateonesbeingdebatedor the broaderonesof democraticprocedureand leadership.The result was a complete defeatfor Nicias. Everyone,Thucydidesadmits, was now more eager than before to go aheadwith the plan-the old and the young, the hoplite soldiers(who were drawn from the wealthier half of the citizenry) and the commonpeoplealike. The few who remainedopposed,he concludes,refrainedfrom voting lest they appearedunpatriotic.4 The wisdom of the Sicilian expeditionis a very difficult matter. Thucydideshimself had more than one view at different times in his life. However, he seemsnot to have changedhis mind about the orators:they promotedthe expeditionfor the wrong reasons and they gainedthe day by playing on the ignoranceandemotions of the Assembly. Alcibiades, he says, pressedhardest of all, becausehe wished to thwart Nicias, becausehe was personally ambitious and hoped to gain fame and wealth from his generalship in the campaign,and becausehis extravagantand licentious tastes were more expensivethan he could really afford. Elsewhere,writing in more generalterms, Thucydidessaysthis:s [Under Pericles] the governmentwas a democracyin name Thuc., 6.1-25. s Thuc., 2.65.9-II.

4

z

ATHENIAN DEMAGOGUES

but in reality rule by the first citizen. His successorswere more equalto eachother, and eachseekingto becomethe first man they even offered the conductof affairs to the whims of the people.This, as was to be expectedin a great stateruling an empire, producedmany blunders. In short,after the deathof PericlesAthensfell into the handsof demagoguesand was ruined. Thucydidesdoes not use the word 'demagogue'in any of the passagesI have beendiscussing.It is an uncommon word with him,6 as it is in Greek literature generally, and that fact may come as a surprise,for there is no more familiar themein the Athenianpicture (despitethe rarity of the word) than the demagogueand his adjutant, the sycophant. The demagogueis a bad thing: to 'leadthe people'is to misleadaboveall, to misleadby failing to lead. The demagogueis driven by self-interest,by the desire to advancehimself in power, and through power, in wealth. To achieve this, he surrendersall principles,all genuineleadership,and he pandersto the peoplein every way-in Thucydides'words, 'even offering the conductof affairs to the whims of the people'.This pictureis drawn not only directly, but also in reverse.Here, for example, is Thucydides' image of the right kind of leader:7 Becauseof his prestige,intelligence,and known incorruptibility with respectto money, Pericleswas able to lead the peopleas a free man should. He led them insteadof being led by them. He did not have to humour them in the pursuit of power; on the contrary, his reputewas such that he could contradictthem and provoke their anger. This was not everyone'sjudgment. Aristotle puts the breakdown earlier: it was after Ephialtestook away the power of the Council of the Areopagusthat the passionfor demagogyset in. Pericles,he continues,first acquiredpolitical influence by prosecuting Cimon for malfeasancein office; he energeticallypursueda policy of naval power, 'which gavethe lower classesthe audacity to take over the leadershipin politics more and more'; and he introducedpay for jury service,thus bribing the peoplewith their own money. Thesewere demagogicpracticesand they brought 6 7

Used only in 4.21.3, and 'demagogy'in 8.65.2. Thuc., 2.65.8.

M. I. FINLEY

Periclesto power, which, Aristotle agrees,he then usedwell and properly.8 But my interestis neitherin evaluatingPericlesas an individual nor in examining the lexicography of demagogy. The Greek political vocabularywasnormally vagueandimprecise,apartfrom formal titles for individual offices or bodies(and often enoughnot even then). The word demos was itself ambiguous; among its meanings,however, was one which came to dominate literary usage,namely'the commonpeople','the lower classes',and that -they became sense provided the overtones in 'demagogues' leadersof the statethanksto the backing of the commonpeople. All writers acceptedthe needfor political leadershipas axiomatic; their problem was to distinguish betweengood and bad types. With respectto Athensand its democracy,the word 'demagogue' understandablybecamethe simplest way of identifying the bad type, and it doesnot matterin the leastwhetherthe word appears in any given text or not. I supposeit was Aristophaneswho establishedthe model in his portrayal of Cleon, yet he never directly applied the noun 'demagogue'to him or anyone else;9 similarly with Thucydides,who surely thought that Cleophon, Hyperbolus,and some, if not all, of the orators responsiblefor the Sicilian disasterwere demagogues,but who never attached the word to any of thesemen. It is importantto stressthe word 'type', for the issueraisedby Greek writers is one of the essentialqualities of the leader, not (exceptvery secondarily)his techniquesor technicalcompetence, not even (exceptin a very generalizedway) his programmeand policies. The crucial distinction is betweenthe man who gives leadershipwith nothing else in mind but the good of the state, andthe manwhoseself-interestmakeshis own positionparamount and urges him to panderto the people. The former may make a mistake and adopt the wrong policy in any given situation; the latter may at times makesoundproposals,as when AlcibiadesdissConst.of Athens, 27-28; cf. Pol., 2.9.3 (1274a3-IO).A. W. Gomme, A Historical Commentaryon Thucydides(Oxford, 1956), ii p. 193, points out that 'Plutarch divided Perikles' political careersharply into two halves, the first when he did use basedemagogicarts to gain power, the second when he had gained it and usedit nobly'. 9 Aristophanesuses'demagogy'and 'demagogic'once each in the Knights, lines 191 and 217, respectively.Otherwisein his surviving plays there is only the verb 'to be a demagogue',also usedonce(Frogs, 419).

4

ATHENIAN DEMAGOGUES

suadedthe fleet at Samosfrom jeopardizingthe naval position by rushing back to Athens in 411 B.C. to overthrow the oligarchs who had seizedpower there,an action to which Thucydidesgave explicit approva1.10 But these are not fundamentaldistinctions. Nor are other traits attributedto individual demagogues:Cleon's habit of shouting when addressingthe Assembly, personaldishonestyin moneymatters,and so on. Suchthings merely sharpen the picture. From Aristophanesto Aristotle, the attack on the demagoguesalways falls back on the one central question: in whoseinterestdoes the leaderlead? Behind this formulation of the questionlay threepropositions. The first is that men are unequal-bothin their moral worth and capability and in their social and economicstatus.The secondis that any community tends to divide into factions, the most fundamentalof which are the rich and well-born on one side, the poor on the other, eachwith its own qualities, potentialities,and interests. The third proposition is that the well-ordered and well-run state is one which overrides faction and servesas an instrumentalityfor the good life. Faction is the greatestevil and the most common danger. 'Faction'is a conventionalEnglish translationof the Greekstasis, one of the mostremarkablewords to be found in any language.Its root-senseis 'placing', 'setting' or 'stature','station'. Its rangeof political meaningscan best be illustratedby merely stringing out the definitions to be found in the lexicon: 'party', 'party formed for seditiouspurposes','faction', 'sedition', 'discord', 'division', 'dissent',and, finally, a well-attestedmeaningwhich the lexicon incomprehensiblyomits, namely, 'civil war' or 'revolution'. Unlike 'demagogue',stasisis a very commonword in the literature, andits connotationis regularlypejorative.Oddly enough,it is also the most neglectedconceptin modernstudy of Greek history. It hasnot beenobservedoften enoughor sharplyenough,I believe, that theremust be deepsignificancein the fact that a word which has the original senseof 'station' or 'position', and which, in abstractlogic, could havean equally neutral sensewhen usedin a political context, in practice does nothing of the kind, but immediately takes on the nastiestovertones.A political position, a partisan position-thatis the inescapableimplication-is a bad 10

Thuc., 8.86.

M. I. FINLEY

thing, leadingto sedition,civil war, andthedisruptionof thesocial fabric.ll And this same tendency is repeated throughout the language.There is no eternal law, after all, why 'demagogue',a 'leader of the people', must become'mis-leaderof the people'. Or why hetairia, an old Greek word which meant, among other things, 'club' or 'society', should in fifth-century Athens have come simultaneouslyto mean 'conspiracy','seditious organization'. Whatever the explanation,it lies not in philology but in Greek societyitself. No one who has readthe Greekpolitical writers canhavefailed to notice the unanimity of approachin this respect.Whateverthe disagreements amongthem, they all insist that the statemuststand outsideclassor otherfactionalinterests.Its aimsandobjectivesare moral ones, timeless and universal, and they can be achievedmore correctly, approachedor approximated-onlyby education, moralconduct(especiallyon thepartof thosein authority),morally correct legislation, and the choice of the right governors.The existenceof classesand interestsas an empirical fact is, of course, not denied.What is deniedis that the choiceof political goalscan legitimately be linked with theseclassesand interests,or that the good of the state can be advancedexcept by ignoring (if not suppressing)private interests. It was Plato, of course,who pursuedthis line of its reasoningto its most radical solutions. In the Gorgias he had arguedthat not even the great Athenian political figures of the past-Miltiades, Themistocles,Cimon and Pericles-weretrue statesmen.They in gratihad merelybeenmore accomplishedthan their successors fying the desiresof the demoswith shipsand walls and dockyards. They had failed to make the citizensbettermen, and to call them 'statesmen'was therefore to confuse the pastrycook with the doctor.12Then, in the Republic,Plato proposedto concentrateall power in the handsof a small, select,appropriatelyeducatedclass, who were to be freed from all specialinterestsby the most radical measures,by the abolition insofar as they were concernedof both private property and the family. Only under those conditions The only systematicanalysisknown to me, and that a brief one, is the inaugural lecture of D. Loenen,Stasis(Amsterdam,1953). He saw, contrary to the view most commonamong modernwriters, that 'illegality is preciselynot the constantelementin stasis'(p. 5). 12 Gorgias, 50zE-519D.

11

6

ATHENIAN DEMAGOGUES

would they behaveas perfectmoral agents,leadingthe stateto its proper goals without the possibility that any self-interestmight intrude. Plato, to be sure, was the most untypical of men. One doesnot safely generalizefrom Plato; not only not to all Greeks, but not even to any other single Greek. Who else sharedhis passionateconvictionthat qualified experts-philosophers-could make(and shouldthereforebe empoweredto enforce)universally correctand authoritativedecisionsaboutthe good life, the life of virtue, which was the sole end of the state?13 Yet on the onepoint with which I am immediately concerned-privateinterestsand the state-Platostood on common ground with many Greek writers (much as they disagreedwith him on the answers).In the greatfinal sceneof Aeschylus'Eumenidesthe chorusexpressesthe doctrine explicitly: the welfare of the state can rest only on harmonyand freedomfrom faction. Thucydidesimplies this more than once.14 And it underliesthe theory of the mixed constitution as we find it in Aristotle's Politics. The most empirical of Greek philosophers,Aristotle collected vast quantitiesof dataaboutthe actualworkings of Greek states, including facts about stasis. The Politics includes an elaborate taxonomyof stasis,and evenadvice on how stasiscan be avoided under a variety of conditions. But Aristotle's canonsand goals were ethical, his work a branchof moral philosophy. He viewed political behaviour teleologically, according to the moral ends which areman'sby his nature;and thoseendsare subvertedif the governorsmake their decisionsout of personalor classinterest. That is the test by which he distinguishedbetween the three 'right' forms of government('accordingto absolutejustice') and their degenerateforms: monarchy becomes tyranny when an individual rules in his own interestrather than in the interestof the whole state,aristocracysimilarly becomesoligarchy,andpolity becomesdemocracy(or, in the languageof Polybius, democracy becomesmob-rule),15Among democracies,furthermore,thosein rural communitieswill be superiorbecausefarmers are too occupied to botherwith meetings,whereasurbancraftsmenand shopSeeR. Bambrough,'Plato'spolitical analogies',in Philosophy,Politics and Society,ed. PeterLaslett (Oxford, 1956), pp. 98-II5. 14 It is developedmost fully in his long account(3.69-85) of the stasisin Corcyra in 427 B.C. IS Arist., Pol., 3.4-5 (I278b-79b),4.6-7 (1293b-94b);Polyb., 6.3-9'

13

7

M. I. FINLEY

keepersfind it easyto attend,and suchpeople'are generallya bad lot' .!6 On this matterof specialinterestand generalinterest,of faction andconcord,the availableexceptionsto the line of thinking I have summarizedare strikingly few and unrewarding. One deserves particularmention,and that, ironically enough,is the pamphleton the Athenianstateby an anonymouswriter of the later half of the fifth century B.C. who now generallygoesunder the too amiable label of the Old Oligarch. This work is a diatribe against the democracy,hammeringat the themethat the systemis a bad one becauseall its actionsare determinedby the interestsof the poorer (inferior) sections of the citizenry. The argument is familiar enough;what gives the pamphletits unusualinterestis this conclusion:!7 As for the Athenian systemof government,I do not like it. However, since they decidedto becomea democracy,it seemsto me that they are preservingthe democracywell by the methodsI have described. In other words, the strengthof the Athenian governmentcomes preciselyfrom that which many merely criticize, namely, the fact that it is governmentby a faction acting unashamedlyto its own advantage. The greatdifferencebetweenpolitical analysisand moral judgment could not be better exemplified. Do not be misled, saysthe Old Oligarchin effect: I and someof you dislike democracy,but a reasonedconsiderationof the facts showsthat what we condemn on moral grounds is very strong as a practical force, and its strength lies in its immorality. This is a very promising line of investigation,but it was not pursuedin antiquity. Instead,those thinkers whoseorientationwas anti-democraticpersistedin their concentrationon political philosophy.And thosewho sidedwith the democracy?A. H. M. Joneshasrecentlytried to formulatethe democratictheory from the fragmentaryevidenceavailablein the surviving literature, most of it from the fourth century.!S Still more recently,Eric Havelockmadea massiveattemptto discover what he calls the 'liberal temper'in fifth-century Athenianpolitics, Arist., Pol., 6.2.7-8 (I319a); cf. Xen., Hell., 5.2.5-7. Pseudo-Xenophon, Const. of Athens, 3.1; see A. Fuks, 'The "Old Oligarch''', Scripta Hieroso/ymitana,i (1954), pp. 21-35. 18 Athenian Democrary(Oxford, 1957), ch. iii.

16

11

8

ATHENIAN DEMAGOGUES

chiefly from the fragments of the pre-Socraticphilosophers.In reviewing his book, Momigliano suggestedthat the effort was foredoomed because'it is not absolutely certain that a wellarticulated democratic idea existed in the fifth century',19 I go further: I do not believethat an articulateddemocratictheory ever existed in Athens. There were notions, maxims, generalitieswhich Joneshas assembled-butthey do not add up to a systematic theory. And why indeedshould they?It is a curious fallacy to supposethat every social or governmentalsystem in history must necessarilyhavebeenaccompaniedby an elaboratetheoretical system.Where that does occur it is often the work of lawyers, and Athens had no jurists in the proper sense.Or it may be the work of philosophers,but the systematicphilosophersof this period had a set of concepts and values incompatible with democracy.The committeddemocratsmet the attackby ignoring it, by going aboutthe businessof conductingtheir political affairs accordingto their own notions, but without writing treatiseson what they were about. None of this, however,is a reasonwhy we should not attempt to make the analysis the Athenians failed to make for themselves. No accountof the Atheniandemocracycan have any validity if it overlooksfour points, eachobvious in itself, yet all four taken together, I venture to say, are rarely given sufficient weight in modernaccounts.The first is that this was a direct democracy,and however much such a systemmay have in common with representativedemocracy,the two differ in certainfundamentalrespects, and particularly on the very issues with which I am here concerned.The secondpoint is what Ehrenbergcalls the 'narrowness of space'of the Greek city-state,an appreciationof which, he has rightly stressed,is crucial to an understandingof its political life. 20 The implications were summed up by Aristotle in a famous 21 passage: A state composedof too many ... will not be a true state, for the simple reasonthat it can hardly have a true constitution. Who can be the generalof a massso excessivelylarge? And who can be herald, except Stentor? E. A. Havelock, The Libera! Temperin Greek Politics (London, 1957), reviewed by A. Momigliano in R.5'I, lxxii (1960), pp. 534-41. 20 Aspectsof the Ancient World (Oxford, 1946), pp. 40-5. 21 Pol., 7+7 (I 326b3-7). 19

9

M. 1. FINLEY

The third point is that the Assemblywas the crown of the system, possessingthe right and the power to makeall thepolicy decisions, in actual practiCe with few limitations, whether of precedentor scope.(Strictly speakingthere was appealfrom the Assembly to the popularcourts with their large lay membership.Nevertheless, I ignore the courts in much, though not all, of what follows, becauseI believe,as the Atheniansdid themselves,that, thoughthey complicatedthe practical mechanismof politics, the courts were an expression,not a reduction,of the absolutepowerof the people functioning directly: and becauseI believe that the operational analysisI am trying to makewould not be significantly alteredand would perhapsbe obscuredif in this brief compassI did not concentrateon the Assembly.)The Assembly,finally, was nothing other than an open-airmassmeeting on the hill called the Pnyx, and the fourth point thereforeis that we aredealingwith problems of crowd behaviour;its psychology,its laws of behaviour,could not havebeenidentical with thoseof the small group, or even of the larger kind of body of which a modern parliament is an example (though, it must be admitted, we can do little more today than acknowledgetheir existence). Who were the Assembly?That is a questionwe cannotanswer satisfactorily.Every male citizen automaticallybecameeligible to attend when he reachedhis eighteenthbirthday, and he retained that privilege to his death(exceptfor the very small numberwho lost their civic rights for one reasonor another).In Pericles'time the number eligible was of the order of 45,000. Women were excluded;so were the fairly numerousnon-citizenswho were free men, nearly all of them Greeks, but outsiders in the political sphere;and so were the far more numerousslaves.All figures are a guess,but it would not be wildly inaccurateto suggestthat the adult male citizenscomprisedaboutone sixth of the total population (taking town and countryside together). But the critical questionto be determinedis which four or five or six thousandof the 45,000 actually went to meetings.It is reasonableto imagine that undernormal conditionsthe attendancecamechiefly from the urban residents.Fewerpeasantswould often have taken the journey in order to attenda meetingof the Assembly.22Thereforeone large sectionof the eligible populationwas, with respectto direct 22

That Aristotle drew very important conclusionsfrom this state of affairs has already been indicated,at note 16. 10

ATHENIAN DEMAGOGUES

participation,excluded.Thatis somethingto know, butit doesnot get us far enough.We canguess,for example,with the aid of a few hints in the sources,that the compositionwas normally weighted on the side of the more agedand the more well-to-do men-but that is only a guess,and the degreeof weighting is beyondeven guessing. Still, oneimportantfact canbe fixed, namely,that eachmeeting of the Assembly was unique in its composition. There was no membershipin the Assemblyas such,only membershipin a given Assemblyon a given day. Perhapsthe shifts were not significant from meeting to meetingin quiet, peacefultimes when no vital issueswerebeingdebated.Yet eventhen an importantelementof predictability was lacking. When he enteredthe Assembly, no policy-makercould be quite surethat a changein the composition of the audiencehad not occurred,whether through accidentor through more or less organizedmobilization of some particular sectorof the population,which could tip the balanceof the votes againsta decision made at a previous meeting. And times were often neither peaceful nor normal. In the final decadeof the Peloponnesian War, to take an extremeexample,the whole rural population was compelled to abandonthe countrysideand live within the city walls. It is beyondreasonablebeliefthat during this period there was not a larger proportion of countrymenat meetings than was normal. A similar situation prevailed for briefer periods at other times, when an enemy army was operating in Attica. We need not interpret Aristophanesliterally when he opensthe Acharnianswith a soliloquy by a farmer who is sitting in the Pnyx waiting for the Assemblyto beginand sayingto himself how he hatesthe city and everyonein it and how he intends to shoutdown any speakerwho proposesanythingexceptpeace.But Cleon could not haveaffordedthe luxury of ignoring this strange element seatedon the hillside before him. They might upset a policy line which he hadbeenableto carrywhile the Assemblywas filled only with city-dwellers. The one clearcut instance came in the year 41 I. Then the Assembly was terrorized into voting the democracy out of existence,andit was surelyno accidentthat this occurredat a time when the fleet was fully mobilized and stationedon the island of Samos.The citizenswho servedin the navy were drawn from the poor and they were known to be the staunchestsupportersof the II

M. I. FINLEY

democraticsystemin its late fifth-century form. Being in Samos, they could not be in Athens,thusenablingthe oligarchsto win the day through a majority in the Assembly which was not only a minority of the eligible membersbut an untypical minority. Our sourcesdo not permit us to study the history of Athenianpolicy systematicallywith suchknowledgeat our disposal,but surelythe men who led Athens were acutely aware of the possibility of a changein the compositionof the Assembly,andincludedit in their tactical calculations. Eachmeeting,furthermore,wascompletein itself. Grantedthat much preparatorywork was done by the Council (boule), that informal canvassingtook place,andthat therewerecertaindevices to control and check frivolous or irresponsible motions, it is neverthelesstrue that the normal procedurewas for a proposalto be introduced, debated, and either passed (with or without amendment)or rejectedin a single continuoussitting. We must reckon,therefore,not only with narrownessof spacebut alsowith narrowness of time, and with the pressuresthat generated, especiallyon leaders(and would-be leaders).I havealreadymentioned the caseof the Sicilian expedition, which was decidedin principle on one day and then planned,so to speak,five dayslater when the scaleand cost were discussedand voted. Another kind of caseis that of the well-known Mytilene debate.Early in the PeloponnesianWar the city of Mytilene revolted from the Athenian Empire. The rebellion was crushedand the Athenian Assembly decided to make an example of the M ytileneans by putting the entire male populationto death.Revulsionof feeling set in at once,the issuewas reopenedat anothermeetingthe very next day, and the decisionwas reversed.23Cleon, at that time the mostimportantpolitical figure in Athens,advocatedthe policy of frightfulness.The secondAssemblywas a personaldefeatfor him -he had participatedin the debateson both days-thoughhe seemsnot to havelost his statuseventemporarilyas a result(as he well might have). But how does one measurethe psychological effect on him of sucha twenty-fourhour reversal?How doesone estimatenot only its impact,but alsohis awarenessall throughhis careeras a leaderthat such a possibility was a constantfactor in Athenianpolitics?I cannotanswersuchquestionsconcretely,but I submit that the weight could have been no light one. Cleon 23

Thuc., 3.27-50. 12.

ATHENIAN DEMAGOGUES

surely appreciated,as we cannot,what it promisedfor men like War, when himself that in the secondyear of the Peloponnesian morale was temporarily shatteredby the plague, the people turnedon Pericles,fined him heavily, and deposedhim for a brief periodfrom the office of general.24 If this couldhappento Pericles, who was immune? In the Mytilene caseThucydides'accountsuggeststhat Cleon's was a lost causethe secondday, that he tried to persuadethe Assemblyto abandona courseof action which they intendedto pursuefrom the momentthe sessionopened,and that he failed. But the story of the meeting in 411, as Thucydidestells it, is a different one. Peisanderbeganthe day with the feeling againsthis proposalthat the introductionof an oligarchicalform of government should be considered,and he endedit with a victory. The actual debatehad swung enoughvotes to give him a majority.25 Debate designed to win votes among an outdoor audience numberingseveralthousandsmeansoratory, in the strict senseof the word. It was therefore perfectly precise language to call political leaders'orators', as a synonym and not merely, as we might do, as a mark of the particularskill of a particularpolitical figure. Under Athenian conditions, however, much more is implied. The picture of the AssemblyI havebeentrying to draw suggestsnot only oratory, but also a 'spontaneity'of debateand decision which parliamentarydemocracylacks, at least in our day.26 Everyone, speakersand audiencealike, knew that before night fell the issuemustbe decided,that eachman presentwould vote 'freely' (without fear of whips or other party controls) and purposefully, and therefore that every speech,every argument must seekto persuadethe audienceon the spot, that it was all a seriousperformance,as a whole and in eachof its parts. I placethe word 'freely' in inverted commas,for the last thing I wish to imply is the activity of a free, disembodiedrational faculty, that favourite illusion of so much political theory since the Enlightenment.Membersof the Assemblywere free from the controls which bind the membersof a parliament:they held no office, they were not elected,and thereforethey could neitherbe Thuc., 2.65.1-4. Thuc., 8.B-54. 26 Seethe valuable article by o. Reverdin, 'Remarques sur la vie politique d'Athenesau Ve siecle', MuseumHelveticum,ii (1945), pp. 201-IZ. 24

2S

M. I. FINLEY

punishednor rewardedfor their voting records.But they werenot free from the humancondition,from habit andtradition, from the influencesof family and friends, of class and status,of personal experiences,resentments, prejudices,values,aspirations,andfears, much of it in the subconscious.Thesethey took with them when they went up on the Pnyx, and with thesethey listened to the debatesand madeup their minds, underconditionsvery different from the voting practicesof our day. There is a vast difference betweenvoting on infrequentoccasionsfor a man or a party on the one hand, and on the other hand voting every few days directly on the issuesthemselves.In Aristotle'stime the Assembly met at leastfour times in eachthirty-six day period. Whetherthis was also the rule in the fifth centuryis not known, but therewere occasions,as during the Peloponnesian War, when meetingstook place even more frequently. Then there were the two other factors I have alreadymentioned,the smallnessof the Athenian world, in which every memberof the Assemblyknew personally many others sitting on the Pnyx, and the mass-meetingbackground of the voting-a situation virtually unrelated to the impersonalact of marking a voting paper in physical isolation from everyothervoter; an act we perform, furthermore,with the knowledge that millions of other men and women are simultaneouslydoing the same thing in many places, some of them hundredsof miles distant. When, for example, Alcibiades and Nicias rosein the Assemblyin 415, the oneto proposethe expedition againstSicily, the other to argueagainstit, eachknew that, should the motion be carried, one or both would be asked to commandin the field. And in the audiencetherewere many who were being askedto vote on whether they, personally,were to march out in a few days, as officers, soldiers,or membersof the fleet. Such examplescan be duplicated in a number of other, scarcelyless vital areas:taxation, food supply, pay for jury duty, extensionof the franchise,laws of citizenship,and so on. To be sure,much of the activity of the Assemblywasin a lower key, largely occupied with technical measures(such as cult regulations)or ceremonialacts (such as honorary decreesfor a great variety of individuals). It would be a mistake to imagine Athens as a city in which week iri and week out great issues dividing the populationwere being debatedand decided.But on the otherhand,therewere very few single years(and certainlyno 14

ATHENIAN DEMAGOGUES

ten-yearperiods)in which somegreatissuedid not arise: the two Persianinvasions,the long series of measureswhich completed the processof democratization,the Empire, the Peloponnesian War (which occupiedtwenty-sevenyears) and its two oligarchic interludes, the endlessdiplomatic manoeuvresand wars of the fourth century,with their attendantfiscal crises,all culminatingin the decadesof Philip andAlexander.It did not often happen,as it did to Cleon in the dispute over Mytilene, that a politician was faced with a repeat performancethe following day; but the Assemblydid meetconstantly,without long periodsof holiday or recess.The week-by-weekconductof a war, for example,had to go before the Assemblyweek by week; as if Winston Churchill were to have been compelledto take a referendumbefore each move in World War II, and then to face anothervote after the move was made,in the Assemblyor the law-courts,to determine not merely what the next step shouldbe but also whetherhe was to be dismissedand his plansabandoned,or evenwhetherhe was to be held criminally culpable, subjectto a fine or exile or, conceivably, the deathpenaltyeitherfor the proposalitself or for the way the previous move had beencarried out. It was part of the Athenian governmentalsystem that, in addition to the endless challengein the Assembly,a politician was faced, equally without respite,with the threat of politically inspired lawsuits.27 If I insist on the psychologicalaspect,it is not to ignore the considerablepolitical experienceof many men who voted in the Assembly-gainedin the Council, the law-courts,the demes,and the Assemblyitself-nor is it merely to counterwhat I havecalled the disembodied-rationalism conception. I want to stresssomething very positive, namely, the intense degreeof involvement which attendanceat the Athenian Assembly entailed. And this intensity was equally (or evenmore strongly) the caseamongthe orators, for each vote judged them as well as the issue to be decidedon. If I had to chooseone word which bestcharacterized the conditionof beinga politicalleaderin Athens,thatword would 27

P. Cloche, 'Les hommespolitiques et la justice populairedansI'Athenes du IVe siecle', Historia, ix (i96o), pp. 80-95, has recently arguedthat this threat is exaggeratedby modern historians,at least for the fourth century. Useful as his assemblingof the evidenceis, he lays too much stresson the argumentfrom silence,whereasthe sourcesare far from full enoughto bear such statisticalweight.

15

M. I. FINLEY

be 'tension'.In somemeasurethat is true of all politicians who are of politics and government' subjectto a vote. 'The desperateness is R. B. McCallum's telling phrase,which he then developedin this way:28 Certainly a note of cynicism and wearinesswith the manoeuvresand posturingsof party politicians is natural and to an extent proper to discerningdons and civil servants, who can reflect independentlyand at leisure on the doings of their harried mastersin government.But this seemsto arise from a deliberaterejection... of the aims and ideals of party statesmenand their followers and the continual responsibility for the security and well-being in the state. For one thing party leadersare in somesenseapostles,althoughall may not be Gladstones;there are policies to which they dedicate themselvesand policies which alarm and terrify them. I believe this to be a fair description of Athenian leaders,too, despite the absenceof political parties, equally applicable to Themistoclesas to Aristides, to Periclesas to Cimon, to Cleon as to Nicias; for, it should be obvious, this kind of judgment is of a independentof any judgmentaboutthe merits or weaknesses particular programmeor policy. More accurately,I should have said that this understatesthe casefor the Athenians.Their leaders hadno respite.Becausetheir influencehadto be earnedandexerted directly and immediately-thiswas a necessaryconsequenceof a direct, as distinct from a representative,democracy-theyhad to lead in person,and they had also to bear, in person,the brunt of the opposition'sattacks.More than that, they walked alone.They had their lieutenants,of course, and politicians made alliances with each other. But these were fundamentally personallinks, shifting frequently, usefulin helping to carry througha particular measureor even a group of measures,but lacking that quality of support,that buttressingor cushioningeffect, which is provided by a bureaucracyand political party, in another way by an institutionalizedEstablishmentlike the Roman Senate,or in still anotherway by large-scalepatronageas in the Roman clientage system.The critical point is that therewas no 'government'in the modern sense.There were posts and offices, but none had any standingin the Assembly.A man was a leadersolely as a function 28

A review in the Listener(z February1961), p. zB.

16

ATHENIAN DEMAGOGUES

of his personal,and in the literal sense,unofficial statuswithin the Assemblyitself. The test of whetheror not he held that statuswas simply whether the Assembly did or did not vote as he wished, and thereforethe test was repeatedwith eachproposal. Thesewerethe conditionswhich facedall leadersin Athens,not merely those whom Thucydides and Plato dismissedas 'demagogues',not merely thosewhom some modern historiansmiscall 'radicaldemocrats',but everyone,aristocrator commoner,altruist or self-seeker,able or incompetent, who, in George Grote's phrase,'stoodforward prominently to advise'the Athenians.No doubt the motives which moved men to stand forward varied greatly. But that does not matter in this context, for each one of them without exception,chose to aspire to, and actively to work and contestfor, leadership,knowing just what that entailed, including the risks. Within narrow limits, they all had to use the same techniques,too. Cleon's platform manner may have been inelegantand boisterous,but how serious is Aristotle's remark that he was the first man to 'shoutand rail' ?29 Are we to imagine that Thucydides the son of Melesias (and kinsman of the historian) and Nicias whisperedwhen they addressedthe Assembly in opposition to Pericles and Cleon, respectively?Thucydides, who broughthis upper-classbackersinto the Assemblyand seated them to form a claque?30 This is obviously a frivolous approach,nothing more than the As Aristotle noted, expressionof classprejudiceand snobbishness. the death of Periclesmarkeda turning-point in the social history of Athenianleadership.Until then they seemto have beendrawn from the old aristocraticlandedfamilies, including the men who were responsiblefor carrying out the reforms which completed the democracy.After Pericles a new class of leadersemerged.31 Despite the familiar prejudicial referencesto Cleon the tanneror Cleophon the lyre-maker, these were in fact not poor men, not craftsmenand labourersturnedpolitician, but men of meanswho differed from their predecessorsin their ancestryand their outlook, and who provoked resentment and hostility for their Arist., Const., 28.3. Plut., Pericles, 11.2. It was againstsuch tactics that the restored democracyin 410 requiredmembersof the Council to swear to take their seatsby lot: Philochorus328 F 140 (in FHG, ed. F. Jacoby). 31 Arist., Const., 28.1.

29

30

17

M. I. FINLEY

presumptionin breaking the old monopoly of leadership.When such attitudes are under discussion, one can always turn to Xenophonto find the lowest level of explanation(which is not thereforenecessarilythe wrong one). One of the most important of the new leaders was a man called Anytus, who, like Cleon before him, drew his wealth from a slave tannery. Anytus had a long and distinguishedcareer,but he was also the chief actor in the prosecutionof Socrates.What is Xenophon'sexplanation? Simply that Socrateshad publicly beratedAnytus for bringing up his son to follow in his tradeinsteadof educatinghim as a proper gentleman,and that Anytus, in revengefor this personalinsult, had Socratestried and executed.32 None of this is to deny that therewere very fundamentalissues behind the thick fas;adeof prejudice and abuse.Throughoutthe fifth centurytherewerethe twin issuesof democracy(or oligarchy) and empire,broughtto a climax in the Peloponnesian War. Defeat in the war endedthe empire and it soon also endedthe debate about the kind of governmentAthens was to have. Oligarchy ceasedto be a seriousissuein practicalpolitics. It is only the persistenceof the philosopherswhich createsan illusion aboutit; they continuedto arguefifth-century issuesin the fourth century, but politically in a vacuum.Down to the middle of the fourth century, the actualpolicy questionswere perhapslessdramaticthanbefore, thoughnot necessarilyless vital to the participants-suchmatters as navy finance, foreign relationsboth with Persiaand with other Greek states,and the ever-presentproblem of corn supply. Then came the final great conflict, over the rising power of Macedon. That debatewent on for somethreedecades,and it endedonly in the year following the death of Alexander the Great when the Macedonianarmy put an end to democracyitself in Athens. All these were questionsabout which men could legitimately disagree,and disagreewith passion.On the issues,the arguments of (say) Plato requireearnestconsideration-butonly insofaras he addressedhimself to the issues. The injection of the charge of demagogyinto the polemic amountsto a resort to the very same unacceptabledebatingtricks for which the so-calleddemagogues are condemned.Suppose,for example,that Thucydideswas right in attributing Alcibiades'advocacyof the Sicilian expeditionto his 32

Xen., Apology, 30-2. Seegenerally GeorgesMeautis, L'aristocratie atlJinienne(Paris, I927).

ATHENIAN DEMAGOGUES

personalextravaganceandto variousdiscreditableprivatemotives. What relevance has that to the merits of the proposal itself? Would the Sicilian expedition, as a war measure,have been a better idea if Alcibiades had been an angelic youth? To ask the question is to dismiss it, and all other such argumentswith it. One must dismiss as summarily the objections to oratory: by definition, to wish to lead Athens implies the burdenof trying to persuadeAthens, and an essentialpart of that effort consistedin public oratory. Onecandraw distinctions,of course.I shouldconcedethe label 'demagogue'in its most pejorative sense,for example,if a campaignwerebuilt aroundpromiseswhich a clique of oratorsneither intendedto honour nor were capableof honouring. But, significantly enough,this accusationis rarelylevelledagainsttheso-called demagogues,and the one definite instancewe know comesfrom the othercamp.The oligarchyof 411 was sold to the Athenianson the appealthat this was now the only way to obtain Persiansupport and thus to win the otherwiselost war. Even on the most favourableview, as Thucydidesmakesquite clear, Peisanderand some of his associatesmay have meant this originally, but they quickly abandonedall pretenceof trying to win the war while they concentratedon preservingthe newly won oligarchyon as narrow a baseas possible.33 That is what I should call 'demagogy',if the word is to merit its pejorative flavour. That is 'misleading the people' in the literal sense. But what then of the interest question, of the supposedclash betweenthe interestsof the whole stateand the interestsof a section or faction within the state?Is that not a valid distinction?It is a pity that we haveno direct evidence(and no indirect evidenceof any value) aboutthe way the long debatewas conductedbetween 508 B.C., when Cleisthenesestablishedthe democracyin its primitive form, and the later yearsof Pericles'dominance.Thosewere the years when class interestswould most likely have been expoundedopenly and bluntly. Actual speechessurvive only from the end of the fifth centuryon, and they revealwhat anyone could have guessedwho had not been blinded by Plato and others, namely, that the appeal was customarily a national one, not a factional one. There is little open panderingto the poor against the rich, to the farmersagainstthe town or to the town againstthe 33

Thuc., 8.68-91.

M. 1. FINLEY

farmers. Why indeed should there have been?,Politicians regularly saythat what they areadvocatingis in the bestinterestsof the nation, and, what is much more important,they believeit. Often, too, they charge their opponentswith sacrificing the national interestfor specialinterests,and they believe that. I know of no evidencewhich warrantsthe view that Athenianpoliticians were somehowpeculiar in this respect;nor do I know any reasonto hold that the argumentis an essentiallydifferent (or better) one becauseit is put forth not by a politician but by Aristophanesor Thucydidesor Plato. At the sametime a politician cannot ignore class or sectional interestsor the conflicts amongthem, whetherin a constituency today or in the Assembly in ancient Athens. The evidencefor Athens suggests'that on many issues-theEmpire and the PeloponnesianWar, for example, or relations with Philip of Macedon-thedivisionsover policy did not closelyfollow classor sectionallines. But other questions,such as the opening of the archonship and other offices to men of the lower property censusesor of pay for jury serviceor, in the fourth century, the financing of the fleet, or the theoric fund, were by their nature classissues.Advocateson both sidesknew this andknew how and when (and when not) to make their appealsaccordingly, at the same time that they each argued, and believed, that only their respectivepoints of view would advanceAthens as a whole. To pleadagainstEphialtesand Periclesthat eunomia,the well-ordered stateruled by law, had the higher moral claim, was merely a plea for the statusquo dressedup in fancy language.34 In his little book on the Athenianconstitution,Aristotle wrote the following :35 Pericleswas the first to give pay for jury service,as a demagogicmeasureto counterthe wealth of Cimon. The latter, who possessed the fortune of a tyrant ... supported many of his fellow-demesmen,everyoneof whom was free to come daily and receivefrom him enoughfor his sustenance.Besides,none of his estateswas enclosed,so that 34

3S

'Eunomia... the ideal of the past and even of Solon ... now meantthe best constitution,basedon inequality. It was now the ideal of oligarchy': Ehrenberg,Aspects,p. 92. Arist., Const., 27.3-4. 2.0

ATHENIAN DEMAGOGUES

anyonewho wished could take from its fruits. Pericles' property did not permit such largesse,and on the advice of Damonides... he distributedamong the peoplefrom what was their own ... and so he introducedpay for the jurors. Aristotle himself, as I indicated earlier, praised Pericles' regime and he refusedresponsibilityfor this silly explanation,but others who repeatedit, both beforeand after him, thoughtit was a telling instance of demagogypandering to the common people. The obvious retort is to ask whetherwhat Cimon did was not pandering in equal measure, or whether opposition to pay for jury service was not pandering,too, but in that case to the men of property. No useful analysis is possiblein such terms, for they serveonly to concealthe real groundsfor disagreement.If one is opposedto full democracyas a form of government,then it is wrong to encouragepopularparticipationin the juries by offering pay; but it is wrongbecausethe objective is wrong, not because Periclesobtainedleadershipstatusby proposingand carrying the measure.And vice versa, if one favours a democraticsystem. What emergesfrom all this is a very simpleproposition,namely, that demagogues-Iuse the word in a neutral sense-werea structuralelementin the Athenianpolitical system.By this I mean, first, that the system could not function at all without them; second,that the term is equally applicableto all leaders,regardless of classor point of view; and third, that within ratherbroadlimits they are to be judged individually not by their mannersor their methods,but by their performance.(And that, I needhardly add, is preciselyhow they were judgedin life, if not in books.) Up to a point one can easily parallel the Athenian demagoguewith the modernpolitician, but theresooncomesa point whendistinctions must be drawn, not merely becausethe work of governmenthas becomeso much more complex,but more basicallybecauseof the differences between a direct and a representativedemocracy. I neednot repeatwhat I have alreadysaid about the mass-meeting (with its uncertaincomposition),about the lack of a bureaucracy and a party system,and, as a result, the continuousstateof tension in which an Athenian demagoguelived and worked. But there is one consequencewhich needsa little examination,for theseconditions make up an important part (if not the whole) of the explanationof an apparentlynegativefeatureof Athenianpolitics, 21

M.

r. FINLEY

and of Greek politics generally.David Hume put it this way:36 To excludefaction from a free government,is very difficult, if not altogetherimpracticable;but such inveteraterage betweenthe factions, and such bloody maxims are found, in moderntimes, amongstreligious partiesalone. In ancient history we may always observe,where one party prevailed, whetherthe nobles or people(for I can observeno difference in this respect),that they immediatelybutchered... and banished.... No form of process,no law, no trial, no pardon. . . . Thesepeoplewere extremelyfond of liberty, but seem not to have understoodit very well. The remarkablething about Athens is how near she came to being the complete exception to this correct observation of Hume's, to being free, in other words, from stasisin its ultimate meaning.The democracywas establishedin 508 B.C. following a brief civil war. Thereafter,in its history of nearly two centuries, armedterror, butcherywithout processor law, was employedon only two occasions,in 4II and 404, both times by oligarchic factions which seizedcontrol of the state for brief periods. And the second time, in particular, the democratic faction, when it regainedpower, was generousand law-abidingin its treatmentof the oligarchs,so much so that they wrung praiseevenfrom Plato. Writing about the restorationof 403, he said that 'no one should be surprisedthat somemen took savagepersonalrevengeagainst their enemiesin this revolution, but in generalthe returningparty behavedequitably'.37This is not to suggestthat the two centuries were totally free from individual acts of injustice and brutality. Hume-speakingof Greecegenerally and not of Athens in particular-observed'no difference in this respect' between the factions. We seemto have a less clear vision of Athens, at least, blocked by the distorting mirror of men like Thucydides,Xenophon and Plato, which magnifies the exceptional incidents of extremedemocraticintolerance-such as the trial andexecutionof the generalswho won the battle of Arginusaeand the trial and execution of Socrates;while it minimizes and often obliterates 'Of the populousnessof ancient nations', in Essays,World's Classics edn (London, 1903), pp. 405-6. Cf. JacobBurckhardt, Griechische Kulturgeschichte(reprint Darmstadt,1956), i, pp. SO-I. 37 Epist., VII 325B; cf. Xen., Hell., 2.4.43; Arist., Const., 40.

36

2.2.

ATHENIAN DEMAGOGUES

altogether the even worse behaviour on the other side, for example,the political assassination of Ephialtesin 462 or 461 and of Androcles in 4I I, eachin his time the most influential of the popular leaders. If Athenslargely escapedthe extremeforms of stasisso common elsewhere,shecould not escapeits lessermanifestations.Athenian politics had an all-or-nothing quality. The objective on eachside was not merely to defeatthe oppositionbut to crushit, to behead it by destroying its leaders. And often enough this game was played within the sides, as a number of men manoeuvredfor leadership. The chief technique was the political trial, and the chief instrumentalitieswere the dining-clubs and the sycophants. These,too, I would argue,were structurallya part of the system, not an accidental or avoidable excrescence.Ostracism, the socalled graphe paranomon, and the formal popular scrutiny of archons,generalsand other officials, were all deliberatelyintroducedas safety devices,either againstexcessiveindividual power (and potential tyranny) or againstcorruption and malfeasanceor against unthinking haste and passion in the Assembly itself.38 Abstractly it may be easy enoughto demonstratethat, however praiseworthyin intention, thesedevicesinevitably invited abuse. The trouble is that they were the only kind of device available, again becausethe democracywas a direct one, lacking a party machinery and so forth. Leaders and would-be leaders had no alternativebut to make use of them, and to seek out still other ways of harassingand breaking competitorsand opponents. Hard as this all-out warfare no doubt was on the participants, unfair and vicious on occasion, it does not follow that it was altogether an evil for the community as a whole. Substantial inequalities,seriousconflicts of interest,andlegitimatedivergences of opinion were real and intense.Under such conditions,conflict is not only inevitable, it is a virtue in democraticpolitics, for it is conflict combined with consent, and not consentalone, which preservesdemocracyfrom eroding into oligarchy. On the constitutional issuewhich dominatedso much of the fifth centuryit was the advocatesof populardemocracywho triumphed,and they did so precisely becausethey fought for it and fought hard. They 38

The fourth-centurylegislative procedureby meansof nomothetaicould properly be addedto this list; see A. R. W. Harrison, 'Law-making at Athens at the end of the fifth century B.C.', ]HS, lxxv (1955), pp. 26-35.

M.

r.

FINLEY

fought a partisanfight, and the Old Oligarch made the correct diagnosisin attributing Athenian strengthto just that. Of course, his insight, or perhapshis honesty, did not extend so far as to note the fact that in his day the democracy'sleaderswere still men of substance,and often of aristocratic background: not only Pericles, but Cleon and Cleophon, and then Thrasybulus and Anytus. The two latter led the democraticfaction in overthrowing the Thirty Tyrants in 403, and in following their victory with the amnestywhich even Plato praised. The partisanfight was not a straightclassfight; it also drew supportfrom amongthe rich and the well-born. Nor was it a fight without rules or legitimacy. The democraticcounter-sloganto eunomiawas isonomia,and, as Vlastos has said, the Athenianspursued'the goal of political equality ... not in defiance,but in supportof the rule of law'. The Athenian poor, he noted, did not once raise the standardGreek revolutionary demand-redistribution of the land-throughoutthe fifth and fourth centuries.39 In thosetwo centuriesAthenswas, by all pragmatictests,much the greatestGreek state, with a powerful feeling of community, with a toughnessand resilience tempered,even grantedits imperial ambitions,by a humanityand senseof equity and responsibility quite extraordinaryfor its day (and for many anotherday as well). Lord Acton, paradoxically enough, was one of the few historiansto have graspedthe historic significanceof the amnesty of 403. 'The hostile parties',he wrote, 'were reconciled,and proclaimed an amnesty, the first in history'.4o The first in history, G. Vlastos, 'Isonomia', AjP, lxxiv (1953), pp. 337-66. Cf. Jones, Democracy,p. 52: 'In general... democratstendedlike Aristotle to regardthe laws as a code laid down once for all by a wise legislator ... which, immutable in principle, might occasionallyrequire to be clarified or supplemented'.The 'rule of law' is a complicatedsubject on its own, but it is not the subject of this paper. Nor is the evaluationof individual demagogues,e.g. Cleon, on whom seemost recently A. G. Woodhead,'Thucydides'portrait of Cleon', Mnemosyne,4th ser., xiii (1960), pp. 289-317; A. Andrewes,'The Mytilene debate',Phoenix, xvi (1962), pp. 64-85. 40 'The history of freedom in antiquity', in Essayson Freedom and Power, ed. G. Himmelfarb (London, 1956), p. 64. The paradoxcan be extended: in reviewing Grote, John Stuart Mill wrote about the years leading up to the oligarchic coups of 41 I and 404: 'The Athenian Many, of whose democraticirritability and suspicionwe hear so much, are rather to be accusedof too easyand good-natureda confidence,when we reflect that

39

ATHENIAN DEMAGOGUES

despiteall the familiar weaknesses, despitethe crowd psychology, the slaves,the personalambition of many leaders,the impatience of the majority with opposition.Nor was this the only Athenian innovation: the structureand mechanismof the democracywere all their own invention, as they groped for somethingwithout precedent,having nothing to go on but their own notion of freedom, their community solidarity, their willingness to inquire (or at least to accept the consequencesof inquiry), and their widely sharedpolitical experience. Much of the credit for the Athenianachievementmustgo to the political leadershipof the state. That, it seemsto me, is beyond dispute.It certainlywould not havebeendisputedby the average Athenian.Despiteall the tensionsanduncertainties,the occasional snap judgment and unreasonableshift in opinion, the people supportedPericlesfor morethantwo decades,as they supporteda very different kind of man, Demosthenes,under very different conditionsa century later. Thesemen, and otherslike them (less well known now), were able to carry througha more or lessconsistentand successfulprogrammeover long stretchesof time. It is altogetherperverseto ignorethis fact, or to ignorethe structureof political life by which Athens becamewhat she was, while one follows the lead of Aristophanesor Plato and looks only at the personalitiesof the politicians,or at the crooksandfailures among them, or at someethical norms of an ideal existence. In the end Athenslost her freedomand independence, brought down by a superiorexternalpower. Shewent down fighting, with an understandingof what was at stakeclearerthan that possessed by many critics in later ages. That final struggle was led by Demosthenes,a demagogue.We cannot have it both ways: we cannotpraiseand admirethe achievementof two centuries,andat the sametime dismissthe demagogueswho were the architectsof the political frameworkand the makersof policy, or the Assembly in and through which they did their work.41 they had living in the midst of them the very men who, on the first show of an opportunity, were ready to compassthe subversionof the democracy... .': Dissertationsand Discussions,ii (London, 1859), p. 540. 41 See most recently W. R. Connor, The New Politicians of Fifth-Century Athens(Princeton,1971). and my De11locracy Ancient and Modern (London, 1973)·

II

ARISTOTLE AND ECONOMIC ANALYSIS*t

M. 1. Finley

For the argumentof this paperit is essentialto distinguish, no matterhow crudely, betweeneconomicanalysisand the observation or description of specific economicactivities, and between both and a conceptof 'the economy'(with which only the final sectionwill be concerned).By 'economicanalysis',wrote Joseph Schumpeter,'I mean... the intellectual efforts that men have madein orderto understandeconomicphenomenaor, which comes to the samething, ... the analytic or scientific aspectsof economic thought'.And later, drawing on a suggestionof GerhardColm's, he added:'economicanalysisdealswith the questionshow people behaveat any time andwhat the economiceffectsarethey produce by so behaving;economicsociologydealswith the questionhow they cameto behaveas they do'.1 Whether one is wholly satisfiedwith Schumpeter'sdefinitions or not,2 they will serve our presentpurposes.To illustrate the difference between analysis and observation,I quote the most ... From no. 47 (1970). This essaywas preparedfor the Festschriftfor ProfessorE. C. Welskopf on her seventiethbirthday, andhas appearedin Germantranslationin the Jahrbuchfur Wirtschaftsgeschichte (1971), ii, pp. 8'7-105. An earlier draft was presentedto the Social History Group in Oxford on 3 December 1969. I have benefitedfrom the advice of a number of friends, A. Andrewes,F. H. Hahn, R. M. Hartwell, G. E. R. Lloyd, G. E. M. de Ste Croix. 1 J. Schumpeter,History of EconomicAnalYsis,ed. E. B. Schumpeter (New York, 1954), pp. 1,21. 2 Seethe review by I. M. D. Little in EHR, 2nd ser., vii (1955-6), pp.

t

9 1 - 8.

2.6

ARISTOTLE AND ECONOMIC ANALYSIS

familiar ancienttext on the division of labour, written by Xenophon before the middle of the fourth century B.C. The contextand this should not be ignored-is the superiority of the meals providedin the Persianpalacewith its staff of kitchen specialists.3 That this should be the case[Xenophonexplains] is not remarkable.For just as the various tradesare most highly developedin the large cities, in the sameway the food at the palaceis preparedin a far superiormanner.In small towns the sameman makescouches,doors, ploughsand tables,and often he even builds houses,and still he is thankful if only he can find enoughwork to supporthimself. And it is impossiblefor a man of many tradesto do all of them well. In large cities, however,becausemany make demandson eachtrade, one alone is enoughto supporta man, and often less than one: for instance,one man makesshoesfor men, anotherfor women, there are placeseven where one man earnsa living just by mending shoes,anotherby cutting them out, anotherjust by sewing the upperstogether,while there is anotherwho performsnone of theseoperationsbut assemblesthe parts. Of necessityhe who pursuesa very specializedtask will do it best. This text contains important evidencefor the economic historian-butnot on division of labourfor which it is so often cited. In the first place,Xenophonis interestedin specializationof crafts ratherthanin division of labour. In the secondplace,the virtues of both are, in his mind, improvementof quality, not increasein productivity. He says this explicitly and it is anyway implicit in the context, the meals servedat the Persiancourt. Nor is Xenophon untypical: division of labour is not often discussedby ancientwriters, but when it is, the interestis regularly in craftsmanship,in quality.4 One needonly glanceat the model of the pin factory at the beginning of Adam Smith's Wealth of Nations to appreciate the leap taken by the latter, from observation to genuineeconomicanalysis. Even as observation,furthermore,Xenophon'sremarksdo not merit the accoladesthey have received. As Schumpeterpointed 3 4

Cyropaedia,viii.2.5. SeeEric Roll, A History of EconomicThought, 3rd edn (London, 1954), PP·27-8.

M. 1. FINLEY

out, economics 'constitutesa particularly difficult case' 1ll any 5 study of the origins of a 'science'because common-senseknowledgegoesin this field much farther relatively to such scientific knowledgeas we have beenable to achieve,than does common-senseknowledgein almost any other field. The layman'sknowledgethat rich harvestsare associatedwith low prices of foodstuffs or that division of labour increasesthe efficiency of the productiveprocessare obviously prescientificand it is absurdto point to such statementsin old writings as if they embodieddiscoveries. The key for antiquity restsnot with Xenophonor Platobut with Aristotle. It is agreedon all sides that only Aristotle offered the rudiments of analysis; hence histories of economic doctrine regularly feature him at the beginning. 'The essentialdifference' betweenPlato and Aristotle in this respect,writes Schumpeter,'is that an analytic intention, which may be said (in a sense)to have beenabsentfrom Plato'smind, was the prime moverof Aristotle's. This is clear from the logical structureof his arguments'.6 Aristotle then becomesdoubly troublesome.In the first place, his supposed efforts at economic analysis were fragmentary, wholly out of scalewith his monumentalcontributionsto physics, metaphysics, logic, meteorology, biology, political science, rhetoric, aestheticsand ethics. Second,and still more puzzling, his efforts produced nothing better than 'decorous,pedestrian, slightly mediocre, and more than slightly pompous common sense'.This judgementof Schumpeter's,7sharedby many, is so wide of the universaljudgementof Aristotle's other work, that it demandsa seriousexplanation. 50p. cit., p. 9. Even if one grants Xenophonthe insight that division of labour is a consequenceof greaterdemand,the observationled to no analysis. To quote Schumpeteragain: 'Classicalscholarsas well as economists... are prone to fall into the error of hailing as a discovery everythingthat suggestslater developments,and of forgetting that, in economicsas elsewhere,most statementsof fundamentalfacts acquire importanceonly by the superstructuresthey are made to bear and are commonplacein the absenceof such superstructures'(p. 54). 6 Ibid., p. 57. Cf. e.g. Roll, op. cit., pp. 3I-5. 70p. cit., p. 57.

28

ARISTOTLE AND ECONOMIC ANALYSIS

I

Thereare only two sectionsin the whole Aristotelian corpusthat permit systematicconsideration,one in Book v of the Nicomachean Ethics, the other in Book I of the Politics.s In both, the 'economic analysis'is only a sub-sectionwithin an inquiry into other, more essentialsubject-matters.Insufficient attentionto the contextshas been responsiblefor much misconceptionof what Aristotle is talking about. The subjectof the fifth book of the Ethics is justice. Aristotle first differentiatesuniversalfrom particular justice, and then proceedsto a systematicanalysisof the latter. It, too, is of two kinds: distributive and corrective. Distributive (dianemetikos)justice is a concernwhen honours, goods, or other 'possessions'of the community are to be distributed. Here justice is the sameas 'equality', but equality understoodas a geometricalproportion(we say 'progression'),not as an arithmeticalone.9 The distribution of equalsharesamongunequal persons,or of unequal sharesamong equal persons,would be unjust. The principle of distributive justice is thereforeto balance the sharewith the worth of the person. All are agreedon this, Aristotle adds,althoughall do not agreeon the standardof value (axia) to be employed where the polis itself is concerned.'For democratsit is the statusof freedom, for some oligarchswealth, for othersgood birth, for aristocratsit is excellence(arete)'.10That Aristotle himselffavouredthe last-namedis not importantfor us, and indeedhe does not himself make the point in this particular The first part of Book II of the pseudo-AristotelianOeconomicais without value on any issue relevant to the presentdiscussion,as I have indicated briefly in a review of the Bude edition in CR, new ser., xx (1970), pp. 315-19. (See also note 51.) 9 This difficult idea of a mathematicalformulation of equality and justice was Pythagorean,probably first introducedby Archytas of Tarentum at the beginning of the fourth century B.C., and then popularizedby Plato (first in Gorgias, 50SA). See F. D. Harvey, 'Two kinds of equality', C/assicaet Mediaevalia, xxvi (1965), pp. 101-46, with corrigendain xxvii (1966), pp. 99-100, who rightly stressesthe point that the mathematical formulation is employedsolely to argueagainstdemocracy.(My translationsfrom the Ethics are basedon H. Rackham'sin the Loeb ClassicalLibrary, 1926.) 10 Ethics, I I 3Ia24-29. 8

M. 1. FINLEY

context, which is concerned only to explain and defend the principle of geometricproportion.ll In corrective justice (diorthotikos, literally 'straighteningout'), however, the issueis not one of distribution from a pool, but of direct, private relations betweenindividuals in which it may be necessaryto 'straightenout' a situation, to rectify an injustice by removing the (unjust) gain and restoring the loss. Here the relative natureand worth of the personsis irrelevant,'for it makesno differencewhethera good man has defraudeda bad man or a bad one a good one, nor whether it is a good or bad man that has committed adultery; the law looks only at the nature of the damage,treating the parties as equal. .. .'12 Corrective justice also has two subdivisions, depending on whetherthe 'transactions'(synallagmata)are voluntary or involuntary. Among the former Aristotle lists sales, loans, pledges, depositsand leases;among the latter, theft, adultery, poisoning, procuring, assault, robbery, murder.13 There is a fundamental difficulty for us here in trying to comprehendAristotle's categories-andno translation of synallagmataby a single English word easesit-but I neednot enterinto the controversyexceptto make one point relevant to some of the discussion that will follow. Underwhat conditionsdid Aristotle envisagean injustice, an unjustgain, in a voluntary transaction,especiallyin a sale?The answeris, I think, beyond dispute that he had in mind fraud or breachof contract,but not an 'unjust' price. An agreementover the price was part of the agreementor 'transaction'itself, and there could be no subsequentclaim by the buyer of unjust gain merely becauseof the price. As Joachimsays, 'the law gives the better bargaineradeia (security)'.14It is necessaryto insist on this It is probablethat for Aristotle distributive justice is also operativein a variety of private associations,permanentor temporary: see H. H. Joachim'scommentary(Oxford, 1951), pp. 138-40, though I see neither necessitynor warrant for his attempt to link distributive justice with the private law suit known as diadikasia. 12 Ethics, II31 bP-32a6. 13 Ethics, II 3Ia3-9. 14 Op. cit., p. 137, with specific referenceto II 32bII-16. I agreewith A. R. W. Harrison, 'Aristotle's NicomacheanEthics, Book v, and the law of Athens', ]HS, lxxvii (1957), pp. 42-7, againstJoachim(see also note II), that 'Aristotle's treatmentof justice in the Ethics shows only a very general,one might perhapssayanacademic,interest in the actual legal institutions of the Athens of his day'.

11

ARISTOTLE AND ECONOMIC ANALYSIS

(leaving aside the unfortunateinjection of bargaining) because efforts have beenmadeto drag this sectionof the Ethics into the argumentabout economicanalysis,for exampleby Soudek,who offers as an illustration of correctivejustice the hypotheticalcase of a house-buyerwho brought suit on a claim that he had been overchargedand who was awardeda refund equal to half the difference betweenthe seller's price and his own proposed'just price'.1sNothing in this or any other text of Aristotle warrants this, nor doesanythingwe know aboutGreeklegal practice.Both arguedecisively the other way. Commentingon the famous passagein the Iliad, 'But then Zeussonof Cronustook from Glaucus his wits, in that he exchangedgoldenarmourwith Diomedesson of Tydeusfor one of bronze,the worth of a hundredoxenfor the worth of nine oxen', Aristotle saystersely, 'one who gives away what is his own cannotbe said to suffer injustice'.16 We shall meet 'what is his own' againin a surprisingcontext. Having completedhis analysisof the two kinds of particular justice,Aristotle abruptlylaunchesinto a digression,17 introducing it polemically: 'Theview is held by somethat justiceis reciprocity (antipeponthos)without any qualification, by the Pythagoreansfor example'.Antipeponthosis a term that hasa technicalmathematical sense,but it also has a general sensewhich, in this context, amountsto the lex talionis, an eye for an eye.1S On the contrary, replies Aristotle, 'in many casesreciprocity is at variance with justice', sinceit 'doesnot coincide either with distributive or with corrective justice'. However, in the 'interchangeof services'the Pythagoreandefinition of justiceis appropriate,providedthe reciprocity 'is on the basisofproportion, not on the basisof equaliry'. 'Interchangeof services'is Rackham'sinadequatetranslationof Aristotle's EV TCXIS KOIVc.vV!CXIS TCXIS aAACXKTIKeXlS, losing the force of the word koinonia, and I am compelled to digress.Koinonia is a central concept in Aristotle's Ethics and Politics. Its range of meaningsextendsfrom thepolis itself, the highestform of koinonia, J. Soudek,'Aristotle's theory of exchange:an inquiry into the origin of economicanalysis',PAPS, xcvi (1952), pp. 45-75, at pp. 51-2,. 16lliad, 6.2,34-6;Ethics, 1I36b9-13. 17 Ethics, 1I32,b2,I-33b2,9. 18 Cf. Magna Mor., 1I94a2,9If.; seeJoachim,op. cit., pp. 147-8, and the commentaryby R. A. Gauthierand J. Y. Jolif (the best commentaryin so far as close readingof the text is concerned),ii (Louvain and Paris, 1959), pp. 372,-3· IS

M. I. FINLEY

to temporaryassociationssuchas sailorson a voyage,soldiersin a campaign,or the partiesin an exchangeof goods.It is a 'natural' is by naturea zoiin koinonikonas well as form of association-man a ZOiin oikonomikon(household-being)and a zoiin politikon (polisbeing). Severalconditionsare requisiteif there is to be a genuine koinonia: (1) the membersmust be free men; (2) they must have a commonpurpose,major or minor, temporaryor of long duration; (3) they must havesomethingin common,sharesomething, such as place, goods,cult, meals,desirefor a good life, burdens, suffering;(4) theremustbephilia (conventionallybut inadequately translated'friendship'), mutuality in other words, and to dikaion, which for simplicity we may reduceto 'fairness'in their mutual relations. Obviously no single word will renderthe spectrumof koinoniai. At the higher levels, 'community' is usually suitable,at the lower perhaps'association',providedthe elementsof fairness, mutuality and commonpurposeare kept in mind. The point to my digressionis to underscorethe overtonesof the section in the Ethics on exchange:koinonia is as integral to the analysisas the act of exchanging.EdouardWill caughtthe right nuancewhen he replacedsuch translationsof the openingphrase as 'interchangeof services'by a paraphrase,'exchangerelations within the framework of the community'(les relations d'echangequi ont pour cadre Ie communaute').19Lest there be any doubt, Aristotle himself promptly dispels it. Immediately following the sentences I quoted before digressing,he goes on to say that the polis itself dependson proportional reciprocity. If men cannot requite evil with evil, good with good, there can be no sharing. 'That is why we set up a shrine to the Charites [Graces] in a public place,to remind men to make a return. For that is integral to grace, since it is a duty not only to return a service done one, but another time to take the initiative in doing a service oneself'.20 And at long last we cometo our problem.The exampleof proportional requital which follows is the exchangeof a housefor shoes.21 How is that to be accomplished?There is no koinonia in E. Will, 'De l'aspectethique des origines grecquesde la monnaie',RH, CCXii(1954), pp. 209-31, atp. 215 note 1. 20 Ethics, II 3P3-5. 21 Aristotle shifts from exampleto exampleand I have followed him, despitethe superficial inconsistencythat entails. 19

32

ARISTOTLE AND ECONOMIC ANALYSIS

this contextbetweentwo doctors,but only between,say,a doctor and a farmer, who are not equals but who must somehowbe equalized.'As a builder is to a shoemaker,so must so many pairs of shoesbe to a house'.The latter must be 'equalizedsomehow', by some common measure, and that is need (chreia),22 now commonly expressedin money. 'There will therefore be reciprocity when(the products)havebeenequalized,so that as farmer is to shoemaker,so is the shoemaker'sproduct to that of the farmer'. In that way, there will be no excessbut 'eachwill have his own'. If one party has no need there will be no exchange, and again money comes to the rescue: it permits a delayed exchange.23 There follows a short repetitive sectionand the digressionon 'this outwork of particular justice' ends.24 Aristotle has been thinking aloud, so to speak,as he often doesin his writings asthey have come down to us, about a particularnuanceor a tangential questionthat is troublesome;he is indulging in a highly abstract exercise,analogousto the passagesin the Politics on the application of geometricproportion to public affairs; here, as often, his reflections are introduced by a polemical statement,and soon droppedas he returns to his main theme,his systematicanalysis. Exchangeof goodsdoes not again appearin the Ethics exceptin two or three casualremarks. That this is not one of Aristotle's more transparentdiscussions is painfully apparent,and we must look at what the most important modern commentatorshave made of it. Joachim, exceptionally, acceptedthat Aristotle really meantit when he wrote 'as a builder is to a shoemaker',andhe promptly added,'How exactly the values of the producersare to be determined,and what the ratio betweenthem can mean,is, I must confess,in the end unI have refrainedfrom the commonrendition, 'demand',to avoid the subconsciousinjection of the moderneconomicconcept;so also Soudek, op. cit., p. 60. The semanticcluster aroundchreia in Greek writers, including Aristotle, includes'use', 'advantage','service',taking us even further from 'demand'. 23 Ethics, II33b6-1Z. In the Politics, I257a31 ff., Aristotle explains that delayedexchangebecamenecessarywhen needswere satisfiedby imports from foreign sources,and 'all the naturally necessarythings were not easily portable'.(My translationsfrom the Politics are basedon Ernest Barker's,Oxford, 1946.) 24 The phrasequotedis that of Harrison, op. cit., p. 45.

22

M. 1. FINLEY

intelligible to me'.25 Gauthierand Jolif make an ingeniouseffort to get round the difficulty by assertingthat the builder and shoemakerare meantto be consideredequal 'as persons'but different (only) in their products.However,I cannotbelievethat Aristotle went out of his way to insist on proportional reciprocity as necessary for justicein this one field, only to concludethat one pair of ratios does not in fact exist, and to make that point in the most ambiguousway possible.26 Max Salomonachievesthe sameresult by more ruthless methods:the mathematics,he says, is a mere 'interpolation', a 'marginal note, so to speak, for listeners interestedin mathematics',and the whole concept of reciprocal proportion must be omitted, leaving Aristotle to say simply that goodsareexchangedaccordingto their values,and nothing more. That then leads Salomonto a seriesof grotesquetranslationsin order to get out of the text what is not there.27 Salomon'sdrastic surgery was not mere wilful caprice. Economics,he writes, cannotbe turnedinto 'a kind of wergeldsystem on a mercantilebase'.28 The first principle of a marketeconomyis, of course,indifferenceto thepersonsof the buyerand seller: that is what troubles most commentatorson Aristotle. Soudektherefore 25 Op. cit., p. 150. 260p. &it., p. 377. They cite in support Magna Mor., II94a7-25, but those lines are only a simplified and more confusingstatementof the argument in the Ethics. For future reference,it should be noted that Magna Mor. saysexplicitly that 'Plato also seemsto employ proportional justice in his Republic'. St GeorgeStock, in the Oxford translation(1915), cites Rep., 369D, but it requiresclairvoyanceto seethe Magna Mor. reference there, since Plato is not discussingat all how the exchangebetween builder and shoemakeris to be equated,and soon goes on to introduce the trader as a middleman(significantly absentin the Aristotelian account).In general,however,his sectionof Book II of the Republicwas obviously influential on Aristotle (including the stresson needand the explanationof money). For what it is worth, in reply to the commentary by Gauthier and Jolif cited above note 18, I note that Plato says (370A-B), to justify specializationof crafts, that 'no two peopleare born exactly alike. There are innate differenceswhich fit them for different occupations'(Cornford's translation,Oxford, 1941). 27 Max Salomon,Der Begriff der Gerechtigkeitbei Aristoteles(Leiden, 1937), in a lengthy appendix,'Der Begriff des Tauschgeschiiftesbei Aristoteles'. My quotationappearson p. 161. Salomonis not alone in dismissingthe mathematicsas irrelevant: see most recently W. F. R. Hardie, Aristotle's Ethical Theory (Oxford, 1968), pp. 198-201. 280p. cit., p. 146.

ARISTOTLE AND ECONOMIC ANALYSIS

suggeststhat 'as a builder is to a shoemaker'must be read 'as the .29 From thereit skill of the builder is to the skill of the shoemaker' is no greatstepto Schumpeter'sinterpretation.The key passagein the Ethics, he writes, 'I interpretlike this: "As the farmer'slabour compareswith the shoemaker'slabour, so the product of the farmer compareswith the product of the shoemaker".At least, I cannotget any other senseout of this passage.If! am right, then Aristotle was groping for somelabour-costtheory of price which he was unableto stateexplicitly'.30 A few pageslater Schumpeter refers to the 'just price' of the artisan's'labour', and still later he assertsthat the 'relevantpart' of Aquinas's'argumenton just price ... is strictly Aristotelian and shouldbe interpretedexactly as we have interpretedAristotle's'.3! However, Aristotle does not once refer to labour costs or costs of production. The medieval theologianswere the first to introducethis considerationinto the discussion,as the foundation for their doctrine of just price, and their alleged Aristotelianism in this respect rested on the ambiguity of the Latin translationsof Aristotle made available to them in the middle of the thirteenthcentury.32 Anyway, none of theseinterpretationsof what Aristotle 'really meant' answersthe question,How are prices, just or otherwise, establishedin the market? More specifically, how are needs,on which Aristotle insists as basic, equatedwith the parties or their skills or their labour or their labour costs,whicheverone prefers? Obviously Aristotle does not say, or at least does not say clearly, otherwisethe modern efforts to discover his concealedmeaning For Karl Marx the answeris that,though would all be unnecessary. Aristotle was the first to identify the centralproblemof exchange value, he then admits defeat'and gives up the further analysisof 2' Soudek,op. cit., pp. 45-6, 60. The samesuggestionis madeby

J. J.

Spengler,'Aristotle on economicimputation and related matters', SouthernEconomicJournal, xxi (1955), pp. 371-89. 300p. cit., p. 60 note I. 31 Ibid., pp. 64, 93. Hardie, op. cit., p. 196, simply assertswithout serious discussionthat 'the comparativevalues of producersmust in Aristotle's view here mean the comparativevalues of their work done in the same time' (my italics). 32 See Soudek,op. cit., pp. 64-5; J. W. Baldwin, The Medieval Theoriesof the Just Price (TAPS, new ser., xlix, part 4 (1959», pp. 62, 74-5; E. Genzmer,'Die antiken Grundlagender Lehre vom gerechtenPreis und der laesio enormis',Z.J ausliindischesu. internat. Privatrecht, Sonderheftxi (1937), pp. 25-64, at pp. 27-8.

35

M. I. FINLEY 33 that 'it is impossiblefor the form of value' when he concedes things so different to becomecommensurablein realiry'.34 Soudek repeatshis error on corrective justice, already discussed,then graspsat the word 'bargain'which W. D. Rossfalsely injects into his translation in one passage(and Rackham in several), and concludesthat the price is determined,and justice satisfied, by mutualbargaininguntil agreementis reached.35That is not a very goodway to describewhat happensin a real marketsituation,and Soudek suggeststhat Aristotle's trouble was that 'he was preoccupiedwith the isolatedexchangebetweenindividuals and not with the exchangeof goodsby many sellersandbuyerscompeting with each other'36-a strange criticism of a discussion that explicitly setsout to look at exchanges'within the framework of the community'.

Ethics, IIHbIS-20. Marx, Capital, trans!' S. Moore and E. Aveling, i (Chicago, 1906), p. 6S. Cf. Roll, op. cit., p. 35; 'What beginswith the promiseof being a theory of value ends up with a mere statementof the accountingfunction of money'. 350p. cit., pp. 61-4. Both Ross(Oxford, 1925) and Rackhamhave 'bargain' in I I Han, Rackhamalso in II64a20; II64a30. (It is worth noting anothermistranslationby Rackham,at IIHbI5: 'Hencethe proper thing is for all commoditiesto have their prices fixed'. What Aristotle actually says is 'Thereforeit is necessaryfor everything to be expressed in money, telimesthai'.)Furthermore,I cannotacceptSoudek'suse of passagesfrom the beginningof Book IX, continuing the analysisof friendship, as relevant. There Aristotle's examplesare drawn from promisesto pay for servicesby musicians,doctors and teachersof philosophy, 'exchanges'in a senseperhaps,but in a sensethat is different in quality from thoseBook v is concernedwith. That should be clear from a numberof passages.In the openingstatement(II63b32-35), Aristotle distinguishes'dissimilar friendships'(which he is about to discuss)from exchangerelations amongcraftsmen,and he soon says explicitly that the value of a philosopher'sservices'is not measurable in money' (II64b3-4). Protagoras,he writes, acceptedwhateverfee his pupils thought proper(II64a24-26),and Aristotle thinks that is on the whole the right procedure(II64b6-S), though he cannotrefrain from the sneer(II64a30-32) that Sophistshad better take their paymentin advance.All this seemsto me to belong to the spirit of gift and counter-gift, of the Charites.There must be reciprocity and proportion here, too, as in all human relations, but I seeno other link to the digressionon the exchangebetweenbuilder and shoemaker. 36 Soudek,op. cit., p. 46. 33 34

ARISTOTLE AND ECONOMIC ANALYSIS

Schumpetertakesthe oppositeline. Startingfrom the erroneous idea that Aristotle 'condemned[monopoly] as "unjust" , he went on to reasonin this way:37 It is not farfetchedto equate,for Aristotle's purpose, monopoly prices with prices that someindividual or group of individuals have set to their own advantage.Prices that are given to the individual and with which he cannottamper, that is to say, the competitiveprices that emergein free market under normal conditions, do not come within the ban. And there is nothing strangein the conjecturethat Aristotle may have taken normal competitiveprices as standardsof commutativejustice or, more precisely, that he was prepared to acceptas 'just' atry transactionbetweenindividuals that was carried out at such prices-whichis in fact what the scholasticdoctors were to do explicitly. We neednot discusswhetheror not it is 'farfetched'to conjecture that all this was in Aristotle's mind, though not expressedin his text; it surely takes us away completely from the starting-point stated in the introduction, with its reference to Pythagorean reciprocity and its consequentmathematics. Scumpeterfurther observedthat the analysiswas restrictedto the artisan, while the 'chiefly agrarianincome of the gentleman' was ignored, the free labourer,'an anomalyin his slaveeconomy', was 'disposedof perfunctorily', the trader,shipowner,shopkeeper and moneylenderjudged only in ethical and political terms, their 'gains' not submitted to 'explanatory analysis'.38 No wonder Schumpeter dismissed the whole performance as 'decorous, Op. cit., p. 61. Both referencesto monopoly which he adducesare incorrect. Politics, I259a5-36,has no condemnationbut rather an implied defenceof public monopoly, whereasthe Ethics, I I 32b21-34aI6, makes no mention of monopoly at all (nor does the Ethics, anywhereelse). Schumpeteris here also repeatinghis error about the scholastic theologians,from whom he takes the unfortunateword 'commutative'. Soudek,op. cit., p. 64, also drags in a condemnationof monopoly, on the untenable(and irrelevant) ground that 'if the seller holds a monopolistic position, then what appearson the surfaceas a "voluntary transaction"is distorted in spirit'. For a correct analysisof the Politics passageon monopoly, see M. Defourny, Aristote, Etudesstir fa 'Politique' (Paris, 1932), pp. 21-7. 380p. cit., pp. 64-5. 37

37

M.

r.

FINLEY

pedestrian,slightly mediocre, and more than slightly pompous common sense'.39An analysis that focuses so exclusively on a minor sector of the economydeservesno more complimentary evaluation.Indeed,the time hascometo ask whetherit is, or was intendedto be, economicanalysisat all. BeforeI proceedto give a negativeanswer,I mustconfessthat, like Joachim,I do not understandwhat the ratios betweenthe producerscanmean,but I do not rule out that 'asa builder is to a shoemaker'is somehowto be taken literally. Marx believedthat there was 'an important fact which preventedAristotle from seeingthat, to attributevalueto commodities,is merelya modeof expressingall labour as equalhumanlabour, and consequentlyas labour of equalquality. Greeksocietywas foundedupon slavery, and had, therefore,for its naturalbasis,the inequality of men and of their labourpower'.40 That naturalinequalityis fundamentalto Aristotle's thinking is beyondargument:it permeateshis analysis of friendship in the Ethics and of slaveryin the Politics. True, his builder andshoemakerin the exchangeparadigmarefree men,not slaves,41but the concurrentexistenceof slave labour would still bar his way to a conceptionof 'equalhumanlabour'.42 Schumpternoticed,but brushedaside,what seemsto me to be central in any judgement,namely, that Aristotle by his silence separatesthe artisanfrom the trader,that he is talking exclusively of an exchangebetweentwo producerswithout the intervention of a middleman.Aristotle knew perfectlywell that this wasnot the way a large volume of goods circulated in his world. He also knew perfectlywell that pricessometimesrespondedto variations in supplyanddemand-thatis the point underlyinghis pagein the Politics on monopoly. In the discussionof moneyin the Ethicshe remarksthat money 'is also subjectto changeand is not always worth the same, but tends to be relatively constant'.43This observationis repeatedin the Politics in a concreteapplication:in the section on revolutions, Aristotle warns againstrigidly fixed 39 Ibid., p. 57. 400p. cit., p. 69. On Marx's views on Aristotle, seeE. C. Weiskopf, Die ProduktionStlerhiiltnisseim alten Orient und in der griechisch-romischen Antike (Berlin, 1957), pp. 336-46. 41 That seemscertainfrom Ethics, II63b32.-35. 42 SeeJ.-P. Vernant, My the el penslechezles Grecs(Paris, 1965), ch. 4. 43 Pol., I2.S9aS-36.Ethics, 1333bI3-14.

38

ARISTOTLE AND ECONOMIC ANALYSIS

assessments in statesthat have a propertyqualification for office, since one should allow for the impact on the assessment'when thereis an abundanceof coin'.44 In short,price variationsaccordingto supplyand demandwere a commonplacein Greeklife in the fourth centuryB.C.45 Yet in the Ethics Aristotle doesnot use any of the normal Greek words for tradeand trader(ashe doesruthlesslyin the Politics), but clings to the neutralword 'exchange'.Thatthis is deliberateI cannotdoubt: in the passagein the Republicon which much of this sectionof the Ethics is a kind of commentary,Plato concedesthat the polis requires petty traders(kapeloi) who will give moneyfor goodsand goodsfor money becauseneitherfarmers nor artisanscan count on finding someonewith whom to exchangewheneverthey bring goods to the market. Aristotle, however, cannot introduce the kapelos, since justice in the exchange(which is not Plato's question) is achievedwhen 'eachhas his own', when, in other words, thereis no gain from someoneelse's10ss.46As part of a theory of price this is nonsense,and Aristotle knew it to be nonsense. Thereforehe was not seekinga theory of market prices.47 The digressionon exchange,I repeat,was placed at the start 'within the framework of the community'. When the digression 44

Pol., 1308a36-38.NowheredoesAristotle explain why moneyis 'relatively

constant'comparedwith other commodities.The generalobservation, it should be noted, had alreadybeenmadeby so shallow a thinker as Xenophon,Wtrys and Means, 4.6. 4S I should perhapsnot have botheredwith theseseemingplatitudes,were it not that Karl Polanyi, 'Aristotle discoversthe economy',in Trade and Market in the EarlY Empires, ed. K. Polanyi, C. M. Arensbergand H. W. Pearson(Chicago, 1957), pp. 64-94, makesthe strangeremark (p. 87) that 'the supply-demand-price mechanismescapedAristotle. The distribution of food in the market allowed as yet but scantroom to the play of that mechanism.... Not before the third century B.C. was the working of a supply-demand-price mechanismin internationaltrade noticeable'.How wrong that is will be evident from Lysias' zznd oration, Against the Corndealers,to be datedabout 387 B.C., on which see R. Seagerin Historia, xv (1966), pp. 17z-84,or from Demosthenes xxxii.Z4-z5 and Pseudo-Demosthenes Ivi.9-IO, half a century later. (Polanyi'schapterhas beenreprintedin the volume cited in note 68, but my referencesare to the original publication.) 46 Rep., 371S-C.Ethics, I I Ba3l-b6. 47 This is also the conclusionof Polanyi, op. cit. Although our analyses diverge, often sharply(seenote 45), I must warmly acknowledgehis having introducedme to the problem nearly twenty years ago.

39

M. I. FINLEY

ends,furthermore,Aristotle resumesthe main threadas follows :48 'We must not forget that the subjectof our investigationis both justice in the absolute senseand political justice'. The phrase 'political justice' is an excessivelyliteral renderingof the Greek, for Aristotle goes on to define it as 'justice among free and (actuallyor proportionately)equalmen, living a communitylife in order to be self-sufficient [or for the purposeof self-sufficiency],. Monetarygain hasno placein suchan investigation:'The moneymakeris someonewho lives under constraint'.49 It is in the context of self-sufficiency,not money-making,that needprovides the measuring-rodof just exchange(and that theproper use of money also becamenecessaryand thereforeethically acceptable).In the Ethics, in sum, there is strictly speaking no economic analysis rather than poor or inadequateeconomicanalysis.

II

It will have beennoticed that in the Ethics Aristotle doesnot ask how farmersor shoemakerscometo behaveastheydo in exchange. In Schumpeter'sterms, then, in the Ethics there is no economic sociologyeither. For that we must turn to Book I of the Politics, and againbegin by carefully fixing the contextin which exchange is discussed.Aristotle first establishesthat both the householdand the polis are natural forms of humanassociation,and proceedsto examinevarious implications, such as the relations of dominance and subjection(including betweenmastersand slaves). Then he turns to property and 'the art of acquiring it' (chrematistike)and asks whether the latter is identical with the art of household management(oikonomike).50His choice of words is importantand has led to much confusionand error. Oikonomike(or oikonomia)in Greek usagenormally retains the primary meaning, 'the art of householdmanagement'.Though that may involve 'economic' Ethics, I 134az4-z6. EthiC!, I096a5-6. For this translationof 6 S~ XP1')l.IcrTlaT';s Ilku6s Tis ~aTIV, see Gauthier and Jolif, op. cit., pp. 33-4, whose commentarycuts through all the unnecessaryemendationand elaborateinterpretationthe text has beensubjectedto. 50 Politics, 1256aI-5.

48 49

ARISTOTLE AND ECONOMIC ANALYSIS

activity, it is misleading,and often flatly wrong, to translateit as 'economics'.51But chrematistikeis ambiguous.(Its root is the noun chrema,'a thing one needsor uses',in the plural chremata,'goods, property'.) We have already met chrematistike(and we shall soon meetit again)in the senseof 'the art of money-making',but hereit hasthe moregenericsenseof acquisition,lesscommonin ordinary Greek usagebut essentialto Aristotle's argument.For he soon concludesthat oikonomia and chrematistike(in the money-making sense) are different though overlapping species of the genus chrematistike.52 Exchange again enters the discussion polemically. What is wealth, Aristotle asks?Is it, as Solon had said, limitless? Or is it a meansto an end and thereforelimited by that end?S3 The answer is categorical.Wealth is a means,necessaryfor the maintenanceof the householdand the polis (with self-sufficiencya principle in the background),and, like all means,it is limited by its end. Of course, he continues,thereis the second,money-makingsenseof chrematislike, and that is what has led to the false opinion that there is no limit to wealth and property.This attitudeto wealth indeedseesit as limitless, but it is against nature and therefore not a proper subject of ethical or political discourse, on his fundamental principle that ethics has a natural basis. ('The money-maker',we Occasionallythe word oikonomia was extendedto the public sphere,and even then it usually refers to administrationin general,as when Dinarchus(i. 97) calls Demosthenes'uselessin the affairs (oikonomiai) of the city' (note the plural). The furthest extensionis to be found in a brief section at the beginning of Book II of the pseudo-Aristotelian Oeconomica(1345b7-46a25),in which four types of 'economy'are said to exist: royal, satrapic,city-state and private. There follow six short paragraphsof excruciatingbanality about the sourcesof revenuein each of the types, and that is the end of the discussion. 52 Beginning with the Sophists,philosopherswere faced with the problem of creatingthe vocabularyfor systematicanalysisout of everydaywords. One increasinglycommon device was to employ the suffix -ikos. There are some sevenhundredsuch words in Aristotle, many first employed by him. See P. Chantraine,La formation des nomsen grec ancien (Paris, 1933), ch. 36. Polanyi, op. cit., pp. 92-3, was right to insist that failure to distinguish betweenthe two meaningsof chrematistikeis fatal to an understandingof this section of the Politics; cf. Defourny, op. cit., pp. 5-7; briefly Barker, Notes E and F of his translation(pp. 22 and 27), though he adds new confusion by suggesting'domesticeconomy'and 'political economy'as English equivalents. 53 Pol., 1256b30-34.

51

4I

M.

r.

FINLEY

rememberfrom the Ethics, 'is someonewho lives under constraint'.)54 Although Aristotle singles out the obolostates,the petty usurer living on small consumerloans,as the most unnaturalof all practitioners of the art of money-making55-money'cameinto exist- , the ence for the sake of exchange,interest makesit increase' type he selectsasthe exemplaris the kapelos,just the manwe noted as missing from the analysisof exchangein the Ethics. Again the choice of words is significant. Greek usagewas not wholly consistent in selecting among the various words for 'trader', but kapelos usually denoted the petty trader, the huckster, in the market-place.In the presentcontext,however,the accentis not on the scaleof his operationsbut on the aim, so that kapelike,theart of the kapelos, must be translated'trade for the sake of gain' or simply 'commercialtrade'.56 Like Plato beforehim, Aristotle now asksthe historical question,how did exchangecometo take place altogether. His answer is that as the koinonia grew beyond the individual household, there were shortagesand surplusesand these were corrected by mutual exchange,'as many barbarian tribesdo to this day ... Whenusedin this way, theart of exchange is not contrary to nature, nor in any way a speciesof the art of money-making.It simply servedto satisfythenaturalrequirements of self-sufficiency'.57 But then,becauseof the difficulties createdby foreign sourcesof supply(a passageI havealreadyquotedin note 23), money was introduced, and out of that there developed kapelike.Its end is not 'the naturalrequirementsof self-sufficiency' but the acquisitionof moneywithout limit. Suchacquisition-we should say 'profit'-is made 'not according to nature but at the S4

Ethics, 1096a5-6.

ss Pol.,

1258b2,-8. Polanyi, op. cit., pp. 91-2" was almostalone in seeingthe point. However, I cannotaccepthis explanations,that 'no namehad yet beengiven to "commercialtrade", (p. 83) and that Aristotle, with a kind of Shavian wit, was exposingthe fact that 'commercialtrade was no mystery ... but hucksteringwritten large' (p. 92,). Polanyi did not take sufficient notice of the Platonic background. S1 Pol., 1257a2,4-30.It is worth noting the contrastwith the 'simplest' model for 'an economictheory of the city state'put forward by John Hicks, A Theory of EconomicHistory (Oxford and London, 1969), pp. 42,-6. That startswith the exchangeby merchantsof oil for corn, 'and the trade is unlikely to get startedunless,to begin with, it is a handsome profit'.

S6

42

ARISTOTLE AND ECONOMIC ANALYSIS

expenseof others',58a phrasethat echoesin reversethe 'eachhashis own' of the Ethics and gives the final proof that commercial exchangewas not the subjectin the Ethics. Aristotle was so rigorousin the ethicalargumentthat he refused to make even Plato's concession.The kape/osis not only un.59 That this was not meant as a natural, he is also 'unnecessary' 'practical'proposalis certain,but that is irrelevantin the present analysis.60 What is relevantis that Aristotle extendedhis ethical judgementsto embracethe highestform of koinonia, thepolis itself. The state, like the householder,must sometimesconcernitself with acquisition.61 Hence, in the discussionof the ideal state,in Book VII of the Politics, he recommendsthat the polis be sited so as to have easy accessto food supplies, timber and the like. That immediately plunges him into another current debate,whether connectionwith the seais a good or badthing, andhe decidesthat 62 the advantagesoutweigh the disadvantages. It should be able to import thosethings which it does not itself produce,and to export the surplus of its own necessities.It should practisecommercefor itself [Aristotle now switchesfrom kapeliketo the commonestword for foreign trade, emporikeor emporia], but notfor others. States which make themselvesmarket-placesfor the world only do it for the sake of revenue;and since it is not proper for a polis to sharein such gain,it ought not have such an emporium.

Therewill haveto be merchants,of course,but 'any disadvantage which may threatencan easily be met by laws defining the persons who may, or may not, have dealingswith one another'.63 Pol., 1258bI-2. Pol., I258aI4-I8. 60 Soudek,op. cit., pp. 71-2, seesa programmaticdifference betweenPlato and Aristotle. Basing himself on Laws, 9I8A-920C, and forgetting both Rep., 37IB-C (which I quotedearlier) and Pol., I327a25-3I (quotedlater in this paragraph),Soudekwrites that 'the author of the Laws . .. had made his peacewith moneymakingand plutocracy, while Aristotle never gave up his opposition to this class'.Beneaththis fundamental misunderstandinglies an equally fantastic picture of a sharp class strugglein Greecebetweenwealthy landownersand merchants. 61 Pol., 1258aI9-34,59a34-36. 62 Pol., 1327a25-31. 63 Plato of coursedrafted the legislation, Laws, 919D-920D.

58

59

43

M. I. FINLEY

Nowherein the Politics doesAristotle ever considerthe rules or mechanicsof commercial exchange.On the contrary, his insistence on the unnaturalnessof commercial gain rules out the possibility of sucha discussion,and also helpsexplain the heavily restrictedanalysisin the Ethics. Of economicanalysisthereis not a trace.

III One could rest the argumentthere, perhapsadding the familiar point that Aristotle, and even more Plato before him, were in many respectsresisting the social, economic,political and moral developmentsof fourth-century Greece. There is the famous analogy of the way Aristotle appearsto ignore completely the for the careersof Philip and Alexander, and their consequences polis, the natural form of political association.He was therefore equally free to ignore the unnaturaldevelopmentsin commercial tradeand money-making,despitetheir growth in the sameperiod and the tensionsthey generated.Schumpeterwas right to comment that 'preoccupationwith the ethics of pricing ... is precisely one of the strongestmotives a man can possibly have for analysingactual market mechanisms'.64It does not follow, however, that ethical preoccupationsmustlead to suchan analysis,and I have tried to show that 'pricing' was actually not Aristotle's concern. In the end, Schumpeteropted for a strictly 'intellectual' explanation. Although he wrote in his introduction that 'to a large extent,the economicsof different epochsdealwith different setsof facts and problems',65he ignored that point when he excused 66 Aristotle for being mediocreand commonsensica1. There is nothing surprising or blameworthyin this. It is by slow degreesthat the physical and social facts of the empirical universeenter the rangeof the analytic searchlight.In the beginningsof scientific analysis,the massof the phenomenais left undisturbedin the compoundof common-sense 640p. cit., p. 60. 65 Ibid., p. 5. 66 Ibid., p. 65. See the generalcriticism by Little, op. cit. in note 2.

44

ARISTOTLE AND ECONOMIC ANALYSIS

knowledge,and only chips of this massarousescientific curiosity and thereuponbecome'problems'. Yet Aristotle's scientific curiosity has rarely been paralleled,and the time has come to ask; the massof what phenomena?Would an economicanalysishave beenpossiblehad his (or anyoneelse's) interest not beendeflected?Indeed, would even a descriptionof 'the economy'have beenpossible? Today we write books with such titles as The Economicsof Ancient Greece, and the chaptersare headedagriculture, mining and minerals, labour, industry, commerce,money and banking, public finance-Schumpeter's 'chips' of the 'massof the phenomena'.67This learned activity presupposesthe existenceof 'the economy'as a concept,difficult as it hasbecometo find a generally acceptabledefinition. The currentdebateabout'economicanthropology', largely stimulatedby Karl Polanyi'sinsistenceon a sharp distinction between what he called the 'substantive' and the 'formal' definitions of the economy,68is a debateaboutdefinitions and their implications for (historical) analysis, not about the existenceof 'the economy'.As Polyani himself said, evenin early societies'only the conceptof the economy,not the economyitself, is in abeyance'.69No one could disagree with his substantive definition; in one of his varied formulations it is 'an instituted processof interactionbetweenman and his environment,which resultsin a continuoussupply of want-satisfyingmaterialmeans';70 his opponentsmerely deny that this is a sufficient operational definition. 'Modern economists make even Robinson Crusoe speculateupon the implicationsof choicewhich they regardas the essenceof economy'.7l The title and chapterheadingsare those of H. l\fichell's book, znd edn (Cambridge,1957). 68 Polanyi's theoreticalessayshave been convenientlyassembled under the title, Primitive, Archaic, and Modern Economies,ed. G. Dalton (Garden City, N.Y., 1968). For a commentaryon the debate,with extensive bibliography, see l\f. Godelier, 'Object et methodede l'anthropologie economique',L'Homme, v. no. 1. (1965), pp. 32.-91, reprinted in his Rationaliti et irratiollaliti en economie(Paris, 1966), pp. 1.32.-93; S. C. Humphreys,'History, eCQnomicsand anthropology:the work of Karl Polanyi', History and Theory, viii (1969), pp. 165-1.11.. 69 Polanyi, op. cit., p. 86. 70 Ibid., p. 145. 71 Roll, op. cit., p. ZI. 67

45

M. 1. FINLEY

Nor were the Greeks themselvesunawarethat men procure their want-satisfactionsby social (Polanyi's 'instituted') arrangements, or that there were things to be said about agriculture, mining, moneyor commerce.Aristotle refers readerswho may be interestedto existing bookson the practicalside. He mentionsby namethe authorsof two agronomictreatises,72andmuchpractical information is scatteredin the surviving botanicalwritings of his pupil Theophrastus.Greekswho thought about the matter were also aware that their want-satisfyingarrangementswere technologically and socially more complexthan had beenthe casein the past. The Homeric poems and the witness of contemporary 'barbarians'wereproof enough.Greek'historical' accountsof the development from early times were largely speculative. One should not attribute to them too much accurateknowledge of the paston suchsubjects;they were, for example,ignorant of the complexpalace-centredorganizationof the late BronzeAge. The significance of the speculation lies rather in its testimony to the valuesof the classicalera,the fifth and fourth centuriesB.C. On this two points are significant for us. The first is that growth of population,increasingspecialization andtechnologicaladvances,the increasein materialresourceswere all judged positively. They were the necessaryconditions for civilization, for the 'natural', that is, the highest, form of social organization,the polis. This was no discoveryof Plato and Aristotle; it was implicit in the Prometheusmyth, it becamemore explicit in the 'prehistory' with which Thucydides opens his History and in other fifth-century writers known to us only from fragments.73 'The ancientGreekworld', writes Thucydides,'lived like the barbariansof today.'74 However, progresswas not an unmixed blessing. It led to bitter class struggles,imperial conquest, and the ethical dangerswe have alreadynoticed. Furthermore, there is an implication that technological and material progresshas come to an end. At least I am' unawareof any text which suggeststhat continuedgrowth in this sphereof human behaviourwas eitherpossibleor desirable,and the whole tenor of Pol., 1258b39ff. Thuc., History, i.2-19. On the fragments,see T. Cole, Democritusand the Sourcesof Greek Anthropology(American Philological Association, Monograph25, 1967). 74 History, i.6.6. 72

73

ARISTOTLE AND ECONOMIC ANALYSIS

the literaturearguesagainstsucha notion.75 Therecanandwill be progress in certain cultural spheres, such as mathematicsor astronomy;therecan, somethought,be improvementsin ethical, socialand political behaviour(moreoften thannot put in termsof a return to older virtues); there can be better (truer) understandingof life and society.But noneof that addsup to the ideaof progresswhich, in my judgement,hasbeenin the backgroundof all modern economic analysesat least since the late eighteenth century.76 For Thucydidesone of the driving forces in 'prehistoric'prothat is gresswas the rise and growth of maritime commerce-and the secondpoint. Given his grandtheme,the Athenianempireand the Peloponnesian War, he wasmoreconcernedwith thecorollary, the navy and the maritime empire,a polemicalsubjectboth in his day and later. But interwoven with this aspectwas always the other, on which I quoted Aristotle earlier, overseastrade as an indispensablesupplementto home production, for foodstuffs, timber, metals and slaves.77 And 'in Athens facts had a way of becoming spiritual problems'.78That is precisely how the discussionturned. I havein mind not maritimepower but the 'probI have examinedsomeaspectsof this themein 'Technicalinnovation and economicprogressin the ancientworld', EHR, 2nd ser., xviii (1965), pp. 29-45; cf. H. W. Pleket, 'Technologyand society in the Graeco-Romanworld', Acta Historiae Neerlandica,ii (1967), pp. 1-25, originally publishedin Dutch in Tijd. v. Geschiedenis,lxxviii (1965), pp. 1-22. 16 The faith of someHippocratic writers, notably the author of On Ancient Medicine (sect. 2), that 'the rest [of medical knowledge] will be discovered eventually',is no exception,though admittedly such progresswould bring 'practical' benefits to mankind:Neglectof the fundamental distinction betweenmaterial and cultural progressin my view vitiates the much praisedpolemic by L. Edelstein,The Idea of Progressin Classical Antiquity (Baltimore, 1967), againstthe 'orthodox' view, summedup by J. B. Bury, The Idea of Progress(New York, 1932 edn), p. 7: ' ... the Greeks,who were so fertile in their speculationson humanlife, did not hit upon an idea which seemsso simple and obvious to us as the idea of Progress'.On Thucydides,seeJ. de Romilly, 'Thucydideet !'idee de progres',Annali della Smola Normale Superioredi Pisa, xxxv (1966), pp. 143-91. 17 On all this seeA. Momigliano, 'Sea-powerin Greek thought', CR, lviii (1944), pp. 1-7, reprintedin his Secondocontributo alia storia degli studi classici (Rome, 1960), pp. 57-67. 18 Ibid., p. 58. 1S

47

M. I. FINLEY

lem' of trade and markets. Herodotus reveals its existence a century before Aristotle. When a Spartanembassycameto warn the Persianking not to harm any Greek city, he tells us, Cyrus replied: 'I have never yet feared men of this kind, who set up a placein the centreof their city wherethey assembleandcheateach other with oaths.' This was addressedto all Greeks,Herodotus explains,'becausethey haveestablishedmarket-placesfor buying and selling', whereasthe Persianshaveneitherthe practicenor the market-place.Xenophon offers partial support in his statement that the Persiansexcludeall huckstersand pedlarsfrom the 'free agora'(hereto be translatedin its original sense,'assembly-place').79 Whatever the truth may be about Persia, the Greek attitude reflectedby Herodotusand Xenophonis evident. Aristotle used the sameterminology as Xenophonwhen he proposedthat 'provision should be made for an agora of the sort called "free" in Thessaly[a district in north-centralGreece].This should be clear of all merchandise,and neither a workingman nor a farmer nor any other such personshould be permitted to enter unless sumso As a final example, there was moned by the magistrates'. Aristotle's contemporary, Aristoxenus,who thoughtit reasonable to claim that the half-legendaryPythagorashad 'extolled and promotedthe study of numbersmore than anyone,diverting it from the businessof merchants'.sI However, neither speculationabout the origins of trade nor doubtsabout marketethics led to the elevationof 'the economy' (which cannotbe translatedinto Greek) to independentstatusas a subject of discussionor study; at least not beyond Aristotle's division of the art of acquisitioninto oikonomiaandmoney-making, and that was a dead end. The model that survived and was imitated was Xenophon'sOikonomikos,a manualcovering all the human relations and activities in the household(oikos), the relations between husband and wife, between master and slaves, betweenhouseholderand his land and goods. It was not from Hausvaterliteratur that modern economic thinking and writing arose in the late eighteenthcentury, but from the radical discovery that there were 'laws' of circulation, of market exchange, of value and prices (to which the theory of ground rent was Herod, i. 152-3. Xen., Cyropaedia,i.2.3. Pol., 133Ia3Q-35. 81 Frag. 58B2 Diels-Kranz. 79

80

ARISTOTLE AND ECONOMIC ANALYSIS

linked).82It is at leastof symbolicinterestthat in preciselythat era David Hume made the brilliant (and still too often neglected) observation:'I do not remembera passagein any ancientauthor, where the growth of a city is ascribedto the establishmentof a manufacture.The commerce,which is said to flourish, is chiefly the exchangeof thosecommodities,for which different soils and climateswere suited'.83 I would be preparedto arguethat without the conceptof relevant 'laws' (or 'statistical uniformities' if one prefers) it is not possibleto have a conceptof 'the economy'.However, I shall be contentheremerely to insist that the ancientsdid not (ratherthan could not) havethe concept,andto suggestwherethe explanation lies. One consequenceof the idea of the koinonia was a heavy encroachment by political andstatusdemandson the behaviourof ordinary Greeks,not just in writings of a few doctrinaireintellectuals. If we considerinvestment,for example, we immediately come up againsta political division of the population that was unbridgeable.All Greek states,so far as we know, restrictedthe right of land ownership to their citizens (save for exceptional individuals who receivedthe right as a personalprivilege). They thereby,in effect, erecteda wall betweenthe land, from which the great majority of the population received their livelihood, and that very substantialproportionof the moneyavailablefor investment which was in the handsof non-citizens.84 Among the most obvious practical consequenceswas a narrowing of choice of investment(whether by purchaseor by loan) for the potential investorson the one hand,and a tendencyon the part of moneyholding citizensto turn to the land from considerationsof status, not of maximization of profits.8s The absencein our sourcesof SeeO. Brunner, 'Das "ganzeHaus" und die alteuropiiischeOkonomik', in his Neue Wegeder Sozialgeschichte(Gottingen, 1956), ch. 2, originally publishedin Z.j. Nationalok., xiii (1950), pp. 114-39. 83 'Of the populousness of ancientnations',Essays(London, World's Classicsedn, 1903), p. 415. How widely and carefully Hume had read ancientauthorsis demonstratednot only in this essaybut also in his notebooks. 84 The important economicrole of the metic (the free resident'alien'), which underliesthis point, will be consideredimmediatelybelow. 8S I have discussedthe evidencebriefly in Studiesin Land and Credit in AncientAthens(New Brunswick, N.J., 1952), pp. 74-8; again in 'Land, debt, and the man of propertyin classicalAthens', Political Science 82

49

M. I. FINLEY

any evidenceof investment(including loans)for improvementson land or in manufactureis noteworthy,especiallyagainstthe considerable evidence of relatively large-scaleborrowing for conspicuousconsumptionand for expensivepolitical obligations.86 No doubt a modern economistcould construct a sophisticated investment model to account for these Greek conditions of choice. But first the usefulness,indeed the possibility, of such a model has to be envisaged,as it was not in antiquity.87 Kept off the land, the non-citizensof necessitylived by manufacture, trade and moneylending.That would be of little interest were it not for the capital fact that this metic activity was not a matter of their being toleratedby the koinonia but of their being indispensable.They weresoughtafter, preciselybecausethe citizens could not carryon all the activities necessaryfor the survival of the community.88(Whetheror not they could not 'only' because they would not is a historically meaningless'psychological'question that seemsto me to divert attentionfrom the central issue.) Slaveswere the sole labour force in all manufacturingestablishQuarterly, lxviii (1953), pp. 249-68. Detailed researchinto the whole

questionof 'investments'is urgently needed. C. Mosse,La fin de fa democratieatbenienne(Paris, 1962), pt I, ch. I, has arguedin great detail that the fourth century B.C. witnessedmore fluidity than my sketch(cited in the previousnote) allowed. Even so, she agreeswith the point at issuehere, e.g. pp. 66-7: 'Certainly such profits [from accumulationof land holdings] were rarely reinvestedin production.... That is why, if there was a concentrationof landholding, it did not bring about any profound transformationin the mode of agricultural production'. 87 Political and status'interference'was equally significant in other aspects, for example,on prices and wageswheneverthe statewas a party, which was often the case.To enter into details would protract this discussion unnecessarily,I believe. 88 This was openly acknowledgedby an anonymousfifth-century oligarchic Const. of Athens, I.II-I2; Plato, pamphleteer,Pseudo-Xenophon, Laws, 919D-920C,made a virtue of the fact; Aristotle was troubled in the Politics by his inability to get round this obtrusiveelementin the koinonia, as J. Pecfrkashowedin a short but important article, 'A note on Aristotle's conceptionof citizenshipand the role of foreignersin fourth-centuryAthens', Eirene, vi (1967), pp. 23-6. On metics generally in fourth-centuryAthens, see Mosse, op. cit., pp. 167-79. Hicks, op. cit., p. 48, seemsto have placedthe accentexactly in the wrong place when he writes of the metics, 'what is remarkableis that there should have beena phasein which their competitionis tolerated,or even welcomed, by thosealreadyestablished'(my italics).

86

ARISTOTLE AND ECONOMIC ANALYSIS

ments exceedingthe immediatefamily circle, right to the managerial level. Without the many thousandsof free non-citizens, mostly Greeks themselves,some transient, others permanently resident (metics), maritime commerce in the more complex urbanized communities would have fallen below the essential minimum for vital supplies,not to mentionluxury goods.Hence fourth-centuryAthensomittedonepiecefrom its networkof laws designedto guaranteea sufficient annualimport of corn-it made no effort to restrictor specifythe personnelengagedin the trade.89 The position is neatly symbolizedby a single pamphlet, the Wqys and Means(or Revenues)written by Xenophonin the period when Aristotle was worrying about oikonomikeand chrematistike. His proposalsfor raising the revenuesof Athensare concentrated on two groups of people.First he suggestsmeasuresto increase the numberof metics, 'one of the best sourcesof revenue':they pay taxes,they are self-supporting,and they receiveno pay from the statefor their services.The stepshe proposesare (I) release meticsfrom the burdensomeobligation of servicein the infantry; (z) admit them to the cavalry (an honorific service); (3) permit them to buy land in the city on which to build residences;(4) offer prizes to the market officials for just and speedysettlementof disputes;(5) give reservedseatsin the theatreand otherforms of hospitality to worthy foreign merchants;(6) build more lodginghousesand hotels in the harbour and increasethe number of market-places.Hesitantly he adds the possibility that the state shouldbuild its own merchantfleet and leasethe vesselsout, and immediatelyturns to his secondgroup, slaves.Startingfrom the observationthat large private fortunes have been made by men who investedin slavesandlet themto holdersof concessions in the Atheniansilver mines, Xenophonproposesthat the stateembark on this activity itself, ploughingback the profits into the purchase of more and more slaves. After some rough calculations and variouscounter-arguments againstpossibleobjections,he writes, '1 havenow explainedwhat measuresshouldbe takenby the state in order that every Athenianmqy be maintainedat public expense'.9O To avoid misunderstanding,I will say explicitly that a count of heads would probably show that even in Athens the citizens who did work of somekind, including agriculture,outnumberedthe others.The point at issueis the location within the economyof the vital minority. 90 Xen., Waysand Means, iV.H.

89

M. I. FINLEY

We need not waste time examining the practicality of these schemes.Many harshthingshavebeensaidaboutthemby modern scholars-all from the wrong point of view, that of modern economicinstitutions and ideas. What matters is the mentality revealedin this uniquedocument,a mentalitywhich pushedto the extremethe notion that what we call the economywas properly the exclusivebusinessof outsiders.91 91

I have discussedthe matter raised in this final section (and elsewhere) more fully in The Anfient Efonomy (London, 1973).

III ROME AND THE CLASS STRUGGLE IN THE GREEK STATES 200-146B.C.* JohnBriscoe

Classdivisions in the Greek world were extremelyclear-cut.The distribution of wealthwas very uneven,andit is possibleto speak of two partiesin everystate,the rich andthe poor, the few andthe many. Thesedescriptionsare commonin ancientwriters.1 There were, no doubt, thosewho did not fit easily into either category but the schematismis far more fruitful than many modernones. Strife betweenthesetwo classeswas a continualfeatureof Greek history, and changesof constitution were the symptom of the victory of one or other of the classes.Oligarchy was the ascendancy of the rich, democracyof the poor. The class struggle in the Greek states,however, was by no meansa purely internal matter for the statesconcerned;it also affectedrelationsbetweenstates.In the fifth centurythe democrats had looked principally to Athens for support, the oligarchs to Sparta.2 In the fourth centuryPhilip II of Macedonsupportedthe upper classes,and this continuedto be the policy of the rulers of Macedondown to AntigonusDoson. The policy was reversedby Philip V, who came to the throne of Macedonin 221 B.C. He is known to have encouragedcivil strife and to have attemptedto woo popular favour by wearing common dress and portraying himself as a man of the people.3 In 200 B.C. Rome went to war *From no. 36 (1967). lOne thinks particularly of the famous passagein Thuc., iii. 82-3. There are innumerablereferencesin Aristotle's Politics. 2 Cf. p. 72 below. 3 Cf. Polyb., vii. II. 10, 12.9,13.6-7,14.2-5, x.26; Livy, xxvii. 31. 3, xxxii. 21. 23; Plut., Aratus, xlix. 2 ff.

JOHN BRISCOE

with Philip, and there followed the series of wars that led to Rome'scompletedominationof the Mediterraneanarea. Rome's policy in this period towardsthe rival factions in the Greekstates is the subjectof this paper.

I

It will be convenientfirst to summarizethe eventsof theseyears. In 2.00 B.C. Rome declaredwar on Macedon,ostensiblyto force Philip to ceasehis attackson otherGreekstates;in fact Romehad seenthe growing power of Macedonand was afraid that Philip, either by himself or in conjunctionwith Antiochus, the king of Syria, would invadeItaly and threatenRome'sdominationof the peninsula.Philip was defeatedby Titus Quinctius Flamininus at the battle of Cynoscephalaein 197 and forced to evacuateall his possessionsoutside Macedon itself. He was, however, left his kingdom andbecamean ally of Rome. The Romansenateissueda famous declaration,proclaiming Greece to be completely free, without garrisons or tribute-though three of Philip's chief garrisontowns continuedto be occupiedby Romantroops until 194. In 195 Flamininusfreed the city of Argos from the domination of Nabis, the tyrant of Sparta. Meanwhile Rome had been troubled by the aggressionsof Antiochus of Syria. He had advancedalong the coast of Asia Minor in 197, andby 196 hadcrossedthe Hellespont.Negotiations with him producedno solution. Before long Romantroops were backin Greece;the Aetoliansweredissatisfiedwith the settlement made after the defeat of Philip and called in Antiochus to 'free' Greece from Roman control. Antiochus arrived inadequately preparedand was defeatedin 191 at the battle of Thermopylae. He retreatedto Asia and was again defeatedat Magnesiain 190. The peacesettlementimposed by Rome drove Antiochus back beyondthe Taurusmountains,and gavelargepartsof Asia Minor to Rhodes and Pergamum, Rome's chief allies amongst the Greeks. In the 180sRomehad to deal with renewedexpansionistmoves by Philip and with a complex seriesof problemsin the Peloponby his sonPerseus.The nese.In 179 Philip died andwas succeeded latter embarkedon a policy of retrenchmentat home and of 54

ROME AND THE CLASS STRUGGLE IN GREEK STATES

renewing friendly relations with other Greek states.Spurredon by reports that Perseuswas making military preparationsfor a new war againstRome, the Romansdeclaredwar on Macedon. Perseus was defeated at Pydna in 168 and the Macedonian monarchywas dismembered.Romewas now undisputedmistress of the Mediterranean.In the next twenty years she was occupied with various disputes in different parts of the Greek world. Military interventionwas avoideduntil the senatedecidedthat the last remaining independentpower of any size, the Achaean League,shouldbe destroyed.Rome went to war in 147; the next yearthe Achaeansweredefeated,the Leaguedismemberedand the city of Corinth razedto the ground. II

Romewas governedby an oligarchy. The peopleas a whole chose the magistrates,but they were organized for this purpose in assemblieswhich gave a dominant influence to men of wealth. The importantpolitical decisionswere madeby the senate,a body which was composedlargely of ex-magistrates,who, once they becamemembers,remainedso for life. Rome'sdominationof the Italian peninsularelied to a large extent on her supportfrom the upper classesin the allied statesof Italy.4 The natural sympathies of the senatorialgovernmentcould thus be expectedto be on the side of the upper classesin the Greek cities. And since Philip V was courting the masses,Rome might well feel that supportfor the rich was a useful weaponin winning the war. Contrariwise, the upperclassesin the Greekstatescould naturallybe expectedto look to Rome for support. It was Fustel de Coulangeswho first expoundedthe view that Rome consistently supported the upper-classelements in the Greek states.Indeedhe saw the existenceof the class strugglein Greeceas the chief reasonfor Rome'seventualdominationof the Greeks.sThis view, acceptedby many later scholars,6was chalOn the Roman constitutioncf. P. A. Brunt, 'The Roman mob', pp. 7680 below. On Italy cf. A. H. McDonald, 'Rome and the Italian Confederation(200-186 B.C.)' fRS, xxxiv (1944), pp. II-33; E. Badian, Foreign Clientelae(264-70 B.C.) (Oxford, 1958), p. 147; P. A. Brunt, 'Italian aims at the time of the Social War', fRS, Iv (1965), p. 92. 5 Fustel de Coulanges,QuestionsHistoriques(Paris, 1893), pp. 12 1-211. 6 G. De Sanctis,Storia dei Romani, iv, 1 (Turin, 1923), p. 98; F. W. 4

55

JOHN BRISCOE

lenged by Passeriniwho arguedthat Roman policy towards the Greekstatescould not be correlatedwith a predilectionin favour of the upper classes.7 It is easyenoughto find prima facie evidencefor the theory of Fustelde Coulanges.In 197, for example,the men of Opouswere divided into two factions. One called in the Aetolians 'but the richer faction shut out the Aetolians, and sendinga messengerto the Roman commander,held the city until he came'.8Aetolia at this time was an ally of Rome, but her constitution was democratic,9 and it was not long before she becameRome's leading opponentin Greece.In 194 Flamininus reorganizedThessalyon timocratic lines:10 the stateswere in completechaosand confusionand had to be broughtinto some reasonable methodof government.... Flamininuschosethe senateand the judgesmainly on the basis of wealth and gave the greatestinfluence to that element in the stateswhich found it to their advantagethat everythingshould remain peacefuland undisturbed. In 192, Livy tells us, the massof the peoplelooked to Antiochus to save them from Roman domination, while the upper classes were on the side of Rome. 'The masseswere eagerfor changeand entirely on the side of Antiochus';l1 and 'it was evidentto all that Walbank, Philip V of Macedon(Cambridge,1940), p. 165; Badian, Foreign CHentelae,p. 78, and 'Rome and Antiochus the Great: a study in Cold War', CP, liv (1959), p. 93 (= Studiesin Greek and Roman History (Oxford, 1964), p. 129); A. Aymard, 'L'organisationde la Macedoineen 167 et Ie regime representatifdans Ie monde grec', CP, xlv (1950), p. 106 (= Etudesd'historie ancienne(Paris, 1967), p. 176); M. Holleaux, Rome,la Grece et les monarchieshellenistiques(Paris, 1921), pp. 228 ff. Holleaux is rather more cautiousin CAH, viii (Cambridge, 1930 ), p. 197· 7 'I moti politico-socia Ii della Grecia e i Romani', Athenaeum,new ser., xi (1933), pp. 309-35. Passerini'sview was acceptedby M. 1. Rostovtzeff, Socialand EconomicHistory of the Hellenistic World (Oxford, 1941), pp. 6II-I2 and 1460 n. 14, and, apparently,by P. Meloni, Perseoe la fine della monarchiamacedone(Cagliari, 1953), p. 254 n. 2. 8 Livy, xxxii. 32. 2-3. 9 Cf. n. 20 below. 10 Livy, xxxiv. 51.4-6. II Livy, xxxv. 33. 1.

ROME AND THE CLASS STRUGGLE IN GREEK STATES

the leadingmen and the aristocracywere in favour of the Roman alliance and were content with the existing situation, while the multitude and those whoseaffairs were not all they could desire were in favour of a complete change'.12In 190 the Phocaeans similarly divided: 'some were trying to sway the minds of the massesin favour of Antiochus ... but the senateand the upper classeswere of the opinion that they should remain loyal to Rome'.13The samedivision of opinion is found in the yearsprecedingthe war with Perseus:'the sympathiesof a largeproportion of the peoplewere on his side'14 and 'the masseseverywhere,as usually happens,were on the worse side, being inclined towards Perseusand the Macedonians'.15Finally one may note the actions of L. Mummius after the defeatof the Achaeansin 146: 'he dissolved democracies,and establishedmagistrateselected on the basis of wealth'.16

III This list of evidence,however, is by no meansthe whole story. There are a number of instanceswhere Rome did not give her support to the right-wing elementsin the Greek states,and any accountof Romanpolicy must explain these.17 Livy, xxxv. 34. 3. Livy, xxxvii. 9; cf. Polyb., xxi. 6. 14 Livy, xlii. 5. 2. 15 Livy, xlii. 30. 1. 16 Paus.,vii. 16. 9. 17 Some of the counter-examples adducedby Passeriniand others are largely basedon Livy's use of the word principes. Passeriniassumedthat this always meant optimates,and thus, for example,when Ismenias,the leader of the anti-Romanfaction in AetoIia in the 17os, is describedas a princepshe can be assumedto have been a supporterof the upper classes (Livy, xlii. 38. 5: in 43. 9 he is called nobilis ac potens.Cf. Meloni, Perseo,p. 147 n. 1.). The assumptionis unjustified. Livy usesprinceps andprincipesin non-Romancontextsvery frequently (a list will be found in Pauly-Wissowa,RE, s.v. princeps, vol. xxii, columns 2004-5) and it is clear that in most casesit meanssimply 'leaders',without implying that they are leadersof the upper-classparty. I can find only four passages (Livy, xxxii. 38. 7, xxxv. 34. 3, xlii. 30. I, xlii. 44. 4) where the class meaningis apparent;and the last two of theseare dubious, for one must always reckon with the leadersof the peoplenot being in complete accordwith their own followers. Seefurther n. 63 below. 12 13

57

JOHN BRISCOE

I have said it was natural that the upper classesin the Greek statesshouldlook to Romeas their naturalchampion.It was they who particularly wantedto resistPhilip's advances,and the latter had tried to gain support among their political opponents.I believe that in the 1905 Rome's natural preference was for oligarchic governments-other things being equal-butshe was prepared to take support from whatever source it came, and never dreamedof pressing her ideological predilections to the point where they endangeredher own bestinterests. The two main powers in Greece, Macedon apart, were the greatconfederationsof the Achaeanand Aetolian Leagues.In the First MacedonianWar, concurrentwith Rome's struggle against Hannibal, the Achaeanshad been allies of Macedon, whilst the Aetolianshad beenRome'sprincipal helper.18 The Aetolians had, however, made a separatepeace with Philip in 206, and their alliancewith Romehad cometo an end. The two Leagueswere of very different character.The AchaeanLeague,though nominally democratic, was in fact controlled by men of means.19 The Aetolian League had a more democratic basis, and it was its leadersScopasand Dorimachuswho were responsiblefor making the alliance with Rome in 212/1.20 In 205/4 Scopasand Dorimachus were given a special commission to enact laws. They proposed the abolition of all debts, but this was successfully resisted by one Alexander Isius. Scopasand Dorimachus then left Aetolia, Scopasto serveas a mercenaryin Egypt.21 Passerini appearsto arguefrom theseeventsthat the Aetolian Leaguein the On the alliance with Rome cf. literature cited by H. H. Schmitt, Die Staatsvertagedes Altertums, iii (Munich, 1969), no. 536. 19 For the AchaeanLeague,cf. F. W. Walbank, Commentaryon Polybius, i (Oxford, 1956), p. 222; K. von Fritz, The Theory of the Mixed Constitution in Antiquity (New York, 1954), pp. 4 ff. The constitution of the League is most fully discussedby A. Aymard, Les assembliesde la conjideration acbaienne(Bordeaux, 1938), J. A. O. Larsen,RepresentativeGovernmentin Greek and RomanHistory (Berkeley and Los Angeles, 1955), pp. 86 ff. 20 On the Aetolian League,cf. Larsen,Representative Government,pp. 70-1; Walbank, Commentary,pp. 453-4. It is true that by the end of the third 18

century important decisionswere taken by a small 'inner cabinet',but it was not controlled by men of wealth, and it is misleadingto call it an oligarchy: thus, e.g., F. W. Walbank, Ara/us of Sicyon(Cambridge, 1933), p. 25· 21 Polyb., xiii. 1-2. On Scopas'later activities cf. Polyb., xv. 25.16 ff., xvi. 39 ff., xviii. 53-5; Livy, xxxi. 43·

ROME AND THE CLASS STRUGGLE IN GREEK STATES

1905 shifted decisively to the right, and that becausethe radicals had beenfriends of Rome, their opponentsbecameher enemies.22 It is true enough that the Aetolians rejected the extreme proposalsof ScopasandDorimachus,andit is also true that Alexander Isius was one of the leading Aetolian opponentsof Rome in the 19os. But the Aetolians after 204 were still far more radical than the Achaeans,andit is wrong to think that Aetolia in the 1905 was ruled by an exclusively upper-classgovernment. Aetolia came back into alliance with Rome in 199.23 The Achaeansjoined a year later. This decision encountered violent oppositionwithin the AchaeanLeague,24but this does not seem to have had any connectionwith differencesover internal policy. The decisionwhetherto supportRomeor Philip was discussedon its own merits. Romewas glad to acceptboth Achaeanand Aetolian supportand was not at all concernedwith the internalpolitics of the stateswho were willing to support her in the war with Macedon. This is madeevenclearerin the caseof Nabis of Sparta.Nabis took to extremesthe radical policies advancedin the third century by Agis IV and CleomenesIII and had establishedan extremeleftwing regime in Sparta.Polybius detestedNabis and all he stood for, and has given us a very lurid picture of his characterand activities;25 For he utterly exterminatedthose of the royal houseswho survivedin Sparta,and banishingthosecitizens who were distinguishedfor their wealth and illustrious ancestry,gave their property and wives to the chief of his supportersand to his mercenaries,who were for the most part murderers, mutilators, highwaymenand burglars. On the evidenceof this passagePasseriniarguedthat Nabis was simply a personaltyrant out for his own ends,and not a man with Passerini,'I moti politico-sociali', pp. 319-20. It is not all that relevant that AlexanderIsius was extremely rich (Polyb., xxi. 26.9). 23 Livy, xxxi. 41. 24 Livy, xxxii. 19 ff. For commentson the policy, cf. Polyb., xviii. 13-15; A. Aymard, 'Le fragment de Polybe "sur les traitres"', REA, xlii (1940), pp. 9-19 (= Etudesd'histoire ancienne,pp. 354-63). 2S Polyb., xiii. 6, cf. xvi. 13. On Nabis cf. V. Ehrenberg,RE, vol. xvi, columns 1471-82; C. Mosse,CaNersd'histoire, ix (1964), pp. 313-23, B. Shimron, 'Nabis of Spartaand the helots', CP, lxi (1966), pp. 1-7.

22

59

JOHN BRISCOE

an advancedsocial programme.26 Polybius, however, is not an unbiasedwitness. He detestedleft-wing political movements,27 andit is not an unusualmethodof attackinga policy one doesnot like to obscurethat policy and claim that its exponentsare merely evil men working for their own ends.Nabis' own claims may be more significant. Theseoccur in a speechto Flamininus in 195: they are reportedby Livy, and though,no doubt, the composition of the speechis Livy's own, the sourceis Polybius.28 'My title of tyrant and my actionsare laid as accusationsagainstme, because I summon slaves to freedom, because I give land to the impoverishedmasses.'And he compareshis policy with that of Rome: 'You chooseyour cavalry on the basisof wealth, and your infantry too; you desirethat a few should be pre-eminentlyrich, and the massof the peopleshould be subservientto them.' We may allow Nabis to have beena geniune,if violent social reformer.His policy towards"RomeandMacedonwasundoubtedly purely expedient.In 197 Philip offered Nabis the possessionof Argos, on condition that if Philip defeatedthe Romans,Nabis would restore the city to him.29 Nabis occupied the town, and introducedthere a social revolution of the samesort that he had instituted at Spartaitself. This achieved,Nabis madea complete volte-face and offered his services to Flamininus.3o Flamininus, despitethe objectionsof Attalus of Pergamum,was only too glad to accept.Once the war with Philip was over, however, Roman policy changed,and in 195 Flamininuswent to war to free Argos from Nabis' control. But even this was not a war undertakento rid the Peloponneseof an unwelcomesocio-political system.The object was, in part, to free Argos from Nabis in order to weaken the growing power of the Spartan leader, and, secondly, to placatethe feelings of Rome'sGreekallies. They were not, in the event, very much placated.The peaceterms imposedon Nabis by Flamininusleft the former in chargeof Spartaanddid not even go so far as restoringthoseexiled by Nabis.31Flamininusdid not 'I moti politico-sociali', pp. 315-18. Cf. Polyb., vi. 43, xx. 5-7; Walbank,Commentary,pp. IZ-13. 28 Livy, xxxiv. 31. II If. Freeingof slavesalso in Polyb., xvi. 13. I. 29 Livy, xxxii. 38 If. 30 Livy, xxxii. 40. 31 Livy, xxxiv. 34 If.; Plut., Flamininus, xiii. On the exiles, cf. A. Aymard, Lespremiersrapports de Romeet de la confederationachaienne(Bordeaux, 1938), pp. 241-4. 26

27

60

ROME AND THE CLASS STRUGGLE IN GREEK STATES

want to eradicateNabis becausethat would haveleft the Achaean Leaguein virtually completecontrol of the Peloponnese.His aim was a balanceof power, not upper-classconstitutionalgovernment, and he preferred to tolerate the continued existenceof a revolutionarygovernmentin Spartaratherthanallow the Achaean Leagueexcessivepower in the Peloponnese. The Achaeansand otherswere l.!pset by Flamininus' refusalto eradicateNabis.32 They had hoped that Rome's victory would meanthe end of a social systemthey detestedandfeared.Passerini, consistentlywith his view that Nabis' regime was simple tyranny and not genuinesocial revolution, held that the Achaeanswere more afraid of Nabis' expansionismthan of his social aims.33 This view is scarcelyborn out by Livy's own words: 'As for the Achaeans,whatever joy the restorationof Argos to the League brought them, was renderedincompleteby the fact that Sparta remainedenslavedand Nabis was left as a thorn in their side'.34 Before the final surrenderof Nabis the thoughtsof the allies had clearly been directed to the political nature of his regime: 'his examplewould incite manyin otherstatesto attackthe freedomof their own citizens',3SForeign domination,it is true, was part of their fear, but only becausethe fear was of the export of revolution. Their main concernwas the dangerof political upheavalin their own cities. Thus Romewas not over-concernedwith the internalpolitics of the Greek stateswhere there were other, and more important, issuesat stake.Where this was not so, shefelt able to indulge her naturalpreferencefor the upperclasses.Flamininus'behaviourin Boeotia in 196 is an interesting casein point.36 In that year he agreedthat thoseBoeotianswho had servedwith Philip shouldbe permitted to return to Boeotia. Their leader Brachylles was immediatelyelectedto the chief magistracy.The pro-Romanparty, led by Zeuxippus, wanted to murder Brachylles and asked for Flamininus' permission. Flamininus replied that 'he himself would not takepart in sucha deed,but he was not standingin the Livy, xxxiv. 4I. 4 If., 4S. 5-6. At Athenaeum,new ser., xxiii (1945), p. IIS Passeriniis less inclined to pressthis point. 34 Livy, xxxiv. 4I. 4. 35 Livy, xxxiv. 33. S. 36 Polyb., xviii. 43, Livy, xxxiii. 27-9 for all this. 32 33

JOHN BRISCOE

way of those who wanted to do it'.37 The murder provoked a wave of anti-Romanfeeling in Boeotia and Zeuxippus fled the country. Roman soldiers in Boeotia were massacred.Flamininus imposedseverepunishmentson the Boeotiansfor this outbreak, but Zeuxippuswas not recalled. Polybius' words suggestthat the pro-Romanparty was upper class. They complain to Flamininus of 'the people's present hostility towardsthem and the generallack of gratitudeshownby the masses'.38It looks as if in this caseFlamininusfelt able to encouragethe intriguesof the pro-Romanupperclasses.But evenso, he was not willing to take this encouragementto the extent of forcing a whole peopleinto open opposition,and it was for that reasonthat he did not attemptto recall Zeuxippus.39 In 194 the immediateproblemshad beensettled,and Flamininus' actionsshortly beforehis departurefrom Greeceindicatehis naturalpreferences.He madeconstitutionalalterations:40 he spenthis time in administeringjustice and altering the arrangementsmadein the statesby the arbitrary conductof Philip or his representatives:they had increasedthe power of the supportersof their own faction by depriving othersof liberty and justice. Who had been suppressedand how becomesclear from Livy's further remarks. Flamininus advocated concordia between the various classesof society. This presumablymeantthat the lower classeswere to acceptgovernmentby the upperclasses.The only placewherewe know the details of Flamininus'reorganizationis Thessaly. There, as we have seen, 'the stateswere in complete chaosand confusionand had to be broughtinto somereasonable methodof government'.Flamininus'solution was to 'choosethe senateand the judgesmainly on the basis of wealth, and to give the greatestinfluence to that elementin the stateswhich found it to their advantagethat everything should remain peaceful and Polyb., xviii. 43. 10. Polyb., xviii. 43. 8. 39 But in 188 he felt safe enoughto attemptto recall Zeuxippusand nearly succeededin causinga war betweenBoeotia and Achaea(Polyb., xxii. 4). 40 Livy, xxxiv. 48. 2. 37 38

62.

ROME AND THE CLASS STRUGGLE IN GREEK STATES

undisturbed'.41It is not completely certain whether Flamininus restrictedthe franchiseitself to the upper classes,or merely the offices of state; or whetherhe introduceda timocratic assembly modelled on that of Rome, whose main principle, according to Cicero, was that the will of the majority should not prevai1.42

IV As we have seen,Flamininus refusedto eradicateNabis because he was afraid of the increasingpowerof the AchaeanLeague.The samedesireto maintaina balanceof power appearsto have motivatedRomanpolicy during the long andcomplexdisputebetween Spartaand the AchaeanLeaguein the 180s.In 192 Nabishadbeen murderedand the AchaeanleaderPhilopoementook Spartainto the AchaeanLeague.Facedwith the invasionof Antiochus,Rome was in no position to object. Then, in 189, the Spartansattacked one of the Laconian coast-townscontaining the Spartanexiles who, though banishedby Nabis or his predecessorMachanidas, had not been restoredby Philopoemenin 192. The matter was referred to the senate,who gavea reply so ambiguousthat both sidesinterpretedit as favouring their own case.Philopoemenproceededto force Spartato dismantleher walls, recall the exiles and abolish the Lycurgan constitution.43 The senateby no means welcomedthis rightwardsmove in Sparta.In 187 someSpartans brought complaints to the senateabout Philopoemen'sactions, and the consul, M. Aemilius Lepidus, wrote to the Achaeans criticizing their conduct.MeanwhilePhilopoemenhaddespatched an embassyto Rome, and this returnedin 185 to report that the senatewas displeasedwith what had happened,thoughthey were not taking any action.44 The senatenow instructedQ. Caecilius Metellus to investigatethe situation. When Metellus arrived the pro-RomanDiophanesattackedPhilopoemen'shandling of both Livy, xxxiv. 5I. 4-6. Cic., De re publica, ii. 39. On whetheror not there was a full assembly in Thessaiy,cf. Aymard, op. cit. (n. 6), p. 105 (=Etudesd'bistoire ancienne, pp. 174-5)· 43 Livy, xxxviii. 30-4; Polyb., xxi. 32C, cf. xxii. 3. 7; Plut., Pbilopoemen,xvi; Paus.,viii. 5I. 3. 44 Polyb., xxii. 7. 6; Diodorus, xxix. 17. 41

42

JOHN BRISCOE

Spartanand Messenianaffairs, but Philopoemenand his supporters Archon and Lycortas defendedtheir policy. As before, the different factionswithin the AchaeanLeagueseemto havehad no relation to class-divisions. Metellus askedfor a specialassemblyto be convened,but was refusedon the groundsthat it was illegal for the Achaeansto summon an assemblyin such circumstancesunlessthey had received written instructionsfrom the senate.Thereare reasonsfor thinking that Philopoemenandhis friends wererathertwisting thesense of the law involved, but the senatehardly clarified matterswhen, after discussingthe rebuff to Metellus,it told the Achaeansto give the same considerationto Roman envoys as Rome did to the representativesof the AchaeanLeague.45 In 184 Areus andAlcibiades,describedas representatives of the 'old exiles', disputed with the Achaeansbefore the senate.The 'old exiles' are clearly thoseexiled by the tyrants and restoredin 189. In the following year, however,we find the 'old exiles' divided into two groups.One section,led by Lysis, wantedcomplete restitution of the exiles' property; the other, led by Areus and Alcibiades, proposedthat only a portion of the property should be restored.They, it seems,were willing to countenancesome redistributionof wealth.46 The result of the mission of Areus and Alcibiades was that the senate instructed Appius Claudius Pulcher to investigate the situation, but before he arrived the Achaeanshad condemned Areus and Alcibiades to death for undertaking an embassyto Romeon their own account.47WhenAppius arrived,a long debate ensued,and Appius appearsto have threatenedthe use of force. Eventually the Achaeansagreedto repealthe sentenceon Areus and Alcibiades,but askedthe Romansthemselvesto be responPolyb., xxii. 10. On the legal issuesinvolved, cf. Aymard, Les assembtees de la confidiration achaienne,pp. 188-204:Contra, Larsen,Representative Government,pp. 90-1. The law probably said that a synkletoswas to be called if written instructionscamefrom the senate.It was interpretedto meanthat it could not be held if there were no such instructions.For the senate'srebuke, cf. Polyb., xxii. 12. 10, Livy, xxxix. 33. 8. 46 Polyb., xxii. 11-12 for the dispute before the senatein 184. On the political position of Areus and Alcibtades,cf. B. Niese, Geschichteder griechischenund makedonischenStaatenseit der Schlachtbe; Chaeronea,iii (Gotha, 1903), p. 42 n. 4, 49. 47 Livy, xxxix. 35. 8; Paus.,vii. 9. 2.

45

ROME AND THE CLASS STRUGGLE IN GREEK STATES

sible for anychangestheywishedto institutein Sparta.48 Thematter was thus referredback to the senate,who found themselvesfaced with a situation more complicatedthan ever. Sparta was now Therewere representedby four different groupsof ambassadors. the two sectionsof the 'old exiles'whom I havealreadymentioned, therewas Serippuswho wantedthe existing situationto continue, and therewas Charonrepresentingthe 'democrats'who hadbeen exiled in 192 or 189.49 The senate,floundering in the confusion,appointeda committee of expertsto deal with the matter: Flamininus,Metellus and Appius Claudius.50 Eventually an agreementwas reachedamong the various Spartangroupsthat the exiles were to return andthe city was to remainpart of the AchaeanLeague,but no agreement wasforthcomingon the questionof the restitution of propertyandwhat this involved, the socialandeconomicpolicy that Sparta was to follow. With some difficulty the Achaeanrepresentatives were persuadedto add their seals to this outline settlement. Q. Marcius PhiIippus was sent out to administerthe settlement. About this time, it seems,someof the 'old exiles' were again exiled.51WhenPhilippusarrivedboth SpartaandMessenewereon the point of revolting from the AchaeanLeague.Messenein fact did so, and in the resultingwar Philopoemenwas killed. 52 PhiIippus appearsto have done his best to embroil both Spartaand Messenewith the League;on his return to Rome, he reportedto the senatethat if the Achaeanscontinuedwith their presentindependentand arrogantpolicy, it would not be surprisingif Sparta joined Messenein revolt. Spartanambassadors were againpresentin Rome,and to these the senatereplied that it did not think that their presentdispute with the Achaeanswas any concernof Rome. This ambiguous reply53 suggestedrebellion to Spartawhilst to the Achaeansit suggestedthat, althoughRomehad no sympathyfor the Achaean Polyb., xxii. 2.; Livy, xxxix. 33, 35-7; Paus.,vii. 9.4. Polyb., xxiii. 4. The senateand Livy (xxxix. 48. 2.-4) were at one in their inability to unravel the confusion. 50 Appius has to be addedin Polybius. 51 Polyb. xxiii. 5. 18,6,9. I, 18. 5. 52 For theseeventsand Philippus' part in them cf. IRS, liv (1964), pp. 66-7. Death of Philopoemen: Polyb:, xxiii. 12.; Livy. xxxix. 48.5 if.; Plut., Philopoemen,xviii; Justin, xxxii. x. 4; Paus.,viii. 51. 5. 53 For other ambiguoussenalusconsulla, cf. pp. 63 aboveand 69 below. 48

49

JOHN BRISCOE

case,shewas not muchconcernedwith what happened.Soonafter this Spartadid secedefrom the League.54 By 18z Lycortas had recoveredboth Messeneand Sparta,55and someof the exiles returnedto Sparta.56 Charonis found in Sparta shortly after this,57andit seemsto follow that the exiles of 19Z and 189 must have been recalled in accordancewith the agreement madein 183, and that it is a sectionof the 'old exiles' who form the subject of further disputes between the Achaeansand the senate.58 A senatorialinstructionin 181 ordering their return was ignored59 and in 180 Lycortas was still sure that the senatewould not enforce its decision. Ambassadorswere sent to Rome, but one of their number,Callicrates,insteadof defendingAchaea,told the senatethat if they wantedtheir will respected,they must give positive support to the pro-Romansin the Greek states.60 The senateagreed,expressedits view clearly, and the exiles returned. The issuesbetweenSpartaand Achaeawere complex,but they were very closely connectedwith decisionsabout the social and political constitution of Sparta. The senate,however, did not concern itself with these questions.Rome's policy had been to createdivisions within the Leagueand to use ambiguousreplies and veiled threatsto keepthe Achaeansin a stateof suspicionand uncertainty. And, as we can see, even the right-wing groups in Sparta could quarrel with Achaea. As in the previous decade Rome saw her main interest as being,the preservationof the balanceof power in the Peloponnese.The internal structureof Spartawas of little concernto her.

v Polybius regardedthe speechof Callicratesat Rome as being of the greatestimportance.61 He took the view that before180 Rome Cf. Polyb., xxiii. 17. I; Niese, Geschichteder griechischenund makedonischen Staaten,iii, p. 49 n. 3. 55 Polyb., xxiii. 16-17. 56 Polyb., xxiii. 18. 1-2. 57 Polyb., xxiii. 18.4, xxiv. 7. 58 Cf. SIG3, no. 634, where Callicratesis honouredfor securingthe restitution of the 'old exiles'. 5. Polyb., xxiv. 2. Paus.,vii. 9. 5-6 is quite unintelligible. 60 Polyb., xxiv. 8. 8 ff. 61 Polyb., xxiv. 10. Cf. Badian,Foreign Clienteiae, p. 91. 54

66

ROME AND THE CLASS STRUGGLE IN GREEK STATES

dealt with the Achaeanson equal terms whilst after that date the senategaveactive supportto her toadiesto get her will enforced. The judgement seems too extreme-aswe have seen Rome's attitude before 180 was nowhere near as open and honest as Polybius appears to have believed. It is true, however, that Callicratesstiffened the senate'sdeterminationto enforceits will by showing that there were peoplein Greecewho would get its wishes obeyed. And in the following years the senatedid give active support to pro-Romansin the Greek states.This is seen particularly clearly in the years following the Third Macedonian War. Charopsin Epirus, Lyciscus in Aetolia, Chremasin Acarnania and Mnasippusin Boeotia were all given support by Rome, and their opponentswere exiled. Polybius' description of these men appearsto indicate that they were proponentsof the sort of tyrannical policy exercisedby Nabis. It is doubtful if the sameis true of Callicrates. His differences with Lycortas and Polybius seemto have beensimply about the attitude to be taken towards Rome, and were not concernedwith the internal policy of the Achaean League.62 But the senate'spolicy after the defeat of Perseuswas ruthless: one thousandsuspectAchaeans,Polybius the most famous of them, were deportedto Italy. Once again the senatelooked to what she now regardedas her own interest-theelimination of her political opponentsin the Greek states and the vigorous championing of her friends, irrespectiveof their internalpolicies. Previouslyshehad refusedto eradicatepopularregimes;now shewent further andactuallysupportedthem. The natureof the regimein eachstatedependedto a great extent on the attitudeswhich the various groups or parties had taken towards Perseusat the time of the Third Macedonian War. It seemsthat in many statesthe conservativeswere,at least, lukewarmtowardsRomeand that their opponentstook the oppor63 tunity to supportRomeand gain political power for themselves. The evidenceof all this is fully set out by Passerini,'I moti politicosociali', pp. 327 ff. When I speakof Rome'spolicy I refer to that of the senatorialmajority. There are reasonsfor thinking that a minority disapproved.Cf. Historia, xviii (1969), pp. 49-70. 63 Livy (xlii. 30. I) says that the plebs were generally on Perseus'side, but the principes were divided. But on the ambiguity of principes, cf. n. 17 above. The plebs were no doubt forced to changetheir attitude towards Perseusby the fact of Romanforce. The demagogicleadersmay have had some difficulty in convincing their natural followers. 62

JOHN BRISCOE

VI The Achaean exiles were at last permitted to return in 150.64 Therefollowed in a shortwhile theAchaeanwar, which led to the destructionof Corinth and the dissolutionof the AchaeanLeague in its existingform. Thewar hasappearedto severalwritersto have beenbroughtaboutby a newly prominentgroup of demagogues who felt that they had to outbid the restoredexiles in nationalistic and anti-Romansentiments.65 Examinationof the details of the war does not, I think, supportthis interpretation. The antecedentsof the war are complex. In about I 56 Athens had attackedOropus. Oropusappealedto the senate,who asked Sicyon to act as mediatorin the matter.66 Sicyonimposeda fine of five hundredtalentson Athens,and the Athenianssentthe heads of the three great Athenian philosophical schools to Rome to pleadtheir case.The senatereducedthefine to onehundredtalents. What followed is obscure,but it seemsthat the Atheniansrefused to paythe fine, and installedcolonistsof their own in Oropus.At this point Oropuspromisedthe SpartanMenalcidasten talentsif he would persuadethe Achaeansto help her againstAthens. Half of this Menalcidasin his turn promisedto Callicrates.67 The Achaeansappearto have beensuccessfulin restoring the Oropiansto their own land. Menalcidas,however,refusedto pay Callicrates,who proceededto accusehim beforethe Achaeansof working in Rome to separateSpartafrom the AchaeanLeague. Menalcidasgave three of his talentsto the AchaeanDiaeuswho helpedhim to get acquittedon this charge. Menalcidasnow madea fresh embassyof his own accordcomPolyb., XXXV. 6 = Plut., Cato maior, ix. For the date, Paus.,vii. 10. 12. G. Niccolini, La confederazioneacaea(Pavia, 1914), p. 182. Cf. Fustel de Coulanges,QuestionsHistoriques, pp. 201 If. 66 Polyb., xxxii. II, xxxiii. 2; Paus.,vii. 1I. 4 If.; SIG3, no. 6n. The embassyof philosophers:Polyb., xxxiii. 2.; Plut., Cato maior, xxii; Cic., Academica,ii. 137=Gellius,vi. 14. 8-10. The narrative given here is that of Pausaniascorrectedto make it consistentwith SIG3, no. 6n. Cf. W. S. Ferguson,Hellenistic Athens(London, 19II), p. 327. 67 Paus.,vii. II. 7-8 (but cf. preceedingnote). For the narrative of the AchaeanWar, seeespeciallyG. De Sanctis,Stor;a dei Romani,iv, 3 (Florence,1964), pp. 127 If. Only fragmentsof Polybius'saccount survive and for most of the details we have to rely on Pausanias (vii. 10-16). 64

6'

68

ROME AND THE CLASS STRUGGLE IN GREEK STATES

plaining about a boundary dispute.68 The senate replied to Menalcidasthat all matters except those involving capital cases were to be dealt with by the Achaean League. Diaeus, now Achaeanstrategos,was confidentthat Rome was not interestedin Sparta and claimed that the senatehad given the federal body jurisdiction in all cases.Spartawas forced by threat of war to condemnto deaththosewhom Diaeusheld to be responsiblefor anti-Achaeanactions in Sparta, and in 14869 Menalcidas again journeyedto Rome to presenthis caseto the senate.Callicrates and Diaeus were sent to representthe League. The senatedespatchedambassadorsand instructedthem to detachseveralimportantstatesfrom the League.But the actualreply that it gavein public was so ambiguousthat both sides were encouragedto think that their requestshad been granted,70and the Roman ambassadorsspent such an inordinate time in reaching Greece that fighting broke out betweenSpartaand Achaea,as a result of which Menalcidas committed suicide.71 When the ambassadors arrivedthey announcedthat Sparta,Corinth, Argos, Heracleaand Orchomenoswere to be detachedfrom the League.There was a riot and the ambassadors were maltreated.72 Despitethis, Romemadestrenuousefforts to avoidgoingto war with the AchaeanLeague. Sex. Julius Caesar,the next Roman ambassador,was instructedto act with moderation.Polybiussays that the senate'saim was to humble the Achaeans,not destroy themaltogether.73 But the senatedid not rescindits decreethat the Leagueshouldbe emasculatedby the separationof so many of its importantmembers.74Critolaus, the Achaeanstrategos,refusedto Paus.,vii. 12. 4. This probably refers to the sameterritory over which there had beena disputebetweenSpartaand Megalopolisin 164/3 (Paus.,vii. II. 1-3; cf. Polyb., xxxi. I. 7; SIG3, no. 665). 69 It may be helpful to enumeratethe slrategoi of the last yearsof the AchaeanLeague: 151/0 Menalcidas,150/49 Diaeus, 149/8 Damocritus, 148/7 Diaeus, 147/6 Critolaus, Diaeus sufJectus. 70 Cf. abovepp. 63 and 65. 71 Paus.,vii. 13. 72 Paus.,vii. 14; cf. Polyb., xxxviii. 9. Polybius adds that they exaggerated what had happened.Dio (fr. 72.) and Livy (ep. li) are probably trying to justify Rome'saction in saying that the detachedcities were thosethat had previously belongedto Philip. Other sources:Justin, xxxiv. 1-2.; Eutropius,iv. 14; Florus, i. 32; Zonaras,ix. 31. 73 Polyb., xxxviii. 9. 6. 74 Cf. Niese, Geschichteder griechischenund makedonischen Slaaten,iii, p. 344. 68

JOHN BRISCOE

come to an agreement,and war was declared.It is noticeablethat Metellus Macedonicusmade considerableefforts to persuadethe Achaeansnot to go to war. In 148 he had askedambassadors on their way to Asia to stop the Achaeansfrom fighting Sparta.75 Two yearslater he himself sentfour ambassadors to Achaea,but they met with nothing but abuse.76 The war thus took place, and the Achaeanforces, though supported by Boeotia, Euboea, Locris and Phocis, were inevitably defeated.77Corinth was razed to the ground and the Achaean Leaguedisbanded.A largepart of Greecebecamean appendageof the Romanprovince of Macedonia.78 Theseeventsdo not supporttheview thatthe returnof theexiles from Rome provoked the democratic party into violent antiRomanfeeling in an attemptto outbid rivals in nationalisticsentiment. In fact we have no information aboutthe political position of the various personalitiesinvolved. The measurestaken by Diaeus after the death of Critolaus in 146-thefreeing of slaves and forced contributionsfrom the rich79-cannotbe used as evidence,for they weredesperatemeasuresin a time of extremecrisis. The only known former exile mentioned is Stratios, who was accusedby Critolaus of secretly intriguing with one of the ambassadorssentby Metellus80 and who laterbeggedDiaeusto accept the terms offered by Metellus after the death of Critolaus.81 Earlier Thearidas,the brother of Polybius, was employed as an ambassadorby the Achaeans,perhapsin an attempt to placate Romanfeeling.82 Thuswhile we canin generaldistinguishbetweenpro- andantiRomans,it is difficult to go much further. Callicratesis linked by Polybiuswith Charopsand other demagogues, but their domestic Paus.,vii. 13.2. Polyb., xxxviii. 12-13. 77 Achaeanallies: Paus.vii. 14. 6, cf. ii. I. 2; Polyb., xxxviii. 3. 8; Livy, ep. lii. For the sourceson the war cf. T. R. S. Broughton, The Magistratesof the RomanRepublic(New York, 1951), i, pp. 465, 467. 78 For the details of the statusof Greeceafter 146, cf. S. Accame,II dominio romano in Grecia dalla guerra acaica ad Augusto(Rome, 1946), ch. i and ii. 79 Polyb., xxxviii. 15. Also postponement of the repaymentof certain debts: Polyb., xxxviii. II; Diodorus, xxxii. 26. 3. 80 Polyb., xxxviii. 13. 4 ff. 81 Polyb., xxxviii. 17. 4. 82 Polyb., xxxviii. 10. I, I I; Paus.,vii. 14. 3. 75

76

70

ROME AND THE CLASS STRUGGLE IN GREEK STATES

policies may have differed considerably.Again Menalcidaswas a pro-Roman,83but that did not stop him quarrelling with Callicratesand ingratiating himself with the supposedlyanti-Roman Diaeus.84 And in 149 Callicratesand Diaeustogetherwere meant to representthe Achaeansbefore the senatein opposition to Menalcidas.85 It looks as if complex personalrivalries played a considerablepart in the events that precededthe war and the principal characterscannotbe fitted into neatcategories.It is not the caseeither that Romeattackedthe AchaeanLeaguebecauseit was showingdangerousleft-wing tendenciesor that a new democratic element attemptedto put an end to Roman domination basedon supportof the upperclasses.The senatehaddecidedthat its interestswerebestservedby removingthe lastvestigesof independentaction by the Greek states.It was not overwhelmingly concernedwith internal politics: it was only after the war that Mummius destroyedthe democraticstructure-suchas it wasof the Achaeancities.86

VII The picture which emergesfrom these fifty years is consistent. Romedid not set out with the intention of establishingoligarchic governmentsin Greece,and she did not have consistentsupport from the oligarchsand consistentoppositionfrom the democrats. The naturalpreferenceof the senateandits representatives wasfor the upperclassesandfor forms of governmentin which the upper classesweredominant.Otherthings beingequal,it was to this end that Romanpolicy was directed. The activities of Flamininus in Thessalyin 194 areperhapsthe bestexample.But in this turbulent period it is only rarely that other things were equal. Rome's object was to win the wars in which she was engagedand to Cf. Polyb., xxx. 16. 2. Paus.,vii. 12. 1. De Sanctis(Sloria dei Romani, iv, 3, pp. 129, 132) argues that Diaeus'father was Diophanes,the pro-Romanopponentof Philopoemen,and that Diaeus himself was originally pro-Roman.But there is no evidencethat Diaeus was Diophanes'son-all we know is that Diophanes'father was called Diaeus: and even if he was, he need not have followed the samepolicy as Diophanes. 8S Paus.,vii. 12. 8. 86 Paus.,vii. 16. 9.

83

84

71

JOHN BRISCOE

maintainthe controloverGreekaffairs which hermilitary successes bestowedon her. To this endthe senatewasglad to acceptsupport from thosewho werewilling to give it to her, irrespectiveof their position in the internal politics of their own states.She had no scruplesin using Nabis againstPhilip or supporting the demagoguesafter 167. We haveseen,too, that differenceswithin the Greekstatesconcerningthe policy to be adoptedtowardsRomedid not necessarily coincide with differenceson domestic policy. There is nothing very surprising about this. Many upper-classpoliticians would havebeenconfidentthattheycouldmaintaintheirpositionwithout Romanprotection,andpreferredto governtheir statesin complete independence ratherthanunderRomancontrol. Otherswould have seenthat Romandominationwasinevitableandpreferredto be on the winning side. Others,again, were willing to use Romansupport as a meansof securingtheir successin political disputeswithin the upper classes. It is interestingto comparetheseconclusionswith thosewhich emergefrom study of the Athenian Empire in the fifth century B.C. Thucydidestells us that in the Peloponnesian War (431-404 B.C.) the democraticelementsin the Greekcities soughthelp from Athens,their opponentsfrom Sparta.87 In an epoch-makingarticle de SteCroix hasshownthat oneof the chiefreasonsfor the success of the AthenianEmpirewasthat Athensgaveactivesupportto the democraticpartiesin theallied states.88 The democratsknew thatit wasonly becauseof Atheniansupportthat they wereable to maintain democraticforms of governmentin their own cities. Athens, as a democracy,had a natural preferencefor democraticgovernmentandthis coincidedwith herself-interestasheadof anempire.89 Thuc., iii. 82. 1. G. E. M. de Ste Croix, 'The characterof the Athenian empire', His/oria, iii (1954-5), pp. 1-40. Cf. also his 'Notes on jurisdiction in the Athenian empire', CQ, Iv (1961), pp. 94-112, 268-80; H. W. Pleket, 'Thasosand the popularity of the Athenian empire', His/oria, xii (1963), pp. 70-7. The objectionsof D. W. Bradeen,'The popularity of the Athenian empire', His/oria, ix (1960), pp. 257-69 and T. J. Quinn, 'Thucydidesand the popularity of the Athenian empire', His/oria, xiii (1964), pp. 257-66 do not seemto me to have invalidatedde Ste Croix's case. 89 We know of a few caseswhere Athens toleratedoligarchic governments: in eachcasethe oligarchsappearto have either secededfrom Athens or 87

88

ROME AND THE CLASS STRUGGLE IN GREEK STATES

Under the Roman Empire the picture is very different. There was now no question of a struggle for leadershipin the Mediterraneanworld-Rome's mastery was unchallenged.It is not surprisingthat undertheseconditionsRome'snaturalpreferences cameto the fore, andthat both in Italy andin the provincesit was the richer classeswho were dominant. The old senatorialaristocracy, it is true, no longer controlled affairs and men from lower orderscould rise and hold positionsof influence.9o But the lot of the mass of the people scarcelyimproved at all. The result of Rome'svictory was indeedto stemthe tide of democracyand the ultimate victory belongedto the upper classes.But it would be wrong to infer that that was Rome's object from the very beginning.91 murderedor oppressedtheir political opponents.Cf. Ps.-Xen.,Ath. Pol., iii. I I , Thuc., i. II5.2-3, iii. 27. 2-3. 90 Cf. K. Hopkins, 'Elite mobility in the Romanempire', chapterv below. 91 It may be useful to refer to the following works which have appeared since the original publication of this article. On relations between Spartaand the AchaeanLeaguethere is a full discussionin R. M. Errington, Philopoemen(Oxford, 1969). On the AchaeanWar cf. A. Fuks, 'The Bellum Achaicumand its social aspect',]HS, xc (1970), 78-89. On the policy of Flamininus, cf. E. Badian,TitusQuinetius Flamininus, Philhellenismand Realpolitik (Cincinnati, 1970), and my own article in Latomus,xxxi (1972), pp. 22-53. Also relevantis 1. Touloumakos, Ver EinflussRomsauf die Staats/ormder griechisehenStadtstaatendes Fest/andes und der Inseln 1m ers/enund zweitenJhdt. v. Chr. (Gottingen, 1967); J. Deininger, Ver politische WiderstandgegenRom in Griechenland2q-86v. Chr. (Berlin and New York, 1970), and my forthcoming review in CR.

73

IV THE ROMAN MOB*l P. A. Brunt

I In February 56 B.C., Publius Clodius, the patrician leaderof the urban proletariatat Rome, had indicted his enemy,Titus Annius Milo, on a chargeof seditiousviolencebeforethe popularassembly. (Milo had successfullydisputedClodius' control of the streets by hiring gladiatorsandotherbravadoes.)Pompeyhadundertaken to appearfor Milo at a preliminary hearing.2 Pompeyspoke [wrote Cicero] or intendedto; in fact, as soon as he rose, the Clodian gang raiseda clamour, and throughout his speechhe was interruptednot only by shoutingbut by loud abuseand insults. When he had finished-in this he certainly showedcourage;he was not frightenedaway, said his piece to the end, and now and again securedsilenceby

* From no. 35 (1966). Many statementscan readily be verified in standardhistories of Rome, or for the period covered,in Greenidgeand Clay, Sourcesfor Roman History IJJ-lo B.C., revisedby E. W. Gray (Oxford, 1960), or under the year namedin T. R. S. Broughton,Magistratesof the RomanRepublic (New York, 1951). All datesare B.C. unlessotherwisestated.For all mattersconcerningthe rural plebs and veteransmentionedsee my article in fRS, Iii (1962), pp. 70-86. Before revising this paperI could read only parts of C. Meier, ResPublica Amissa(Wiesbaden,1966), esp. pp. 95-115. ]. W. Heaton,Mob Violence in the late RomanRepublic (Illinois Studiesin Social Science,xxiii, 4, 1939) is inaccuratein details and superficial in interpretation. 2 AdQuintumfratrem, ii. 3. 2. 1

74

THE ROMAN MOB

his authority-upgot Clodius. Our peoplemadesuch a clamour-wehad decidedto show him the samecourtesythat he could not control his mind, tongueor expression. Pompeyhad barely finished at noon; this went on till two o'clock; every kind of insult and the most bawdy verseswere shoutedat Clodius and his sister. Livid with fury, Clodius askedhis followers who was starving the peopleto death. The gang replied: "Pompey". Who wantedto go to Alexandria?"Pompey."Whom did they want to go? "Crassus".... At about three o'clock, as if at a signal, Clodius' peoplebeganto spit in unison at ours. A crescendo of anger. They beganto shoveour peopleout. We charged; the gangstersfled; Clodius was thrown off the platform, and I took to flight; there might have beenan accident. This was a relatively peaceful scenein the 50S. In 58, when Clodius was driving Cicero into temporaryexile, a senatorwas killed in streetfighting. The day after Cicero left Rome,beforehe hadbeencondemnedin law, his houseon the Palatinewas sacked and burned,and the mob marchedout to treat his Tusculanvilla in the sameway. Later that year, Pompeykept to his housein fear for his life. In 57 the efforts of Milo and Sestiusas tribunesto restoreCicero were met by violence; Sestiuswas left for deadin the street;Clodius broughtgladiatorsinto the senate-house. Milo and Sestiusrepelledforce with force, until at last the gentry and bourgeoisieof all Italy came in to vote for Cicero's return. In November,an armedband drove off the workmenwho were rebuilding his house,demolisheda neighbouringportico and setfire to the mansionof his brother 'with the city looking on'. A week later, Cicero was going down the SacraVia, the principal streetin the city centre,which ran from where the Colosseumnow stands to the foot of the Capitol and was lined with great housesand luxury shops,when Clodius' gang attacked:'there were shouts, stones,clubs, swords,all without a moment'snotice'. Cicero was savedby his escort.Next day Clodius tried to storm Milo's house in a fashionableresidentialquarter.'Quite openly in the middle of the morning, he broughtup men with shieldsand drawn swords and otherswith lighted torches'.A successfulcounter-attackwas madeand Clodius fled for his life. 3 3

T. Rice Holmes, RomanRepublic(Oxford, 1923), i, pp. 330-3; ii, pp.

75

P. A. BRUNT

Such violence reacheda climax in early 52., whenMilo at last in murderingClodius outsideRomeanda frenziedmob succeeded brought the body into the senate-house, tore down tribunal and benches,seizedthe clerk's papersand burnedeverythingup, the senate-houseitself and the adjacentPorcian basilica, in a great funeral pyre. A rather similar scenerecurredin 44, whenCaesar's body was burned,and the mob tore to piecesthe poet, Helvius Cinna, underthe misapprehension that he was a praetorwho had publicly sympathizedwith Caesar'sassassins.4But, though the proportionsof violence were unprecedented,violence itself was not somethingnovel in Rome; for almost a century it had been growing more frequent. I proposehere to examine the conditions which favoured or causedit (II-IV), to sketch its progress(v), and to considerthe compositionof the mobs and their aims (VI-VII); I shall conclude by assessingwhat the mob achieved(VIII).

II

The true governingorganof the RomanRepublicwasthe senate, which actedthroughannualmagistrateselectedby the peoplebut drawn from its own ranks. The senateitself was dominatedby a few noble families whosepower reposedon their wealth and on the numberof their dependants,and on the prestigethey derived from their past servicesto the state.Candidatesfor office seldom stood on programmes,and organizedparties did not exist. Men were returned to office occasionallyfor personalmerits (talent could carry outsiderslike Cicero to the highestplace),more often by reasonof their munificenceandlavish bribes,in generalbecause of their family and connections.Birth and wealth usually went together. Cicero describesLucius Domitius Ahenobarbusas a man destinedfor the consulshipsincehe wasborn; in 49 he could 54-61; esp. Cic., ad Alt., iv. 3. 2-3' SacraVia, S. B. Platner,Topographic Dictionary of AncientRome,revisedby T. Ashby (Oxford, 1929), pp. 456 If. 4 For 52 Ascon., 32-3 (Oxford text); Appian, Civil Wars, ii. 20 If.; (Cassius)Dio, xl. 48 If.; for 44 Appian, ii. 143-8; Dio, xliv. 35-51; Plut., Caesar, 68; Antony, 14; Brutus, 18; 20. The peoplewere inflamed by the readingof Caesar'swill in which he left his gardensfor public use and a sum of money to every citizen domiciled at Rome.

THE ROMAN MOB

offer farms of thirty acresapieceto some thousandsof soldiers. or clientswho weremorally Suchnobleshad numerousdependants expectedand often economically compelled to support them.s They usedtheir power to grow richer from the profits of war and empire, and to oppose every measureto relieve the poor, the provisionof cheapgrain, the distribution of land or the remission of debt. Here they had the backing of the upper classin general, whosespokesman,Cicero,declaredthat the prime duty of government was to ensure'that every man kept his own'. And public largesses,which did not infringe propertyrights, could be rejected on the ground that they were more than the treasurycould bear, the treasuryfrom which senatorsdrew handsomeallowancesfor 6 themselves. In theory the people at Rome possessedgreat power. They electedthe magistrates,declaredwar and ratified treaties,passed laws, and until the creationof standingcourts in the late second centurydecidedthe most importantcriminal cases;to the end of the Republic somepolitical chargescamebefore them. From the late secondcenturythey votedby ballot; this naturallydiminished aristocraticcontrol. Therewasmorethanonepopularassembly.Of thesethe comitia centuriata was timocratically organized.Decisionswere taken by a majority not of heads,but of voting units called centuries;the well-to-do, if they were of one mind, could decidethe issues;the citizens with no property at all, and who are said to have outnumberedall the rest put together by the time of Augustus, formed only a single century, which might neverevenbe called.7 The rural poor therefore had little influence in this body the importanceof which was great,for it electedthe chiefmagistrates; and it was men who had held the highestoffices who dominated the senateitself. This assemblywas also competentto legislate, but laws were H. Jolowicz, Historical Introduction to the Study of RomanLaw, znd edn (Cambridge,1952), chapterii and iv, gives an excellentintroduction to the Romanconstitution. For the working of the political systemsee L. R. Taylor, Party Politics in the Age of Caesar(Berkeley, 1949). Domitius: Cic., ad Alt., iv. 8a, z. Caesar,Civil War, i. 17. Clients; see especiallyM. Ge1zer,Kleine Schriften, i (Wiesbaden,196z), pp. 68 ff. 6 De officiis, ii. 7z-end,cf. Brunt, fRS, Iii (196z), pp. 69 ff. 7 Dionysius of Halicarnassus, RomanAntiquities, iv. zo. 5; Zl. I (basedon personalobservation,Zl. 3, after 30 B.C., i. 7. z). S

77

P. A. BRUNT

more generally passedby a less cumbrous body, in which the voting units werethirty-five tribes. The tribes werelocal divisions of the people; thirty-one were rural and four urban, though all freedmen (except such as were substantial landowners) were registered,whereverdomiciled, in the urban tribes. In the tribes, rich and poor had equal votes.S At all times by far the greaternumber of citizens lived in the country, and it might seem that the organization of the tribal assemblyensuredthat the wishes of the rural majority would prevail, perhapseven to an,undueextent; thirty-one to four was not the true proportion between town and country dwellers.9 However, as ProfessorToynbee has recently pointed out, the systemof primary democracy,in which the citizen canexercisehis voting rights only by attendingthe sovereignassemblyin person, can only work democraticallyif voters have not to spend more than two nights away from home.1o Even in the third century many citizenswere a hundredmiles distantfrom Rome, and after 80 they comprisedthe free populationof Italy southof the Po. It was only on rareoccasionsthat the peasantscamein to vote. If the censorswho held office every five years were careful to register everycitizenwho movedfrom the countryinto the city in an urban tribe insteadof a rural, the votesof the rural tribes musthavebeen exercisedby the minority of their memberswho had the leisure and meansto visit Rome for the purpose,the very sameclass of wealthy landownerswho controlledthe centuriateassembly.ll It seems,however, that the censors did not do their work thoroughly. Dionysius of Halicarnassussharply contrasts the centuriateassemblycontrolled by the respectableclassesand the tribal, composedof artisanswith no hearthsof their own.12 He purports to be describing the early Republic, but the picture is imaginaryand drawn from the conditionsknown to the annalists of the first century. In the Principateurban dwellers are attested L. R. Taylor, Voting Districts of the RomanRepublic(Rome, 1960) is fundamentalon the tribes. 9 In 70, 910,000adult male citizens were returned;there might have been much failure to register,but I hope to argueelsewherethat about 1,200,000is a realistic figure; of thesein the 40S about a quarterlived in Rome (III below). 10 Hannibal's Legacy(Oxford, 1965), i, p. 297. 11 Taylor, Party Politics, 57 fr.; cf. Brunt, IRS, Iv (1965), pp. 103 fr. 12 iv. 16-21; viii. 6; x. 17. 8

THE ROMAN MOB

in rural tribes. One piece of evidence suggeststhat this was possibly as early as 133. Tiberius Gracchus,who had hitherto relied on the rural voters (one of the few known instancesin which they swarmedin to vote), beganto court the urbanplebs, as his followers were occupiedwith the harvest.His action would havehadlittle purpose,if the urbanplebshadbeenconfinedto the four urban tribes. It may be indeed that even a few immigrants who had moved into Rome since the last censusand had not yet beenreregisteredmight have balancedor outvotedthe wealthier membersof their tribes, and that it was to such a handful of citizens that he appealed.Even so, urbandwellers were evidently influential in the rural tribes.13 And between70 and 28 it is not clear that any censuswas completed.l4 It seemsprobablethenthat normally the urbanplebshad a majority in the tribal assembly. However, the assembliescould do nothing except with the collaboration of a magistrate. They could meet only on his summons,and only vote 'Yea' or 'Nay' on his proposals;a private citizen could not even speakexcepton his invitation. The plebs could not obtain redressof its grievances,unless a magistrate drawn from the upper classeswas preparedto take the initiative. Genuinesocial concernor personalambition led nobles like the Gracchi,Caesarand Clodius to comeforward as 'popular'leaders from time to time, but there was no consistentand continuous opposition,no organizedand enduringpopular party.lS Even if a magistratesubmitteda popular proposal,it did not follow that it would go through. It could be obstructedon religious pretexts,or vetoed. A single tribune could veto what all his nine colleaguesproposed.The tribunatehad arisenin the class struggles of the early Republic for the protection of popular H. Dessau,[LS, 168; 176; z86; 6045 f.; 6063 f.; Appian, i. 14. Taylor, Parry Politics, p. 53 cites Cicero, Sest.,109, but the assertionhere that not more than five men might vote in a tribe could refer to uncontroversiallegislation.Cic., de lege agraria, ii. 71 treats the voting rights of city dwellers as an important privilege, though motions to redistributefreedmenamongthe rural tribes in 88-7, 84 (Periocheof Livy, lxxxiv) and 66 (cf. Clodius' plan in 5Z, Asconius, 5z) suggestthat votes in the urban tribes were not much regarded. 14 G. Tibiletti, Studia et DocumentaHistoriae el [uris, xxv (1959), pp. 94 ff. thinks that Sulla and Caesaras dictators revisedthe lists. I do not feel certain that use was not madeof the incompleterevisions effectedby somecensorsin the interveningyears. 15 C. Meier, RE, Supplement,x (1965), pp. 550-67.

13

79

P. A. BRUNT

interests,and in the secondcenturyPolybiuscould still say that it was the tribune'sduty to do alwayswhat the peopleapproved.To the end most of the championsof the commonsactedas tribunes. None the less, Polybius' statementdid not correspondto the constitutionalpracticethat had evolved by his time. The senate could almost always find at least one tribune to act on its behalf and (as Livy put it) to use the tribunician veto to dissolve the tribunician power. Tribuneswere often noblesthemselves,or in Livy's words 'chattels of the nobility'. Marcus Octavius who vetoedTiberius Gracchus'agrarianbill (seebelow, page92) was, for the middle and late Republic, the more typical tribune of the twO. 16

According to Burke 'a statewithout the meansof somechange is without the meansof its conservation'.At Rometherewere too many checksand balancesin the constitution,which operatedin practiceonly in the interestof the ruling class.Reformershad to useforce, or at leastto createconditionsin which the senatehad reasonto fear its use (seebelow, pages92 ff.). This was the first factor which favouredthe growth of violenceat Rome.

III In the second place, Rome was even by modern standardsa populouscity, in which there was no garrison and no police to control the multitude. To the total size of the populationthereis no direct testimony. But the numberof recipientsof free grain had risen to 320,000in the 40S.17 Only adult maleswerenormally eligible,18andwe therefore have to estimatethe numberof women and children in this class.The grain recipientswere partly free-born,partly freedmen. Polyb., vi. 16; Livy, v. 2. 14; x. n. 8 (both texts that reflect later conditions). Cic., de legibus, iii. 24 describesthe tribunateas 'temperamentumquo tenuiorescum principibus aequarise putarent'. On its role in the middle Republic see]. Bleicken, Vas Volkstribunatder klass. Republik(Munich, 1955). 17 Suet.,Caesar, 41; cf. Dio, xlii. 21. 4. 18 Trajan included some children, and Augustusmademoney gifts to children (Pliny, Pane.wricus,26; Suet., Aug., 41); theseseemexceptional, contra D. van Berchem,Les distributions de ble et d'argent,) fa p/eberomaine sous/'empire (Geneva,1939), pp. 32 if.

16

80

THE ROMAN MOB

Appianimplies thatin 133 thepoorwereunableto raisechildren.19 Abortion andinfanticidewerenot forbiddenby the law, andmany parentsmust have exposedtheir babies, some of whom might then be brought up as slavesby the finders. The infanticide of female infants must have been common even in the senatorial class, among whom in Augustus' reign men outnumbered women; if we make the reasonableassumptionthat it was still moreprevalentwith the poor, the birth-ratewould also havebeen depressedby a scarcity of reproductivewomen.20 There is some ground, however, for thinking that the urban plebs consisted preponderantlyof freedmen(seebelow, pages89 f.), and particularly after Clodius made grain distributions free in 58 masters were very readyto manumitslaves,who could still be requiredto work for them, while obtaining rations from the state.21 Now it seemsto me unlikely that there were so many female slaves or freedwomenas male slaves or freedmen. In this period slavewomen were not neededto keep up the stock of slaves,most of whom were 'made' by capturein war or kidnapping. And they were employableonly for householddutiesand to someextentin spinning,weavingandmaking clothes,occupationsperhapsmore commonon country estatesthan in town houses.22 Slavesmight enterinto a quasi-marriage,but both spouseswere not necessarily freed together,and any children bom in slavery,who were slaves themselves,might be manumittedonly at a later date. In many thousandsepulchralinscriptions of freedmen at Rome (mainly imperial) under thirty per cent record offspring, and still fewer marriage.23 For thesereasonsI doubtif we needmorethandouble the figure of 320,000to includebothwomenand childrenof comrecipients. Civil Wars, i. 7. T. Mommsen,RomischeStrafrecht (Leipzig, 1899), pp. 617-20, 637; Dio, liv. 16. 2. 21 Dionysius of Halicarnassus, iv. 24. 5; Dio, xxxix. 24. The services freedmenowed patronsare best discussedby C. Cosentini, Studi sui liberti (1948-50). 22 T. Frank, EconomicSurveyof Ancient Rome(Baltimore, 1933-40),i, pp. 373 f.; v, pp. 199 if.; H. J. Loane, Industry and Commercein the City of Rome,10 B.C.-200 A.D. (Baltimore, 1938), pp. 69 if.; cf. ILS, 8393 (30), 19

20

where assiduityin wool-making is commemoratedamongthe virtues of a great lady in Augustus'time; Ascon., 43 (Oxford text) for weaving 'ex vetere more' in a town house. 23 T. Frank, AHR, xxi (1915/6), pp. 689 if.

81

P. A. BRUNT

Well-to-do residentswere presumablynot numerically significant. Thereremainthe slaves.A rich man requireda large staff of domestic servants,secretaries,etc; and his standing might be measuredby the number of his attendantsand flunkeys. There might also be women engagedin textile work. Under Nero an eminent senatorhad four hundred slaves in his town-house.24 However, in the 5os the scale of manumissionsshould have diminishedthe slavepopulation.I guessthat 100,000would be a liberal estimate.The city populationmight then have beenmore or less than 750,000. Clodius' bill probably acceleratedthe drift from the country, but it had beengoing on before, and the number of slaves and freedmen had been progressivelyincreasing. However, no numericalestimatecan be venturedfor any earlier date.25 In the early Principate the governmenthad at least 12,000 soldiersin Rome,not to spe.akof sevencohorts,which ultimately and perhapsfrom the first comprised7,000 men, raised to deal with fires; they were military units and could also be used as police. Evenso, it washardto keeporder.In 39 B.C., thoughthere were troops at hand which savedhim in the end, Octavianwas almost lynched in a riot, and Claudiuslater was only rescuedby soldiersfrom a famishedmob. The narrow, winding streetsand high buildings (see below, page 85) did not help in suppressing riots. In A.D. 238 the populace,armedby the senate,besiegedthe depletedpraetorianguard in its camp; when the soldiers sallied out and pursuedthem into the streets, the peopleclimbed up into the housesand harassedthe soldiersby throwing down on them tiles, stonesand pots of all kinds; the soldiersdarednot go up after them, not knowing their way about the houses;but as the housesand workshopshad their doors barred,they set fire to the many woodenbalconies;the tenementswere set close together,and large parts of the buildings were wooden; so the flames soon devastateda very greatpart of the city, one sectionafter another. 24 25

Tac., Ann., iii. SS; xiv, 43; cf. n. 22. Brunt, IRS, Iii (1962), pp. 69 f.; here I underestimatedthe opportunities of employmentfor free men (cf. p. 89 below), and went too far in minimizing the drift before S8. The building of the Marcian aqueduct in 144 is a notable indication of increasedpopulation.

82

THE ROMAN MOB

Somethingof the same kind nearly occurredin 88, when Sulla marchedinto the city. No doubtit was suchdangersthat madethe emperorsready to spendlarge sums on 'bread and circuses'.In other towns they had no suchmotive to carefor the poor and did not do SO.26 The aristocratic governmentof the Republic had no police available; the magistrateshad but a few attendants.Nor were troops normally found in the city, though in 121 the consul happenedto haveat his disposalCretanarcherswhom he usedin 27 How could the nobility ever hold suppressingthe Gracchans. the mob in check,when it was inflamed againstthe government? The mob wasgenerallyunarmedandrelied on sticksandstones. To carry arms was a capital offence,28and in any event the poor would possessnone, except knives. Moreover, as legions were recruited in the country, not the city, the urban poor were not trained in the use of arms. The well-to-do would have their own equipment,including body-armour,and had mostly seenmilitary service; and the senate could authorize the arming of their followers. On occasionspopularleadersdistributedarms illegally to the mob, but even then, man for man, their followers were probably unequalto their opponents.Though armed, the partisansof Gaius Gracchusin 121 put up but a feeble resistance.The numerousclients of the greathousesin the city itself often enabled the governingclassto make a standagainstthe mob, reciprocating or eveninitiating violence(seebelow, pages90-1, 94-5). Given time, the senateor magistratesor individual noblescould call up clients with military experiencefrom the country. In 100, Imperial police: A. Passerini,Le coorti pretorie (Rome, 1939), pp. 44-66 (esp. on Dio, Iv. 24. 6 and Tac., Ann., iv. 5); P. K. Baillie Reynolds, Vigiles of Imperial Rome(Oxford, 1926); for their police duties see Dig., i. 15. 3. Riots: Appian, v. 68; Tac., Ann., xii. 43; Herodian,vii, 12. 5; cf. Appian i. 58. C. Meier, Res Publica Amissa,pp. 157 ff. is interesting on the lack of a police force in the Republic. 27 Plut., C. Gracchus, 16; Orosius, v. 12.7. 28 W. Kunkel, Untersuchungenzur Entwicklungdes rom. Kriminalverfahrensin vorsullanischerZeit (Munich, 1962), pp. 64 ff. shows that Sulla's law (cf. Cic., pro Milone, I I) has much earlier antecedents(Plautus,Aulularia, 415 ff.). Arms were given to the 'mob' in 121, 100, 88, 87, 62 (Plut., Cato Minor, 27 f.) and of courseby Clodius; by the senatein 121 to senators,equitesand their servants(Plut., C. Gracchus, 14), to the plebs or 'populus'in 100 (Cic., pro C. Rabirio, 20; Oros., v. 17. 7), presumably only to reliable elements.

26

P. A. BRUNT

men from Picenum took part in the suppressionof Saturninus. The armed followers with whom Cicero surrounded himself during the Catilinarian conspiracyof 63 included chosenyoung men from Reate.In 59 he was hoping to resist Clodius by force; his friends and their clients, freedmen and slaves would band togetherin his defence.A great concourseof substantialcitizens from all over Italy ensuredthat he was recalledfrom exile in 57, although his enemy, Clodius, remained dominant over the city proletariate.In 56 Pompey summonedfollowers from Picenum andthe Po valley for his protection.To endthe uproarensuingon Clodius' murderin 52, the senateauthorizeda levy all over Italy, and soldiersrestoredorder in the city. But between59 and 52 the senatewas generallyimpotent,becausePompeywith his veterans and Caesarwith his greatarmy in the north could marshalforces strongerthan the senatecould command.29

IV The third factor in the turbulenceof the city populationmay be found in the misery and squalor in which they lived, which naturally made them responsiveto politicians who promised to improve their conditions and engenderedhostility (if only intermittent) to the upper classeswho showed little care for their interests.3o For lack of modernmeansof transport,the peoplewere crammed into a small built-up area, not much larger than that of 29 Cic., pro C. Rabirio, 22 (100); in Cat., i. II; ii. 5; iii. 5; pro Murena, 52;

cf. Sail., Cat., 26. 4; 30. 7; 50. 4 (63). Cicero was accusedof using armedslavesbut claims to have mobilized in Decemberall the upper classes,and all true citizens, indeed 'omnis ingenuorummultitudo, etiam tenuissimorum'(implausible), Phil., ii. 16; cf. in Cat., iv. 15 f.; they took a military oath, Dio, xxxvii. 35. Rice Holmes, RomanRepublic, ii, p. 60 (57); Cic., adQuintumfratrem,i. 2. 16 (59); ii. 3· 4 (56); Holmes, ii, p. 167 (52). 30 In this section where referencesare omitted, the texts are cited by J. Carcopino,Daily Life in AncientRome(London, 1941), mainly basedon imperial evidence;in the Republic things were worse. Z. Yavetz, Latomus,xvii (1958), pp. 500 ff. gives further details for Republic (e.g. on fearful conditions createdby plagues);his referencesand interpretationof texts are not all reliable.

THE ROMAN MOB

modern Oxford, with a density seven or eight times as great. The streetswere winding and narrow, evenmain thorough-fares under twenty feet wide. While the rich had their luxurious mansionson the Palatine or spaciousgardensin the suburbs, most inhabitantswere pennedinto tiny flats in tenements,which had to be built high; Augustusimposeda limit of seventyfeet (which suggeststhat this had been exceeded),and Trajan pronouncedthat dangerous,reducingit to sixty. Cicero constrastsa newly plannedcity with Rome 'situatedon hills and in valleys, lifted up and suspendedin the air, with no fine streetsto boastof but only narrowpaths'. The lower partsof the city were subjectto periodic floods, and the collapseand conflagrationof buildings were commonoccurrences.In the Principateit is said that not a day passedwithout a seriousfire, yet thentherewere7,000vigUe! to put themout; in the Republiconly a small force of publicly ownedslaves.Crassushad a gang of five hundredbuilders, and boughtup housesthat were afire or adjacentto a blaze at knock-downprices with a view to rebuilding on the sites.31 Thesedangerswere aggravatedby bad methodsof construction.Ownerswould not or could not afford to employ skilled architects or suitable materials. The local travertine crackedin fires, but it was too costly ·to bring better stoneevenfifty miles by land. A thin facing of stonemight conceal a filling of soft rubble. To conservespace,party walls had to be not more than a foot and a half thick; given this limit, only bakedbrick was strong enoughfor high buildings, yet sun-dried brick was often used. Walls were sometimesof wattlework, the more dangerousas it was too expensiveto bring larchwood, relatively imperviousto fire, all the way from the Adriatic. In 44 Cicero reportedto Atticus that two of his tenementshad fallen down, and that crackswere showing in others;the tenants-and the mice-hadall fled.32 The housesof the poor must also havebeenill-lit, ill-ventilated and unwarmed;facilites for cooking were inadequate;water had to be fetched from the public fountains, and the supply cannot 31

32

Catullus, 23. 9; Dig., i. 15. 2; Plut., Crassus,2. Vitruv., i. 3. 2; ii. 2; 7.3 f.; 8. 7-9; 8.16 f.; 8. 20; 9.14-17; vi. 8. 9; x. pro 2; Cic., de divinatione, ii. 99; ad Att., xiv. 9. I. Yavetz thinks that many of Vitruvius' preceptsare directedagainstcommonbad practices in building.

P. A. BRUNT

havebeenabundantuntil the old conduitswere repairedand new. aqueductsbuilt underAugustus;further, the tenementswere not connectedwith the public sewers. We may fairly supposethat most of the inhabitantsof Rome lived in appalling slums. They offered shelter,but little more. As for furniture, Cicero speaksof the poor man as having no more than a stool and a bed wherehe lived, worked and slept.33 From suchtenementsmen like Cicero drew as landlordsa good income. Cicero'spropertyon the Aventine and in the Argiletum, probably two lower-class districts, was in 44 bringing him in 80,000 HSS, enoughto have paid 160 legionariesfor a year under the ratesthat had obtaineduntil recently;he appropriatedit to the allowancefor his undergraduateson at Athens, and was anxious to havetenantswho would pay on the nail,34 Perhapsthat was not so easyto ensure.Then, as later, it is probablethat the return on investment in house-propertywas high precisely becausethe risk was great.35 In the 40S therewas a prolongedagitationabouturbanrents.In 48 the praetor,Marcus Caelius,who proposeda year'sremission, was driven out of the city by the consul,but only after bloodshed. Caesar,however, granted the remission in the same year, and perhapsextendedit in 47, after further tumults, when barricades were raised,soldierscalled in and eight hundredrioters killed. It appliedto rentsup to 2,000 sestercesin Rome,and 500 elsewhere, an indication that the cost of living in Rome was exceptionally high.36 (A generationearlier, Cicero gave the daily wage for an unskilled laboureras threesesterces;obviouslyhe could not have afforded 2,000 for a year'srent. We cannotsaywhetherwageshad risen in the interim, or whether the remission was intended to benefit people at a rather higher level, such as shopkeepers.) Cicero'scommentis characteristic.'Thereis no equity in abolishing or suspendingrents. Am I to buy and build and repair and spend,and you to havethe benefit againstmy will? Is this not to Cic., in Cat., iv. 17. Cic., ad Alt., xii. 32. z; xv. 17. 1; ZOo 4; xvi. 1. 5. 3S Gellius, xv. 1. 3. But F. Schulz, ClassicalRomanLaw (Oxford, 1951), pp. HZ ff. shows how the law favoured owners againsttenants. 36 Caesar,Civil Wars, iii. zo f.; Dio, xli. 37 f.; Appian, ii. 48 (Caelius); Dio, xIii. z9-33 (riots of 47). The Fasti Ostiensesunder 48 record a year's remissionof rents; Dio, xlii. 51 puts remissionin 47; other details in Suet.,Caesar, 38; 4z.

33 34

86

THE ROMAN MOB

take away the propertyof someand give to otherswhat doesnot belong to them?'37 How did the people of Rome live? Rome was never a great industrial city; indeedthereneverwas any large-scaleindustry in the ancientworld of the kind familiar since the industrial revolution: the high cost of transportalone forbade the productionof factory goodsfor a world-wide market.38 Adjacentto Romethere were no abundantsuppliesof fuel or raw materials.The Tiber is not well suited to navigation, and the port of Ostia had not yet beendeveloped;the largershipshadto dischargein an openroadsteadinto lighters. None the less far more use was made of the river and its affiuents(for downstreamtraffic as well as for transport from the mouth) than we would expectfrom presentconditions; the growth of the urban population left no alternative.39 The supply of this populationcreateda greatdemandfor wholesaleand retail traders,dock labour, cartersand so on. So too large numbersmust have been employedin the building trade: more fine public edifices were now being put up; the rich were continually erecting more luxurious town-housesand villas in the vicinity of Rome, and the increaseof the population in itself required more tenementsand shops, a demand augmentedby the frequencyof fires and collapses.Evidencefrom pre-industrial cities in other times may help to supply the lack of ancientstatistics. In 1586 up to 6,000 workmen were engagedon public buildings at Rome,of whom 800 with I 5°horseswere neededto movethe obeliskinto thePiazzaof St Peter's;at the time the total populationseemsto have beenunder 100,000.In 1791 a third of all Paris wage-earnerswere occupiedin the building trade.40 In addition,therewereartisansandshopkeepers of all kinds, manyof whom musthavesold goodsthey madethemselves,perhapsto the order of clients. Beggars,curiously, are hardly ever mentioned, perhapsbecausethe Romans(unlike the Jews,and the Christians Cic., pro Rosciocomoedo,28; de ofjiciis, ii. 8, f. Seee.g. A. H. M. Jones,Later RomanEmpire (Oxford, 1964), ii, pp. 841 if. 39 SeeR. Meiggs, RomanOstia (Oxford, 1960), chaptersii, iv and viii; J. Ie Gall, Le Tibre, fteuve de Romedansl'antiquiti (Paris, 195,),passim;L. Casson,IRS, Iv (1965), pp. 31 if. 40 J. Delumeau,Vie economiqueet sociale de Romedans la secondemoitie du XVI' siMe (Paris, 1959), i, pp. ,66 f.; cf. p. 281. G. Rude, The Crowd in the French Revolution(Oxford, 1959), p. 19. 31 38

P. A. BRUNT

after them) recognizedno specialobligation to relieve the poor as such; it was another matter if the great housessupportedidle dependants,whosevotes and strong arms they could employ; on themthey conferredbenefitsin accordancewith the usualprinciple of Romanmorality: 'do ut des'.41 According to tradition king Numa had organized craftsmen into collegia or corporationsof flautists, goldsmiths, carpentersthe word fabri came to mean builders in all sorts of materialdyers, shoemakers,coppersmithsand potters. If only theseparticular corporationswere in fact ancient, they go back to a very remotetime, when for instancethe useof iron was still unknown; in the historic period there must have been many ironworkers, especially to make arms for the legions which were regularly enrolled and equippedjust outsidethe city. The list also doesnot include bakers;accordingto Pliny there were none down to the middle of the secondcentury; the womenusedto grind and bake at home; presumablythey ceasedto do so, when so many of the poor were lodgedin houseswithout suitableovens.In the course of time many more corporationscameinto existence.The fishermen who fished in the Tiber had an old festival. Fulling ceasedto be a domestic craft. Plautus casually mentions a score of other trades.Cato in the secondcentury recommendedbuying at Rome tunics, togas,cloaks,patchworkcloth and woodenshoes(though someof thesethings were also madeon his estates),and in addition jars, bowls, ploughs, yokes, locks and keys and the finest baskets.42 As in medieval towns men of one craft tendedto congregate. H. Bolkestein, Wohltatigkeit und Armenspflegeim vorchristlichen Altertum (Utrecht, 1939),passim; for beggars,pp. 339-41; add Seneca,de beata vita, 25. 42 For tradesat Rome see Loane (n. 22); for collegia J-P. Waltzing, Etudes historiquessur les corporationsprofessionelleschez les romains, i-iv (Louvain, 1895-19°0),esp. i, pp. 62-92 (Republic); W. Liebenam,Zur Gesch.u. Organisationdes rom. Vereinswesens (Leipzig, 1890); F. M. de Robertis, II Diritto AssociativoRomano(Bari, 1938). Numa, Plut., Numa, 17; arms factories, Livy, xxi. 57. IO; Cic., Phil. vii. 13; bakers,Pliny, Nat. Hist., xviii. 107; fishing, Festus,232; 274L; Horace,Sat., ii. 2, 31 ff.; fulling, Pliny, xxxv. 197 (220 B.C.); Vitruv., vi. pro 7 (no longer domestic); Cato: see his de agricultura, 135. Waltzing gives a list of known collegia at Rome; for Republicancollegia there and in Italian towns attestedin inscriptions, see A. Degrassi,InscriptionesLatinae liberae reipublicae (Florence,1963), ii, pp. 476 ff.

41

88

THE ROMAN MOB

There was a pottery district, and streetswere named after the silversmiths,grain merchants,sandal-makers,timber merchants, log-sellers,perfumersand scythemakers,probably many more.43 Collegia of artisanswould thus be composedof neighbours. Many tradersand artisanswere not of free birth. Slaveswere employed in every trade, craft and profession. Freedomwas a necessaryincentive to good work and seemsoften to have been grantedfairly soon,or boughtby the slavefrom the wageor share of the profits he was allowed. The freedmannaturally worked at his old trade and was probably often still financed by his old master.Most of our evidencecomesfrom epitaphs,which tendno doubt to give the impression that more craftsmen were free than was the case; most slaves who appearhad probably been unlucky enoughto die early. Of jewellersandgoldsmithsat Rome, to take one instance,35% are slaves, 58% freedmen,only 7% of free birth. The last figure is astoundinglylow. But the inscriptions only fortify the evidencewe havefrom literary sourcesthat in the urban populationas a whole, as well as in the crafts and trades, men of servile origin preponderated.The statistics may exaggeratethe preponderance;the freebornmay have beenless ready to indicate manual employments in which they took no pride or even to commemoratetheir lives at all (a freedman could be proudof havingbeenfreed); they may evenhavebeentoo poor to leave a record, having been confined to unskilled and unremunerativework.44 Many freedmen(perhapsmostin Rome)camefrom the eastand probably brought with them new skills; with the capital their patronsprovided, they thus had an advantageover native workmen.4S FreebornItalians, someof whom were displacedpeasants, would then havehad no meansof employmentexceptcasual,unskilled labour. They could go out into the countryfor the harvest, vintage and olive-picking, just as Londoners go out today to pick the Kent hops. This is well attested,and can be explained. The Romanlandownerpreferredto rely on a permanentlabourLiebenam,Zur Celeh. u. Organization, p. 9 f. L. R. Taylor, AlP, lxxiii (1961), pp. 113 ff., with earlier literature; add A. M. Duff, Freedmenin the Early RomanEmpire (Oxford, 1928) ch. vi. 45 Also, like Jewsand Quakersat other times, they were barredfrom many other activities and theirenergieswere directedinto economic advancement.

43

44

P. A. BRUNT

force of slaves,but as Catomakesclear,he did not wish to feed idle mouths. For seasonaloperations,therefore, he required supplementary labour provided by free hired men.46 On the same principle we mustsupposethat most dock labour and the ancillary carting of supplieswas free; there was little sailing for half the year,andwork musthavebunchedin afew monthsor weeks.47 And it required no special skill. Similarly building contractors,whose businessis likely to have fluctuated, would not have found it profitable to keep enoughslavesthroughoutthe year for all their work. The builders on Cicero'sTusculanvilla went back to Rome to collect their free grain rations as citizens. It has beenplausibly conjecturedthat the distressTiberius Gracchussoughtto alleviate had beennewly aggravatedby unemploymentresulting from the completionof the Marcian aqueduct.The emperorVespasianwas to refuse to adopt a labour-savingdevice; if he did so, he asked, how could he feed his poor commons?48 Sallust and others tell of the drift of countryfolk into Rome; Sallust speaksof young men who had barely made a livelihood with labour in the fields and were attractedby the private and public largessesin the city, and Cicero could urge the urbanplebs with somesuccessin 63 not to forsakethe advantagesof life there, their votes (which could of coursebe sold), games,festivals and so on, for land allotments in barren or malarial places. What Sallust says of the private largessesis probably important; the greathousescould afford to maintainclients, and they might even be given rent-freelodgings.49 Sometimesmagistrates,to enhance their popularity, distributedgrain or oil at low prices, bearingthe cost themselves. 50 Above all there were the cheap or free public Brunt (n. I), p. 72. Toynbee,Hannibal's Legary, ii, pp. 296 ff. on Cato. J. Rouge, REA, liv (1954), pp. 316 ff. Piracy or hope of great profits made merchantssail in winter; Claudius had to assumethe risk of storm damage,to induce shipownersto bring grain to Rome in winter (Pliny, Nat. Hist., ii. 125; Suet.,Claud., 18). 48 Cic., ad Att., xiv. 3. I; H. C. Boren, AJP, lxxix (1958), pp. 140 ff; AHR, lxiii (1957-8), pp. 890 ff.; Suet., Vesp., 18. See Loane, Industry and Commerce,pp. 79 ff.; builders naturally had permanentgangsof slaves too. 49 SaIl., Cat., 37 (cf. Varro, de re rustica, ii. pro 3; Appian, ii. 120; Suet., Aug., 42); Cic., de lege agraria, ii. 72. Rent paid, Trebatiusin Dig., ix.

46 47

3· 5· 50

1.

E.g. Cic., in Verrem, ii. 3. 215, de offic., ii. 58; Pliny, Nat. Hist., xv. 2; xviii. 16.

THE ROMAN MOB

corn-dolesinstitutedgenerallyby popularleaders,partly perhaps to reducethe dependence of the plebson noblepatrons.However, the distributionswere not free until 58, the liberality of the cheap distributions provided under Gaius Gracchus'law in 123 was soon reducedand not restoredtill 100, and distributionswere in abeyancefrom 80 to 73 andrestrictedto only some40,000recipients from 73 to 62.51 Moreovermen could not live on breadand shows alone; therewas other food, and clothesto be paid for, and rent. Augustuswas to introducea quicker methodof distributing free grain which did not take the recipientsaway from their work so long as in the past.52 The people of Rome had to earn much of their living, and for many of them casual employmentwas the only means. Gaius Gracchusmust have won much support by his programmefor building roads and granaries.53 The feeding of the city populationwas also a grave problem. Therewere largeimports from Sicily, Sardiniaand Africa, but the supplywas precarious,liable to be interruptedby piracy and wars. Much grain must still havecomefrom Italy, or elsethe population could not have survivedthe years43-36, for most of which it was cut off from overseasupplies.54 The public rations did not suffice for a family, 55 and some,if only a minority, of the recipientsmust havehad wives and children. Somegrain had to be boughton the market, evenin the yearswhen therewere public distributionsto most of the free population.In 57-6 it seemslikely that therewas Brunt, fRS, Iii (1962), p. 70 n. 10. R. J. Rowland Jr, Acta Antiqua, xiii. 81 deniesthe restriction in 73 without explaining the evidencefor it (Sallust,Oralio Macri, 19 with Cicero in Verrem, ii. 3. 72) on the ground that Sicilian surplusesin the late 70S acquiredby the government sufficed for 180,000(in Verrem, ii. 3. 163); but some of this grain was probably neededfor large armies in Italy and Spain, and the rest sold at market prices, which fluctuated(ibid., 215). 52 Suet., Aug., 42. 53 Plut., C. Gracchus,6. About 85 a praetor, Marius Gratidianus,gained great popularity by trying to eliminate debasedcoinage,seeBroughton, Magistrates,ii, p. 57; this seemsto imply that the masseswere interested in stoppingan inflationary rise in prices. 54 M. Rostovtzeff,RE, vii (1910), pp. 126 ff., who like all writers (esp. Toynbee,Hannibal's Legacy,ii, pp. 296 ff.; pp. 585 ff.) unduly depreciates the continuanceof cerealcultivation in Italy; I hope to show this elsewhere. 55 5 modii (about 41 lit res) a month. Cato gave his slaves 3-41. accordingto the heavinessof their work; de agricultura, 56. 51

P. A. BRUNT

not enoughin the public granariesto honour the state'sobligation; the marketprice was a matterof generalconcernand might soar to famine rates(seebelow, pages99 f.). And market prices fluctuated sharply, soaring when the harvestswere poor and when hoardingby growersand merchantsaggravatedthe shortage. It is an illusion that in the late Republic the urbanplebs was usually well and cheaplyfed by the state.As for modernscholars who repeat ancient gibes that the doles corrupted the urban population, one must wonder if they would also condemn all modern measuresof social welfare; in Rome there were no charitable foundations for the poor, and no unemployment benefits.

v The progressof violence.may now be sketched.In 133 Tiberius Gracchusproposedto redistributeamong the poor public lands which the rich hadoccupied.His colleague,Octavius,interposeda veto; Gracchushad him deposedby vote of the assembly,an unprecedentedact which set aside the most important of the constitutionalchecks.His bill was then carried. Actual violence was not used,but the menacingattitudeof the peasantrywho had flocked in to back Gracchusmay explain why Octavius did not dare to veto the motion for his own deposition.Later in the year the senators charged Gracchuswith aspiring to tyranny and lynchedhim in public. Thefirst openact of illegal political violence camefrom the nobility. In 123-2 GaiusGracchusastribunecarriedmanyanti-senatorial measures.(In 123 no other tribune had the will or courageto opposehim; he had the backingof both urbanandrural plebsand of the equites,rich men outsidethe senate,on whom he conferred important benefits;he did not needto use force.) But eventually he lost popular favour and office, and as a private personin 121 armedhis followers to obstructthe repealof one of his laws; he and they were massacredby senatorialforces quite legally.56 56

The biasedand contradictoryevidencehardly enablesus to decidethe extent to which either of the Gracchi was to blame. Amid the tumults the true facts may never have been known.

THE ROMAN MOB

In 103 and 100 the tribune, Saturninus,who alsoproposedlanddistribution and revived the grain dole on the Gracchanscale,did not scrupleto murderopponentsandrivals; he too was suppressed by the senate.In 88 the tribune Sulpicius,promotingthe interests of the newly enfranchisedItalians,and also of the freedmenwhom he proposedto redistributeamongall the tribes, drove his opponentsfrom the forum by force; the consul, Sulla, appealedto his army (where his ability and generosityassuredhim of support), marchedon Rome and proscribedSulpicius and his friends. This was the first occasionon which the army was employedto overturn decisions made at Rome; once again, it was a noble and conservativewho took the fatal step. Sulla's successor,Cinna, revived Sulpicius'proposals;the streetsran with blood in conflict betweenhim and his colleague,Octavius. Defeatedin the city, Cinna imitated Sulla in appealing to the army, and with like success.Only a greatcivil war concludedthis phaseof the revolution and enabled Sulla to restore and consolidate the senate's control of the state. So far it is not clear that the urbanproletariate,even thoughit owedcheapgrain to popularleaders,took a strongpart againstthe senate,which in 100 and 87 is said to havehad the supportof the townsmen.The Gracchiand Saturninusrelied chiefly on the rural poor, Cinna and perhapsSulpicius on the new Italian citizens. Sulla, however,severelylimited the powersof tribunesandput an end to corn doles. The latter measuredirectly injured the urban poor, and the former denied them hope of redress for their grievances. In the 70S the prevalenceof piracy began to affect the cornsupply. In 75 the price of grain was cruel, and a mob attackedthe consulsproceedingalong the SacraVia and put them to flight; 57 this riot does not seemto have been 'incited by demagogues'. The senateitself re-institutedcorn doles in 73, but on a miserably limited scale(seeabove,note 5I). Pompeyin 70 forced throughthe restorationof the tribunes' powers; he probably envisagedthat tribunicianlegislationcould beadvantageous to him (asit proved); and his wishes could not be denied, as he had a large and loyal army outsidethe city. Threeyearslater, the tribune Gabiniushad 57

SaIl., Hist. fragments,iii. 45 f. Shipping had been diverted from the corn-trade,and the treasurywas short of money, Oralio Cottae, 6 f. Cf. also Oratio Macri, 19.

93

P. A. BRUNT

a great commandconferredon Pompeyto put down the pirates. Almost all the senatorsopposedthe bill; the mob stormed the senate-house and put them to flight. Tribuneswho tried to interpose their veto were overawed by a threat of deposition. The peoplewould not tolerateany oppositionto a measurethat might endthe scarcity.Pompey'smereappointmentresultedin fact in an immediateand abrupt fall in the price of grain, and within a few weeks he clearedthe seasof pirates.58 His prestigewas such that he could not be debarredfrom another great commandin the east. It could be foreseenthat on his return with a large army he would be potentially masterof the state.This was why the senate had resistedthe proposalin 67 to granthim extraordinarypowers. The yearsfrom 67 to 62 (when Pompeycameback) were full of violenceand threatsof violence.In 63 Catiline rosein armsagainst the governmentwith a band of discontentedpeasants.The urban plebs had at first favoured him, perhapsbecausehis proposalto canceldebtswould have relieved them of somepaymentsof rentarrears.Cicero won them over to the governmentby alleging that Catiline'sfriends in the city intendedto burn it down and deprive them of their miserable shelter and few personal belongings.59 But his executionwithout trial of Catiline's accomplicesviolated the principle on which the humblestRomanrelied for the protection of his own person.Cicero incurred the lasting hatred of the masses.When Clodius hadhim banishedin 58, he erecteda shrine to Liberty on the site of Cicero'stown house;he hadvindicatedthe 60 freedomof citizensagainstarbitraryill-treatmentby magistrates. Early in 62 Marcus Cato greatly extendedthe scaleof distribution of cheap grain. He was the staunchestchampion of the senate'spower. It seemsparadoxicalthat he shouldbe the author of this measure.But the urban masseswere volatile, and it was necessaryto assuagetheir discontents,when Catiline was still in arms and therewas a proposalto bring Pompeyback to deal with the crisis.61 Cic., de imperio Cn. Pompeii, 31-5; 44, cf. Rice Holmes, i, pp. 167 fr. Cic., in Cat., iii. 15; 21; iv. 17; Sail., Cat., 48.2. 60 C. Wirszubski, Libertas as a Political Idea at Rome(Cambridge,1950), pp. 24-7; 55-6I ; Rice Holmes, RomanRepublic, i, pp. 82 f. Cicero's unpopularity, ad Alt., i. 16. I I ; ii. 3.4; viii. I I D. 7; Ascon., 37 (Oxford text); in Phil., vii. 4 he refers to himself as a well-known adversaryof the multitude. Temple of Liberty, Plut., Cic., 33. 61 Plut., Cato Minor, 26-9.

58

5'

94

THE ROMAN MOB

The fears entertainedof Pompey proved unjustified. On his return he disbandedhis army. But he needed to reward his veteranswith land-allotments.Senatorialobstructionthrew him into alliance with Caesar,who had consistentlyidentified himself with popularaims,and asconsulin 59 Caesarcarriedagrarianlaws by the help of the strongarmsof Pompey'sveterans.In return, he receivedthe greatcommandin Gaul. To checksenatorialreaction onceCaesarhad left for his province,Pompeyand Caesarpromoted the election to the tribunate of Publius Clodius, and it was Clodiuswho finally madethe grain distributionsfree. This was the prime source of the enormouspopularity he enjoyed with the plebs so long as he lived. Another measure,to be considered presently,ensuredthat he, unlike previousdemagogues,remained powerful in the city even when out of office. VI This sketchwill have shown that violence at Rome did not proceedfrom any single sectionof the people.Before Sulla 'popular' leadersdrew supportmainly from citizens who camein from the country to vote and fight in the streets;in 70 it was Pompey's army (recruited in the country) that made the restoration of tribunician power irresistible; in 59 it was again his verterans who forced through Caesar'sbills. On the other hand, in 67 it was the urban plebs which broke the oppositionto Gabinius'law, and in most of the post-Sullanperiod it is their riots that we hear of. But the senatealso, or some of its members,initiated illegal violencefrom time to time, or at leastmet force with force. They could mobilize their clients not only from other partsof Italy (see above,pages83 f.), but within the city itself. The urban plebs was not a united body, and sometimeswe are not told what sectionof it took this or that action.62 In annalisticaccountsof theclass-struggles in theearlyRepublic, which are colouredin detail by the experienceof the secondand first centuries,we hear much of the dependants(clients) of the nobility supporting them againstplebeianleaders.63In 133 the Meier, ResPublica Amissa,pp. I 12 f: is too ready to assumethat where the people is mentioned,the urban plebs is meant. 63 E.g. Livy, ii. 35.4; 56. 3; 64.2; iii. 14.4; 16.4;v. p. 8; 30.4; vi. 18·5; H. 6 f.; Dionysius, vii. 18.2; viii. 71. 3; ix. 41.5; 44. 7 etc. 62

95

P. A. BRUNT

assailantsof Tiberius Gracchusincluded, besidesmembersof the upperclasses,'the plebsuncontaminatedby pernicio~s schemes'.64 The nobility drew supportwithin the city against Saturninusin 100 and Cinna in 87; and it may be that we should think of this coming ratherfrom their own clients than from the urbanmasses in general(though Saturninus'followers were countrymen,and Cinna's new citizens from Italy, and neither is known to have had much urban backing). Cicero's claims that his return in 57 was popular, if true at all, may be so only in the sensethat the dependants of the nobility demonstrated in his favour.65 Tacitus' distinction for A.D. 69 between'the soundsectionof the populace,attached to the great houses'and the 'sordid plebs, habituesof the circus and theatres'may be relevant.66 But perhaps some Republican acclamations of 'anti-popular' figures in the theatresmight be explained by the hypothesis that they were crowdedwith clients,for whomtheirpatronshadprocuredplaces.67 Sallustassertsthat in 63 the whole plebswasat first on Catiline's side against the government,which he explains by saying that invariably men who have nothing are enviousof the 'good'-the term is in practiceindistinguishablefrom 'rich'; 'they hatethe old order and yearn for a new; in detestationof their own lot they work for total change;to them turmoil and riots are a sourcenot of anxiety, but of nourishment;for the destitute cannot easily suffer any loss'. Cicero too more than once saysthat the property and fortunes of the rich were endangeredby Clodius' gangs;and the existenceof class-hatredin Rome can hardly be doubted;it is significant that in 52 the mob killed anyone they met wearing gold rings or fine clothes.68 But it was not felt or evincedby all the poor there; a large numberdependedon the upper classes. Velleius, ii. 3. 2. ad Alt., iv. I. 5. His claim that all collegia supportedhis return cannotbe accepted(de domo, 74); somemust have done so, perhapsthosewith upper-classofficers (for whom seeadQuintumfratrem, ii. 6. 2; ILS, 2676). Cicero also boastsof the popularity of his policy in 43 (Phil., vii. 22; xiv. 16); improbable,as it was likely to result in corn-scarcity(xiv. 5); viii. 8 is significant: 'omnesidem volunt ... cum omnis dico, eos excipio quos nemo civitate dignos putat'. Cf. pro Milone, 3: 'reliqua multitudo, quae quidem civium est, tota nostra est'; Ascon., 32; 37; 40; 42 shows that the masseswere againstMilo. 66 Hist., i. 4. 67 E.g. Cic., ad Alt., ii. 19· 3. 68 Sail., Cat., 37; Cic., Sest.,49; III; de domo, 12 f.; pro Plancio, 86; pro

64 65

THE ROMAN MOB

Sallustthought that the plebs was at a disadvantageagainstthe nobility in that it was less organized.It could do nothing except with leadershipfrom inside the ruling class.69 It was also notoriously volatile, and could be persuadedto desertits leadersby the plausible demagogyof senatorialspokesmen,as in 122 and 63.70 And no popular leader before Clodius sought to organize his supportersin such a way that they would effectively supporthim beyondthe brief period for which he held office. The Twelve Tables,the ancientcodeof Romanlaw, apparently allowed freedom of association,if there was no conflict with public law.71 Many collegia of artisansas such or of personsliving in the samedistrict (vicus) thus arose, some at a very early date. Evidently someof them were implicatedin riots in the 60S, and in 64 the senatedissolvedall 'excepta few namedcorporationsrequiredby the public interest'.At the time Catiline was standingfor the consulship,and it was probably feared that they would exert themselveson his behalf. In 58 a law of Clodius restoredthe right of association,and he himselforganizedcollegia, old and new, on a local basis in para-military units and provided a supply of arms. The proximity of Caesar'sarmy and the backingthe consuls,who also had some soldiers, gave Clodius, madeit impossiblefor the senate to resist; and henceforth Clodius was an independent power in Rome, even when a private individual, thanks to his control of the collegia.72 Only from Cicero do we know anything of the compositionof Clodius' bands. He speaks of slaves, including runaways and thugs whom Clodius had brought himself for the purpose of terrorism,criminals-'assassins freedfrom the jail', which Clodius Milone, 95; cf. Appian, ii..22 with modern parallels in G. Rude, The Crowd in History (New York, 1964), pp. 224 f. 69 fllgllrthine War, 41. 6 as interpretedby]. Hellegouarc'h,Vocablliaire Latin des relations et despartis . .. (Paris, 1963), p. 101; Cic., pro Mllrena, 50. 70 Cf. Livy, vi. 17. 71 Dig., xlvii. 22. 4. 72 Ascon., 7; 45; 59 f.; 75 (Oxford text); Cic., pro Mllrena, 71. If the CommentariolllmPetitionis is by Q. Cicero, or at least well-informed, the suppressionmust be later in 64; cf. sect. 30. Clodius' law: e.g. Ascon., 8; Cic., Sest., 33 f.; 55; de domo, 54; in Pisonem,8-11. Caesar'sarmy, Sest.,40 f.; de domo, 131; cf. E. Meyer, CaesarsMonarchie II. das Principat des Pompeills(Stuttgart, 1922), 3rd edn, p. 94. Consuls' soldiers, de domo, 55; 119·

97

P. A. BRUNT

'emptiedinto the forum'-foreigners;at best they were hirelings (operae, conducti, mercennarii). Clodius was a rich man, and according to Cicero he acquiredillicit funds to distribute; no doubt he could afford to buy or hire armed escorts.Freedmenand indeed slaveswere admitted to collegia in large numbers(as inscriptions show), and such people, foreigners by extraction, naturally formed a substantialelementin his gangs. Wherever slavery is found, there are always runaways,and in the unpoliced purlieus of Rome they could easily lurk. Rome must also have provided armedrobberswith ample opportunities,though it may be noted that in Romanlaw imprisonmentwas not a penalty,and if Clodius freed prisoners, they may have been not only persons merely awaiting trial but also men seizedfor debt.73 Cicero'sdescriptions are, however, suspect;he admits himself that it was a common rhetorical device to vilify all who attendedpolitical meetingsas 'exiles, slaves, madmen',and we know of at least one occasion whenhe choseto speakof freedmenas slaves.74 In his view Clodius was Catiline's heir and enjoyedthe supportof survivorsfrom his movement; we may recall that Catiline had originally had the favour of the whole urban population. Cicero writes of Clodius' followers much as contemporariesof the better classeswrote of the mobs which rioted in Paris in 1789-95 or 1848, or in English towns of the eighteenthand early nineteenthcenturies;they were, it was said,banditti, desperadoes, ragamuffins,convictsand the like. ProfessorRudehasshownthat whereverrecordsexist to check thesedescriptionsthey prove to be largely false. Men with criminal convictionswere never more than a minority among the rioters; mostly they were men of 'fixed abode and settled occupation'; for instance all the 662 'vainqueurs de la Bastille' were small workshop masters and journeymen, retailers, artisans and labourers of all kinds.75 In Seee.g. Sest.,6; 27; 34; 38; 53; 57; 59; 65; 75; 78; 81 f.; 84 f.; 89; 95; 106; 112; 126 f.; de domo, 5-7; 13; 45; 53 f.; 75; 79; 89 f.; 92; 129; in Pisonem,8-11. Clodius' funds, de haruspicumresponsis,28. Runaways, W. Buckland, RomanLaw of Slavery(Cambridge,1908), pp. 257 £1'. Dionysius, iv. 24. 5 atteststhe practiceof liberating thugs; Augustus was to bar such freedmenfrom citizenshipor from living within 100 miles of Rome. For riots ascribedto slaves,freedmenand hirelings before 58, cf. Ascon., 45; 66; Cic., ad Alt., i. 1. 13; 14. 5; ii. 1. 8. 74 Cic., Academica,ii. 144; Ascon., 52, cf. 8. 23. 75 Rude The Crowd in History, pp. 7 £1'.; pp. 195 £1'.

73

THE ROMAN MOB

Rome the Catilinarianstried to raise 'the artisansand slaves',and Cicero lets out that Clodius' following included shopkeepers; when he wished to gather a mob, he had the shops closed, a practicecommonwith seditioustribunes.76 We shouldnot assume on his biasedtestimonythat artisansand shopkeepersneededto be incited or hired on every occasionto give up their day's earnings and risk their lives andlimbs in a demonstration,without real grievancesto demonstrateabout. In 41, when famine was raging, 'thepeopleclosedtheir shopsanddrovethe magistratesfrom their places,thinking that they had no needof magistratesor crafts in a city suffering from want and robbery [by soldiers]'.77 Then at leastthey actedwithout any demagogueto instigateandpay them. makesa carpenterand a cobbler I suspectthat when Shakespeare typical membersof the Romanmob, he was, by intuition, right, andthat Clodiuswould havehadlittle powerover suchpeoplebut that they had complaintsand looked on him as their champion. But even if most of them (freedmenincluded) were artisansand shopkeepers,that would not have endearedthem to Cicero; he hadoncespokenof 'artisans,shopkeepers andall the scumin cities whom it is so easy to excite'. He characterizesthe Clodians as 'destitute'(egentes);but their plight did not evokehis compassion; the word is almosta synonymwith the epithetwhich often accompaniesit-'scoundrels'(perditi). He recognizedthat the plebswas 'wretchedand half-starved',but added at once that it was 'the bloodsuckerof the treasury'.78It was such attitudeson the part of the governingclasswhich gave Clodius his opportunity.

VII Violence was actuatedby many different aims. The clients of the greathousesusedit simply in their patrons'interest,the followers of popularleaderssometimesmerely from loyalty to their leaders. But they were attachedto the 'demagogues'becausethe demagogueswere active for their welfare. Country people, including SaIl., Cat., 50. I; Cic., de domo, 13; 54; S9 f.; Academica,ii. 144; Ascon., 40 f. 77 Appian, v. IS. 78 Pro Fiacco, IS; ad Alt., i. 16. II; 19. 4; ii. 1. S. 'Egentes'and 'perditi', e.g. de domo, 45.

76

99

P. A. BRUNT

the veterans,usually sought land distributions. The burden of rent, indignation at arbitrary punishments,proposalsto redistribute freedmen among all the tribes could sometimesraise an urban mob. But in 75, 67, in the heyday of Clodius' ascendancy and again in 41 and 39 hunger seems to have been the chief motive force.79 When Cicero was banished,there was a scarcity; his sarcasm that the bands who pulled down his house were not going to satisfy their appetite on tiles and cementimplies that they were hungry.so Clodius' grain law may have increasedthe effective demand,which certainly outranthe supply. In July 57 therewas a food riot. A few days later, when the senatevoted for Cicero's restoration,the price of grain providentially sank. It was but a temporaryimprovement.For days togetherthe senatedebatedthe corn supply. Cicero gave three possible explanations for the shortage:exportingprovinceshad no surplus,or they sentit elsewhere to get higher prices, or the suppliersheld grain in store in the expectationof famine rates.On the 5th Septemberhe boasted that plenty had returned with him. This was an illusion. Prices continuedto oscillate(a familiar phenomenonin many ages).On the next two days they went sky-high, and the mob rose; Cicero acknowledgedthat therewas suffering and hunger.He and others did not venture to the senate-house.But a day or two later he risked attendance;the streetswere evidently quiet again. If the rioters had beenmerely Clodius' hirelings, out for Cicero'sblood, this would be strange;if they were exasperatedartisansand shopkeepers,with work to do, they could not be kept in the streets continuously. On Cicero's motion Pompey was now invested with the procurementof grain and given wide powers, probably enabling him to requisition grain from recalcitrant suppliers. Plutarchthoughthe securedabundanceas by magic, but soon all was not well again, and now the blamecould be laid on Pompey. The locust plague in Africa in 125 (Orosius, v. 11. 2) may also have paved the way for Gaius Gracchus;for hunger in the Gracchanperiod: cf. Lucilius, fragment 214 (Loeb edn). In the 40S too the rent-burden must have been the greater,as Africa was under Pompeiancontrol and grain must have been scarceand dearer,leaving less money to pay the rent. 80 De domo, 61. For what follows see de domo, 9-17 with Ascon., 48; Cic., post reditum in senatu, 34; adQuirites, 18; ad Att., iv. 1. 6; adQuintum jratrem, ii. 5. I; de haruspicumresponsis,3I. 79

100

THE ROMAN MOB

Hence, in the scenewith which I opened,the mob shoutedthat Pompey was starving them. In April 56 there were renewed debateson the high price of grain, and Pompeywas voted more money. In August Cicero deploredhigh costs, the infertility of the fields, the poorharvest.Persistentscarcitywas the background to continual violence. Rude has shown that in eighteenth-and nineteenth-century Franceand Englandriots were often(not always)provokedby, or associatedwith scarcities,whatevertheir avowedaims. In October 1789, when the Paris mob went to bring the royal family from Versailles,they said that they would fetch 'the baker, the baker's wife and the little baker boy'; they thought that there would be plenty with the king in their midst.81 If we haddatafor thefluctuating grain prices of Rome, it might well be that we could plot a correlation with the outbreaksof mob violence. But this must remain a speculation. VIII If we look beyond the ambitions and machinationsof the great figures of the late Republic, the main causeof its fall must in my view be found in agrarian discontents;it was the soldiers, who were of peasantorigin, whose disloyalty to the Republic was fatal. The role of the urban mob was more restricted.Still, it was their clamourthat gavePompeyhis extraordinarycommandin 67 and set in motion the eventsthat led to his alliancewith Caesarin 59. And the violence in the city from 58 to 52, which was itself one result of that alliance, produced such chaos that it finally brought Pompey and the senatorialleaderstogether again, and helpedto severhis connectionwith Caesar;hencethe civil wars in which the Republic foundered. Popular leaders sometimesproclaimed the sovereigntyof the people. But the people who could actually attend meetings at Romewere not truly representativeand wereincapableof governing an empire. The only workablealternativeto the government of the few was the governmentof one man. The interventions of the peoplein affairs led on to monarchy.82 81

82

Rude, The Crowd in History, ch. 14. Wirszubski, Liberlas, pp. 47 If. Cf. Sallust,Jugurlhine War, 31: 'sane fuerit regni paratia plebi sua restituere'. 101

P. A. BRUNT

To the urban proletariatethis was no disadvantage.It was the aristocracywho suffered from loss of liberty. Tacitus says that Augustus won over the people with bread, and this was the greatestneed.83They also benefited from improvementsin the supply of water, from betterfire-protection,betterpreservationof water,more splendidshows,moreexpenditureon buildingswhich gave employment.The emperorsfor their own security had to keepthem content,and their misery was somewhatreduced.This was all they could expect in a world whose material resources remainedsmall.84 83 84

kn., i. 5; cf. Augustus'Res Gestae,5. I have made no changein the text in this article. Since writing it, I have sought to justify the statementson population (n. 4 and pp. 80-1) in my Italian Manpower (Oxford, 1971), Part I and chapterxxi. The statementon p. 80 that only adult males were eligible for grain rations is wrong; cf. Italian Manpower, p. 382, though it would perhapsbe best to amend Suet., Aug. 41, to mean that eligibility begannot at the age of ten but at that of fourteen, the conventionalage for male puberty. On free and slave labour, J. A. Crook, Law and Life of Rome (London, 1967), pp. 191 ff. supplies further information, and G. E. Rickman, Roman Granaries and Store Buildings (Cambridge, 1971) shows the great number of porters required (pp. 8-11,79,86);on my view they would have been mostly free. In my Social Conflicts in the RomanRepublic (London, 1971) further examplescan be found of the importanceof public works (see the index s.v.) in the activity of popular leaders. Italian Manpower also contains material promisedin n. 54 on the continuanceof grain cultivation (index s.v.) in Italy. On p. 80 or 92 I should have made it plain that Octavius' veto of a tribunician bill was as bizarre constitutionally as Tiberius Gracchus'riposte. The text of Cicero cited in n. 32 (Alt. xiv 9, I) may refer to slum property at Puteoli, but conditions will have been no better at Rome.

102

V

ELITE MOBILITY IN THE ROMAN EMPIRE*t Keith Hopkins

I THE SYSTEM OF STRATIFICATION

Rome was throughoutthe empirean 'estate'society,thatis, there were legal distinctions of status. The aristocratic elite was the senate,which in the first threecenturiesA.D. hadaboutsix hundred members. There was a general expectationthat the sons of a senatorwould also be senators-butthere was also the proviso that eachmembershouldhavepropertyworth at leastone million sestercesand be electedto office. In principle electionwas by the senate,in effect often by the emperor,who also appointedhis own nominees.The secondestatewas the equestrianorder (equites), whose minimum property requirement was 400,000 sesterces. The formal method of entry was by imperial grant. A third privileged group consistedof the decurions,in general the top one hundredmen of eachcity. Their wealth varied immensely,as did the size of cities. Below them came the urban plebs and the rural peasantry.Juridically at the bottom camethe ex-slavesand the slavesthemselves. The senatorsand equestrianseach wore distinctive dress,and betweenthem filled most important and honorific governmental posts. It seemsespecially significant that both orders depended '" From no. 32 (1965). shouldlike to thank Dr M. 1. Finley, ProfessorA. H. M. Jonesand Dr A. N. Little for their help.

tI

KEITH HOPKINS

upona minimum propertyholding and election.In the aristocracy theserequirementswerereinforcedby the competitivedemandsof ostentatiousexpenditure.The major sourceof incomewas land; a lot of nonsensehas beenwritten about the equitesas a merchant class.Fortunesmay sometimeshavehad their origin in trade,but were soon convertedinto land which was safer than trade and gave higher prestige. ProfessorA. H. M. Joneshas shown for the later empire that mercantile fortunes were small compared with thoseof noble landowners.l The systemof formal stratificationwas supplemented,evencut across, by other groupings. In the first century A.D., Roman citizens had important social privileges denied to non-citizens, thoughwithin both groups,especiallyamongcitizens,therewere greatdifferences,for examplein wealth. In 2 I 2 as the climax of a long processof assimilation, Roman citizenship was granted to virtually all provincials. But already, and perhapsin its place, a new division of society, more in alignment with the formal stratification, had become established. This distinguished the humiliores-or lower class-from the honestiores,who comprised the top threeprivilegedgroups,senators,equitesanddecurionsand all legionaries of whatever rank-a symptom of the army's political power. The processof political unification which marksthe empirehad great consequences for social mobility. The Roman empire had beenwon by Romansand their Italian allies who initially monopolized all high status positions in the central government. Correspondinglywealthy provincialsin the early period had been 'content' with local or provincial honours. As the government becamemore centralizedthe spheresof its control broadened, leading provincials became Roman citizens and competed for traditional Roman honours. By the end of the first century, for example,there was a SpanishRoman as emperor,by the end of the secondan African, by the beginning of the third a Syrian. The crisesof continuousbarbarianinvasionsin the third century investedthe armies and their professionalofficers with supreme power. They brokethe aristocraticsenate'smonopolyof privilege, and in fact excluded senatorsfrom all important or powerful positions. Under Diocletian (284-305) and Constantine(306-337), the 1

A. H. M. Jones,The Later RomanEmpire (Oxford, 1964), pp. 870-1.

ELITE MOBILITY

IN THE ROMAN EMPIRE

army was doubled in size, additional taxes were raised, bureaucratic control much tightened. Under Constantine,aristocrats were readmittedto high positions, but the equestrianand senatorial orders were fused. The new expandedorder (clarissimi), based upon the same admixture of hereditary expectationand actualtenureof office or imperial grant, was the highestestatefor the whole empire.As taxesandbureaucraticinterferenceincreased, decurionscould less and less opt out of the single statussystem; they tried to becomesenators-clarissimi.It is arguable,therefore, that the centralization of the administration and the political unification of the empirepromotedmobility in the fourth century to an unprecedenteddegree. The spate of fourth-century laws againstsocial mobility has traditionally beenregardedas evidence of immobility.2 But the laws were only spasmodicallyenforced; their repetition is evidenceof their failure.3

II GENERAL PROPOSITIONS

I have presumptuouslycompressedthe developmentsof four centuriesinto two pages.The historian may rightly object that in doing so I have neglectednot only numerousexceptionsbut also importantareasof differentiationwithin the estates.He might also feel uncomfortablewith the transition from a formal analysisof stratificationwhich coincideswith the Romans'own terminology and institutions(senate,equites,etc.), to an historicalinterpretation of the pressureswhich producedchangesin the formal system.All these objections seem to me to be valid and germane;germane becauseit is exactly in theseareasof objection that we shall find social mobility concentrated. This situation arises becausethere are in any complex society several criteria of status; for example: wealth, birth, formal education,learnt skill, ability, achievement,styleof life. Theformal and dominant estatesystem of stratification assumestheir conCf. for example,O. Seeck,Geschichtedes Untergangsder antiken Welt (Berlin, 1901), ii, p. 301. 3 A random collection of casesof upward mobility may be found in R. MacMullen, 'Social mobility and the Theodosiancode', fRS, liv (1964), pp. 49-53·

2

105

KEITH HOPKINS

gruence,that is that people who rate highly on one criterion or dimension will rate highly on others. There is a general and strong belief in the legitimacy of the hierarchyand a generaland strong expectationthat status will be inherited. Various institutions allow the aristocratic child, for example, privileged access to situationsin which he can distinguish himself and succeedhis father. Social mobility, whether upward or downward, by its very nature confounds these expectations of inherited status and straddlesthe formal systemof stratification.It canbe usefully seen as a processof statusdiscrepancyor dissonance,that is a situation in which people rate highly on one or more dimensions,but not on others. The upwardly mobile man, for example, rating relatively low on birth may acquirea good formal education,a highly valued professionalskill, and be successfulin his profession.He may still be relatively poor and not on equal terms of social friendship with born nobles.In time, he may acquirethis too, and this may especiallybe the caseif we look at mobility as a function of family, rather than of individuals. A nouveauriche father may have a more socially acceptableson. This processoccurredin Rome as well as in modern Britain, but thereis, I think, one major difference.In a complex industrial societylike Britain, occupationalgroupsare in someways insulated from one another; each has its own specific criteria of excellence;a good historianin Britain is contentby and large to be a professor: he rejects birth or income as an overall criterion by which his standing relative to other groups should be judged. Rome, on the otherhand,was dominatedby its own overall estate system of stratification; it really did matter whether one was a senatoror not. A man'srank (e.g. senator,eques,servtls)was noted in the ancient sourcesin the context in which a modern writer would name his occupation.But even if this formal systemwas dominant,it was not the only sourceof status.Thereare numerous examplesof senatorsor equestrianswho ratedhigh on somedimensions of status,but not on others. This is exactly the point; the more differentiateda societybecomes(that is, the more professions and occupationsbecomeinsulated),the more mobility is possible; and the morepossibleit becomesto acceptor assimilatethesocially mobile-menor womenwho ratehigh on somestatusdimensions, but low on others. 106

ELITE MOBILITY IN THE ROMAN EMPIRE

My first two generalpropositionsare thus that:

(I)

social mobility may usefully be seenas a processof status dissonance-especially in a societywith a strongideology of hereditarystatus; (2) social mobility was a productof the structuraldifferentiation of institutions(for example,bureaucracy,army, law). The third generalpropositionrelatesto the processesat work in the political unification of the empire. Put baldly, a major element of this process was the tension or conflict primarily betweenthe emperorand the aristocracyand secondarilybetween eachof thosetwo andthe structurallydifferentiatedinstitutions.In spite of the massiveevidenceof Tacitus, Suetonius,Dio Cassius and the ScriptoresHistoriae Augustae,there is a tendencyamong modern historians to minimize this conflict, and of courseit is difficult or impossibleto prove its importance.It was certainly not the only factor. But equally, leading ancientwriters thought the conflict important,and as I shall arguecertainpatternsof mobility can be most easily understoodin terms of it. Lastly I shall take a brief look at elementsof the structureof property and inheritance,and at high aristocratic consumption and low fertility: the combined effect of these was to reduce tension causedby upward mobility. For by downwardmobility, and biological non-replacement, the aristocracy left vacant establishedand highly valued administrativepositions. Thus I also proposethat: (3) certain patternsof mobility may be best understoodas the productof conflict betweenthe emperorand the aristocracy, and betweeneachof thesetwo and the structurally differentiatedinstitutions; (4) resistanceto upward mobility was diminishedby the downward mobility of the aristocracy,and their biological nonreplacement. In the remainderof this paperI shall expandand illustrate these propositions.

KEITH HOPKINS

III STRUCTURAL DIFFERENTIATION AND STATUS DISSONANCE

In early Rome, say before 300 B.C., the chief magistrateswere nobles in charge of military, civil, religious, judicial, legislative and executive actions. The aristocracy was the locus of most virtue. Its pride was its noble birth and ancestors;its expectations were hereditary. It had a near monopoly of power. It was protected againstoutsidersby institutions, but where thesefailed it could always rely upon a pervasivebelief in superior 'nature' or 'blood'. By what other belief can an aristocracyset itself apart, when the technical thresholdsof an aristocrat'sduties are low? The final defenceis the defenceagainstacceptance;the arriviste is pinned by the hostile stereotypeof the nouveauriche. As Romeexpandedto a hugeempirewith a populationof some fifty millions (first century A.D.), the complexity of government and its centralization under the emperors brought about the developmentand separationof, for example, military, bureaucratic, legal, educationaland economicinstitutions. One can trace the growth of a professionalarmy, the separationof military from civil bureaucraticcareers,the creation of a hierarchy of bureaucratic ranks, the codification of laws, the creationof law schools, the establishmentof publicly supportededucationand of capitalistic market production. The differentiation of thesevarious activities creatednew professions,new careers,new criteria of status.Each professionhad its own criteria of excellence,its own rules of entry and advancement. The aristocracycould not monopolizeall valued skills or fill all important executivepositions. To do so they would have had to sacrificetheir valuedleisureand metropolitanpleasures.At the sametime if the aristocracywas not to becomea placid backwaterreflecting only a glorious pastit had to keepits finger on the pulse of all the new elementsof the complex society. The aristocracy was alreadyacknowledgedas the 'betterpart of the human race','the flower of the whole world'.4 It was alsoideally and often in fact the most wealthy, the most literary elementin society and 4

Symmachus,Epist., 1. 52; Paneg. Lat., iv. 35.

108

2.

ELITE MOBILITY

IN THE ROMAN EMPIRE

held the highest offices and the highest military commands.Becauseof this, it survived as the supremestratum; but becauseof this also it openedits own ranks to arrivistes. So long as an aristocracydependsupon birth aloneit can remainexclusive;when it admits complementarycriteria of achievement,whethermoneyor professionalskill, it opensthe way to arrivistes. If aristocratswant to be literateurs,literateurshave a credit which helps to disguise them as aristocrats.If aristocratswant to be generals,generals who are not aristocratshave a fulcrum by which they can lever themselvesinto acceptability.This is not to say that allliterateurs or military leaderswere equally acceptableor acceptedas equals by aristocrats. Far from it. But at least we can see that social mobility is a processof gradualacquisitionof statuson a variety of fronts; and the processis made much easier if the aristocrats themselvestry to acquirestatuson a numberof fronts too. Yet the patternsof mobility, that is the patternsof acquisition of highly valued marks of status, were by no means random. There were a few channelsof mobility which seemto havebeen mostoften used.I shall discusstwo briefly: the military andliterary spheres.Traditionally the aristocracyhad seenitself as embattled heroescompetingfor the conqueror'sglory, of coursefor Rome's benefit. Its culmination was the award of a Triumph. When the governmentcame under the control of a single emperor, such pursuits of military glory through the unified commandof an army presenteda threat to the emperor'sposition. Victory, as Tacitus noted, was a virtue suited to an emperor.Without hope of a Triumph a military careerhad only tradition to recommendit to the senator;it was part of his normal careerto hold a coupleof short-termcommandsof military units. The armies were, moreover, stationedon the frontiers, and it was hardly worth being an aristocratpermanentlystationedat a distancefrom Rome.S The long-termprofessionalofficers, formally subordinatesto senatorial commanders,were never aristocrats. When the barbariansinvaded continuouslyin the third century it seemseither that the necessityand technicaldemandsof winning excludedamateurish aristocrats,or more likely that the professionalgroups of underofficers were in a position to reject their amateurisharistocratic commanders.Certainly they did; for from the middle of the third 5

E. Birley, 'Senatorsin the Emperors'service', PBA, xxxix (I95 3), pp. 207-8.

KEITH HOPKINS

century, aristocratswere excluded first from military and then from other important governmentposts. Literary skill was part and parcelof aristocraticdaily life; it was used as a secondaryreinforcementof aristocraticnature. It was a mark of noblepolish to write skilfully and allusively. Thosewho taught nobles,especiallythosewho helpedadult noblesmaintain their skills, could with practicepassthemselvesoff as aristocrats. Libanius, for example, a teacher of rhetoric in Antioch in the fourth century, was madean honorarypraetorianprefect, a very high position; and there are several similar cases. A study of thirty-four teachersat Bordeauxin the fourth century showsthat at leasttwelve were upwardlymobile, that is becameprofessorsin the capitalsConstantinopleor Rome, marriedrich or noble wives or held provincial governorships.6While literateurs were more attractiveto aristocratsthan skilful but boorishgenerals,they also had less power to force their way into the highestranks. Both literateursand military officers were subjectedto repeated public tests of competence,the one in rhetorical exhibitions, the otherin battle. It is not that aristocratswereper se incompetentin thesefields-severalwere clearly mastersof literary style (Tacitus, Gellius). But mostly that they hadno ethosof gainingprofessional competence.If one is superior by nature, one is good without seemingto try. The aristocraticethospervadedthe Romanelite. Even the now differentiated professions(for example, military, legal, literary) owed their high statuspartly to the fact that in the old days their best or only practitionerswere aristocrats.But on occasionprofessionalismoutweighedthe aristocraticethos. This happenedin the military and bureaucraticspheresin the third century. Yet although one can point to the exclusion of the aristocracyfrom military and from bureaucraticoffice at that time and call it a victory for professionalism,the end of the story points another moral. For Constantinere-engagedaristocratsin high positions and like his predecessorsrewarded equites (non-aristocrats)in terms of the traditional honour system. The legitimation of the emperorwas too insecure,evenwhenhe was supremegeneral,for him to be able to ignore the social power of aristocratsor the traditional systemof honour. Therewas no possibility of revolu6

M. K. Hopkins, 'Social mobility in the later Romanempire: the evidenceof Ausonius',CQ, xi (1961), pp. 246-7. lIO

ELITE MOBILITY

IN THE ROMAN EMPIRE

tion basedupon a professionalbourgeoisie.'Such bourgeoisas there were, were too imbued with the sense of aristocratic superiorityand wantedto be rewardedby being madearistocrats. It is not surprising then that the military government of the latter part of the third century capitulated to the aristocratic system, even if it compromisedpartly by employing aristocrats and partly by ennobling generals.The whole society was pervaded by standards which emphasizedpersonal attachments rather than objective professional achievements.The conflict betweenthem can be seen in the regulations of the praetorian prefect'scourt, for example,which insistedupon legal knowledge as a prerequisitefor enrolment,but also upon priority and privilege for sonsoflawyers.7 Outburstsof professionalismwere the exception;the rule was a victory of the aristocratic ethos, in one of two ways. Firstly aristocrats,though amateurs,headedor directedthe professional institutions. They held most high government offices, commanded armies, wrote belles lellres, history and philosophy. Secondly they absorbedthe aspiring professionals.We know very little of the ways in which aristocrats and professionals mixed; but almostcertainlyit wasnot as equals.What survivesis a cruel stereotypeof the social riser, and the undercurrentof an ideology that ability shouldbe rewarded.As far as formal rank is concerned, generals, literateurs, lawyers and financiers were promotedto the senate.

IV SOCIAL MOBILITY

AS A FUNCTION OF POLITICAL CONFLICT

Just before his death, Augustus, the first emperor, discussed possiblecontendersfor the throne. 'All those mentioned,except one, were soon struck down on one charge or another, at the instigation of Tiberius [his successor]'. Domitian, emperor A.D. 81-96, invited leading senatorsto dinner-theycameto find eachplacedecoratedwith a tombstoneengravedwith the guest's name,the hall deckedout as a tomb; the servantswereblack boys, 7

ClassicalJournal, ii. 7.

II

(460). III

KEITH HOPKINS

and the food was that traditionally given to the dead. Only the emperor'svoice was to be heard and the guestswere in fear for their lives.s Admittedly a macabrerarity, but examplesof hostility between the emperor and the senatorial aristocracy are innumerable.Many emperorson their accessionswore not to kill any senators,or not without proper trial. Few, if any, kept their promise. Historians,ancientand modern,havegenerallyconcentratedon aspectsof behaviourwhich differentiateemperorsfrom eachother, and have tried to explain the assassinationor persecution of aristocrats in terms of the emperors' psychopathology.Nero, Domitian, Commodusand Elagabalusgive considerablejustification to such an approach. Nevertheless,at a different level of analysis,it seemsappropriateto seeall the emperorsas autocratic, as necessarilyengagedwith the aristocracyin a strugglefor power. Partly aristocrats'fear of disgraceor assassination promptedtheir rebellion, or fear of rebellion promptedthe emperorto assassinate aristocratsor confiscatetheir property.The sameconflict is visible in other spheres.For example,it was to the emperor'sadvantage to limit the aristocratic governors' rapacious exploitation of provincials. This can be used as a paradigm. By tradition, the emperorusedaristocratsas governors,but in his own interestand that of the provincials, he limited their power. The techniquesand institutions which developedin responseto this conflict involved social mobility. Firstly, since it was aristocraticexerciseof powerwhich threatenedtheemperor'ssupremacy, the emperorcould and did employ non-aristocratsin positionsof power. Secondly, emperors helped in the development of differentiated institutions. These gave rise to interest groups, which limited aristocratic power (for example, the professional under-officersin the army). In the same way, the differentiated institutions dependedupon bureaucraticrules, for example, or legal institutions or norms (for example, a belief that laws were valid and shouldbe obeyed),which also limited aristocraticpower. These differentiatedinstitutions, as we have seen, also provided channelsof upward mobility. Thirdly, emperorswere interested in the uniform or maximum exploitation of the empire-hence census-taking,for example, and bureaucratic control of taxraising. To the emperorsthe privileges of Romans or Italians, 8

Tac., Ann., i. 13; Dio, lxvii. 9. lIZ

ELITE MOBILITY

IN THE ROMAN EMPIRE

which were the original basis of the empire, becameof less importancethan its political and administrativeunification. Because of thesecentralizingtendencies,accessto the bureaucraticmachine rather than birth-status by itself became the major source of power. One can see this process in the expansion of Roman citizenship, in the reduction of Italy to the statusof an ordinary province, and in the creation of an aristoracywhose rank rested formally uponits holding bureaucraticoffice. It involved mobility in the assimilationof provincials into the Romanhonour system. In the next three sectionsI shall illustrate theseprocesses.

The employmentof non-aristocratsin positionsofpower In the first century A.D. some of the highestadministrativepositions of the empireand someof the largestpersonalfortunes ever amassedin Rome were in the hands of ex-slaves,the notorious imperial freedmen. In the later empire, in the fourth and fifth centuries,court eunuchs,who wereex-slavesandbarbarians,were acknowledgedto be at the centreof power.9 The chief eunuch,ex officio, rankedafter the prefectsand the very highestgenerals.If we consider freedmenand eunuchsas groups rather than as individuals, then the continuity is too great to be explicable only in terms of emperors'characters.Powerful eunuchs,for instance, were loathedby aristocrats,often sacrificedto their baying, exiled, even burnt-only to be succeededby other eunuchs. Imperial freedmenand court eunuchsprovide dramatic examples of upward mobility. Their elevationis most easily explicable in terms of the conflict betweenthe emperorand the aristocracy; for the aristocracywas angeredby their power, yet the emperors, who could hardly be unawareof this hatred,repeatedlyusedthem. The conventionalexplanationthat freedmentraditionally did such jobs in large Romanhouseholdsis hardly a sufficient explanation; it certainly does not explain the transition from freedmen to equites. From aboutthe beginningof the secondcentury,imperialfreedmen in the palaceadministrationwere supervisedby equiteswho 9

A. M. Duff, Freedmenin the Early RomanEmpire (Cambridge,1958); M. K. Hopkins 'Eunuchsin politics in the later Roman empire', PCPS, clxxxix (1963), pp. 62 ff.

KEITH HOPKINS

also held the chief positionsin the imperial fiscal service.Equites were also exclusivelyin chargeof the highly centralizedgovernment of Egypt. Indeed, no senator was ever allowed to visit Egypt without specialpermission.The crack troops stationedin or nearthe court were commandedby equitesalso, and this office, the praetorianprefecture,developedfrom a military commandto a post almost of vice-emperor,with wide administrative,judicial and supervisorypowers.But beforethe fourth century,when the equestrianand senatorialorderswere amalgamated,this post was never held by a senator;even in the fourth century, praetorian prefectswere quite often men of humble origin. Emperorsused men of lowly origin in key positions of the administrationbecausethey were not identified with aristocratic interests,becausetheir mobility madethemmoredependentupon, evengrateful to, the emperors,andbecausethey might not be too easily assimilatedto the aristocracy.In this respectit is easy to understandwhy eunuchs managedto occupy such a tactically important position amongpalaceservants.For others, like most emperorstoo, were influenced by the existing honours system. They saw rewardsin termsof ennoblement.Freedmenwere made praetors,equestrianprefectswere given the insignia of consuls.1o Formally, even if not in their associations,they had become aristocrats.Their sonsmight not be of the sameusefulnessto the emperorsas their arriviste fathers. It was not only individuals who were assimilated to the aristocracy.The very channelsof mobility went throughthe same process; the climbers climbed and drew up their ladders after them. For example, the imperial cabinet secretariesin the early fourth century were shorthandwriters, who were, as their skill dictated,of humbleorigins. Their accessto statesecretsprompted the emperorsto usethemon privateandimportantmissions.They won prestige,and as a group gained privileges. Their job now becameboth possiblefor and attractiveto aristocrats.By the end of the fourth centuryit was an aristocraticsinecurerequiring no skill.ll Somethingof this sameprocess,the upgrading of a job becauseit gaveaccessto power,occurredwith the displacementof freedmenby equites. 10

11

Cf. A. Stein, Der romischeRitterstand(Munich, 1927), pp. 245 ff. Jones,op. cit., pp. 572-4.

114

ELITE MOBILITY

IN THE ROMAN EMPIRE

Differentiatedinstitutions It is, of course,very difficult to investigatethe consciouspolicy of the emperors.They certainly aided the developmentof differentiated institutions,for exampleby establishingprofessorialchairs with statefunds or by employinglawyersas legislativeor judicial advisers.But it would be rashto seethis as exclusivelydictatedby a desireto control the aristocracyratherthanby the complexneeds of administeringa large empire. We have already seenhow the differentiatedinstitutions providednew criteria of statusand new careers.What concernsus here.is rather to point out that laws and bureaucraticrules, as well as courts and bureaucraticpractices(for exampletax-gathering)also servedto control aristocrats. But besides providing an extra element of power which the emperorcould use in his battle againstaristocrats,theseinstitutions providedsourcesof powerwhich limited him, and which he soughtto control. Emperorsmadelaws, but they could not break them too often without robbing the legal systemof some of its power. They had to raise most money by taxation rather than by confiscation.They were caughtby the systemthey had helpedto create.One exampleof this can be seenin their ambivalenceover promotion within the bureaucracy.They might have liked to promote by merit but this would have openedthe way to subjective(that is nepotisticor patronal)and thereforeuncontrollable estimates of talent by subordinates.By and large, therefore, emperorsfavoured seniority as the principle of promotion; they met the constantpressuresfor specialtreatmentby the cheapening of privilege and by the creation of extra places. Positions were and even sold.12 There was then pre-emptedby supernumeraries a consistentpressuretowards hereditaryprivilege, stability, even inertia. Spasmodically the system was attacked by energetic emperors, or beaten by the whimsical promotions of their favourites. But it is wrong to say that the system was beaten; both the tendency to inertia and the spasmodicwhimsy were themselveselementsof the system. 12

Ibid., pp. 602-6, especiallyn. 94.

II5

KEITH HOPKINS

Thepolitical unification of the empire One of the primary forces which brought about the political unification of the empire was the increasingbureaucraticinterference in previously autonomouscities. It was borne in upon municipalleadersthat their chancesof resistingthe exploitationof the central bureaucracy,and their social standingwith its representatives,dependedupon their own place in the Romanhonour system.It was no longer enoughto be a leading memberof the local community, with nothing but local standing. Certainly leading provincials were gradually assimilatedto the Romanaristocracy.In accordwith prevailingculturalvaluesit was easierfor Latinizedthanfor Hellenizedprovincialsto be accepted. The figures for the known Roman senatorsare striking: by the end of the first century A.D. (in the time of Domitian), 23 % were of provincial origin; by the middle of the secondcentury A.D. (in the time of Antoninus Pius), the percentagehad risen to 42%; by the third centuryit was 56%.13Thesefigures give a very rough idea of the degreeto which the originally Italian aristocracywas not in fact hereditary. In the fourth century, rates of mobility seem to have increased.For as bureaucraticcontrol reachedan unprecedented intensity, the positive disadvantagesof remaining only a municipal decurion(the old third estate)becameapparent. The decurions pushed upwards into the expandedfirst estate (the clarissimate)or into the expandedbureaucracy. Only occasionallycan one see traces of the conflict between aristocracyand emperorin this type of mobility. The elevationof Gauls by the emperor Claudius (41-54), and of Pannoniansby ValentinianI (264-76) obviously angeredthe entrenchednobility, and was obviously intended as a breach of vested interests.1 4 But mostly the assimilation of the provincial leadersproceeded gradually and without fuss. This assimilation forms one of the mostimportantpatternsof elite mobility in Rome,yet it is significantly different from what is generally understoodby mobility. M. Hammond,'Compositionof the SenateA.D. 68-235',IRS, xlvii (1957), p. n 14 Tac., Ann., ii. 23; ClL, xii, 1668. A. AlfOldi, A Conf/iet of Ideas in the Later RomanEmpire. The C/asb bet)veenthe Senateand Va/entinian I (Oxford, 1952), pp. 13 ff. 13

ELITE MOBILITY

IN THE ROMAN EMPIRE

For these men were mostly 'of good family' and high standing within a local or provincial statussystem;they did not so much increasetheir fortunes as change their point of referencefrom their locality to the empire. The small rangeof their socialmovementmust havedonemuch to removearistocraticobstructions.The more dramaticexamples of mobility catch the eye and are of great political significance: equestrianimperial freedmenin the first century,soldier-emperors in the third century, praetorianprefects,bureaucraticinnovators like John the Cappadocian,are notable examples.But the predominant form of mobility is a priori likely to have been the movementof a wealthy equestrianor provincial dignitary to the lesseraristocracy,and thenceperhapsto higher things. Most often it must havebeenthe slow climb of a family over generations,not the rocketing climb of individuals. Certainly this gradualmobility existedunsupportedby any majorideologyof advancement.There are occasionalstatementswhich praise ability rather than birth, but the dominant ideology praised aristocratic birth, blood and natureaboveall else. The rapid climber gainedformal status,but was wardedoff from assimilationby the stereotypeof the nouveau riche.

v DEMOGRAPHIC FACTORS

I havetried to show that therewas a steadyand significant stream of mobility into the Romanaristocracy.This resultedpartly in its expansion;in part its effects were muted by the low fertility of Romanaristocrats.15 Augustusoffered rewardsto aristocratswith three children, and later emperorsre-enactedhis laws. There is plenty of evidencethat aristocratsregardedthree children as an upper limit or as an unreachabletarget. This low figure was achieved mostly by abortion, much less by contraceptionand perhapsby infanticide. Given the high mortality then prevalent, three children born was far below the level of replacement. 15

The detailed evidenceand argumentof this section is taken from M. K. Hopkins, 'The later Roman aristocracy:a demographicprofile' (unpublishedfellowship dissertation,King's College, Cambridge,1963).

II7

KEITH HOPKINS

Therefore as aristocratic families died out, posts once held by them were availableto arrivistes. Neverthelesssuccessionin the male line was highly valued, so that low aristocraticfertility needssomeexplanation.It might be tempting to point to the inheritancesystem;for it was customary in Rome to split propertymore or lessequally amongall children irrespective of sex. More than two children surviving their parents' death would therefore result in a diminution of the children's wealth. But Lorimer has suggested that neither primogeniturenor partible inheritanceper se have any definite effect upon levels of fertility. 16 This doesnot meanto say that the structureof inheritanceis irrelevant, but rather that it cannotby itself explain low fertility. Severalother influential factorsare worth mentioning, though there is no certain way of assessingtheir relative importance. Firstly, Roman aristocratswere engagedin a bitter struggle of ostentatiousexpenditure.For somethis resultedin their downfall; there are several references to impoverished patricians. The ostentationand wealth of others arousedthe jealousy or fear of emperors.For yet others the demandsof expenditureinvolved a limit on children,becausechildrenwereexpensive.Daughtershad to be given dowries commensuratewith their father's status. Spinsterhoodwas almostunknownbeforeChristiantimes (fourth century). Sonshad to be launchedinto public life, and the games celebratingtheir first high office sometimescost the whole of an aristocrat's annual cashincomeor eventwice that.17 Secondly,the socially approvedways in which an aristocratcould get money were restrictedto inheritance,dowry and governmentoffice. This was not an iron rule, but it was general practice. The typical tenureof governmentoffice was short. This was in the emperors' interest, becausethey wanted to restrict aristocraticpower; and it was in accord with the aristocraticideal of otium cum dignitate. Thirdly, aristocraticwomenhad sufficient social power,for example control over their own property and rights of initiating their divorce, to be able to refuse to have children. They lived in a competitivesalon culture, and did not want to spoil their figures with large numbersof children. Lastly, we, with the benefit of modern statistics and statistical concepts, can talk of rates of 16 17

F. Lorimer et al., Culture and Human Fertility (Unesco,Paris, 1954), p. 165. Olympiodorus,frag.44 (FGH, iv).

118

ELITE MOBILITY IN THE ROMAN EMPIRE

mortality and averageexpectationof life at birth; such concepts were not availableto the Romans.They saw deathas capricious. If one had five children all might survive and the family would sink in the socialscale,or beruined.Aemilius Paullus,for example, had four sons. While they were still young he gave two away to be adoptedby noble families. His own two remaining sonsthen died prematurely.Many noblesapparentlyhad one son only and preferredto risk his survival, ratherthanhavemore and risk their survival and the consequentlowering of the family's wealth and status.Whateverthe result in individual cases,the result for the aristocracy as a whole was the continual dwindling of their numbers.

VI SUMMARY

A brief summarymaybeuseful.Thereis little quantitativeevidence of social mobility in Rome.In this paperthereforeI haveattempted a structuralanalysisof someaspectsof elite mobility. The task was madesimplerbecauseRomehadan estatesystemof stratification, with legal distinctions betweenstrata. In accord with this, Romans expected a high degree of status congruence.Social mobility may usefully be seenas a processof statusdissonance, thatis a processin which the socialriserrateshighly on somestatus dimensionsbut not on others.This in turn raisesthe problem of his acceptabilityto the elite. I have tried to explain the increased chancesof such acceptancewhen differentiated or specialized social institutions replace undifferentiated ones: for example schools, the army. Traditionally, whether in myth or reality, Roman aristocratshad been good at everything that mattered. They strove to live up to this ideal and recognizedachievement; they thus openedthe way for the assimilationof the upwardly mobile. Not all achieverswere immediately acceptable,and some of thesewere usedby emperorsin their rivalry with the established aristocracy.Sucha contestseemsa function of autocracy.Objection to social risers might have beengreaterbut for two factors. Firstly, there was a reservoir of wealthy provincials 'of good 119

KEITH HOPKINS

family', who becameassimilatedto the old nobility as the empire becamemore of a political unity. Secondly, the old aristocracy left vacantplacespartly throughits own downwardsocialmobility, but mostly throughits keepingthe numberof children below the level of replacement.

12.0

VI SOCIAL MOBILITY IN THE EARLY ROMAN EMPIRE: THE EVIDENCE OF THE IMPERIAL FREEDMEN AND SLAVES* P. R. C. Weaver

Roman Imperial society, both of the early and late empire, exhibits a highly developedsystemof formal stratification.It was madeup of a rigid hierarchyof ordersor 'estates'with legal distinctions of statusbetweenthem. In the early empiretheseestates were the senatorialand equestrianorders,the free-bornplebsand rural peasantry,the slave-bornfreedmen(liberti) and thosestill of slave status(servI). However, juridical statusis far from being a reliable indication of social status. A considerabledegree of flexibility in the working of the systemof stratification was permitted, and was indeedinescapable,if the basicstructurewas not to be strainedandbreakdownin socialdiscontentand revolution. Social mobility, or social movementin the restricted senseof changesin classor status,the processby which discrepancies arose betweenthe legal systemof stratificationand actual social status, and were tolerated, is important enough in the history of the Roman empire to demandmore attention than it has hitherto received.1It cannot be measuredstatistically. In the absenceof anything approachingcomplete data or statistically significant samples,or devices such as indices of association,or even an agreedlist of power and occupationalrankings,a different, non• From no. 37 (1967). 1 Foremostin this field is M. K. Hopkins: 'Social mobility in the later Romanempire: the evidenceof Ausonius',CQ, new ser.,xi (1961), pp. 239-49; 'Eunuchsin politics in the later Romanempire',PCPS,new ser., ix (1963), pp. 62-80; 'Elite Mobility in the RomanEmpire', chapterv above; cf. also Hopkins, 'Contraceptionin the Romanempire', ComparativeStudiesin Societyand History, vii (196S), pp. I24-S I. IZI

P. R. C. WEAVER

numericalapproachhas to be adoptedfor the problem of social mobility in the Roman empire. M. K. Hopkins has recently suggestedsuch an approach.2 It involves tracing what he calls patternsof mobility, thatis the orderin which highly valued status symbolsareacquiredby membersof a groupor classto which they were not appropriatein the hierarchical society of the empire. Social mobility in Romeis thus seenas a processof statusdissonanceby which personsratehighly on somecriteria of status,such as ability, achievement,wealth, but low on others, such as birth or legal condition. Taking Hopkins's article as a starting point, I propose to considerto what extentthe evidencefrom the Familia Caesaris(the slaves and freedmen of the Imperial household) supports or illustratesthe propositionsthat:

(I) social mobility can usefully be seenas a processof status dissonance;

(2) social mobility was a product of the structural differentiation of institutions (including an investigation of the importanceof bureaucracyas a differential sourceof mobility for the lower ranks of society); (3) certain patternsof mobility may be best understoodas the productof conflict betweenthe emperorandthe aristocracy, and betweenthese two and the structurally differentiated institutions. First, what is meantby 'structurallydifferentiated'institutions? The greater the size, complexity and centralizationin the administrativeapparatusof the empire,the greaterthe tendencyfor (for example) the military, bureaucratic,legal, and educational institutions to separatefrom one another.Theserequiredspecialized training and acquiredan increasinglyprofessionalstatus,as shown in the recorded military and bureaucraticcareers. The bureaucracyin particular rapidly becamea 'structurally differentiated' institution. In effect thesechangesproceededto the point where considerable differentiation, both administrative and social/legal, developed not only within the bureaucracyitself (equites, freedmen, slaves), but also within the Familia Caesar;s (domestic-administrative;sub-clerical-clerical-procuratorial). 2

Seechapterv, above. 122

SOCIAL MOBILITY

IN THE EARLY ROMAN EMPIRE

I STATUS DISSONANCE IN THE FAMILIA CAESARIS

Some of the most spectacularexamplesof social mobility in the early empireare Imperial freedmen.Many rosefrom humbleslave statusin a junior postin the emperor'sservice,to freed statuswith a responsibleposition in the bureaucracy;and some, becauseof their ability, which was usually financial, and their legal and often personalrelationship to the emperor,reachedsenior posts from which they exercisedgreatand, in a few cases,undueinfluencein the Imperial power structure.The casesof Licinus under Augustus, PallasandNarcissusunderClaudius,EpaphroditusandHelius under Nero, the father of Claudius Etruscusand Hormus under Vespasian,andPartheniusunderDomitian illustratethe continuity of this phenomenonthroughoutthe first century.Nor did it stop there, as I will show later. However, it is not the individual examples,significantas they are, that are important here; it is the Familia Caesarisas a whole, as a status-groupwhosememberscan be readily identified and which is of sufficient size and importance to be of generalsignificancein Romansocietyof the early empire. As a group the Familia Caesarisconstitutesoneof the mostnotably 'unstable'elementsin Imperial society.3 An examination of the causesof their social 'instability' may, amongother things, throw light on the processof social mobility in general. The Familia Caesaris,the slavesand freedmenof the emperor's household, were the elite status-groupin the slave-freedman section of Roman Imperial society. This was partly due to the pre-eminentstatusof their masteror patron,the emperorhimself. But even more important was the nature of their duties. As assistantsto the emperorin the performanceof many of his manifold magisterialdutiesthey hadaccessto positionsof powerin the statewhich were totally inaccessibleto other slavesand freedmen outside the Familia, exceptin the caseof the servi publici (public slaves)(much less significant under the empire than the republic both in prestige and numbers) who in many aspectsserve as forerunnersof the servi Caesaris(Imperial slaves)as an elite group. Theseoccupationalfunctions gave the Familia Caesarisstatusin 3

Cf. D. V. Glass, Social Mobility in Britain (London, 19H), p. 286.

P. R. C. WEAVER

Roman society as a whole and not only within the slave and freedmanclasses.From the legal point of view the slave-born orders (servi, libertt) were inferior to free-born society (plebs, equites, senators).However, from the social point of view many liberti enjoyedhigher statusthanmany of the plebs.In the Familia Caesaris,most servi and liberti were higher in statusthan most of the plebsand indeedsomehad statusequalto that of someequestrians. They thus serve as a clear example of statusdissonance, rating highly on some status criteria, such as acquired skills, ability and possibly intelligence, power ranking, and to a considerableextent,wealth and style of life, but low on others,such as birth or legal status. Within the Familia Caesaristhe distinction betweenthe familia urbana (the urban household) and the familia rustica (the rural household)which is normal in the largehouseholdof any wealthy Romandid not apply. The basic criterion for membershipof the Imperialfamilia is use of the Imperial status-indication:Aug(usti) lib(ertus) for freedmenandCaes(aris)ser(vus)or Aug(usti)vern(a)for slaves. But this status-indication,with some exceptionslargely confined to the African inscriptions, tended not to be used by womenor childrenor by low statusagriculturalworkers.It came to indicate not birth from Imperial slave parents or actual or former ownershipby the emperor, but rather the holding of a postin the emperor'sservice.This could be eitherin the domestic service,properly so called, of the emperor'shouseholdconcerned with the upkeep of the Imperial establishmentsand gardenson the Palatine and elsewherein Rome or in the Imperial villas throughout Italy. The duties ranged from the menial (porters, gardeners,ushers)to the managerial(procuratores[supervisors]of individual villas, including the procurator castrensis[head of the Palatinedomesticservice]) up to the highly influential a cubiculo [supervisorof the emperor'sbed-chamber]).But, apart from the unofficial influenceof the last-named,the power of this sectionof the Familia Caesariswas small comparedwith that of the bureaucratic section. The administrativeserviceitself had a wide rangein the status of the postswhich were includedin it. Correspondingto the low status domestic staff were the sub-clerical administrative staff, such as tabellarii (postmen), pedisequi (ushers, attendants), custodes(watchmen)and so on. They rarely roseto the postswhich 124

SOCIAL MOBILITY IN THE EARLY ROMAN EMPIRE

can properly be regardedas forming part of the slave-freedman administrativecursus or 'career'.4Theserangedfrom the junior postsof adiutor (assistant)held at the ageof twenty to thirty, to the intermediateslave and freedmanposts,dispensator(paymaster),a commentariis(record officer), tabularius (accountant),held roughly betweenthe ages of thirty and forty, and the senior gradesof proximus (deputy-head),procurator, and the secretariesof the Palatinebureaux,a rationibus (the financial secretary),ab epistulis (the secretaryin charge of correspondence),etc., not normally appointedbeforethe ageof forty or later. Thosewho were fortunate enough to gain entry to the administrative cursus did so normally as slavesby the age of twenty, after preliminary attendance at one of the Imperial administrative training establishments.5 They were selected at a comparatively early age for a professional administrative career, and thus were the most favoured sectionof the Familia Caesaris. In the caseof the Caesarisservi, the basic causeof mobility is the dissonanceor discrepancybetweentheir high occupational prestigeand power and their low legal statusand birth. One of the clearestindicatorsof social mobility is changein the marriage patternof the group concerned.The legal and social statusof one marriagepartneris importantfor the statusof the other. A change in the patterninvolvesupwardmobility for onepartneranddownward mobility for the other, unlessthereis a compensatingfactor accordingto someother statuscriterion, such as power-ranking or wealth.Marriageis rarely the initiating factor in socialmobility, but as a secondarycharacteristicit is relatively easyto identify. In Imperial society there were legal restrictionson the freedom of intermarriagebetweenmembersof someclasses.Under the Lex Iulia of 18 B.C. (part of the Augustansocial legislation) freedmen wereforbiddento marry womenof senatorialfamilies;6 andunder the SenatusconsultumClaudianum of A.D. 52, legal penaltieswere imposedon freebornwomen who cohabitedwith slaves,with or without the consentof the master: a free-born woman (ingenua) SeeP. R. C. Weaver,'The slave and freedmancursusin the imperial administration',PCPS, new ser., x (1964), pp. 74 If. 5 SeeS. L. Mohler, 'Slave educationin the Romanempire', TAP A, lxxi (1940), pp. z6z If., esp. on the Imperialpaedagogiumad Caput Africae on the Caelian,pp. Z71 If. 6 Dig., xxiii. z. 44.

4

P. R. C. WEAVER

was reducedto the statusof eitherservaor liberta, and the children also becameslaves.Theserestraintswere mostly rigidly enforced: it was rareevenfor an Imperial freedmanto marry the daughterof a senatorialfamily. In the well-known caseof ClaudiusEtruscus' father, there is no direct evidencethat Etrusca,his wife, was of senatorialbirth despitethe fact that her brother was of consular status (he is usually identified with L. Tettius Iulianus, consul A.D. 83). Adoption of the brother from an equestrianinto a senatorialfamily is just as likely as (or evenmore likely than) the marriage of his sister to a freedman, if she were of senatorial birth.7 Even amidst the magnificentsuccessesof Antonius Felix, the brotherof Pallas,who married two princessesand one queen, we must assumethat those wives whom he married before his elevation to equestrianstatus were all non-citizens and not of senatorialrank.8 As for marriagebetweenpartnersof slaveand freebornstatus, this too was comparativelyrare. Beryl Rawson9 has found little evidenceof sucha marriagepatternamongthe sepulchralinscriptions from Romein which the namesof childrenand both parents occur. In an analysis of 700 slave or freedman marriagesfrom Rome, outside the Familia Caesaris, where the slave or ex-slave statusof at least one of the partnersis certain, I found that for liberti in this group a maximumof only 15 per cent of wives may havebeeningenuae,andfor servia maximumof only 10 per cent. In fact the actual proportion is likely to be even lower, perhaps considerablylower. This is the normal slaveor freedmanmarriage pattern:to an overwhelmingextentthey marriedpartnersof their own statusand origin, asis to be expected,given the provisionsof the ClaudianSenatusconsultum. Thus, for slavesand freedmennot belongingto the emperor's familia, thereis little, if any, evidence of significant mobility through marriage. By contrast,the upward mobility of the Caesarisservi in respect of the status of their wives is very marked, if not astonishing. SeeP. R. C. Weaver, 'The father of Claudius Etruscus',CQ, new ser., xv (1965), pp. 150 If. a Antonius Felix: ProsopographiaImperii Romani, 2nd edn (Berlin, 1933), A 828; Suet.,Claud., 28. • B. Rawson,'Family life amongthe lower classesat Rome in the first two centuriesof the empire',CP, lxi (1966), pp. 71 If.; cf. her Bryn Mawr dissertation,'The namesof children in Romanimperial epitaphs' (1961). 7

IZ6

SOCIAL MOBILITY

IN THE EARLY ROMAN EMPIR'E

From a detailedanalysisof the nomina(family names)of all wives in this group, it can be clearly shown that for the Julian period (Augustus to Gaius) the marriage pattern of members of the Familia Caesarisconforms fairly closely with that of other slaves and freedmen, that is a lulius Augusti libertus is usually found married to a lulia Augusti liberta; but from the Claudianperiod (Claudius and Nero) right through to the end of the Aurelian period (M. Aurelius to SeverusAlexander) the pattern is quite reversed.The proportion of Caesarisservi who marry ingenuaeis at least 66 per cent, and more likely 75 per cent or even more.10 The only othergroupof slavestatuswith a comparablemarriage patternare the servipublici, especiallyunderthe late Republic.It is certain that servipublici frequently married ingenuae.l l The reasons are the sameas for the Caesarisservi: high social statusresulting from their holding official positions and hence exercising some degreeof power.TheCaesarisservi,at leastin their lower andmiddle ranks, were the successorsof the servi publici, but in their higher echelonsthe functions and responsibilitiesof the Augusti liberti were greatly extended,up to and overlappingwith the equestrian procuratoriallevel.12 Their power ranking and social statuswere correspondinglyhigher. Henceit is to be expectedthat the members of the Familia Caesaris,and certainly thosein the administrative service and favoured posts in the domestic service, would increasinglymarry freebornwomen once the tradition and status of theseserviceshad beenfirmly established.It would, indeed,be surprisingif this werenot the case.It is not possiblehereto discuss the detailedbreakdown,period by period, of the nomina of wives in the Familia Caesaris,but one feature to be noted is the higher proportion of non-imperialnomina belongingto wives of Augusti Note that as the averageage at marriagefor slaveswas someten to fifteen years earlier than their averageage at manumission-normally not before the age of thirty, accordingto the Lex Aelia Sentia of A.D. 4 -the marriagepatternof liberti does not differ greatly from that of servi; it merely representsthe sameset of social facts at a later stage.A great number of the sepulchralinscriptions owe their existenceto the fact of manumissionof one or other of those namedon them; a greater number, of course,owe their existenceto an even later fact-death. 11 T. Mommsen,RiimischesStaatsrecht,3rd edn (Leipzig, 1887), i, p. 324; L. Halkin, Les Esc/avespublics chezles Romains(Brussels,1897), pp. 118 ff. 12 For other privileges of the servi publici, with regard to salary,peculium and will: Frontinus,de Aqu., 100; Pliny, Epist., x. 3I; Dig., xvi. 2. 19; Ulpian, Frag., 20. 16; elL, vi. 2354. 10

127

P. R. C. WEAVER

liberti and Caesaris servi in the higher status career-groups. Becauseof the namesand statusof the children (apart from the intrinsic improbability of sucha pattern),thesewives can scarcely havebeen,at the time of their marriage,slavesor (evenlesslikely, becauseof the age factor) freedwomenof masters/patronsunconnectedwith the Imperial family. If the freeborn statusof the wives is also taken as a criterion of their husbands'occupational status within the Familia Caesaris, it appears(what might also havebeenconfidently surmised)that by and large the administrative service enjoyed higher prestigethan the rest of the Familia Caesaris,and that postsin Rome, within a given service,were of higher status than those in the provincial centres. In fact, the family data of these officials becomesone fruitful method of approachingthe relationshipof the governmentin Rome to the administrativeorganizationin the provinces,the questionof the centreand the periphery, and the degreeof bureaucratizationin the early empire. More surprising, however, is the Claudian Senatusconsultum, both in its timing, A.D. 52, and in the mannerof its passing.By the middle of the first centuryA.D., in the early Claudianperiod, the definitive marriagepattern in the Familia Caesaris discussed abovehad becomefirmly established.By penalizingthe unionsof freeborn women with slaves, the Smatus consultum brought precisely the marriagesin the Familia Caesaris itself within its scope.I 3 Moreover, the inspiration for the measure,accordingto its proposerClaudius,the dominus(master)of the Familia Caesaris, came from Pallas, who was himself an Augusti libertus, however little his actual power correspondedwith his legal status. His name is placed prominently and ironically by Tacitus at the beginning of the sentencein Annals,xii. 53, following the account 13

Strictly speaking,it is not possibleto speakof slave marriageat all, but only of contuberniumor concubinage.This applied when either one or both partieswere of slave status:e.g. Ulpian 5. 5, 'cum servis nullum est conubium';Sent. Pauli, ii. 19. 6, 'inter servoset liberos matrimonium contrahi non potest'.But the terminology of legal marriage(iustum matrimoniumbetweenpartnerswith conubium),e.g. uxor, maritus, filius, etc., is so constantlyusedof contuberniumin the inscriptions,and even in the legal texts themselves(cf. W. Buckland, RomanLaw of Slavery (Cambridge,1908), p. 76, n. 15), that it is convenientto speaknormally of slave marriageand to use terms such as contuberniumand matrimonium only when the distinction is relevantto the argument.

128

SOCIAL MOBILITY IN THE EARLY ROMAN EMPIRE

of the Senatuseonsultum:'Pallaswas namedby the emperoras the authorof this proposal'.Therearecomplicationsaboutthe precise provisionsand subsequentchangesin the law, enteredin Gaiusi. 84 ff., which neednot be discussedhere;14but it seemsto me clear that at leastone, and probably the main, purposeof the Claudian Senatuseonsultum was to regulate in the interests of the fiseus (imperial Treasury) the rights of inheritanceaccruing from the wives and children of membersof the Familia Caesaris which might otherwisehave beenlost to it if the Caesarisservi had been allowed to continue marrying women of freeborn status. The practicaleffect of the measureon the real marriagepatternwas nil, as the marriageswith ingenuae continuedunabated;but as such ingenuaeand their childrenhenceforthbecametechnicallyservae(or libertae) and servi, the fiscus, of which Pallaswas the head,did not suffer financially. The needfor the measureis in effect a tribute to the social mobility of the Caesaris servi as expressedin their marriages. Not unconnectedwith their marriage-patternis the economic statusof membersof the Imperial Familia. The discordancehere betweentheir high level of wealth and their low legal statusis most marked; it is also well known and need only be briefly mentioned.The sourcesof this wealth can only be a matter for speculation:regular salary cannotfully accountfor the fortunes which were frequently made;financial opportunitiesand acumen as well as corruption on the usual scalealso played a significant part. Certainlytheir official positiongavethem a strongeconomic advantageand was the basis of their power. Economically, not only the Imperial freedmenbut the youngerImperial slavesaswell were a favouredsectionin Romansociety. It has beenspeculated that the colossal fortunes of Pallas and Narcissuswere greater than those of anyoneelse, bond or freeborn, in the empire with the exceptionof the emperor.Even humbler officials than these, notably dispensatores,were very wealthy; for example,Rotundus Drusillianus in Hither Spain, who, according to Pliny, had a specialfactory built to fabricatesilver plate-oneweighed500 lbs. and eight others 250 lbs. each;ls there was also Musicus ScurFor referencesand literature, seeCR, new ser., xiv (1964), p. 138, n. I; CQ, new ser., xv (1965), p. 324, n. 4. Seemost recently J. Crook, CR, new ser., xvii (1967), pp. 7 fr. 15 Pliny, Nat. Hist., xxxiii. 145. 14

P. R. C. WEAVER

ranus,16and severalwho offered large sums for manumissionor promotion.I? In addition to the status relationship between the Familia Caesaris and slave-bornsociety on the one hand, and free-born society on the other, a further questionis whetherthere is status dissonancewithin the Familia Caesarisitself: that is, do the categories of juridical and occupationalstatus coincide within the Familia as a whole and evenwithin the elite of the administrative service itself? Thus, do freedmenregularly hold posts of higher status than slaves? In terms of the slave-freedmancursus or career grades in the administration, is promotion to the intermediateclerical gradesdependenton manumission? Firstly, it would be foolish to underestimate the role of patronage in determiningadvancement in the sub-equestrian careersas much as in the equestrianand senatorialcareers.For the slave-freedman servicesthis might be dispensed,in varying degreesof effectiveness, by the emperor, by senators,equites or Imperial freedmen and sometimesslaves themselves,in the latter casesoften with financial implications. An amusing example of the emperor (typically, Vespasian)getting the better of one of his own slaves in this role is in Suetonius,Vespasian,23 (which should be comparedwith Suetonius,Otho, 5): When one of his favourite slaveswas asking for a financial post for someonehe pretendedwas his brother, Vespasian put him off and summonedthe applicantin person;and after extractingfrom him the sum which he had contractedto pay his sponsor,the emperorappointedhim without delay. The systemof official patronageand promotion for freedmenis illustrated by Pliny's testimonialfor Maximus, Trajan'sfreedman and assistant procurator of Virdius Gemellinus in Pontus et Bithynia;18 and by Fronto'sto M. Caesaron behalfof the latter's freedmanAridelus.19 The careerof Vespasianhimself is said to havebeenhelpedin the reign of Claudiusby the patronageof the See below n. 29. Pliny, Nat. Hist., vii. 129; Suet.,Otho, 5; Vesp., 23. 18 Pliny, Epist., x. 85; cf. A. N. Sherwin-White,The Letters of Pliny (Oxford, 1966), pp. 681 f. 19 Fronto, ad Marcum Caesarem,v. 37 (ed. Nab, p. 87). Cf. H. G. Pflaum, Les Procurateursequestressous Ie Haut-empireRomain(Paris, 1950), pp. 198 f. 16

17

SOCIAL MOBILITY

IN THE EARLY ROMAN EMPIRE

freedmanNarcissus,in his appointmentas legionary commander (legatus /egionis) in Upper Germany.20 However, patronage,at least in the administrativeservice of the Familia Caesaris,is exercisedmostly within the framework of a set of bureaucraticrules. There was a regular system of annual promotion, probably in November.21The principle of seniority was favoured over that of unrestrictedpatronage,both in the early andlater empire.22 The clearestevidencethat this was regular in the careeradministrativeservice of the Familia Caesarisis to be found in the age-at-deathfigures. In the many caseswhere an official died in office and his age at death was recordedon his betweenthe tombstone,there is a high degreeof correspondence gradeof post held and the ,agerecorded.Thus, the senior grades are not held by slavesor even by freedmenunder forty to fortyfive years-theyoungestprocurator whoseage at deathis known died aged fifty-five years-andfor the intermediate grades of tabu/arius, dispensator, a commentariis the regular age range is thirty to forty, with very few, if any, under the age of thirty.23 The conclusionis that, generallyspeaking,the ageand promotion structurecorresponded,and that just as it was rare for a Caesaris servusto gain early manumissionbefore the age of thirty, so also was it rare for promotion from a junior to an intermediategrade and thenceto a seniorgradeto occurbeforethe agesof thirty and forty yearsrespectively.This canbe seento apply to suchsuccessful careersas thoseof Pallasand the father of ClaudiusEtruscus. Thus, mostfreedmenwere of higher statusthanmost slavesin the Familia purely becauseof the applicationof the Lex Aelia Sentia of A.D. 4 which limited formal manumissionto slavesagedthirty and over, unless one of the valid grounds for manumission (iustae causae manumissionis)could be upheld before the council (consi/ium) as provided for by the law. With the Caesarisservi the acknowledgediustae causae, such as blood relationship to the Suet., Vesp., 4. 1. Cf. also the senatorialVespasian'scontemptuous rebuff by the slave/freedmanofficial of Nero's officium admissionis: ibid., 14· 21 SeePflaum and Sherwin-White,loco cit. (notes 18 and 19 above). 22 Cf. Hopkins, p. II5 above; A. H. M. Jones,The Later Roman Empire (Oxford, 1964), pp. 602-6. 23 On the age-at-death figures and for referencessee Weaver, PCPS,new ser., x (1964), pp. 76 f. On the father of Claudius Etruscus,seealso CQ, new ser., xv (1965), pp. 146 f.

20

P. R. C. WEAVER

manumittorand intention to marry, could scarcelyapply. In fact, the law was quite strictly observedin their case,no doubt in the interestsof the fismspeculiorum(treasuryfor the receipt of manumission taxes) whither went the proceedsof manumissionpurchasedby Imperial slavesat the regularage. But some slaves were of higher status than some freedmen. Despite the rules of manumissionand promotion, there are two notable exceptions which constitute clear cases of status dissonanceof this kind within the core of the administrativeservice. Among the most powerful and wealthy officials of intermediate gradewere the dispensatores.Their responsibilitieswere exclusively financial-they had personalsupervisionover cash transactions, payments,etc. in eachprovincialfiscus and in many of the smaller departmentsin Rome. Thedispensatoreswereall of slavestatus.24 The reasonlies in their responsibilities,the desire to avoid peculationand in the closer degreeof control over them by the emperor which their slave status afforded. The manipulation of the financial strings of administrationwas always of vital concernto the emperor. For slavesin this grademanumissionwas normally delayedfor ten to fifteen years, till the age of forty to forty-five, after which time they were manumittedand moveddirectly into the seniorgrades. Thereaftertheir ascentwas rapid. They were in fact a favoured sectionof the slave-freedmanelite. The other group in which statusdissonanceis found are the vicarii, the assistantsor deputiesof the dispensatores.Their functions were also financial and as they were directly responsibleto their dispensatorin each case,who was in turn responsibleto the emperor; the vicarius, who had the legal status of slave of his dispensator,stood also in the same legal relationship to the dispensatoras the latter did to the emperor.Hencethe anomalyof the lowest legal status,that of slave of a slave(vicarius), attachingto officials who, although junior, were of considerableinfluence in the financial administration, and who were mostly destined to becomedispensatoresin their turn on the retirementor promotion of their master.25 On the other hand, there are also the occasionalexamplesof 24 2S

Cod. lust., xi. 37. I. For further details see P. R. C. Weaver, 'Vicarius and vicarianusin the Familia Caesaris',IRS, liv (1964), pp. 117 ff.

SOCIAL MOBILITY

IN THE EARLY ROMAN EMPIRE

holdersof junior posts,suchas adiutores(normally held by slaves), who have beenmanumittedat the normal age but not promoted to a post of intermediaterank. In the domesticservice of the Imperial household,of course, becauseof the much closer degreeof personalcontact with the emperor, and greater flexibility in the promotion structure,the scope for personal influence was much enlarged. Hence the unscrupulous,and often unworthy slave-born palace officials, usually cubicularii, who won their way to irregular and unofficial, but none the less real, positions of power.26 II STRUCTURAL DIFFERENTIATION WITHIN THE BUREAUCRACY

Tacitus,by a naturalreflex action, was alwaysinclined to seevice, corruption and disgrace whenever the Imperial freedmen and slaves are mentioned. For him, as for any Roman nobilis, they representedone of the most hateful and degradingaspectsof the Principate.Where freedmenflourish in high places,there liberty is extinguished.This feeling is nowheremore explicitly expressed than in Germania 25, where he contrasts the natural situation under primitive but more honourableconditions:27 Freedmenare not of much higher statusthan slaves,they seldomhave much influence in the household,and never in the state,with the exceptionof courseof thosepeopleswho are under a monarchicalform of government.In thesecases freedmenrise aboveboth thoseof free and thoseof aristocraticbirth. But amongstother peoples,the lower statusof freedmenis a sure sign of freedom. This passageillustratesclearly the tensionunderthe Principate betweenemperorand nobility. The more autocraticthe regime the less dependentwas the emperor on the traditional Roman institutions of government.Even under the 'best' emperors,the Senatewas of much less significanceas a sourceof legislation or SeeE. Fairon, 'L'Organisationdu palais imperiale ~ Rome', Musie Beige, iv (1900), pp. 5 ff.; and for the later empire, Jones,op. cit., pp. 566 ff., and Hopkins, PCPS, new ser., ix (1963), pp. 62 ff. 27 Cf. Ann., xiv. 39.

26

P. R. C. WEAVER

for the performanceof any constitutional function than as the chief repository of ability and experiencein military and civil administrationin the empire. The commandersof legions, the governors of provinces, consular or praetorian, armed or unarmed, the membersof the Imperial consilium (council), were all, with few exceptions, drawn from the higher echelons of the senatorialorder. But while nobilesmonopolizedmostof the honorific positions in the state,and novi homines-thefirst of consular standingin their families-occupiedmore than their shareof the top executive posts, especially in the provinces, even together they did not by any meanshave a monopoloy of power in the administration. The emperor from the beginning was at great pains to createinstitutionsof his own devisingand underhis own control as a bulwark for his own position, a sourceof power to counterbalance that of the senate,andasa channelfor the reservoir of talent and energyin all partsof the empireandin all sectionsof society to be usefully employedin administeringand defending the empire.Augustuswas the creatorof the equestrianorderin its vastly changedImperial form; he was also the creator of the Imperial civil service,which fromthe beginningof the Principate employed not only equestriansin responsiblepositionsbut also freedmen and slaves. Highly significant (and notorious) is C. Iulius Licinus, the slave and freedman of Caesar,employed by Augustus as his procuratorin Gaul. Licinus' avarice and abuse of his position led to his recall, but not beforehe had laid the basis of a fortune that was to be a by-word in Imperial literature of the first century. The power of Licinus was real and an object of envy to the noble aristocratswho observedits effects.28At a lower level of responsibility the Imperial freedmen and slaves in the administrationexercisedpowerandcontinuedto accumulatewealth, thoughlessspectacularly,undereveryemperorfrom Augustuson. For example, Musicus, the slave of Tiberius, chief dispensatorin the treasury at Lugdunum, came to Rome (where he died) accompaniedby no fewer than sixteenpersonalslavesof his own, who stayedto dedicatea monumentto him.29 Licinus: Dio, 54. 21; Suet., Aug., 67. Cf. ProsopographiaImperii Romani, 1St edn (Berlin, 1897), L 193; Stein, Pauly-Wissowa,RE, xiii, cols 501 fr.; V. Gardthausen,Augustusund seine Zeit (Leipzig, 1896), i, pp. 616 f., ii, P·336. 29 elL, vi, 5197.

28

SOCIAL MOBILITY IN THE EARLY ROMAN EMPIRE

The creationand extensionof the slave-freedmanservicein the bureaucracydemandsan explanationat the generalinstitutional level, andnot as a moreor lesswhimsicalor accidentalby-product of the personalitiesof particular emperors.The hey-day of the freedmenin the middle of the first century was no doubt partly due to the laxity of Imperial control by Claudius and Nero, but Claudius did not call into being a monsterwhich he could not control. The bureaucracywas deliberatelyestablishedby Augustus. It kept expandingunder Tiberius, Claudius, Nero and Vespasian,and the slave-freedmanshareof it, if anything, increased proportionally. The cohesion of the bureaucracysurvived unimpairedthe upheavalof civil war in A.D. 68-69.30Vespasiannot only appointeda former slaveof Tiberius to control the financial bureauwhich Pallas had notoriously controlled under Claudius, but he also elevatedthis official to full equestrianstatus,the kind of practical reward that even Claudius and Nero were chary of dispensing.The slave-freedmanbureaucracypersistedwith undiminished numbers, status, and to a large extent power, into the secondcentury, and in the courseof the secondcentury, in the periodafter Hadrian,the numberof postsavailable,including senior posts, even expanded considerably. The permanent place which the slavesand freedmenfilled in the administration cannot be explainedon the assumptionthat this was simply an extensionof the kind of employmentslavesandfreedmencouldbe expectedto havein the householdof any wealthyRomanmagnate. This hasas much validity as a belief in the continuedexistenceof the 'restoredrepublic' after it had servedthe endsof Augustusin the decadebefore 19 B.C. The bureaucracyin its financial and managerialaspectswas staffedby freedmenandslavesaswell asby equestriansas a matter of policy. An instructive exampleof how this policy could be applied is found in Tacitus, Annals, xiv. 39, already quoted.3! In A.D. 61 Nero'sfreedmanPolyclitus was sentby the emperoron a mission to Britain to resolve the differencesbetweenthe Imperial legate, SuetoniusPaulinus, and the Imperial procurator Iulius Classicianus. The four main institutional elementsin the power situation and the tensions betweenthem are here represented:the emperor,the senatorialaristocracy,the equestrianhierarchyand 30

31

Cf. Statius,Si/IJae, iii. 3. 83-4. Above, note 27.

P. R. C. WEAVER

the freedman administration in the bureaucracy:'An imperial freedmanPolyclitus was sent,by whoseauthority Nero fervently the legateand hoped... that harmonycould be restored between the procurators'.The other three are, with varying degreesof patronage,all appointeesof the emperor. This policy was commonto all emperorstill after the Severan period. The motives behind it are clear enough as regards the equestrians.By reorganizing a second order within the upper class, dependenton himself for professionaladvancementand patronage,the emperorwas creatinga sourceof power on which he could rely in the institutional tug-of-war betweenhimself and the senatorial aristocracy. The equestrian bureaucracywas a differentiated institution which also served the ends of social mobility. It provided the main channelof social advancementfor membersof the secondorder into the aristocracyunder Imperial patronage.Such diversification, with a separatehierarchy in the senatorialcareeras well as in the equestriancareer,also tendedto producea greaterstability in the higher ranks of the army, and to provide regular channelsof promotion from the junior through to the senior officer grades.32 However, the interesting thing about the bureaucracyof the early empire is not its differentiationfrom the aristocracybut the degreeof differentiationwithin its own ranks. This was of a much more fundamentalnature.The legal distinctionsbetweenthoseof slave and freedman status who made up the lower and middle gradesof the bureaucracy(as well as someof the higher gradesat leastfor a time) and thoseof equestrianstatusin the uppergrades of the bureaucracywere much greater than those between the equestrianbureaucracyand the senatorialaristocracy.Upward mobility betweenthe equestrianand senatorialorderswas regular andcommon,indeedinevitable;that betweenthe equestrianorder and the Imperial freedmen, for all the high status of the latter relative both to all other slavesand freedmenand to the eques32

In the languageof S. N. Eisenstadt,The Political Systemsof Empires (New York, Free Press,1963), pp. 273 If., the equestrianbureaucracy was moderatelyservice-orientatedboth to the ruler and to the aristocratic elementsin society but more especiaIIy"tothe former. The bureaucracy's goals were some degreeof autonomyfrom the ruler and also service goals partly in common with the aristocracyto whom they tendedto assimilatethemselvesthrough formal elevation in status,intermarriage, and accessto positions of equal power.

SOCIAL MOBILITY

IN THE EARLY ROMAN EMPIRE

trians themselves,was the meresttrickle. Emperorsof all periods, and endowed with whatever degree of historical approbation, declinedto endanger their supportin the upperranksof societyby massivepromotion of freedmento equestriandignity.33 Examples of freedmen in financial posts rising to equestrian status, such as the father of Claudius Etruscus,are very rare.34 Most of the casesof such elevation to equestrianrank are to be explained as due to personal influence with the emperor (for example Antonius Felix, Icelus Marcianus, L. Aurelius Nicomedes,tutor of L. Verus), often throughthe post of a cubiculo (for exampleCleander,Theocritus). Nor was occupationaldifferentiation betweenthe two sections of the bureaucracyany less rigidly observedthan the social and legal differentiation. The flow of freedmen to posts carrying equestrianresponsibilitiesand hencegreaterpower, without the appropriaterise in legal status,was also very slight. The idea that has recently been canvassed,35that the majority of civilian posts opento equestrianscould also be held by freedmenand that there was no type of procuratorialpost from which the freedmenwere rigidly excluded,could scarcelybe further from the truth. Roman Imperial society was extremelystatus-conscious, and if freedmen in the upper ranks of the administrativehierarchyhad full parity of occupational rank with their equestrian counterparts,this remarkablestateof affairs could scarcelyhave escapedthe notice of the rathersensitiveliterary authoritiesof the first and second centuries.In fact, the whole relationshipbetweenthe equestrian and slave-freedmansectionsof the bureaucracyas it developedin the first and secondcenturiesA.D. needsreappraisal,paying close attentionto the massiveepigraphicevidencealong strictly chronologicallines.36 A. Stein, Das riimische Ritterstand(Munich, 192.7), pp. 107 ff. Stein, op. cit., p. I12., assumesequestrianstatusfor Licinus, but this is very doubtful. Cf. O. Hirschfeld, Die Kaiserlichen Verwaltungsbeamten bis auf Diocletian, 3rd edn (Berlin, 1963), pp. 377, n. 7, 468, n. I. 3S By F. Millar, Historia, xiii (1964), p. 187, and fRS, liii (1963), p. 196. 36 The starting point, and often the finishing point, of many enquiriesinto the developmentof the Imperial administrationin this period is bound to be the Historia Augusta,Hadrian, 2.2.. 8: '[Hadrian] was the first to appoint equestriansas secretariesin chargeof the correspondence (ab epistulis) and petitions (a libellis) bureaux'.That the Historia Augustain generaland in many particularsis about as thoroughly discrediteda 33

34

P. R. C. WEAVER

Attention hasnaturally centredon the equestrianhierarchy,the elite of the professionaladministrators,the prefectsand the procurators. Such scholarsas Last, Rostovtzeffand Hirschfeld have to a large extentconsecratedthe view that the key phenomenonin the developmentof the administrativehierarchyis 'the oustingof the freedmenby equites'(Last).37 The secretof the administration's successis revealedas a kind of class struggle: the true Roman equestriansagainstthe upstartoriental freedmen,with virtue and superiorbreeding,but not necessarilysuperiorintellect or education, inevitably winning the day. This allegedsecond-century A.D. conflict of the orders-between the equestrianprocuratorson the onehand,andthe Palatinefreedmen officials on the other-is, for this period at least, under the five 'good' emperorsNerva to M. Aurelius when the decisive equestrianvictory is allegedto haveoccurred,a misinterpretation of the evidence. It arises from the traumatic experienceof the senatorialhistorians and other writers in the first century under such notorious and arrogant upstarts as Pallas, Narcissus and Callistus under Claudius,Phaon,Epaphroditusand Helius under literary sourceas can well be for the serioushistorian of the secondand third centuriesA.D. is no obstacle.There are always to hand persuasive argumentsto show why any particular passage,phrase,or even word should be singled out for respectfultreatment,like a pearl in a pig-trough, by referring it back to its original, and usually hypothetical, author. The words quotedabove enshrineHadrian'swell-known 'reform' of the Roman administration-hisreplacing of freedmenby equestriansin all the important departmentsof the Palatinesecretariate. In the first place, Historia AngustaHadrian, 22. 8 is factually wrong. The first equestrianab epistulis was Cn. Octavius Capito, the friend of Pliny the Younger; he held th!! post from the reign of Domitian till early in the reign of Trajan: H. G. Pflaum, Les carrieres procuratoriennesiquestres sousIe haut-empireromain (Paris, 1960), no. 60. The first equestriana libellis was Sex. CaesiusPropertianuswho was appointedby Vitellius in A.D. 69: ibid., no. 37. In the secondplace, amongthe Palatinesecretarieswho were equestrianby the time of Hadrian, no mention is made of the most important of them all, the secretarya rationibus. And appropriately enough,becauseTrajan regularly employedequitesin the post, and even as early as Vespasianthe first equestriana rationibus is to be found in the personof the father of Claudius Etruscus,raised to equestrianstatusin A.D. 72-73. 37 E.g. H. Last, in CAH, xi, pp. 430 f.; Hirschfeld, Die Kaiserlichen Verwaltungsbeamten, pp. 429, 459; M. Rostovtzeff, Dizionario Epigraftco (Rome, 1895-), iii. 137.

SOCIAL MOBILITY

IN THE EARLY ROMAN EMPIRE

Nero, and Partheniusand other cubicularii under Domitian. But for the secondcenturyA.D., much the most importantevidenceis not literary but documentary-thehundredsof inscriptions recording the careersof officials at all levels in the administration. This evidence has the advantageof being for the most part emotionally restrainedor even neutral. Formerly the freedmenin chargeof the Palatinebureaux(ojjicia Palatina) were notorious partly becauseof their arroganceand corruption, but also partly, and even mainly, becausethey had remainedat the headof the largestdepartments,in positionsof the greatestpower, long after the importanceand size of thesehad warrantedtheir inclusion in the equestriancursus. Equestrians,if differentiationwithin the bureaucracywas to meananything atall, had to have a monopoly of the postscarrying the highestofficial status.In generalit can be said that equestrianprocuratorsmoved from one post to the next with greater frequency than their freedmensubordinates.Theselatter, by their lessfrequentoccupational mobility, often filled the role of permanentsecretaryand thus exercisedan effective power. But from the point of view of official status-ranking,the equestrianand freedmancareerscould not be mixed. The differentiation became, if anything, more marked as the secondcentury progressed.Hadrian was not the first to employ eqttites as secretariesab epistulis or even a libellis, despitewhat is said in the Historia Augusta.Yet thereare plentiful signs that the Augusti/iberti continuedin honourableand indeed senior employmentin the administrativeservice throughoutthe secondcentury.The new freedmenproximi and the new freedmen procuratoresprovinciae illustrate this both for the central departments in Rome, the ojji'cia Pa/atina, and for the provincial administration throughout the empire, both in Imperial and in senatorialprovinces.38 The purposeof thesenew seniorfreedmanpostswas to provide a full senior careerfor the Augustiliberti distinct from the equestrian cttrsus in terms of official titles and duties. The talent of the freedmenofficials was not wastedor discarded.It was fruitfully assimilateduntil the middle of the third century,when chaosand disasterappearin the administrativeas well as the military sphere. 38

Thesechangesbelong substantivelyto the early period of M. Aurelius' reign, as I have arguedin PCPS, new ser., x (1964), pp. 8S fr., and in His/oria, xiv (1965), pp. 460 If.

139

P. R. C. WEAVER

This exercisein structural differentiation within the bureaucracy was not merely to provide a counter-check,firmly under the emperor'scontrol, againstthe potential movementof the equestrian bureaucracytowards greater autonomy or independence from the emperor.Nor doesit appearto havebeenprecipitatedby any significant weakening in the power of the senatorial aristocracy. The freedmanbureaucracynot only kept a watchful eye on their equestriancounterparts,but also provided deputiesor temporaryreplacementsfor all important posts in all important departments.They thus also servedthe interestsof efficiency and administrative continuity.

VII LEGAL PRIVILEGE IN THE ROMAN EMPIRE*t PeterGarnsey

I INTRODUCTION

For the equality aimedat by the many (sc. arithmeticalequality) is the greatestof all injustices,and God has removedit out of the world as being unattainable;but he protectsand maintainsthe distribution of things accordingto merit, determiningit geometrically,that is, in accordancewith proportionand law. Thus a speakerin Plutarch's Dinner-Table Discussionstried to interpretPlato'sstatement(authenticor apochryphal)that 'God is always busy with geometry'.l Whatevermay be thought of his ingeniousexplanation,it does not misrepresentPlato'sviews on the subject of equality. Both Plato and Aristotle held that the equal distribution of things to persons of unequal merit was unequa1.2 The Roman governing classesshowed by their administrationof the law that they sharedthis conviction. So much can be deducedfrom the operationin Rome, Italy and the pro• From no. 41 (1968).

t I wish to thank Mr G. E. M. de Ste Croix for his valuableadvice and criticism. Plut., Mor., § 719. b-c, quotedB. Farrington,Scienceand Politics in tbe Ancient World (London, 1939), pp. 2.9-30. 2 Plato, Laws, §757. a. 1£.; cf. § 744. b-c. Seealso Arist., Nic. Etb., § IIp. a. 15 1£.

I

141

PETER GARNSEY

vinces of a thorough-goingsystemof legal discriminationto the advantageof certain privileged classes. The legal systemof any state may be biased in at least three main ways. The law may favour certainpartiesin a relationshipfor example,creditorsover debtorsor landlordsover tenants,the poor being usually on the losing side in such relationships. Secondly, inequality may result from the incapacity of some through their social or economicposition to enjoy the full benefits of the law even where theseare not officially denied themthus the rich can afford bail, while the poor must generallygo to prison. Finally, the law as written or the law as administeredmay deny equal benefits and protection to different sections of the population, again in accordancewith differing social and economic situations-thusa commonermight be whipped and an aristocratfined for the like offences.3 Only examplesof the second and third varietiesof inequality, de facto and de iure inequality, are exploredin any detail in this paper.For theperiodthatis examined, developmentsin procedureand in the administrationof the law are of specialinterest. For practical reasons,this investigationhas beenlimited to the period from about the beginningof the Empire to about the end of the Severandynasty, that is, from 27 B.C. to A.D. 235. These years, following the civil wars of the first century B.C., saw the emergenceof a disguised monarchy which claimed to be the restored Republic. The political revolution brought in its train institutional changesin the sphereof law. New proceduresgrew up alongsidethe old ones(which, like the old political structures, were not destroyed, but allowed to wither away), and were characterizedby new types of discrimination. One aim of this paper is to describe and contrast the different patterns of discrimination associatedwith Republicanand Imperial procedures in civil and criminal law. Threeexcerptsfrom literary sourcesare quotedfirst to indicate the kinds of peoplewho gainedpreferentialtreatmentin the law courts and to give an idea of the privileges they enjoyed. The details of the systemof privilege are then assembled,and an analysis attemptedof aspectsof the Romancivil suit and criminal trial. 3

This schemeof classificationis used in ). E. Carlin, J. Howard and S. Messinger,'Civil justice and the poor: issuesfor sociologicalresearch', Law and SocietyReview,i (1966), pp. 9-90.

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

Next, problemsconnectedwith the growth of the systemand its basisin law are considered.Finally, the relationshipbetweenthe upper-dass/lower-dass distinction (the so-called honestiores/humiNores distinction)4 and the citizen/aliendistinction is discussed.

II THREE ILLUSTRATIONS

(I) L. CalpurniusPiso causedthe EmperorTiberius considerable embarrassmentbecauseof his independentattitudes and free speech.Once,accordingto Tacitus,Piso threatenedto retire from politics in protestagainstthe prevalenceof judicial corruptionand other abuses.Tiberius was able to mollify him temporarily:5 The samePiso before long gave just as vivid proof of his free and passionatespirit by summoningto court Urgulania, whosefriendship with Augustahad lifted her beyondthe reachof the laws. Urgulania did not obey the summons,but, scorningPiso, rode in a carriageto the Emperor'shouse.But Piso held his ground, despiteAugusta'scomplaintsthat her majestywas being draggedin the dust. Tiberius, thinking he should so far gratify his mother as to say that he would go to the praetor'stribunal and supportUrgulania, set out from the Palatium. His soldiers had orders to follow at a distance.The mob rushedup to look, as Tiberius, his face composed,and pausingevery now and then for conversation,madehis leisurely way, until, Piso's relations being unableto restrain Piso, Augustaissuedordersfor the paymentof the money he was claiming. This endedan affair from which Piso emerged with glory and Caesarwith an enhancedreputation. Tacitus told this story for the moral it conveyedaboutthe way in which senatorsshould stand up to emperors. Our present interestis lessin Piso'sboldnessandthe way the casewas resolved The honestioresJhumiliores formula is only one of severalthat occur in the legal sources.It is in fact confined to the 'Sentencesof Paulus',a late third-centurycompilation, which useshardly any other terms. An earlier juristic writer once employshonestiorJhumiliorisloci, a close approximation. SeeDig., xlviii. 5. 39. 8. 5 Tac., Ann., ii. 34; cf. iv. 21 fr.

4

143

PETER GARNSEY

thanin the attitudesofUrgulaniaandLivia and, in particular,in a structuralweaknessof the Romancivil-law procedurewhich the incident reveals.It was for a plaintiff, and not the authorities of thestate,to seethatthedefendantappearedbeforethepraetorat the start of a judicial action. However, the would-be plaintiff might find himself in difficulties if his opponentwere strongerthan he (or had strong supporters)and resistedor ignored the summons. Similarly, executionof sentencelay with the individual ratherthan the state. Later we will ask how far the upper classesmade use of the biased nature of the systemof self-help as outlined here.6 It is doubtful whether Urgulania's behaviour was typical of that of membersof the privileged classeswhenthey were facedwith legal actionsbroughtagainstthem by their social inferiors. It might be claimed that there was no need for an upper-classdefendantto flout thelaw-andit mightbeprudentof him not to do so-asthere was a good chancethat the natural prejudicesof the judge, who invariably camefrom the samesocial milieu, might operatein his favour.

(2.) A letter of Pliny the Younger gives someforce to this last suggestion.The setting of the letter is provincial rather than Roman,which servesas a reminderthat preferentialtreatmentin the law courts was given not only to officials of the central administration(senatorsandequestrians)but also to the provincial and Italian aristocracy,the 'curial' class.7 Pliny was writing to Calestrius Tiro, who was on the point of taking up office as governorof Baeticain Spain(in successionto anothergovernorship). As governor,Tiro would be personallyresponsiblefor the administrationof justice in his province. The letter runs:8 You have done splendidly-andI hope you will not rest on your laurels-incommendingyour administrationof justice to the provincials by your exerciseof tact. This you have shown particularly in maintainingconsiderationfor the best men,9but, in doing so, winning the respectof the lower Seepp. 147-8 below. The legal sourcesdo not give a completelist of those who benefited from legal privilege; but theseincluded senators,equestrians,decurions, veteransand soldiers(and their families). 8 Pliny, Epist., ix. 5. 9 Honestissimum quemque.The adjectiveis ambiguous,implying both high 6 7

144

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

classeswhile holding the affection of their superiors.Many men, in their anxiety to avoid seemingto show excessive favour to men of influence, succeedonly in gaining a reputationfor perversityand malice. I know there is no chanceof your falling prey to that vice, but in praising you for the way you tread the middle-course,I cannothelp soundingas if I were offering you advice: that you maintain the distinctionsbetweenranks and degreesof dignity. Nothing could be more unequalthan that equality which resultswhen thosedistinctionsare confusedor broken down. Tiro had achievedwhat might seemto us the remarkablefeat of payingduerespectto dignitas(socialposition)while avoidinggratia (excessivefavour).l0 Tiberius was lesssuccessfulin observingthis precariousdistinctionin his supervisionof judicial affairs in Rome. This at any rate was Tacitus' judgement:l l The investigationsin the Senatewere not enoughfor Tiberius. He sat in the courts, at the side of the tribunal, so as not to drive the praetorfrom his chair. In his presence, many judgmentswere reachedwhich disregardedthe bribes and pressureof the powerful. But in his concernfor truth, Tiberius underminedliberty. (3) In his treatise on criminal investigations, Callistratus, a jurist of the Severanage(A.D. 193-235),dealt with the questionof the trustworthinessof witnesses:12

It is especiallyimportant to examinethe statusof eachman, to seewhetherhe is a decurionor a commoner;to ask whetherhis life is virtuous or marredby vice, whetherhe is rich or poor (for poverty might imply that he is out for gain), and whetherhe is personallyhostile to the man againstwhom moral characterand social and political ascendance.SeeCicero, Brutus, 282, and n. 21 below. 10 On gratia, seeJ. N. L. Myres, 'Pelagiusand the end of Roman rule in Britain', IRS, 1 (1960), pp. 24 If. 11 Tac., Ann., i. 75. 12 Dig., xxii. 5. 3. preface(decurionswere the membersof the councils of cities in Italy and the provinces).The key word is fides, trustworthiness. Cf. Gellius, NoctesAtticae, xiv, 2. 10, where the claimant is said to be well-endowedwith fides ('fidei ... plenum'). Seep. 149 below.

145

PETER GARNSEY

he is witnessingor friendly to the man whosecausehe is advocating.... The questions range widely, covering not only the witness's interestin the trial, but also his characterand socialand economic condition. By contrast,in a moderncourt of law in a democratic society it would be thought relevant (in theory at any rate) to establishonly a witness'simpartiality-unlessa previous convic13 tion for perjury should bring his characterinto consideration.

III THE CIVIL SUIT

In Rome the private claims of possession,breach of contract, damage,fraud or injury (to mention only the most important) were settled by the civil law according to the 'formulary' procedure.The civil suit (or actio) consistedof two stages,actio in itlre before the praetor,and actio apud iudicem before a private judge. The praetor was responsiblefor appointing the judge, with the agreementof both parties,and for passingon to him aformula, or a list of instructionssetting out the factual and legal groundson which the casewas to be decided.14 A man who sought a civil action, therefore, approachedthe praetor. The praetor, however, was not obliged to grant him an action, and indeedwas expectedto withhold from a man of low rank (hllmilis) an action for fraud againsthis social superior.15 The authority for this is Labeo, the Augustanjurist, whosecommentary on the praetor'sedict, in which a statementto this effect appears,was intendedas a guide to magistratesand legal practitioners. The commentary probably contained little original material,for the statementsof jurists were commonlybasedupon Callistratusgoes on to refer to Hadrianic rescripts:see p. 149 below. A rescript was a reply by the Emperor to the consultationof a governor or anotherofficial. It was meant as an order, and taken as such, probably from the early Empire. 14 SeeF. Schulz, ClassicalRomanLaw (Oxford, 1951), pp. 19 If. IS Dig., iv. 3. II. I. The praetor'spermissionwas neededfor the issuing of any summonsagainstparents,patrons,magistrates,priests, etc.: see ibid., 11. 4. 2 and 4. The recurrenceof the word reverentia in ibid., 13 and Dig., xlviii. 19. 28. 4 is suggestive.

13

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

or were explanatoryof tendenciesor normsalreadypresentin the legal system. It is probablethat the samerule which existed in relation to the action for fraud appliedin respectof other actions which carriedinfamia, or loss of status. The mostdangerousof theseactionsfor the higherordersin the Roman state was the action for debt. The consequences of this action, which was in effect a bankruptcy suit, included sale of property,loss of status,andpolitical extinction. The ruling classes appearto have relied on the aristocraticprejudicesof the praetor to ward off from themselvesthis disastrousactionuntil the time of Julius Caesar,when a law was passedpermitting impoverished aristocratsto cedepart of their land to their creditorswithout loss of status.Moreover, under the early Empire, a milder alternative to the venditio bonorum(sale of property) which was the lot of the ordinarydebtorwas setup or confirmedby senatorialdecree.This was distractio bonorum (division of property). By this procedure, a part of the propertywas sold by a specialagent,again without loss of statuson the part of the former owner. Distractio bonorum was available to men of high status (clarae personae),especially membersof the senatorialclass,16 The plaintiff surmountedone barrierif the praetorgrantedhim an action; the next obstaclewas provided not so much by the magistratewho administeredthe civil law as by the civil law itself. The formulary procedureoperatedon the assumptionthat any plaintiff could get his opponent to court. It is true that some resistanceon the part of defendantswas bargainedfor, but the measureswhich were intended to cope with this eventuality (praetorianinterdicts and the order for the seizureof the defendant's property preparatoryto sale) did not include the use of physical force by the o"rgans of the state. It is not difficult to imagine situations where a defendant, relying on his greater strengthor influence,might defy the would-beplaintiff: it will be recalledthat Urgulaniarefusedto go to courtandsoughtsanctuary in the palace,in the knowledgethat Livia would protecther.17 For the backgroundand context of Caesar'slaw de bonis cedendis,see M. W. Frederiksen,fRS, lvi (1966), pp. 128 If. For the (first-century) decree,Dig., xxvii. 10. 5. 17 The problem discussedin this paragraphis posedforcefully by J. M. Kelly, in RomanLitigation (Oxford, 1966), ch. I; reviewed by J. Crook and R. Stone,CR, xvii (1967), pp. 83 If.

16

147

PETER GARNSEY

On the other side it must be acknowledgedthat the plaintiff of low statuswas not necessarilyon his own in his efforts to get his opponentto justice. He might have a patron of greaterinfluence and more substantialphysical resourcesthan himself who could perhapsbring pressureto bear on his opponent.In the case of Urgulania it was the defendantwho possessed a powerful patron. Livia's friendship raised Urgulania 'above the laws'. If the defendant still failed to respond, the plaintiff could look for assistancefrom a different quarter: with the aid of the local auctioneer,who had a financial interest in the matter, he might enforcea seizureof the defendant'sproperty.IS A third ally of the plaintiff was the Roman social conscience.Romans in general consideredit important to maintain their good namein the community and their standing with the magistrates.Probably few men were preparedto flout conventionand the law by refusingto obey a summons.But there were exceptions:why else did the Julianlaw on violencecoverthe caseof thosewho resisteda courtsummonswith the aid of a gang of thugs?19 Even if the plaintiff of low rank had beengrantedan action, and had securedthe appearanceof his opponentbeforethe praetor,he could not havehad much confidencein the outcomeof the action. For a man of influencewould standa good chanceof winning his case,even without corruption or the threat of force. Judgesand juries (wheretherewere juries)20wereeasilyimpressedby qualities such as social prominence,wealth and good character,21and this 18 The latter could be rich (Cic., pro Caecina, 10) and strong (ibid., 2.7). 19 Dig., xlviii. 7. 4 (Julian law on violence-?of Julius Caesar).Later it becamea criminal offence to disobey the injunctions of magistrates:see ibid., xlviii. 19. 5. preface. 20 The juries for the public criminal courts of the Republic were drawn from the upper classes.The adviserswho assisteda judge or magistrate in reachinga decision in an exira ordinem court characteristicof the Empire were similarly drawn from the higher orders. 21 For the Romans,good characterwas in part a matter of birth and social position. It is significant that a word with a primarily moral meaning, honeslus,should have been chosento describemen of distinguishedsocial position, honesliores.(Note that moral words such as boni and optimales were used to describepolitical groups under the Republic.) Quintilian, Instil., v. 10. 2.4, noted that there is sometimesa causalconnection betweenbirth (genus) and mannerof life: ' ... et nonnumquamad honesle turpiterquevivendum inde causaefluunt'. Cf. Dig., xlvii. 2. F. 2. I, where honeslusvir is synonymouswith vir locuples('rich man').

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

was thought perfectly proper. When Aulus Gellius judged a money-claim, he could not bring himself to decide against a plaintiff who was upright andtrustworthyin favour of a defendant who was in his eyesa rascal,and 'not possessed of good fortune' (non bonae rei), even though the former's case was quite unconvincing. He took the problem to the philosopher Favorinus. Favorinus quoted words of Marcus Cato to the effect that, if a causewas evenly balanced,the judge should decidewhich of the two parties was'better' (melior), a term ambiguous as between characterand property, and should come down for him. Cato added that this was not a private opinion of his own, but a traditional Romanattitude. But Gellius could not declarefor the claimant, and backedout of the case,pleadingyouth and lack of merit.22 Cato had held that the outcomeshoulddependon the character of the partiesonly if witnesseswerelacking. Gellius'plaintiff could produce 'neither documentsnor witnesses'.The excerpt from Callistratus quoted above23 shows that, where witnesseswere called, the judge reckonedtheir social position and characteras relevant as the quality of their evidence. Nor was this merely Calli stratus' private judgment. He went on to cite a series of rescripts of Hadrian (A.n. 117-138) in support of his statement. The burdenof the rescriptsis that a witness with dignitas, existimatio and auctoritas was especially acceptable.These are three upper-classvirtues: socialstanding,goodreputation,andprestige. To sum up, as long as plaintiffs could haverecourseonly to the praetorand a private judge,both of whom sharedthe feelingsand prejudicesof the upper classes,and as long as private summons andprivateexecutionwerethe rule, actionsby commonersagainst membersof the upper classeswere not likely to be frequent.24 The end of the Republic saw an improvementin the lot of the humbleplaintiff in severalrespects.A clauseon violencefrom the Lex Julia, already referred to in anotherconnection,shows that Gellius, Nocle! Atticae, xiv. 2. Above pp. 145-6. 24 Kelly, op. cit., pp. 62 ff. On private executionand its difficulties, ibid., pp. 12 ff. In addition, many matterswould have been settledout of court. Cicero often brought pressureto bear on judgeson behalfof clients, who as a result were sometimesable to avoid court-cases.See e.g. adfamiliares, xiii. 54. Cf. the Piso/Urgulaniaaffair: as far as we can tell this never reachedthe courts. 22 23

149

PE'TER GARNSEY

forcible resistanceto a summonswas viewed as a crime in the early Empire.25 Crimes were punishable by the state, so that neither summons nor execution was the responsibility of the individual. Further, it becamepossible under the Empire for a plaintiff to approachthe city prefectand to seekthe settlementof a claim outside the formulary procedure.26 Again, summonsand executionfell to the state-andthe plaintiff might gain a fairer hearing. Before concludingthis sectionon the civil law, we may return oncemore to the role of the praetor.The praetorwas in a position to discriminateagainstthe weakerparty both as the controller of the sanctionsbehind the judicial summonsand execution,and as the granteror withholder of actions. In addition, during the first stagesof somesuits, notably suits for injury, he fixed the amount of compensationpayable in the event of the court's deciding againstthe defendant.Now, accordingto Labeo,thepersonaof the injured party might, presumablyby the praetor'sdecision,convert an action for injury into an action for 'grave' injury, with higher penalties.27 The word personahas abroadconnotation,as is shown by statementsof later jurists which expandLabeo'ssentence.We read,for instance,in the 'Sentencesof Paulus',a late-third-century compilation, the following :28 Injury is regardedas grave... with respectto person, wheneverthe injured party is a senatoror equestrianor decurion,or someoneelse of conspicuousprestige:for example,if a plebeianor a man of low birth has injured a senatoror equestrian,or if a decurionor magistrateor aedile or judge has sufferedat the handsof a plebeian. Other passagesshow that injury by a child againsta parentor by a freedmanagainsta patronwere also automatically'grave'. If all the texts dealing with this matter are put together,it can be seen that in Romansociety,somepersonsweregiven specialprotection becauseand in so far as they performedcertainsocial, political or judicial roles (as parents,magistrates,and judges, respectively), and othersbecauseof personalstatus. See n. 19 above. See e.g. Dig., i. 12.2: the intrusion of the prefect'scourt into the sphere of pecuniarycases,at least from Hadrian'stime. 27 Dig., xlvii. 10. 7. 8. 28 Sent. Pauli, v. 4. 10; cf. Gaius, Instit, iii. 225; Instit. Just., iv. 4· 9.

25

26

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

IV THE CRIMINAL TRIAL

In the late Republic, laws were passedwhich definedas 'public', andtherefore'criminal', offencessuchas extortion, treason,homicide, forgery, peculation,electoralbribery andviolence,andwhich set up permanentjury-courts to try them. The jury-courts (or quaestiones)survived into the Empire, and indeedAugustushimself was behind the legislation which establisheda court for adultery.However,the form of criminal proceedingmost characteristic of the Empire was the cognitio or extra ordinem procedure, which placedthe.whole trial, including the passingof sentence,in the hands of the judge. (The praetor in charge of a jury-court merely chairedthe proceedingsand was boundby the verdict of the jury.) The jury-court system proved less durable than the formulary systemin civil law, althoughboth were Republicanin origin. As late as the Severanperiod, those seeking redressfor private injury or loss could still turn to the formulary systemas a viable alternativeto cognitio. By contrast,thereis no evidencethat 'public' offenceswereeverinvestigatedby a jury-court after about the turn of the first century A.D.29 The civil action by the formulary processand the criminal trial by the cognitio procedurewere very different in structure,and this differenceis reflectedin the patternsof discriminationwhich were characteristicof each. But two qualifications should be kept in mind. It is probable that criminal judges, especially provincial governors, used their discretion in granting and withholding trials, as praetorsdid whenadministeringthe civil law. Again, the influencesto which a private judgewas subjectedin the courseof a trial were felt equally by a criminal judge. On the other hand, discriminationin the apportionmentof casesto courts was a distinctive featureof the Imperial criminal law. In addition,therewas much more scopein a criminal trial by cognitio for the variation of penalty according to the social class of the defendant. The Republicansystemof criminal law knew neitherof thesemethods of discrimination. Only one set of courts was provided for the 29

SeeIRS, Ivii (1967), pp. 56-60.

PETER GARNSEY

trial of all defendants,and no penaltieswere inflicted other than thosefixed by the laws which set up the courts.30 Under the Empire, however, several courts were capableof hearing most criminal cases,and the choice of court was sometimes of consequence for the defendant.The senatorialcourt was notoriously'soft' on senatorialdefendants.Chargesagainstequestrian officials were heardby the Emperor,to whom theseofficials were personally responsible,and he might be expectedto have their interests at heart.3! Again, by the late-second century, governorswere not allowed to executea sentenceof deportation or capital exile. In practicethis meantthat capital casesinvolving membersof the provincial elite were automaticallyreferredto the Emperor.32 Other courts in Rome, the jury-courts, for as long as they existed, and the court of the urban prefect, must have dealt principally, though not exclusively,33with lower-classoffenders. The cognitio procedurewas notablefor the degreeof freedomit left to the judge. Not only the examinationof the partiesand their Defendantsbefore jury-courts might be favoured(and it is commonly assumedwere favoured, if of senatorialif not equestrianrank) in two principal ways. Either the court could bring in a verdict of not guilty when an acquittal was not justified (this form of favouritism was not, of course,peculiar to a jury-court); or the magistratescould permit a defendantto go into voluntary exile and so avoid the necessityof having him executed(in capital cases). 31 Senators:see Pliny, Epist., ix. 13. 21. Pliny claims to have causedthe indignation of 'the other orders' to subside.The indignation arosefrom the fact that 'the Senatewas harsh to others and lenient only to senators, as if by mutual connivance'. Equestrians:for an early exampleof the trial of an equestrianby the Emperor(Claudius), see Dio, Ix. 33. 6. In a later incident (Pliny, Epist., vi. 31. 7 if.), Trajan angrily rejectedthe imputation that an Emperor was likely to be lenient to his own officials. In this casethe defendantwas a freedman-procurator(it is unclear whether the co-defendant,an equestrian,was an Imperial official or not). But another'good' Emperor, Augustus,had shown favouritism to Licinus, a freedman-procurator (Dio, liv. 2 I). (A procuratorwas a financial agent responsibleespecially for tax-collection.) 32 Dig., xlviii. 22. 6. 1. 33 See Tac., Ann., xiv. 40-4r, for both prefect and jury-courts. The defendants,who included senators,might have been tried either by the prefect or by the jury-court for fa/sum (forgery)-their crime was testamentaryforgery. The prefect also had the right to deport, the ius deportandi: see Dig., i. 12. 1. 3; xxxii. 1. 4; xlviii. 22. 6. r; etc. Deportationbecamean upper-classpenalty (seep. 153 below).

30

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

witnessesand advocates,but also the choiceof penaltywas in his hands.The laws which setup jury-courtshadboth definedcertain crimes and prescribedfixed penalties for anyone convicted of them in thosecourts. The cognitio judge was not restrictedin this way. 'Today',wrote Ulpian early in the third century,'he who sits in judgmentin a criminal caseextra ordinemmay issuethe sentence which pleaseshim, be it relatively severeor relatively mild, so long as he stayswithin the limits defined by reason.'34 Thereis a further point. The judge,in choosinga penalty,could go beyondthosefew which had gainedrecognitionin the law, the straightdeathpenalty,interdiction from fire and water(a form of exile), and the monetaryfine. He could apply sanctionswhich had beenusedunder the Republic mainly as administrativemeasures againstfree aliens or slaves. The 'dual-penaltysystem',familiar to jurists of the age of the Severans,recognizedthe distinction between 'legal' and 'nonlegal' penalties.In practice, each group of penaltieswas aligned with a broad social category, such that membersof the upper classes,or honestiores,sufferedonly penaltiesdrawn from the first group, and membersof the lower classesonly penaltiesfrom the secondgroup.35'Deportation'and 'relegation',two forms of exile, were standardpenaltiesin the first group.36The former deprived the condemnedof citizenship but not freedom, and the latter of neither.Execution,which was rare for honestiores,was by decapitation. The money-fine,oncethe only penaltyin civil law, was used as a commonminor sanction.Expulsion from the senate(in the case of a senator) and from a local council (in the case of a decurion),and prohibition from office-holding,were also known. The most serious 'lower-class penalty' is called by the jurists summumsupplicium('the highestpunishment').The term stoodfor aggravatedforms of th~ deathpenalty,including exposureto wild Dig., xlviii. 19. 3. The judge'sfreedom in choosingpenaltieswas to a certain extent restrictedby Imperial rescripts,at least from the turn of the first century. SeeF. M. de Robertis, Zeitschrift Savigny-Stiftung,lix (1939), pp. 219-60. 3S Most of the penaltieslisted below are mentionedand gradedin an edict ('upper-classpenalties',Dig., xlix. 19. 28. 13) and a rescript ('lower-classpenalties',ibid., 28. 14) of Hadrian. 36 Relegatiobecamethe 'legal' penalty for adultery under Augustus. See Collatio, ii. 26. 14. Deportatio, strictly speaking,was not a 'legal' penalty, but was similar to, if it did not succeed,interdiction.

34

153

PETER GARNSEY

beasts,crucifixion, and deathby fire. Next, condemnationto hard labour in the mineswas for life, and the condemnedwas reduced to a statusakin to slavery. Condemnationto live and fight as a gladiatorwas just as degradingand carrieda greaterrisk of death. A lessseverepenaltyof the sametype was labouron public works. Corporal punishmentwas also reservedfor humiliores. Torture was, by tradition, applied only to slaves. But legal texts which forbid the use of torture for certain classesof free men indicate both that free men were not immunefrom torture in the middleand late-secondcentury, and that only well-connectedfree men were consideredworthy of protection against it.37 As for the treatmentof accusedmen before trial, honestiorescould generally avoid imprisonment.They might be entrustedto guarantors,or soldiers,or 'to themselves',thoughnot if the chargewas serious.38

v LEGAL PRIVILEGE: EVOLUTION AND BASIS IN LAW

Much of our knowledgeoflegalprivilege in Romeis derivedfrom the writings of the Severanjurists. But legal privilege was not a phenomenonpeculiar to the Severanage. The systemwhich the jurists describeevolved in my view gradually in the course of the first century, and was well-establishedin legal practicein the second.This was true as much of the 'dual-penaltysystem'as of other forms of discrimination. In an important article39 published in 1950, G. Cardascia expresseda rather different opinion on the developmentof one Seeespec.Cod. Just., ix. 41. II; Dig., xlviii. 18. 15. preface;1. 2.. 14; cf. xlviii. 18. 9. 2.. All texts but the secondrule out torture for various privileged groups; the secondimplies that a freemancan be tortured if his statementsas a witness are inconsistent.It is unlikely that lower-class defendantsgenerallysuffered judicial or inquisitorial torture (as distinct from 'third-degree'torture, torture as a punitive measure)in the first century and early second(seee.g. Pliny, Epist., x. 96. 8 (slave women were tortured, but not thoseprisonerswho were free). But the Christiansat Lyons in A.D. 177(?) were certainly tortured: Euseb., Eccl. Hist., v. I. 38 Dig., xlviii. 3. I and 3. 39 'L'apparitiondansIe droit desclassesd' "honestiores"et d' "humiliores", Revuehistorique du droit franfais et itranger, xxvii (1950), pp. 305-37, 461- 85.

37

154

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

aspect of the system of privilege, differential punishment. He grantedthat variation of penaltywas a reality in legal practiceas early as the reign of Augustus(2.7 B.C.-A.D. 14), but claimedthat a judgewasunderno obligation to makeconcessionsto rank until the time of Pius (A.D. 138-160),when, accordingto him, distinctions began 'to pass into the law'. Subsequentlyvariation in penaltybecamefrequentin 'legislation',until under the Severans it took on 'a generalcharacter'.40 By 'the law' and 'legislation' are apparently meant Imperial constitutions,the rescriptsand edicts of Emperors.It is not clear whether Cardasciaholds that the rescripts of Pius and later Emperors'established'the systemof privilege in a newer, more highly-developedform; or whetherthe rescriptsare supposedto have given a systemwhich was alreadycompletea new validity, by the fact that they refer to it and therebyindirectly sanctionit. Both ideas deserveinvestigation. Thoserescriptsissuedby Pius and his successorswhich are at all relevant to the question of differential punishmentscan be divided into threegroups.The first groupcomprisestwo rescripts, both of Pius,which setforth how two crimeswereto be punished, the murder of an adultererby the aggrievedhusband,and thefts from imperial mines.41 The secondgroup consistsof four rescripts banningthe useof certainpenaltiesagainstchildrenor descendants of certainprivileged groupS.42Finally, four rescriptsdirectly concern decurionsor soldiers.43 It can be seenthat the rescriptsfail to cover the whole field of penalties,crimes and privileged categories.Insteadof a general edict launching the dual-penalty system, there are two judicial decisionsin which a differential penalty-scaleis employed.Again, the fact that veterans,for example, are favoured has to be inferred from the sentencingof their sonsto exile ratherthanforced labour of any kind.44 Further,it is nowhereexplicitly statedby an Emperorthat decurionswere not to be condemnedto the mines. But the Emperor Severus Alexander (A.D. 2.2.2.-2.35) informed Ibid., p. 471. Dig., xlviii. 5. 39.8; cf. xlviii. 9. 1.5; xlviii. 13. 8. 42 Cod. Just., ix. 41. II. preface(includestorture); ix. 47.5; 9; 12. We might add ix. 41. 8, preface;but the referenceto soldiersis primary, so it is included in the third group. 43 Seenotes 47-50 below. 44 Cod. Just., ix. 47.5. 40 41

155

PETER GARNSEY

Demetrianusthat, as his mother was the daughterof a decurion, she was not to suffer that fate.45 There is a second, related, difficulty. The rescripts, far from establishinga system of privilege, were apparently designedto preserveone which alreadyexisted.For example,all the rescripts in the third group reaffirm known regulations, or simply make referenceto them.46 The first of the SeveranEmperors,Septimius Severus(A.D. 193-2II), in a reply to the petition of Ambrosius, cited the rule(prohibitumest)that decurionswere not to bebeaten.47 His son and successorCaracalla(A.D. 197-217)wrote to Geminius that decurionspatently(manifestumest) should not be sentencedto forced labour.48 There was plenty of precedentfor Diocletian's protection of soldiers from torture and 'plebeian' penalties.49 Finally, Marcus and Verus (A.D. 161-9), when consulted by a governor on the proper penalty for a certain Priscus,wrote that decurionsought to be relegatedor deportedfor capital crimes.50 Nor was this a novelty: a ruling of Hadrian which prohibited executionand prescribedexile for decurionswho were murderers (exceptfor parricides)had implied as much.5! The first relevant 'legislation', indeed, was issuednot by Pius, but by Hadrian. In addition to the ruling just mentioned,there is record of a judicial decision of A.D. I 19-of the sametype as the rescripts of Pius in the first group referred to above-in which Hadrian stated that exile (relegatio) was the proper penalty for splendidiorespersonae(which standsfor honestiores),and two years' labour on public works (opuspublicum) and a beatingfor alii ('the rest', equivalentto humiliores),for the offenceof moving boundarystones.52 We must therefore ask whether either or both of Hadrian'srescripts53 were innovatory. Cod. just., iX-47.9. In the secondgroup, only the edict of Marcus (Cod. jllSt., ix. 4I. I I. preface)might have addedto the ranks of the privileged class (in a very small way)-unlessthe edict was intendedto re-statean acceptedrule which had been disregarded,or to make explicit an exemptionwhich had until then been only implicit. 47 Cod. just., ii. I I. 5. 48 Ibid., ix. 47· 3. 49 Ibid., ix. 4I. 8. preface; cf. Dig., xlix .. 18. I and 3. 50 Dig., xlviii. 22. 6. 2. 51 Ibid., xlviii. 19. 15. 52 Ibid., xlvii. 2I. 2; cf. Collatio, xiii. 3. 1-2. For Cardascia'sdiscussionof the text, seeart. cit., pp. 468-9. 53 Dig., xlvii. 2I. 2, and xlvii. 19. 15. 45

46

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

The rescriptover boundary-stones, if it was the first of its kind, may have actually 'set up' the penalty differential. The grounds for assertingthis are purely and simply that no earlier rescript is recordedin the Digestand other legal compilations.54 But chance playedan importantpart in the selectionand preservationof such texts as have survived. Another band of compilers, working in A.D. 530-3 (when the Digest was compiled) or at another time, might have come up with a different set of rescripts, including somefrom an earlierperiod. It is obviousthat neitherthe classical jurists from whose works excerptswere drawn, nor the Justinianic compilers, had the intention of describingthe evolution of class discriminationin the law.55 As for the secondrescript, it was genuinely innovatory if it elevateddecurionsas a class.But it must be said that it does not look like an edict which deliberatelyconferreda new statusand a new set of privileges,andit was very limited in applicationsinceit concernedonly murder cases.56 It cannotthereforebe seenas, for instance, one of a series of edicts exempting decurions from specific penalties.Nor is thereany indication that it announcedor constituted a departurefrom previous policy. Pliny's letter to CalestriusTiro, quoted in full above,57belongs to the previous reign (it was perhapswritten in 107-8), and is importantevidence for the care with which the Romanadministrationdistinguished betweenthe local aristocracyand the restof the provincial population. Nor was this a new attitude, originating in Trajan's reign (A.D. 97-117) and voiced first by his officials. The wealth, social status and political usefulnessof the provincial gentry had long beenevident.58 Thesewere the qualities which madeit inevitable It is held by d'Ors in Les empereursromains d'Espagne(Paris, 1965), pp. 147 ff. and Gaudemetin FestschriftE. Rabel(Tiibingen, 1954), ii, pp. 169 ff. that Hadrian'srescriptswere the first 'true' rescripts,'des rescrits proprementdit'. This doctrine rests on one text, Vita Macrini, 13. I, which shows that Trajan was averseto sendingrescriptsin reply to the petitions of private individuals (libelli), but not that he was unwilling to send rescripts(of the Hadrianic type) to functionaries. There are in fact numerousexamplesof the secondtype of rescript from the pre-Hadrianicera. ss For the aims of the latter, SeeConst. Tanta, 10. 56 This is an inferencefrom the referencesin the text (Dig., xlviii. 19. 15) to the 'poenalegis Corneliae(sc., de sicariis)" and to parricide. 57 Above, pp. 144-5. 58 On class division in Greek cities under the Republic and the Roman attitude to it, see J. Briscoe, 'Rome and the class struggle 200-146 B.C.' chapterIII above. 54

157

PETER GARNSEY

that their most conspicuousrepresentativeswould gain the citizenship and advanceinto the higher orders. The actions and attitudesof courts had long respectedthesesamequalities.It was expected of judges that they would discriminate in favour of upper-classprovincials. Thus, Hadrianmay havebeennot so much raising decurions,or leadingprovincial notables,to a new status,as confirming themin one they alreadypossessed.This would havebeenan appropriate measureat a time whenpenaltieswerebecomingmore severe,and the newer,degradingsanctionswere coming into wider use-and when governorswere no less inclined to arbitrary behaviourthan their predecessorshad been. The Emperor'sbiographerreports that Hadrian, in his travels throughoutthe Empire, made it his business to correct abuseswhere he found them. Governors, amongstothers,were punished.59 Their misdemeanorsmay have included acts implying an unwillingnessto make any distinction between decurions on the one hand, and the rest of the free populationin the provinceson the other.6o The 'dual-penaltysystem',then, doesnot seemto havebeenset up by formal enactments,or, if it was (and this must be counted unlikely), those enactmentsare lost to us. Nor, for that matter, doesa law survive which gavesenatorsthe prerogativeof trial by their peers or which maintainedthat privilege. Further, to my knowledge, the inequitable aspects of the formulary system, referredto above,61were not set up or sanctionedby any edict. It remainsfor us to considerbriefly the moremodestsuggestion that the rescriptsgavethe systemof discriminationa new validity in law. This theory has unsatisfactoryimplications. Our informaVita Hadriani, xiii. 10. H~drian was a notorious 'busybody':see Dio, lxix. 5. 1. 60 Here is a summaryof the evidencefor the dual-penaltysystemin the time of Hadrian: the arrangementof penalties'legal' and 'non-legal' into two distinct groups is not attestedbefore Hadrian: see Dig., xlviii. 19. 28. 13-14. Hadrian further recognizeda correspondence betweena categoryof penaltiesand a social class: seexlvii. 21. 2. Dig., xlviii. 19. 15, indeed,implies the existenceof a whole systemof differential punishments.The fact that Hadrian'ssuccessorsmadefew additions to the systemindirectly confirms that the systemwas well-establishedunder Hadrian. Dig., xlvii. 14. 1. prefaceand xlviii. 8. 4. 2 show, at the most, that the dual-penaltyscale was not yet applied universally, in the punishmentof every crime (cf. ibid., xlvii. 9. 4. 1, for Pius). 61 Above, III, The Civil Suit.

59

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

tion on the systemof differential penaltiescomes to us through both Imperial rescripts and juristic generalizations.But, as we have seen,no Imperial pronouncementsare preservedwhich so much as mentionthe bulk of upper-classprivileges.It would seem to follow that only a few of theseprivileges were well-basedin law. The difficulty can be resolvedif it is askedwhat was in the mind of a jurist when he cited an Imperial rescript,and also what was the Emperor'smotive in issuing the rescript. Callistratusdid not cite Hadrian's rescripts on witnesses,referred to above,62 becausethey authorized judges in a new and superior way to favour upper-classwitnesses;nor was that either the purpose,or the effect of the rescripts.Again, Hadrian'srescripton boundaries (quoted by Callistratus) shows that he believed in the principle that criminals of different social status should be punisheddifferently; but he did not intend, by issuingthe rescript,to give that principal a legal standingwhich it did not have before, nor was that an indirect result of the publication of the rescript. The rescript set up new penaltiesfor an old crime, penaltieswhich still held good in Callistratus'day.63 In short, differential punishment was already a feature of the judicial system,and had been tacitly if not overtly approvedand sanctionedby judges, jurists and Emperors.

VI THE ROMAN CITIZENSHIP

The Romansrecognizedin their society distinctions other than that between the upper class and lower class (honestiores and humiliores); for example that between freemen and slaves, and between citizens and aliens. The relationship of these to that betweenhonestioresand humilioreswill now be considered. For writers of juristic handbooks,the distinction betweenthe freeman and the slave was 'the basic division in the law of persons'.64But it is evidently not assimilable to that betweenthe honestioresandthe humiliores.Humiliores, the lower classes,included Above, p. 149. Dig., xxii. 5. 3. 1 If. (witnesses);xlvii. 21. 2 (boundaries). 64 Gaius, Instit., i. 9. For a reflection of the division in penal law, Dig., xlviii. 19. 28. 16. 62

6J

159

PETER GARNSEY

slaves,freedmen,and also any men of free birth who lacked the criteria for legal privilege, that is, the dignity and prestigeassociatedwith good birth and characterand the possessionof wealth and office. The distinction betweencitizen and alien requirescloseranalysis. It has been suggestedthat it was replacedby that between honestioresand humiliores, which is said to have arisen in the late first or early second century.65 Citizens held several important advantagesover aliens. They could seekthe help of a tribune, or exercise their right of appeal, against the arbitrary actions of magistrates.Aliens, strictly speaking,had no standingwithin the civil law, and hencelacked theserights. In Jerusalemin the reign of Nero (A.D. 54-68), St Paul stavedoff a beatingand struck fear into a tribune with the words: 'Is it lawful for you to scourgea Roman citizen, uncondemned ?'66 St Paul had previously embarrassedthe magistratesof the colony of Philippi by disclosing his citizen-statusafter he and his companionshad beenbeatenand cast into prison.67 Considerablylater, peasantson an Imperial estatein Africa protestedto the Emperor Commodus(A.D. 180193) that, even though some of them were Romancitizens, they had been beatenby a procurator and various overseersof the estate.68 This incidentindicatesthat appealwas not a deadletter in the Antonine period. About a generationlater, Ulpian cited the section of the Julian law on public violence which dealt with appealin his treatiseon the duties of a proconsul,69and several sectionsof the Digest are devotedto appeal.70 Further, a citizen might expect to gain a less severesentence than an alien if both were defendantson the samecharge.In A.D. 17, the Senatetook firm measuresto stampout astrology:among the astrologers,citizens were exiled while foreignerswere put to death.71In A.D. 177(?), Christiansarrestedat Lyons were either A. N. Sherwin-White,RomanSocietyand RomanLaw in the New Testament (Oxford, 1963), p. 174. Seealso A. H. M. Jones,Studiesin Roman Governmentand Law (Oxford, 1960), pp. 64-5. 66 Acts 22 :24. 67 Ibid., 16:37. 68 Riccobono,Fontes Iuris RomaniAntejust., 2nd edn, i, n. 103, p. 496, lI. 10 ff. 69 Dig., xlviii. 6. 7. 70 On appeal,seefRS, lvi (1966), pp. 167 ff. 71 Col/atio, xv. 2. I; cf. Dio, lvii. 15.8, and Tac., Ann., ii. 32. 5.

65

160

LEGAL PRIVILEGE IN THE ROMA,N EMPIRE

beheadedor sent to the beasts,dependingon whetherthey were citizens or aliens.72 The preferentialtreatmentof citizens as opposedto aliens was thereforea long-standingform of legal discrimination,which persistedin the first two centuriesof the Empire. It was basedon the exclusionof aliens from the civil law (ius civile), but was practised also in other spheresof the law; for examplein the criminal law, which underthe Empire was subjectto the cognitio of magistrates, and which coveredboth citizens and non-citizens. The dividing line betweenthe honestioresand humilioreswas not identical with that between citizens and aliens, for there were citizens (and non-citizens)on both sidesof it. First, those with legal privilege included non-citizens,because decurionswere a privileged group, and not all decurions were citizens. Decurionsof cities of 'Roman'statushad citizenship.In 'Latin' cities,73however,citizenshipwas won by performanceof a magistracy rather than by mere membershipof a city-council. From the time of Hadrianor Pius it was possiblefor a 'Latin' city to obtain a higher grade of Latinity, Latium maius, and so gain citizenship for all its decurions.74 But the epigraphic evidence suggeststhat this had to be suedfor; apparentlyLatium maiuswas not conferred on all 'Latin' cities by an act of the central administration.Moreover,while Latium maiusundoubtedlyreduced the numbersof decurionswho lacked citizenshipin the West, it madeno impact on the Easternhalf of the Empire. Cities in the Greek Easthad neverbeenanxiousfor either 'Roman'or 'Latin' status;there,citizenshipwaspossessed by individualsandfamilies rather than by whole city-populationsor ruling castes.75 Secondly,the sourcesimply that citizenshipdid not excludethe humblefrom cruel andhumiliating penalties.TheAfrican peasants were, after all, beaten,and though this may have been strictly Euseb.,Erel. Hist., v. 1. 47. But Attalus, a citizen (§ 44), was sent to the beasts.Decapitationwas the least unpleasantand least degradingform of the deathpenalty. 73 'Latinity' in the late Republic and early Empire was a half-way stage betweenthe position of an alien and full Roman citizenship. 74 Gaius, Instit., i. 96; ILS, 6780 (Gigthis, N. Africa). " SeeA. N. Sherwin-White,RomanCitizenship(Oxford, 1939), pp. 194-257 esp. pp. 236 ff. It should be addedthat in no sourceis it statedor implied that only those decurionswho were citizens were entitled to milder punishments. 72

161

PETER GARNSEY

illegal, it could have beenexpected.The peasantscall themselves 'ordinary men of the countryside' (homines rustici tenues), a descriptionwhich finds an echo in Calli stratus:76 It is not normal for everybodyto be beaten,but only free

men whose standingis relatively low (hi . .. qui /iberi sunt et quidemtenuioreshomines).Membersof the upper classes (honestiores)are exempt,and this is emphasizedin Imperial rescripts. Callistratusmay havehad in mind a rescript of SeptimiusSeverus which pointed out that decurionswere not to be beaten.77 This was no new development.Severuswas citing an establishedrule (prohibitum est), and long before, in A.D. II9, Hadrian had ruled that the offenceof moving boundary-stones was to be punishedin part by beating,exceptwhen the culprits were splendidiorespersonae or honestiores.78 Citizens were in theory protectednot only againstbeating,but also against the executionof a death sentencein the face of an appeal.But decurions,andothermembersof theprivilegedclasses, were in the more enviableposition of knowing that the highest punishmentto which they were liable was exile, unlessthey were guilty of parricide or treason. In general, an explanation is required for the failure of the jurists to mention citizenship in those passagesin which they dealt with the criteria on the basis of which the privileged class was to be identified. The omission may be due to systematicrewriting of the classicallegal texts by later jurists andespeciallythe sixth-centurycompilers, for whom citizenship held little significance.But perhapsthe classicaljurists themselvesconsideredthat the distinction between humble citizens and free aliens was unimportant compared with the distinction between senators, equestriansor decurions on the one hand, and membersof the lower classes,whethercitizens or aliens, on the other. If this was the case,and the evidencealready adducedseemsto point that way, then Justinian'smen would have had no cause to make substantialalterationsin the relevanttexts. Dig., xlviii. 19. 28. 2. Cod. Just., ii. I I. 5 (A.D. 198). 78 Dig., xlvii. 2 I. 2.

76

77

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

The distinction between honestioresand humiliores, then, cuts acrossthat betweencitizens and aliens. It is, of course,a further stepto say that the former replacedthe latter. That would seemto imply that the former distinction was not important in the preHadrianic period; and, conversely,that the latter was once (perhaps in the first century) as fundamentalas the former became (perhapsin the courseof the secondcentury). There may be a close associationbetweenthe 'replacement' theory and the idea that the differential-penaltysystembeganto emergein the reigns of Hadrian and Pius. If so, it should be statedthat thereis good reasonfor believing that the differentialpenalty systemexistedin its essentialsby the reign of Hadrian;79 and that discrimination in favour of the upper classesinvolved more than the assigningof milder penalties. To reinforce the secondpoint, it may be helpful to pick out someforms of upper-classdiscriminationwhich were practisedin the courseof the first centuryA.D.SO The formulary procedurewas in its hey-day in the late Republic and early Empire. It will be recalled that it was administeredby praetorsand private judges with wide discretionarypowers,which they usedto the advantage of men of wealth and social position. Again, senatorsappearto have had the right to a senatorialtrial in the first century, while equestrianswent before the tribunal of the Emperor or the Senate,and the public courts and the urban prefect dealt mainly with lower-class criminals at Rome. Thirdly, in the matter of punishments,variation of penalty was practisedas soon as extra ordinem courts beganto operate,and it appearsthat while citizens were favoured above aliens, citizens of low rank lost in comparison with upper-classcitizens.s! This last statementis necesSeen. 60 above. The 'honestiores'/'humiliores' formula will not be found in first-century sources.This is doubtlesspartly due to the fact that those sourcesare largely non-legal: they are concernedwith individual events,affecting individuals, and do not generalize(except about politics). It should also be emphasizedthat even the classical jurists fail to use the formula (see n. 4 above). There is a perfectly good equivalentin an excerptfrom the Augustanjurist Labeo, in Dig., iv. 3. 1 I. 1: it is unnecessaryto hold that the words are Ulpian's and not Labeo's.Nor is it likely that for Labeo the phrase'qui dignitate excellet' applied only to ex-consuls. 81 Citizens/aliens:See n. 71, above; Pliny, Epist., x. 96. 4. Equestrians/others:ibid., ii. 1 I. 8: an equestrianis exiled, his friends (sc., of lower status)are executed:anotherequestrianis strangledafter 79

80

PETER GARNSEY

sarily cautious, given the nature of the evidence. The Digest is preoccupiedwith the post-Trajanic period, and, with few exceptions, the historical and biographicalauthorsare Rome-basedand caughtup in the political strugglebetweenEmperorand Senate. Hence information is lacking about the way in which penalties were normally applied to lesser men, including citizens of low station, and to provincials of all ranks.82 The conclusionis not that citizenshipwasworth nothing. There is someevidencethat the Emperorwas interestedin the welfare of citizens in the provinces,83although this does not mean that he was not preparedto come to the assistanceof free alienS.84 The rights of citizenswere respectedby someofficials at least,and the authors describe with disapproval occasions when they were the application of servile sanctions-butMarius Priscusapparently demandsa higher sum for this. Senators/others:Suet., Aug., v: a patrician(sc., a senator)asks the senatefor a milder penalty becauseof his age and birth; Pliny, Epist., iv. I I. 10 fr.: an equestrianis beatento death,an ex-praetorallowed to take much of his property into exile (on the latter's complicity, see A. N. Sherwin-White,Commentary(Oxford, 1966), on iv. II. II); ibid., ix. 13.21 (see n. 31 above). Foreign prince/citizen: Tac., Ann., vi. 40: even Tigranescannot escape 'supplicia civium' (citizen-punishments). 82 Penaltiesabnormally applied to equestriansor to men of rank (grade unspecified):Suet., Tiber., Ii (treadmill); Caius, xxvii. 3-4 (mutilation, exposureto beasts;condemnationto mines and road-building); ] os., Bell. Iud., ii. 308 (GessiusFlorus, Nero; flogging and crucifixion); Pliny, Epist., ii. II. 8 (n. 81 above). Epist., x. 58 is more problematic: Flavius Archippus was sent to the mines for forgery; there is no hint of irregularity; he was a citizen, but otherwisenothing is known about his statusat the time of the trial-his later prosperity may be solely due to Domitian's favour. Suet., Calba, ix is also difficult: a citizen is crucified, again for a genuine,seriouscrime. Was this atypical?Florus' crucifixions (above) undoubtedlywere. Suetoniusimplies the penalty was harsh, and the man pleadshis citizen-status,as if expectinga milder punishment. 83 Cyzicus lost its freedom once (perhapstwice) and Rhodesonce for violence againstRomansor for putting them to death: Dio, liv. 7. 6; lvii. 24. 6; Tac., Ann., iv. 36. 2-3; cf. Suet., Tiber., xxxvii. 3. 84 Seee.g. Riccobono,op. cit., ii, n. 185, p. 582 (Cnidus; Augustusassists free aliens). Only someof the African peasantswho petitioned Commodus(seen. 68 above) were citizens. Dig., xlviii. 6. 6 mayor may not be relevant: Pius instructs a proconsulto investigatea savageattack on a youth, whosename indicateshe was a citizen; yet Pius describes him simply as ingenuus('free-born').

LEGAL PRIVILEGE IN THE ROMAN EMPIRE

ignored.8s The relevantreferencesareconcentrated within aperiod of about 175 years,from the 70S B.C. to the reign of Trajan. This suggeststhat the value of citizenship may have declined in the secondcentury,perhapsbecauseof the increasein the numbersof citizens.86 Alternatively, the imbalancein the evidencemay be due to mere chance.Citizenship was still soughtafter in the second century,8?and the verdict at Lyons in A.D. 177(?)alreadyreferred to88 suggeststhat the reasonswerenot entirelysentimental.But the fact remainsthat the distinctionbetweencitizensand alienswas at all times only oneof severalwhich the Romansrecognized,andit skated over the realities of Roman politics and social life. The Romans rejected juridical equality, the equality of all citizens before the law, as easily as they rejectedpolitical equality. Cicero viewedas unequalthat kind of equalitywhich 'doesnot recognize grades of dignity'.89 This attitude must have been universal among those who dominatedpolitics and the administrationof justice in both Republicanand Imperial Rome.9o 8, Respected:Acts, 22: 24; cf. 25: 9 fr. (and seejRS,lvi (1966), p. 167);

Pliny, EPist., x. 96. 4; Euseb.,Ecd. Hist., v. I. 47. Ignored: Cic., adfamiliares, x. 32. 2 (Balbus); 2 in Verrem, v. 162 fr. (Verres); Suet., Galba, ix (Galba); Dio, lxiv. 2. 3 (Capito). 86 If Tacitus was any judge, the 'snob-value'of citizenship had declined anyway by the late first century. SeeAnn., iii. 40. 2: the citizenship was once 'rare, and given solely as a reward for virtue'. This is a conservative, upper-class-Roman view. 87 Sherwin-White,RomanCitizenship,chaptersix-x. The citizen/alien distinction lost most of its meaningwhen Caracallamade virtually the whole free populationof the Empire Roman citizens in A.D. 212(?). 88 Above, pp. 160-1. 89 Cic., de re publica, i. 43: ' ... tamen ipsa aequabilitasest iniqua, cum habetnullos gradusdignitatis'. 90 The mattersdiscussedin this paperare treatedmore fully in Social Statusand Legal Privilege in the RomanEmpire (Oxford, 1970). On increasingseverity of penalties,see Garnsey,'Why penaltiesbecame harsher:the Romancase,late republic to fourth-centuryempire', Natural Law Forum, no. 13 (1968), pp. I41 fr.

VIII

GREEKS AND THEIR PAST IN THE SECOND SOPHISTIC*t E. L. Bowie

INTRODUCTION

This paperattemptsto gathertogetherand interpretthe principal manifestationsof archaismin the Greekworld of the late first, the second and the early third centuries A.D., a period known to literary historians by Philostratus'sname 'the secondsophistic'. The archaism has been observedby historians of society and literaturealike, but the mostrecentwork on the sophistsdoeslittle more than allude to it, while literary historians have seen it primarily in terms of linguistic AtHeism, that imitation of fifthand fourth-century Attic prose writers which was already the subjectof controversyin the first centuryB.C. and which affected all Greek writers of the imperial period to a greater or lesser extent.This linguistic Atticism hasbeenexhaustivelydocumented by Wilhelm Schmidin his greatwork Der Atticismus.But Schmid seesthe developmentof Atticizing fashions almost entirely as a movementwithin literature.! This paper tries to show that the • From no. 46 (1970).

t An earlier draft of this paperwas read to the Oxford Philological Society on 1 December1967. I am very grateful to its membersand to the editorial board of Past and Presentfor their helpful criticism. 1 G. W. Bowersock,Greek Sophistsin the RomanEmpire (Oxford, 1969), pp. 15-16 toucheson the matter of archaism,but althoughhe excludes the explanationsthat 'the sophistswere eatenup by nostalgiafor the old times' and that they were 'affirming the independentgreatnessof the Greeksagainstthe Romans'he offers no positive elucidationof the phenomenon.For the linguistic aspectsW. Schmid, Der Atticismus, 4 vols (Stuttgart, 1887-96),remainsfundamental.Literary archaismis

166

GREEKS AND THEIR PAST IN SECOND SOPHISTIC

archaismof languageand style known as Atticism is only part of a wider tendency,a tendencythat prevailsin literaturenot only in style but also in choice of themeand treatment,and that equally affects other areasof cultural activity. If archaismaffects the cultural activity of upper-classGreeks over a wide range, a purely literary explanationshould not be sufficient. After presenting the evidence for archaism in the sophists,in the historiansand relatedliterary practitioners,and in other aspectsof life, this paper offers an explanationin wider terms, relating the Greeks'preoccupationwith their pastto their dissatisfactionwith the political situation of the present.This is not to deny that otherfactors, suchas the classicizingemphasisof Greek educationor the personaltastesof influential individuals like Hadrian, contributedto the ultimate amalgam.In particular it is clear that the archaism of the second century A.D. is not solelY a function of the Greeks'relation to their presentfrom the fact that archaismis evidentin the Latin as well as in the Greek world. Although the Latin and the Greek archaismmust have influenced each other-theirexponentswere often the sameand bilingual, like Hadrian or Fronto, and they belongedto a world of Greco-Romanupper-classculture that was essentiallya unityboth chronologyand Greekarchaism'spuristic rejectionof Latin exclude the possibility that Latin archaism can be a sufficient explanationof the growth of Greek archaism.It can only have helpedit on its way. Accordingly this paperdoesnot take account of archaismin Latin writers, althoughit is worth observingthat there too there were political overtonesthat correspondmutatis mutandisto thosesuggestedfor Greek archaism,a hankeringfor the libertas of the respublica. It shouldalso be saidthat althoughthe themeherediscussedis an importantaspectof Greekattitudesto describedand its extensionbeyondliterature noted, but not fully explained,by B. A. van Groningen,'Generalliterary tendenciesin the secondcentury A.D.', Mnemovne,xviii (1965), pp. 41 ff. F. Millar, 'P. HerenniusDexippus: the Greek world and the third-century invasions',fRS, lix (1969), pp. 12.-29, appearedwhen this paperwas alreadyin proof. Millar's contentionthat constantallusion to the past acted as a 'frame of referenceor a channelof communication'is illuminating, but does not excludethe question'why this frame of reference?'and the answer(which he seemsto regardas an a/ternatille, not a further, explanation)that it was 'a meansof flight from an oppressiveand inglorious present'(p. 12.). Seealso below, nn. IIO, III.

E. L. BOWIE

the contemporaryRomanworld, it is only one aspect:the rangeof attitudeswithin the Greekupperclassesis wide and complex,and this paper is intended to explore only one constituent of the amalgam. SOPHISTS

The most characteristicand influential figures of the age, not least in their own eyes, were the sophists. Oratory had always been importantin Greek society, before even it had beendevelopedas an art to help the individual to survive or succeedin the fifthcentury democraciesof Sicily and Athens. Training in rhetoric becamea major part of Greekhigher educationand was for many synonymouswith it, contestingthe role of educatorwith philosophy. Hencerhetoric continuedto dominateeducationafter the opportunitiesof achievingPericleanor Demosthenicgreatnessby successfuloratory had been considerablyrestricted by the overshadowingof the city statesby Hellenistic monarchies.It is to the time of this changein the centresof political power that we can trace the earliest attested declamation on a fictitious theme, attributed to Demetrius of Phaleron.2 It is doubtlessno coincidence that the practice seemsto have been instituted at a time when Alexander'sconquestshad madeit clear that the era of city statepolitics was coming to an end. By the time Octavian'svictory at Actium guaranteedthe absorption of the last great Hellenistic monarchy,PtolemaicEgypt, into the Romanempire,declamationson fictitious topics, both judicial and historical, had long beena standardelementin Greek rhetorical education,though it is only in the elder Seneca'srecords of declaimers under Augustus and Tiberius that we first get any substantialinformation on their contentand treatment.3It is not clear at what point such declamationsbecamemore than simply a part of rhetorical training and joined panegyricand commemorative speechesin the role of public entertainment:certainly by the secondhalf of the first century A.D. declamationseemsto have moved into the first rank of cultural activities and acquired an unprecedentedand almost unintelligible popularity. Its practi2

3

Quintilian, Institt/tio Oratoria, ii. 4. 41. See H. Bornecque,Les declamationset /es dec/amateursd'apres SeneqlleIe pere (Lille, 1902).

GREEKS AND THEIR PAST IN SECOND SOPHISTIC

tionerswereoftenfrom the wealthiestandmostinfluential families in their city's aristocracy,and they displayed their skill to enrapturedor critical audiences,not only in their native places,but throughoutthe Greek world, in the great cities that vied with eachotherin honouringor acquiringas residentsthemostbrilliant exponents-Athens, Ephesus,Smyrnaand evenRome.The name of rhetor (initially 'orator', but by now connoting a teacherand exponentof epideictic rhetoric) was sufficiently grand to appear on sepulchralor honorific inscriptions,but even greaterwas the name sophist(O"oqnO"TT)s). The fifth-century sophists had taught rhetoric as well as conductingmore seriousenquiries,and it was from them, as we learn from Philostratus,their biographer,that the 'sophists'of the secondsophistic derived their tradition of epideictic oratory. The term 'sophist' appearsto be contained within the term rhetor andto apply particularlyto thoseteachersof rhetoric (rhetors) whoseattainmentwas of such a level as to give public performances.Its link with the useof the term 'sophist'in the fifth centuryB.C. will hardly standclose examination,but the claim of classicalprecedenton however slendera basis is itself symptomaticof the times.4 The world in which sophistic oratory moved was not purelY artificial. The place of rhetoric in the Hellenistic and Roman worlds can be distortedif it is forgotten that membersof the city aristocracieswere still faced by many occasionson which real speecheswere required and could be influential. Embassiesto monarchsor to othercities, seekingprivilege or offering honour; debatesin the council (boule), still important within 'city politics evenwhen(as in the Romanperiod) the assembly(ecclesia)hadlost all powers; exhortatoryaddressesto the citizen body at crisesof riot or famine (such as we find among the speechesof Dio of Prusa);and, requiring not persuasionbut dignity and brilliance, ceremonialspeechesat the dedicationof public buildings,festivals or the welcomeof visiting potentates.All these-quiteapartfrom 4

SeeBowersock,Greek Sophistsin the RomanEmpire, pp. 12-14. That public performancemarkedthe differencebetweensophistand rhetor seemsclear from Galen'sphrase(of Hadrianusin his early career)6 ~itTWP O\hTW U0'lnUTEVWv (14. 627 Kiihn)-the use of the verb UOCPlaTEVWV, denotingan activity rather than simply claim to a status,is significant. UOCPlaTEVWV doesnot, of course,mean 'giving public performances', but refers to the pursuit of a careerinvolving these:cf. the context of ~V raAaT(a UOCPlaTEVWV in Plut., Mar., 131A.

E. L. BOWIE

law-courtoratory,usuallylookeduponwith contemptby sophists -offereda valid groundfor the acquisitionand practiceof the art of public speaking by ambitious members of the Greek city aristocracies.What is surprisingis that they were not enough,and that the samemen who were sent on embassiesto the emperor would sweat out their intellectual talents on perfecting declamations set in a fictitious classicalcontext for audienceswho were more interestedin tricks of style and delivery than in content.5 The choiceof thesethemesmay have somethingto do with the causesof the rise of this fashionas a whole amongthe Greekcity aristocracies(though theseare outsidethe scopeof this paper). It would be intelligible that Greeksof the first and secondcentury, deniedthataspectof public oratorywhich hadacquiredpowerand immortality for a Periclesor Demosthenesin the classicalperiod, speechesof persuasionto sovereignassembliesin autonomous cities, should chooseto display their intellectual superiority to their fellow citizens and enjoy the rhetorician's control of his audience'semotionsin treating themesthat precisely substituted for the occasionsthat werelacking. What is certain,at least,is that the favouredthemesof the sophistsharkedback constantlyto the classical period. The classicismof theme is as much evident in those orators whose rhythms have been labelled 'Asian' (as opposedto 'Attic').6 The most prominentthemesderive from the history of Athens, the greatestof the classicalGreekcities, or Alexander,the greatest individual Greek. Alexanderfigures in two of the orationsof Dio of Prusa,written in the first decadeof the secondcentury: in one (the second)he converseswith his father Philip, in the other (the fourth) with the cynic Diogenes.Dio also wrote eight books On the virtuesof Alexander,now lost.7 Plutarch,slightly his junior and For the strain of declaiming cf. Polemo'scomparisonwith the situation of a gladiator and use of the verb aywvl'E~OV i\ KaTe. aocpiav M;O~OTEpOl) are most easily accessiblein the Loeb edition of Plutarch'sMoralia, vol. iv, ed. F. C. Babbitt (1936). 9 Most recently Hamilton, op. cit., p. xxii, n. I. 10 Philostratus,Lives of the Sophists,I. 21, p. P9. 11 Ibid., I. 22, p. 522; I. 25, p. 542. 12 Ibid., I. 23, p. 527. The Persianwars were alreadya favourite with Nicetesof Smyrna,the earliestof Philostratus'sophistsand already active under Nero: seeLives of the Sophists,I. 20, p. 513. For Chaeronea cf. Syrianus,ii. 165; Apsines,ix. 471. For Leptinescf. Apsines,ix. 496.

171

E. L. BOWIE

themes:13 Demosthenesswearsthat he did not take the bribe of fifty talents (the charge brought against him by Demades);an argumentthat the trophieserectedby the Greeksbe takendown at the end of the Peloponnesianwar; and an exhortationto the Atheniansto return to their demesafter the battle of Aegospotami. Other of his declamationsincludedthe argumentsof Xenophon refusing to survive Socrates;Solon demanding that his legislationbe annulledafter Pisistratushad obtaineda bodyguard; and threefurther Demosthenicpieces.No wonderthat one of his witticisms, when he encountereda sophist buying sausagesand sprats, was to remark 'My dear chap, you can't give a good representationof the arroganceof Darius and Xerxes if you live on that'.14 Athens was not the only favourite. Plutarch'sMoraNa show as much interestin ancientSpartaas his Lives, and his coevalIsaeus from Syria (or, as the archaizing sophistsliked to say, Assyria) debated the hackneyed topic of whether the Lacedaemonians shouldfortify Spartawith walls.IS Marcusof Byzantium,a decade or two later, arguedthat the Spartansshould not take back the hoplites who had surrenderedthemselvesand their weaponsat Sphacteria.16 Sometimesthe theme was Homeric, as in Dio of Prusa'sTrrijan Speech(arguingthat Troy was nevercapturedby the Greeks), Aristides' Speechfor an embas!J to Achilles (from Iliad, Book 9), or Philostratus'own Heroic Tale.17 A very high proportionof the themesmentionedin the Livesby Philostratus derives from Greek history, and all these from the period before 326 B.C. That this fairly reflects sophisticbehaviour and is not simply the consequenceof deliberateselectionby the archaizing Philostratusis guaranteedby a comparisonwith the extant corpusof Aristides' works. Aristides (chiefly active under Pius and Marcus) sets all his purely declamatory works (as opposedto speechesfor real and contemporaryoccasions)in the Philostratus,op. cit., I. 25, p. 538. Ibid., I. 25, p. 541. 15 Ibid., I. 20, p. 514; cf. also Aristides, ibid., 2.9, p. 584. 16 Ibid., 1. 24, p. 528. 17 Dio's TpWIKOS A6yos (Or., 11 in the traditional order) is in the first volume of the Loeb's edition of J. W. Cohoon(1932). For Aristides' rrpEO"l3evTIKos rrpos AXlhhEO, Dindorf's edition (Leipzig, 1829), Or., 52; and Philostratus'sHPWIKOS, C. L. Kayser'sedition, vol. ii (see n. 5), on which seebelow, p. 197.

13

14

GREEKS AND THEIR PAST IN SECOND SOPHISTIC

classicalperiod. We are not in a position to judge how different this choice of material was from that of Greek rhetoricians of earlier periods. Senecathe elder is some help: historical themes from the classical period are well represented,but far from universal,in the selectionhe gives in his suasoriae(deliberationsby historical figures at momentsof crisis), whereasthere are only a few-about half a dozen-in the controversiae (fictitious court arguments).It would seem that in the early first century A.D. historical themeswere less popular, but Seneca'stestimonyis not sufficiently comprehensivefor us to judge how much SO.18 What is clear, however,is that in our period the tendencywas at its height. Some of its practitioners might ridicule it, as did Lucian, who askedwhy one shouldimitate Aeschinesin a time of peace when there was no Philip attacking and no Alexander issuing orders. But Lucian too betrays himself as a child of his time, not only by his Atticizing idiom but by the setting of his dialogues.They often take themesfrom earlier writers and are set in Athens, sometimesclassical Athens with antiquarian details carefully sketchedin. When he turns to satirical dialogue it is Plato and Old Comedythat he proclaimsas his principal models. It would not have beenhard to guess,if Lucian had not told us, that he startedhis literary careeras a practisingsophistand commandedhigh fees.19 The same phenomenonappearsin some of the minor genres which the sophists, self-appointed experts in all branches of literary activity, did not hesitateto cultivate. In epistolography, for example, Alciphron's letters are set in Athens of the fourth century B.C., made amply clear by choice of names and local descriptions.The erotic novel is less helpful, for its stereotyped plot demandedfantastic adventureslittle conduciveto a precise historical setting ancientor modern. Chariton, however, writing in the first or secondcenturyA.D., set his romanceexplicitly in the fifth century B.C. Longus's Daphnis and Chloe (? late second century A.D.) shows the same hankering for the primitive and For classicalfigures in Seneca'scontroversiaecf. 3. 8 the Olynthians; 6. 5 Iphicrates;8. 2 Phidias; 9. I Miltiades; 10. 5 Parrhasius. 19 Lucian on imitating AeschinesKa\ TcriiTa EV dpiJV') IlT\TE C!>IAimrov emoVTos IlT\TE AAE~ov5pov hrlTOTTOVTOS, Rhetorumpraeceptor, 10. For his relation to classicalAthens, J. Bompaire,Lucien ecrivain (Paris, 1958) and J. Delz, LukiansKenntnisder AthenischenAntiquitaten(Basel diss., 1947; Fribourg, 1950); Plato and comedyas models Lucian, Bis accusatus,32.

18

173

E. L. BOWIE

'unspoilt' life of the Greekcountryside(now considerablyaffected by imperial estates)that we find in Dio ofPrusa'sEuboeanTale and in HerodesAtticus's cultivation of the rustic Agathion.20 The most illuminating genre, however, is that of historiography, likewise attempted by sophists, but reaching greater heightsin other hands.To this I shall now turn. HISTORIANS

A valuablecommentaryon the mentality of an ageis usually to be found in the sort of history it choosesto write and read and the mannerin which the chosenthemesare treated.In attemptingto apply this criterion to the Greekworld of the secondsophisticwe are fortunatein possessingsubstantialproportionsof Arrian and Appian from the secondcenturyandof CassiusDio and Herodian from the early third century. We have also a wealth of historical biographyin Plutarch and a representativeof antiquarianliterature in Pausanias.But the materialthat has beenpreservedis only a small proportion of what was written. Little value could attach to analysisof it alone, and someattemptmust be madeto assess such tendenciesas can be detectedin the massof lost historiography.Onmuchof it ouronly informationis late (often the tenthcenturySudalexicon) and sometimesaddsimprecisionto dubious reliability. Accordingly any reconstructionof the generalpattern of historiographyis precarious.It must be checkedagainstthe behaviourof the better documentedwriters who have survived. I shall therefore follow a generaland in many ways superficial conspectuswith a closer examinationof one writer, Arrian, who will be found to exemplify severalof the proclivities observable elsewhere.At the sametime I shall try to guaranteethe soundness of the generalconspectusby working as far as possiblefrom those writers of whom there are considerableremains. 20

Seebelow, p. 197. For Alciphron seethe Loeb edition of Alciphron, Aelian and Phi/ostratus'Letters(1949), ed. A. R. Bennerand F. H. Forbes (with a discussionof the vexed questionof date). Dio's EVt!OlKOS (Or. 7) is in the Loeb volume cited n. 17 above. For the novelists seeB. E. Perry, The AncientRomances(California, 1967).

174

GREEKS AND THEIR PAST IN SECOND SOPHISTIC

Universal Histories I shall first consider the writing of 'universal' histories, that is historiescovering severalnations or empiresover a considerable spanof time.21 The consolidationof a world empire by Rome in the first centuryB.C. had called forth world histories.Nicolaus of Damascusand Diodorus from Agyrion in Sicily had written the first such sinceEphorus:Nicolaus had naturally broughtit down to his own time; Diodorus put considerableemphasison the archaicperiod, but there is no hint that it might have beenconceivableto stop there.The historicalwork of Strabo,althoughnot starting till the Hellenistic period, had the characterof a world history and predictablycamedown to his own time. For Strabo, the contemporaryGreek world was vital and interesting,as we canjudgefrom the wealth ofbiographicnoticeson his coevalsand immediateseniorsthat he gives us in his geography.Moreover, his positive attitude to the uses of history, that it should be 'of political and public utility',22 requireda knowledgeand assumed an interest in the affairs of the recent past. When a writer like Dionysius of Halicarnassusdid devotehimself exclusively to the past as in the Roman Antiquities (they terminatedwith the first Punic War), it was to the past of the great world power of his time and not to the past of Greece. Thereafter,for more than a century,the Greekworld seemsnot to have produced histories covering a great span. When they reappear,they are notably different. The difference does not, of course,extendto chronography,for it might be arguedthat the work of chronographersmust, by necessityof the genre, come down to their own time. That had been the courseadoptedby Apollodorus of Athens and Castor of Rhodes at a time when Greekcontemporaryhistory still had vestigesof life, and seemsto For categoriesof Greek historical writing cf. F. Jacoby,'Dber die Entwicklung der griechischenHistoriographie',Klio, ix (1909), pp. 80-123 (repr. in F. Jacoby,Abhandlungenzur griechischen Geschichtschreibung,ed. H. Bloch (Leiden, 1956). Jacoby'scollection of the fragmentsof the Greek historians,FGH (Berlin, 1923-) is fundamentalto any study of this sort, For Greek historiographyin the secondand third centuriesA.D., seenow F. Millar, op. cit. (n. 1 above), pp. 14 If. 22 Strabo, Geography,i. 1. 22-3: lTOAlTIKOV Kat BTJ.,"'wv 'Eq>ealos:cf. Schmid-Stiihlin,op. cit., p. 810. But for doubts about this man's acquaintancewith the city of Ephesus,B. E. Perry, The AncientRomances,p. 170 (following Lavagnini).

51

E. L. BOWIE

'historian and rhetor' and ascription of epistles strongly suggest. Timogeneswrote a work in threebooksOn Heracleia Ponticaandits Great Men, and since the great men of Heracleia distinguished themselvesin the cultural rather than the political arena,we seem here to have a local cultural history.56 Only conjecture puts three books of Corinthian Histories (KOplV61CXKO:) by one Theseusin the Roman period. The author's namefits the ascription,and what evidencethere is suggeststhat all three books dealt with the archaicperiod.57 Telephus of Pergamumis fortunately securely dated to the Roman period, even if we do not accept the statementof the Historia Augustathat he was tutor to the emperorVerus(and there is no particular reasonto reject it). Apart from many philological and literary treatisesTelephuswrote a Periegesisof Pergamum,On the templeof Augustusat Pergamum,and five booksOn the Pergamene Kings. The former two works will have beenfirmly in the guidebook class. The last was clearly historical, and shares with Arrian's work on Bithynia, to which I shall return, the terminusof the transition from independentkingdom to Romanprovincia.58 The other great city of the Asian seaboard,Smyrna, found its historian in the primarily medical writer Hermogenes,son of Charidemus,whose literary fertility has been commemoratedfor us on stone.To seventy-seven medicalworks he addedA historical ess'?)'on Smyrnain two volumes;a volume on the sophia of Homer and anotheron his birthplace; two on Asian Settlements(KTicYEls), four on Settlementsin Europe and a similar volume for the Islands; two geographicalvolumesand two of stratagems,and a catalogue of (great?) Romansand Smyrnaeans.The approachis that of a scholar, if not a litterateur, and the works on Settlementswill naturally have concentratedon, if not beenlimited to, the archaic period of cities' foundations. The catalogue of Romans and Smyrnaeansagain suggests cultural history, but the interpretation of this title is ambiguous. We have no independentmeans of Timogenes,Jacoby,FCH, no. 435. Theseus,ibid., 453. The name does not exclude his being a Corinthian, cf. SEC, xi. 77+Add. p. 216, claiming Attic stock. (lowe this referenceto !\Irs S. C. Humphreys.) . " Telephus,Jacoby,FCH, no. 505; the Historia Allgllsta, Vems2.5 includes Tclephusin a reputablelist of tutors. For the reliability of this notice, T. D. Barnes,'Hadrian and Lucius Verus', fRS, lvii (1967), p.67·

56

57

186

GREEKS AND THEIR PAST IN SECOND SOPHIS'I'IC

inferring the content of the work on Smyrna, but the author's other antiquariantendencieswould makea work devotedlargely to the distantpastmoreprobablethan any other.59 Fortunatelyhe can be datedfairly securely,to the middle of the secondcentury, as his interestsmight anyway haveindicated. Alexandriawas the subjectof a work addressedto Hadrianby the scholarNicanor (his chief work was on punctuation,but he also tackledthe Homeric catalogueof ships),and anothersuchby Aelius Dios should belong to the same century. We have no indication how they treatedthe subject.6o Already under Augustus, Tarsus had been written up by its distinguishedcitizen Athenodorusin a work whoseonly surviving fragment dealswith the time of its foundation. Its rival and neighbour Aegae had to wait for the work On my country (rrepi Tiis 1TCXTpi50s) of the late second-centurysophistP. Anteius Antiochus. We have no direct testimonyon the content of this history, but it is perhapsrelevantthat inscriptional evidencefor Antiochus' researchesshowshim to have been interestedin the early period that we would label mythica1.61 The distantpastwill also havemost probablybeenthe content of Claudius Iolaus's Phoenician History-the date should be late first century-as of, in the second century, Philo of Byblos's nine books purporting to translatethe Phoenicianhistorian of pre-Trojanvintage, Sanchuniathon.A work of Philo on Hadrian has alreadybeenmentioned62and it is also worth remarkingthat he produceda monumentalencyclopediaof cultural history On Cities and their Famous Men in thirty books. A similar but more limited approachis likely for Aspasius of Tyre's work on his native city entitled On Tyre and its Citizens, while the more disHermogenes,Jacoby,FGH, no. 579; cf. Schmid-Stahlin,op. cit., p. 925. Nicanor, Jacoby,FGH, no. 628; Dios, ibid., no. 629. The existenceof a political history of Alexandria in the Romanimperial period in someform is a strong possibility: it would provide the contextfor the many fragmentsof 'paganmartyr acts' that have turned up on papyrus, professingto relate contumaciousencountersof leading Alexandrians with Roman officials and emperors.On the problem,H. Musurillo, The Acts of the Pagan Martyrs (Oxford, 1954). 61 Athenodorus,Jacoby,FGH, no. 746. Antiochus, ibid., no. 747; Philostratus,Lives of the Sophists,2. 4, p. 586; epigraphicevidence, Bulletin de CorrespondanceHe/Unique, xxviii (1904), pp. 421 ff. 62 Above, p. 181.

59

60

E. L. BOWIE

tinguishedAspasiusof Byblos wrote On Byblos, with no further clue to content. The latter certainly, and the former probably, belongsto the secondcentury.63 Although in many casesscantyanduninformativeon detail, the evidencehere assembledsuggeststhat in local history the same avoidanceof the morerecentpastis to be found aswas detectedin world histories: the exceptionsappear to have been periegetic literature(where,as we shall seein Pausanias,theremay also have been opportunity for consciousarchaism)and cultural histories whereGreececould be seenstill to be producinggreatfigures who dominatedthe intellectual world (Greek and Roman alike). The bias is confirmedif we analyseour only full exampleof periegetic literature to survive and comparetwo other by-forms of history, collectionsof stratagemsand collectionsof myths.

Periegesis,Stratagemsand Myths Pausanias's Periegesisof Greecedisplaysarchaismin both treatment and content.His languageis Atticizing, while his turns of phrase recall Herodotus,and the form of his work can be relatedto early Ionian periegeticliterature. His interest in antique local history and mythology and in classicalworks of art cohereswell with the middle of the secondcenturyto which he belongs.He was not an Athenian(Magnesiaad Sipylum seemsto be his native place, but, the problemis not simple) yet he sharedcontemporaryadmiration for Attica and not unnaturallystartshis periegesiswith it.64 It is not surprising, then, that he almost completely neglects monumentsand dedicationslater than c. 150 B.C. There are indeed referencesto the benefactionsof contemporaryphilhellenic emperors,Hadrian and Antoninus Pius, but notablemonuments of the interveningperiodarenot mentioned,andhis list of dedications by Olympic victors stopsat the samepoint. The traditional explanationof the phenomenonthrows the chief blame on his sources:he was using writers of the secondcentury B.C. and did not trouble to supplementthem by later information. But this Iolaus, Jacoby,FGH, no. 788; Philon, ibid., no. 790; Aspasii, ibid., nos 792 (Byblos) and 793 (Tyre). 64 On Pausanias cf. Schmid-Stahlin,op. cit., pp. 755 fI. For admiration of Attica cf. Paus.,i. 17; i. 24. 3; i. 29. 6. 63

188

GREEKS AND THEIR PAST IN SECOND SOPHISTIC

cannotbe the whole truth of the matter. His omissionof the later period must have beenpatentto his readership,and both readers and writers must have acquiescedin the neglect. This acquiescenceis only intelligible if it was taken for grantedthat the intervening years were of no interest to a Greek, and their lack of interestwas due to the political (rather than cultural) decline that they represented.It is significant for Greek attitudes that the political decline seems to have been allowed to influence the treatmentof what was essentiallya cultural history. The political and cultural achievementsof classical Greece, and particularly Athens, were very closely woven togetherin the Greek memory of the past, and this may well have fosteredthe illusion in some that a cultural resurgencewould somehowbring with it a restoration of political power and independence.Monarchistthough he is, Pausaniascannot help praising the archaic institution of Athenian democracy.65 The eight books of Stratagemsdedicatedin 162 to the Emperors Marcus and Verus by the rhetor from Macedon,Polyaenus,show the same archaism of style and content. Polyaenustries, if not always successfully,to Atticize (using dual and optative forms) 66 He and he recognizesHomer as the first war correspondent. concentratesalmostentirely on the classicaland Hellenistic world of Greeceand devotes only the first twenty-five sectionsof his eighth book to Romanincidents(the remainderbeing allotted to stratagemsinvolving womenI). The latest reference is to the year 43 B.C. It may be that he omits the ensuing two hundred yearsbecausehe has no easily accessiblehandbookto turn to, but as in the caseof Pausanias,the omissionpresumesacquiescence on the part of writer and readerin so glaring a lacuna.67 The Bibliotheca of Apollodorus is in a somewhat different category.The date of its compositionis uncertain,but almost all the evidencepoints to the late first or the secondcentury A.D. Democracy,Paus.,iv. 35. 3. On Pausaniasand his sources,cf. SchmidStahlin, loc. cit. 66 1. pref. 4 f. 67 On Polyaenus,Schmid-Stahlin,op. cit., p. 754. Polyaenus'swork In defenceof the assemblyof the Macedoniansshould not be classifiedas local history (as Jacoby,FGH, no. 739); it is more probably a speechon behalfof the Macedoniankoinon (so, independently,J. Deininger, Die Provinziallandtageder romischenKaizerzeit(Munich, 1965), pp. 91 n. 8, 95 n.8). 65

E. L. BOWIE

The subject-matter,a collection of the Greek myths, allows no room for anything other than archaism,so it cannotbe taken as evidence of reluctance to treat contemporaryevents. What is remarkable,however, is that Roman versions of Greek myths, most notably, for example,the foundation of Rome by Aeneas, are entirely absent.It is instructive to quote the reactionof one scholarto this phenomenon,J. G. Frazer,in his introduction to the Loeb edition, 192I : From this remarkablesilencewe can hardly draw any other inferencethan that the writer was either unawareof the existenceof Rome or deliberatelyresolvedto ignore it. He cannothave beenunawareof it if he wrote, as is now generallybelieved,under the RomanEmpire. It remainsto supposethat, living with the evidenceof Romanpower all aroundhim, and familiar as he must have beenwith the claims which the Romansset up to Trojan descenthe carefully abstainedfrom noticing theseclaims, though the mention of them was naturally invited by the scopeand tenor of his work. It must be confessedthat such an obstinate refusal to recognisethe mastersof the world is somewhat puzzling, and that it presentsa seriousdifficulty to the now prevalentview that the author was a citizen of the Roman empire. On the other hand it would be intelligible enoughif he wrote in somequiet corner of the Greek world at a time when Rome was still a purely Italian power, when rumours of her wars had hardly begunto trickle acrossthe Adriatic, and when Romansails had not yet shown themselvesin the Aegean(p. xii). There is no possibility of dating the Bibliotheca so early as the last whimsical fantasiesof Frazer'sparagraphwould suggest,nor is there any need. Apollodorus's attitude is not puzzling. It supportsrather than conflicts with a second-centurydate. Like many other culturedGreeksof his time the writer liked to forget from time to time the ubiquitous dominanceof Rome,andwhere better to exercisethat amnesiathan in a work devoted to the safely antiqueand establishedGreek myths?68 68

On Apollodorus cf. Schmid-Stahlin,op. cit., vol. ii.

I,

pp. 428 if.

GREEKS AND THEIR PAST IN SECOND SOPHISTIC

Arrian It has beensuggestedthat a generaltendencycan be detectedin historiansof the late first and secondcenturiesto turn to themes from the classicalageof Greeceandneglectthe present,a tendency that is presenteven in men who held high positions in contemporary society. In the personof Arrian of Nicomediathis type of behaviourcan be examinedmore closely and assessed againstour not inconsiderableknowledgeof the man'slife. 69 Arrian's cultural backgroundin Bithynian Nicomedia and his education,culminating in attendanceat the lecturesof Epictetus in the first decade of the second century, were undoubtedly Greek. His name, Flavius Arrianus, lacking any Greek cognomen, suggeststhat his family may havebeenpartly ofItalian origin; by the end of the century, however,it will have beenabsorbedinto the life of the Greekpolis. Arrian's entry into imperial service,in which he roseto a consulate(almostcertainly a suffect of the year 129) and then was given the importantlegateshipof the large and strategically vital provincia of Cappadocia,points to a realistic appraisalof the contemporaryworld and proved ability in grappling with its problems.Thereis nothing to confirm, and certain considerationscast in doubt, the later tradition that he was promoted by reasonof his literary prominence(paideia), the major monumentsto which can be securelydatedafter his consulate. It is clear, however, that his literary interestsdated from his youth,70and his approachto literary activity seemsalreadyto be .9 A satisfactorystudy of Arrian is lacking, and the brief sketchgiven here is

an abbreviationof a paperwhich will be publishedelsewhere.A. B. Bosworth, CQ 22 (1972), pp. 163-85, offering a different chronology, is now fundamental:I cannot agreewith his conclusions.The best treatmentremainsthat of E. Schwartzin.RE, ii (1896), cc. 1230 ff., Arrianus (9). Seealso H. Pelham,English Historical Review,xi (1896), pp. 625 ff.; G. Wirth, Historia, xiii (1964), pp. 209 ff.; E. Gabba,RSI, lxxi (1959) pp. 361-81. The suggestionthat his full name was Q. Eppius Flavius Arrianus (ProsopographiaImperii Romani2 , F 219, fol.lowed in Past and Present, 1970) is no longer acceptable.A new fragment of an inscription from Delphi (Fouilles de Delphes,III. iv, p. 57 n. 294 last line = inv. no. 4961, illustrated PI. IX: ed. A. Plassart) shows Q. Eppius and FI. Arrianus to be different persons,and a new inscription from Athens gives his praenomenas Aulus or Lucius (D. Peppa-Delmouzou,AAA 3 (1970); J. H. Oliver, GRBS,xi (1970), pp. 335 ff.); E. Borzz, AAA 5 (1972). 70 Anabasis,1. 12. 5, quotedbelow n. 75.

E. L. BOWIE

fixed by the time of his earliestfirmly datedwork to survive, the Circumnavigationof the Black Sea. This takes the ostensibleform of a Greek letter to Hadrian that accompaniedand coveredthe same ground as an official (Latin?) report from the legate of Cappadociato his emperor.The date is early in Arrian's legateship. It is a literary game, and in it Arrian already plays Xenophon,the Athenianwhosegeneralshipled the Ten Thousandback across Anatolia to safety and the sea and whose literary skill recordedthat achievementand the ways of Socratesin simple and lucid Attic Greek. Xenophonis already the model for Arrian's 71 style and is constantlyreferredto for comparisonor precedent. The form does not in fact correspondwith any of Xenophon's works but rather to the Ionian genreof periegesis. The samegameis also being playedin the surviving fragment of the slightly later Order of battle against the Alans: here Arrian actually refers to himself as Xenophon.72 In the Essqyon Tactics, which proclaimsits dateas the 20th year of Hadrian(136/7) there is fulsome praiseof the emperorand the regime, and the impression is given of an attemptto harmonizea love of the Greekpast with an involvement in the Greco-Romanpresent. So far, at least, Arrian's admiration for the classicalpast has not excluded application to things contemporary,and if in peregrinationof Anatolia he kept a text of Xenophonby his side,it did not prevent him from checkingthe Alans. We haveno evidenceof his imperialcareerbeingprolongedafter his legateshipof Cappadocia,and it is a reasonableinferencethat his political participationwas reducedif not abandonedafter the death of his friend Hadrian. As his permanentresidenceArrian appearsto havechosennot Rome, wherehe might haveattended meetingsof the senateand influencedpolicy through less formal channels,but Athens. The main reasonfor the choice of Athens rather than his native Nicomedia will doubtlesshave been that other literary people had settled there. After Hadrian's benefactions to the city it might justly be consideredthe centreof the Greekworld. But the causeof this resurgenceof culturein Athens 71 72

Chs I, 2, 11-14, 16, 25. Arrian, Expeditiocontra A/anos,10. 22. The epigraphicevidenceis sufficient (cf. ProsopographiaImperii Roman;2F 219) to excludethe suggestionofP. A. Stadter,GRBS, viii (1967), pp. 155 ff. that Xenophonwas ever part of Arrian's official nomenclature.

GREEKS AND THEIR PAST IN SECOND SOPHISTIC

and of Hadrian'strend-settingpatronageof the Athenianswas the city's associationwith the classicalpast: this too, and in particular the fact that it was Xenophon'snative city, will have determined Arrian's choice. He may have returnedto his native Nicomedia to hold a priesthood,but all the evidencesuggeststhat he did not stay there. He becamean Athenian citizen, enrolling, like Hadrianusof Tyre about the sametime,73 in a tribe and deme.In 145/6he held the archonship,and the Esseryon hunting(KVVTlYETIKOS) makesit clear that he regardedAthens as his home.74 This work also providesone detail which showsthe sentimentalextremesto which evenan intelligent man like Arrian could take his admiration of the past: he namedhis dogs after thoseof Xenophon. Imitation of Xenophonpercolateshis literary production.First, probably, he turned to writing up the lecturesof Epictetusin a mannerthat might recall Xenophon'sMemorabilia. Then (but the chronology cannot be firmly established)he published monographson Dio of Syracuseand Timoleon, now lost: Xenophon's Agesilauswill perhapshavebeenthe model. But the magnumopus was the work on Alexander, entitled the Anabasis to recall Xenophon's account of the march of the Ten Thousand. He stressesthat it has been important to him from his youth, and valuesit higher than his birth and successfulpolitical career:'But this I wish to record, that to me this work representsmy country and my family and my public offices, and has done right from my youth. And on this accountI think myself not unworthy of the first rank in Greekletters,just indeedas Alexanderwas first in military achievement.'75 The Anabasiswas almost certainly publishedearly in the reign of Marcus, in the first half of the 160s, and may have owed some of its inspirationto Trajan'swars in Arrian's youth andto Arrian's own experienceon Rome's Eastern frontier. But the primary inspirationis Alexanderhimselfandthe absenceof a greatwork on him in the classicaltradition. This is madeclear by the preface.The title, division into sevenbooksand constantstylistic echo remind Philostratus,Lives of the Sophists,2. xviii (1969), pp. 503 ff. 74 See especiallych. 5.

73

75

10,

p. 588; cf. M. Woloch, Historia,

aAA' EKeivo avaypacpw,chI Ej.10t naTplS Te Kat yevos Kai apxai olSe 01 Myol etal Te Kat ano veov gTI EyevoVTo. Kai Ent T