This page intentionally left blankCAMBRI DGE CLASSI CAL TEXTSAND COMMENTARI ESedi torsJ. DI GGLE N. HOPKI NSON J. G. F. POWELLM. D. REEVE D. N. SEDLEY R. J. TARRANT43THEOPHRASTUS: CHARACTERSTHEOPHRASTUSCHARACTERSEDI TED WI TH I NTRODUCTI ON,TRANSLATI ON AND COMMENTARYBYJAMES DI GGLEProfessor of Greek and LatinCambridge UniversityCAMBRIDGE UNIVERSITY PRESSCambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So PauloCambridge University PressThe Edinburgh Building, Cambridge CB2 8RU, UKFirst published in print formatISBN-13 978-0-521-83980-8ISBN-13 978-0-511-26468-9 Cambridge University Press 20042004Information on this title: www.cambridg e.org /9780521839808This publication is in copyright. Subject to statutory exception and to the provision ofrelevant collective licensing agreements, no reproduction of any part may take placewithout the written permission of Cambridge University Press.ISBN-10 0-511-26468-2ISBN-10 0-521-83980-7Cambridge University Press has no responsibility for the persistence or accuracy of urlsfor external or third-party internet websites referred to in this publication, and does notguarantee that any content on such websites is, or will remain, accurate or appropriate.Published in the United States of America by Cambridge University Press, New Yorkwww.cambridge.orghardbackeBook (EBL)eBook (EBL)hardbackCONTENTSPreface page viii ntroducti on 1I Theophrastus and his times 1II The nature and purpose of the Characters 4i Title 4ii Antecedents and relations 5iii Later Peripatetics 9iv Other developments 11v The purpose of the Characters 12vi Authenticity and integrity 16vii Integrity and style 19viii Literary inuence 25III Date 27IV Transmission 37i Preliminaries 37ii The tradition in a nutshell 41iii The manuscripts 43V Some texts and commentaries 52text and trans lati on 59commentary 161abbrevi ati ons and bi bli ographyI Select abbreviations 523II Select bibliography 524i Editions 524ii Other works 525vCONTENTSi ndexesI Index uerborum 533II Passages 557III Subjects 567IV Greek 587viPREFACEI owe an especial debt to three scholars. Jeffrey Fish transcribedfor me in the most minute detail the section of P. Herc. 1457 con-taining sketch no. V. I am deeply grateful to him for selesslyundertaking and so meticulously executing this long and de-manding task. Howmuch it has beneted me will be apparent toreaders of the commentary. Ioannis Stefanis generously suppliedinformation about the readings of the later manuscripts, loanedme his photographs of A and B, and sent me a copy of an un-published text and apparatus criticus of his own. I found that ina few places he and I had independently hit upon the same con-jecture. I should have assigned sole credit to Professor Stefanis,had he not requested that I publish these conjectures under ourjoint names. Paul Millett, from whom (as the commentary at-tests) I had already learned so much, read the whole typescript,saved me from several slips, and at other points sharpened myargument.I am grateful to Martin Ruehl for procuring photocopies ofmore than a score of older books and pamphlets from librariesin Germany; and, for a similar service in Greece, to DimitriosBeroutsos, Georgios Christodoulou, Daniel Jakob, and AntoniosRengakos. Nigel Wilsonkindly lent me his photographs of Vandsent me some comments on its script. Geoffrey Arnott answeredquestions on pheasants and monkeys, Sir James Beament onbotany and entomology, and Paul Cartledge on historical prob-lems. I am also indebted, for advice or help of various kinds, toJohn Dillery, Bruce Fraser, Nikolaos Gonis, Ioannis Konstanta-kos, Luigi Lehnus, Marianne McDonald, Stephen Oakley, DirkObbink, Michael Reeve, Jeffrey Rusten, and Anne Thompson;to Muriel Hall, copy-editor, for her care and vigilance; and forgenerously undertaking to read the proofs, to Stephen Oakleyand Frederick Williams.Two matters of numeration. First, fr. 100 Fortenbaugh isshorthand for fr. 100 in W. W. Fortenbaugh, P. M. Huby, R. W.viiPREFACESharples, D. Gutas (edd.), Theophrastus of Eresus: Sources for his Life,Writings, Thought and Inuence (Leiden etc. 1992). Second, I havenumberedthe sections of the Greek text afresh. Section-numberswere rst added by the Leipzig editors (1897), and these weremodied by Diels (1909). My numbering reects what I take tobe the main divisions within the text.CambridgeSeptember 2003viiiI NTRODUCTI ONI THEOPHRASTUS AND HI S TI MESThe sources for the life of Theophrastus are collected in W. W.Fortenbaugh, P. M. Huby, R. W. Sharples, D. Gutas, Theophras-tus of Eresus: Sources for his Life, Writings, Thought and Inuence(Leiden 1992) frs. 136. The primary source is D.L. 5.3657(fr. 1). Some modern discussions: O. Regenbogen, Theophras-tos, RE Suppl. vii (1940) 135561 (ii.1 Vita. Lebensumst ande),M. G. Sollenberger, The Lives of the Peripatetics: An analy-sis of the contents and structure of Diogenes Laertius VitaePhilosophorum Book 5, ANRW ii.36.6 (1992) 37933879,J. Mejer, A Life in fragments: the Vita Theophrasti, in J. M. vanOphuijsen and M. van Raalte (edd.), Theophrastus: Reappraisingthe Sources (New Brunswick and London 1998) 128.Theophrastus was born at Eresos on Lesbos (D.L. 5.36 = fr.1.2) in 372/1 or 371/0.1His name, originally Tp:cuc, waschanged by Aristotle to Otcgpc:c, in recognition (so laterwriters believed) of his divine eloquence (D.L. 5.38 = fr. 1.301oic :c :n gptc ttticv, Suda O 199 = fr. 2.4 oic :ctic gptiv).2His association with Aristotle will have begin atAthens, if we accept that he studied with Plato (D.L. 5.36 = fr.1Regenbogen 1357, Sollenberger 3843.2Cf. Str. 13.2.4 = fr. 5A.3 :cv :n gptc co:c0 ncv ttinucivcutvc,setting his seal of approval onhis style of speech (LSJ nc iii.2; ttinucivciv.3, as in Char. II.4), not signifying the fervour of his speech (H. L. Jones,Loeb ed. 1929) nor signifying his keenness for speech (Fortenbaugh et al.),Cic. Orat. 62 =fr. 5b.2 diuinitate loquendi nomen inuenit, Plin. Nat. praef. 29hominem in eloquentia tantum ut nomen diuinum inde inuenerit, Quint. Inst. 10.1.83in Theophrasto tam est loquendi nitor ille diuinus ut ex eo nomen quoque traxisse dicatur.Anecdotal tradition (Cic. Brut. 172, Quint. Inst. 8.1.2 = fr. 7ab; cf. Mejer1516) suggests that he was proud of his command of Attic but that othersregarded it as over-correct. The name Otcgpc:c is common in Attica(LGPN 2.223) and is attested elsewhere (LGPN 1.219, 3a.2067). Cf. Regen-bogen 1357, J. H. M. A. Indemans, Studien over Theophrastus (Nijmegen 1953)36, Sollenberger 38335.1I NTRODUCTI ON1.4; cf. D.L. 3.46).3Otherwise it will have begun at Assos (onthe coast of Asia Minor opposite Lesbos), where Hermias, rulerof Atarneus, former fellow-student of Aristotle in the Academy,gathered together a group of philosophers after the death ofPlato in 348/7. The association continued in Macedonia, whereAristotle was invited by Philip II in 343/2,4and in Athens, whenAristotle returned there in 335/4 and founded the Lyceum.The vicissitudes of the period which follows, and some of itsleading gures, are reected in the Characters.5Lycurgus, duringwhose period of political inuence Athens had retained a demo-cratic constitution and a measure of independence from Mace-don, died c. 325/4. Alexander (XXIII.3) died in 323. During theuprising against Macedon which followed, Aristotle left Athensfor Euboea, where he died in 322/1, and Theophrastus becamehead of the Lyceum (D.L. 5.36 =fr. 1.57). Antipater (XXIII.4),regent of Macedonia, defeated the Athenians and their allies in322, placed Athens under the control of Phocion, and imposedan oligarchic constitution and a Macedonian garrison. He des-ignated Polyperchon (VIII.6), general of Alexander, to succeedhim in preference to his own son Cassander (VIII.6, 9), withwhom Theophrastus was on friendly terms (D.L. 5.37 = fr. 1.13,Suda O199 =fr. 2.89). Antipater died in 319. Astruggle ensuedbetween Polyperchon and Cassander. Polyperchon offered theGreek cities autonomy in return for their support. Athens ralliedto him and executed Phocion. Cassander defeated Polyperchonand captured Athens in 317 and placed it under the controlof Demetrius of Phaleron, pupil of Theophrastus (D.L. 5.75).6Through his inuence Theophrastus, though a metic (like Aris-totle), was allowed to own land (D.L. 5.39 = fr. 1.3840), and so3Regenbogen 13578, W. K. C. Guthrie, A History of Greek Philosophy 6(Cambridge 1981) 345, K. Gaiser, Theophrast in Assos: zur Entwicklungder Naturwissenschaft zwischen Akademie und Peripatos (Heidelberg 1985) 247,Sollenberger 38067, Mejer 1719.4Cf. Ael. VH 4.19 = fr. 28.5For fuller discussion of historical allusions see the section on Date (pp. 2737).6W. W. Fortenbaugh and E. Sch utrumpf (edd.), Demetrius of Phalerum: Text,Translation and Discussion (New Brunswick and London 2000) 39 (no. 8).2THEOPHRASTUS AND HI S TI MESto establish the Lyceum in buildings of its own.7Demetrius wasexpelled in 307. The restored democracy passed a law requir-ing heads of philosophical schools to obtain a licence from thestate, and Theophrastus (along with other philosophers) brieywithdrew from Athens (D.L. 5.38 = fr. 1.229).8On his return(the law was soon repealed) he remained head of the Lyceumuntil his death at the age of 85 (D.L. 5.40 = fr. 1.46) in 288/7 or287/6.He is reputed to have had some 2,000 students (D.L. 5.37 =fr. 1.16, Suda O 199 = fr. 2.7).9He bequeathed his books to hispupil Neleus of Scepsis (D.L. 5.52 = fr. 1.31011). The narra-tive of their subsequent history should be treated with reserve:together with the books of Aristotle, which Theophrastus hadinherited, they were stored underground, suffered damage, andwere sold to Apellicon of Teos, who issued unreliable copies; thelibrary of Apelliconwas carriedoff toRome whenSulla capturedAthens, and acquired by Tyrannion the grammarian, who, withAndronicus of Rhodes, put further unsatisfactory copies intocirculation (Str. 13.1.54, Plu. Sull. 26.13 = fr. 378).107J. P. Lynch, Aristotles School (Berkeley etc. 1972) 97105, Guthrie 3940,Sollenberger 38223, C. Habicht, Hellenistic Athens and her philoso-phers, inAthen in Hellenistischer Zeit: Gesammelte Aufs atze (Munich1994) 23147(at 236), Mejer 20, L. OSullivan, The law of Sophocles and the beginningof permanent philosophical schools in Athens, RhM 145 (2002) 25162.8Lynch 1034, Sollenberger 38212, Habicht 2367, W. G. Arnott, Alexis:The Fragments (Cambridge 1996) Appendix ii, H. B. Gottschalk in J. M. vanOphuijsen and M. van Raalte (edd.), Theophrastus: Reappraising the Sources(New Brunswick and London 1998) 2823, OSullivan (n. 7 above).9Probably during his whole career (Regenbogen 1358, Habicht 2334, Mejer21, Gottschalk 283) rather than at any one time (advocates of this view arelisted by Sollenberger 3828; add Lane Fox 134 and n. 69, misrepresentingHabicht).10Guthrie 5965 is less sceptical of this story than H. B. Gottschalk, Hermes100 (1972) 33542. For its possible relevance to the early distribution of thephilosophical works of Aristotle and Theophrastus see Regenbogen 13759,Mejer 257. It is unwise to found on it any theory concerning the earlyhistory of the text of the Characters (as does Navarre (1931) 224; contra,Ussher (1960) 1415, Rusten 33). See p. 38 below.3I NTRODUCTI ONII THE NATURE AND PURPOSEOF THE CHARACTERS(i) TitleABV entitle the work Xcpcs:npt. Diogenes Laertius, in hiscatalogue of Theophrastus writings,11lists it twice, rst asHisci ycpcs:npt c, second as Xcpcs:npt nisci (5.478= fr. 1.201, 241 = fr. 436.4a).12The history of the noun ycpcs:np is discussed by A. K orte,Hermes 64 (1929) 6986 and B. A. van Groningen, Mnemosyne58 (1930) 4553. It describes the stamp or imprint on a coin,a distinguishing mark of type or value (Arist. Pol. 1257a41 c,cp ycpcs:np t:tn :c0 tcc0 nutcv; cf. E. El. 5589 :iu totocpstv cttp p,pcu sctcv | cutpcv ycpcs:np ;n tpctisti ut :ci;).13It is also used guratively, to describethe stamp of facial or bodily features, by which kinship or raceare distinguished (Hdt. 1.116.1 :c0:c t,cv:c :c0 tcioc :cvA:u,tc tnit v,vci co:c0 sci c c . . . ycpcs:np :c0tpcctcu tpcgtptci tocstt t tcu:cv, Hyp. fr. 196 Jensenycpcs:np cooti ttt:iv tti :c0 tpcctcu :n oicvcic :cvpctci; cf. A. Su. 282, E. Med. 51619, Hec. 379, El. 572),14and the stamp of speech, as marked by local dialect (ycpcs:np,cn Hdt. 1.57.3, 1.142.4; cf. S. fr. 176) or by a style of speech(Ar. Pax 220 c ,c0v ycpcs:np nutoctc :cv pnu:cv) or(in later literary criticism) by a style of writing (LSJ ii.5, K orte7983). Into this pattern ts Men. fr. 72 vopc ycpcs:np ts11On the nature and sources of this catalogue see H. Usener, AnalectaTheophrastea (Leipzig 1858), Regenbogen 136370, Sollenberger 38545,Mejer 224.12Two late manuscripts which have the title Xcpcs:npt nisci are copiedfrom printed editions (Torraca (1994a) xii n. 8). For the suggestion (unac-ceptable) that the repeated title refers to a second book of Characters seep. 18.13R. Seaford, JHS 118 (1998) 1379; also F. Will, The concept of ycpcs:npin Euripides, Glotta 39 (1960) 2338.14Similarly Shakespeare, The Winters Tale II.3.989 although the print belittle, the whole matter / and copy of the father.4THE NATURE AND PURPOSE OF THE CHARACTERSc,cu ,vcpit:ci, the stamp of a man is recognised from hisspeech: speech typies him, makes him a distinct and recognis-able individual.A work entitled Xcpcs:npt advertises nothing more specicthan types, marks, distinctive features, or styles. This is notan adequate advertisement of Theophrastus work. Denition isneeded, and is provided by nisci, which the manuscripts havelost, but Diogenes Laertius has preserved. The title Characters,hallowed by usage, is both misleading and incomplete. The truetitle means something like Behavioural Types or Distinctive Marks ofCharacter.15We hear of a few other works which may have been enti-tled, in whole or part, Xcpcs:npt: (i) ltpi ttc n ttpiycpcs:npcv by Antisthenes (D.L. 6.15);16(ii) Xcpcs:npt cby Heraclides Ponticus (D.L. 5.88 = fr. 165 Wehrli), perhaps onstyle;17(iii) Xcpcs:npt n 1icscucioci by an unknown tragicpoet Dionysiades of Mallos (TrGF 105), tv ci :cu ycpcs:npc(styles?) tc,,tti :cv tcin:cv (Suda A1169);18(iv) :upctv :c ttpi ycpcs:npcv (Ath. 168C = FHG 3.164 fr. 20), dis-cussed below (p. 11).(ii) Antecedents and relationsThe Characters, in conception and design, is a novel work: noth-ing like it, so far as we know, had been attempted before. Butantecedents and relations can be recognised.Descriptions of character-types had appeared sporadically inother genres. Homer describes the otic and the csiuc in15Addition of nisci is commended by K orte 77 n. 3, P. Steinmetz, AUS 8(1959) 2246 = Kleine Schriften (Stuttgart 2000) 1302 (and his commentary,2 (1962) 78), W. W. Fortenbaugh, RhM 118 (1975) 812, id. Quellen zur EthikTheophrasts (Amsterdam 1984) 934. Contra van Groningen 523.16The nature of the work and the authenticity of the title are disputed: G.Giannantoni, Socratis et Socraticorum Reliquiae 4 (Naples 1990) 2401.17F. Wehrli, Die Schule des Aristoteles, vii: Herakleides Pontikos (Basel 21969) 119.18R. Pfeiffer, History of Classical Scholarship, from the Beginnings to the Hellenistic Age(Oxford 1968) 160.5I NTRODUCTI ONambush, the former pale and dgety, his heart thumping and histeeth chattering, the latter never blanching, eager for the ghtto start (Il. 13.27886). Eustathius recognised in this a foreshad-owing of Theophrastus: oicstucv:c :c0 tcin:c0 pyt-:utisc c tv :tci ycpcs:npc, ctcicu on :ivc 0:tpcvsci Otcgpc:c tt:utcc:c, cc utv c csiuc tv scipcicycu, cc ot c otic (931.223 = 3.469.35 van der Valk).19Semonides describes ten types of women (fr. 7).20Herodotus(throughthe mouthof a Persian) describes the ucvcpyc (3.80.36), and Plato describes the :iucspc:isc (R. 548d550b), theci,cpyisc (553a555a), the onucspc:isc (558c562a), andthe :upcvvisc (571a576b). Aristotle in the Rhetoric describes atlengththe characters (nn) of vtci, tpt:tpci, andsucv:t(1389a31390b13), and more briey of to,tvt, tcici, andouvutvci (1390b161391a29).In the Nicomachean Ethics Aristotle distinguishes and analysesmoral virtues and vices, nisci (as opposed to c,isci) pt:ciand scsici. Virtue is a mean between two opposing vices, one ofdeciency, the other of excess, in emotions and actions (1106b1618). First he lists 13 pairs of vices, with their mean (1107a321108b6).21Theophrastus has 9 (here asterisked) of the 26 vices.Deciency Mean Excessotiic voptic pcvcinic cgpcvn sccicvttutpic ttutpic:n c:ictipcvtic ntic ccvtic,pcisic to:pcttic cuccyic19For a modern misunderstanding which has been built on the passage seep. 19.20H. Lloyd-Jones, Females of the Species: Semonides on Women (London 1975) 29(he may be considered an ancestor of Theophrastus), 323.21Cf. EE 1220b211221b3 (a rather different list), W. F. R. Hardie, AristotlesEthical Theory (Oxford 21980) 12951, R. Bosley, R. A. Shiner, J. D. Sisson(edd.), Aristotle, Virtue and the Mean (Edmonton 1995).6THE NATURE AND PURPOSE OF THE CHARACTERSoutpi:ic giic ptsticouscic giic sccsticvciyuv:ic cionucvn sc:tniAristotle develops the analysis of individual virtues and viceslater (1115a41128b33).22Although he personalises their bear-ers (exemplifying the otic and the voptc, and so on, justas in the Rhetoric he exemplies vtci and tpt:tpci), his per-sons exist, for the most part, out of time and space, moralparadigms, not esh and blood. And so it is with the ucvcpycof Herodotus and the political characters drawn by Plato.But Aristotle provides the seed from which Theophrastussdescriptions grow. He often indicates, in abstract and gen-eral terms, the circumstances or behaviour which are asso-ciated with each virtue and vice. For example, Rh. 1379b1719 :c ttiycipcui :c :uyici sci cc touucuutvci tv:c co:cv :uyici n ,cp typc0 n ci,cpc0v:c nutcv(taking pleasure in the discomforts of others is the nutcv,i.e. ycpcs:np, of a hostile or scornful man), 1383b1920 ccv:c tcctv tioc n gu,tv tc otiic ,p. sci :ctc:tpnci tcpcsc:cnsnv tc oisic ,p, 1383b225:c stpocivtiv tc uispcv n ciypcv n tc ouv:cv . . . tcciypcstpotic ,cp sci vttutpic.Instead of an abstract circumstance Theophrastus gives us areal occasion, and instead of an anonymous agent, a real individ-ual. So, while Aristotle says that :c ttpi co:c0 tv:c t,tiv scittc,,ttci is typical of ccvtic (1384a46), Theophras-tus lets us hear an Accv making just such grand claims forhimself before visitors in the Piraeus (XXIII). The voptc,according to Aristotle, will best display his fearlessness at sea orin war (EN 1115a34b1). Theophrastus shows us the Atic on aship and on the battleeld (XXV). Aristotle is even capable ofanticipating Theophrastuss technique. The vcuc (VulgarMan) makes a tasteless display of his wealth on unimportant22Cf. EE 1228a231234b13, MM 1190b91193a38.7I NTRODUCTI ONoccasions, for example by entertaining his dining club on thescale of a wedding banquet or, when acting as choregus fora comedy, bringing on the chorus in purple (EN 1123a223ccv tpcvi:c ,cuisc t:icv sci scucioc ycpn,cv tv :nitcpcoci tcpgpcv tigtpcv). Witha minimumof change (cctpcvi:c ,cuisc t:icv sci . . . tigtptiv) this becomes indis-tinguishable from Theophrastus in content and style.Like Homer, in his description of the otic and the csiuc,Theophrastus locates his characters in a specic time and place.The time is the late fourth century. The place is Athens. Andit is an Athens whose daily life he recreates for us in dozens ofdramatic pictures and incidents. If we look elsewhere for suchscenes and such people, we shall not nd them (until we cometo the Mimes of Herodas)23except on the comic stage. Plurimainuenias in his breuibus reliquiis, observed Casaubon, quaeueluti tabulae e naufragio superstites utcunque remanserunt,ex quibus huius operis cum poetis, scenicis maxime et comi-cis, quos esse optimos exprimendorum morum artices scimus,afnitas percipi queat.24Comedy furnishes much the same castof players. Five characters of Theophrastus give their namesto plays: the A,pcisc (Antiphanes, Menander, Philemon andothers), Ati:c (Menander), Atiiociucv (Menander), Kcc(Menander and others), Mtuiucipc (Antidotus). Another, theAccv, appears regularly on stage.25Alate and dubious source(Pamphile, FHG3.522 fr. 10 ap. D.L. 5.36 =T. fr. 1.1112 =Men.Test. 8) claims Menander as a pupil of Theophrastus.2623Cf. L. A. Llera Fueyo, Teofrasto y Herodas, Minerva 12 (1998) 91102, andn. 77 below.243rd edn. (1612) 88. 25See the Introductory Note to XXIII.26For suggested afnities with Old Comedy see R. G. Ussher, G&R 24 (1977)759; with later Comedy and Menander, J. van Ijzeren, Theophrastus ende nieuwe comedie, NPh 8 (1923) 20820, P. Steinmetz, Menander undTheophrast: Folgerungen aus dem Dyskolos, RhM 103 (1960) 18591 =Kleine Schriften (Stuttgart 2000) 1528, A. Barigazzi, La Formazione spiritualedi Menandro (Turin 1965) 6986. The subject is handled judiciously by K.Gaiser, Menander und der Peripatos, AA 13 (1967) 840 (esp. 15 n. 36),R. L. Hunter, The New Comedy of Greece and Rome (Cambridge 1985) 1489,8THE NATURE AND PURPOSE OF THE CHARACTERSAnd so a new type of work came into existence, owing some-thing to the ethical theorising of the Lyceum and something tothe comic stage.(iii) Later PeripateticsLater Peripatetics attempted character-drawing of this kind, butto what extent and for what purpose is unclear. Lycon, who suc-ceeded Theophrastuss successor Straton as head of the Lyceumc. 269 bc, wrote a description of a drunkard, preserved in theLatin translation of Rutilius Lupus (Lycon fr. 26 Wehrli ap. Rut.Lup. 2.7, 1st cent. ad). Rutilius adduces it as an example of char-acterismos, the schema by which an orator depicts virtues and vices,and he compares it to a painters use of colours. The opening(Quid in hoc arbitrer bonae spei reliquumresidere, qui omne uitae tempus unaac despicatissima consuetudine producit?) betrays a moralising purpose.The sketch is composed not of illustrations loosely linked but asa coherent narrative, which follows the drunkard through theday, a technique used only once by Theophrastus (the exploitsof the Atic in XXV). In style, it is far from Theophrastus:colours garish, rhetoric over-dressed, cleverness unremitting.27A papyrus of Philodemus preserves parts of a series ofcharacter-sketches, perhaps from a work ltpi :c0 scugitivottpngcvic, OnRelief fromArrogance,28by Aristonof Keos,who was probably Lycons successor (c. 225 bc). The charactersdepicted in the parts we have (they represent aspects of ottpn-gcvic) are the Aoon, Aotsc:c, lcv:tionucv, and Lpcv,of whom the rst and fourth are also depicted by Theophras-tus; and perhaps also the tuvcsctc, Lo:ti:n, andH.-G. Nesselrath, Die attische mittlere Kom odie (Berlin 1990) esp. 1501, LaneFox 13940. See also W. W. Fortenbaugh, Theophrast uber den komischenCharakter, RhM 124 (1981) 24560. For suggested afnities with mime seeH. Reich, Der Mimus (Berlin 1903) 30720.27There is a good appreciation of the piece by G. Pasquali, RLC 1 (1918)1434 = Scritti Filologici (Florence 1986) 568.28For this translation of the title see M. Gigante, CErc 26 (1966) 132 n. 16(cf. 27 (1997) 1534).9I NTRODUCTI ONOootvc:n.29Although the form of the original sketcheshas been obscured by introductory matter, commentary, andparaphrase from Philodemus, it is clear that Ariston followsTheophrastus closely in style, technique, and content. He usesthe introductory formula :cic0:c . . . cc or something like it,30builds his sentences around innitives constructed with that for-mula, makes much use of participles, and normally links clausesand sentences with a simple sci. And he uses the same kind ofillustrative vignettes fromeveryday life: a manasks for hot or coldwater without consulting his fellow-bather (fr. 14, i p. 36.1719tv :ni uspci tpu[c]v [n u]ypcv ci:tv u[n t]pccvcsp.[iv]c.

:cv uutnsc: (uu- Kassel and Austin on Eup. 490)ti ssti[vci uvcptsti) and does not reciprocate a rub withoil (fr. 14, ii p. 36.212 :cv uvcticv:c un v:iuvctigtiv)or is decient in epistolary courtesies (fr. 14, ii p. 36.256,p[g]cv tti:cnv :c yciptiv un tpc,pci (Diggle: tpc-l) uno tppcci :ttu:ccv)31or postures Socratically (fr. 14,vii p. 39.1314 L,c,cp coc:i t[nv ,t] :c:cu, c:i [co]otvcoc;). In style and wit there is nothing to distinguish these fromTheophrastus.3229Text in F. Wehrli, Die Schule des Aristoteles, vi: Lykon und Ariston von Keos (Basel21968) frs. 1416, also in Rusten 18295. Wehrlis view that the character-sketches belong to a separate work, not the work onottpngcvic, is contestedby M. Gigante, Kepos e Peripatos (Naples 1999) 12333. See also W. Kn ogel, DerPeripatetiker Ariston von Keos bei Philodem (Leipzig 1933), Regenbogen 15089.Further bibliography in E. Kondo, CErc 1 (1971) 87 n. 9.30See the commentary on I.2.31Cf. Pl. Bac. 1000 non priu salutem scripsit?, Plu. 1035 b-c (Chrysipp. SVF 2 fr. 30)ti un, scttp c :c ngiuc:c :c tctiv tigtpcv:t tpc,pgcuivA,cnv Tynv, c0:csci co:c tpc,ptit :cv Aics:., Luc. Laps. 5 c0:t:c yciptiv c0:t :c to tp::tiv tpc,pcgtv. The prex tpc- is (i) apt with:cyciptiv, (ii) neededtoprovide a temporal contrast with:ttu:ccv. Thereis a mild zeugma: with uno tppcci :ttu:ccv understand otc,pci(Luc. Laps. 10 tti :tti . . . v:i :c0 tppcci otc,pc :c yciptiv). Seealso XXIV.13.32A good appreciation of his style by Pasquali, RLC 1 (1918) 1447 = ScrittiFilologici (1986) 5962.10THE NATURE AND PURPOSE OF THE CHARACTERSA single sentence is preserved from a work, possibly but notcertainly entitled ltpi ycpcs:npcv,33by Satyrus (Ath. 168c =FHG 3.164 fr. 20), presumably the Peripatetic biographer(3rd/2nd cent.).34It describes the behaviour of cc:ci, ina series of asyndetic participial clauses: tctuici :n coicotpycv:t, c :upc tv :c ttpi ycpcs:npcv tpnstv,sc:c:ptycv:t :cv c,pcv, oicptcv:t :nv cisicv, cupc-tcc0v:t :c otpycv:c, sctc0v:t co :i otoctvn:cic:i octcvnnt:ci, coot :i ttpit:ci c:i cottpit:ci,tv :ni vtc:n:i :c :c0 ,npc tgcoic tpcsc:cvciscv:t,ycipcv:t :ni t:cipci, co :c t:cipci, sci :ci cvci, co :cuutc:ci. The style, all rhetorical balance and antithesis, isunlike Theophrastus, but is not unlike some of the spuriousaccretions (VI.7, VIII.11, X.14).35(iv) Other developmentsThe Stoic Posidonius (fr. 176 Edelstein-Kidd ap. Sen. Ep. 95.657) proclaims the utility of ncc,ic, his termfor ycpcs:npiuc:to display a model of virtue is to invite its imitation. We havealready seen Lycon, with his model of vice, serving the samemoral purpose (p. 9).In the Roman period character-drawing becomes rmlyassociated with rhetoric. The author (1st cent. bc) of Rhetor-ica ad Herennium illustrates the technique of what he calls nota-tio (i.e. ycpcs:npiuc) with a richly textured sketch (4.634),for delivery in court, of The Man Who Shows Off PretendedWealth (ostentatorem pecuniae gloriosum),36at rst in the manner33See p. 5.34Gudeman, Satyros (16 and 17), RE ii.1A (1921) 22835.35Cf. Pasquali (1918) 144 = (1986) 589.36I adopt pecuniosum (Kayser) for pecuniosi (u.l. -sum), since the con-struction ostentatorem pecuniosi (endorsed by TLL and OLD) is unbelievable.Cf. 4.65 huiusmodi notationes . . . totam . . . naturam cuiuspiam ponunt ante oculos,aut gloriosi, ut nos exempli causa coeperamus, aut inuidi etc., Cic. Flac. 52 gloriosaostentatio ciuitatis.11I NTRODUCTI ONof Theophrastus, but soon developing into anecdotal narrativemore inthe manner of Lycon(p. 9).37Cicerouses the termdescrip-tio (Top. 83 descriptio, quam ycpcs:npc Graeci uocant . . . qualis sitauarus, qualis adsentator ceteraque eiusdem generis, in quibus et naturaet uita describitur). Such character-drawing was practised in theschools of rhetoric (Quint. Inst. 6.2.17 illa in scholis nn . . . quibusplerumque rusticos superstitiosos auaros timidos secundumcondicionemposi-tionum efngimus).And character-types are sketched by the satirists: the bore(Hor. S. 1.9), the bellus homo (Mart. 3.63), the miser (Juv. 14.10934).(v) The purpose of the CharactersThe work has been tailored, by more than one hand, to serve anethical purpose. The prooemium introduces it as a work of moralguidance for the young. The epilogues advise or moralise. Thedenitions have links with ethical theorising.38When we are ridof these accretions, the work lacks all ethical dimension. Nothingis analysed, no moral is drawn, no motive is sought.39If thework has a purpose, that purpose must be sought elsewhere. Butpurpose cannot be separated from form. And we do not knowwhether what remains, after the ethical accretions are removed,has the form which Theophrastus gave it.It has been suggested that the Characters are a collection ofextracts fromone or more works of Theophrastus. But the coher-ence and stylistic unity of the collection prove that its parts arenot derived from unconnected works. And, if they are derived37He is comparable to Theophrastuss Accv (XXIII). There is anothershared motif at XXI.4.38See p. 17.39For these as features which fundamentally distinguish the work from Aristo-tles ethical writings see D. J. Furley, The purpose of Theophrastus Charac-ters, SO30 (1953) 5660, W. W. Fortenbaugh, Die Charaktere Theophrasts:Verhaltensregelm aigkeiten und aristotelische Laster, RhM 118 (1975)6282.12THE NATURE AND PURPOSE OF THE CHARACTERSfrom a single work, it still remains to explain what the purposeof that other work might have been.40It has been suggested that the Characters were conceived witha rhetorical purpose.41They are models for orators, a paintboxout of which an orator may draw the shades to suit him.42Or that they have connections with the theoretical writingsof Theophrastus and others on comedy, such as Theophrastussltpi ,tcicu and ltpi scucioic (D.L. 5.46, 47 = fr. 1.184,208 = fr. 666 nos. 23 and 22), or the Tractactus Coislinianus,which has Peripatetic associations and has even been taken to40Extracts from a variety of works were rst suggested by K. G. Sonntag, Inprooemium Characterum Theophrasti (Leipzig 1787); extracts from a work onethics by Schneider (1799) xxv, H. Sauppe, Philodemi de Vitiis Liber Dec-imus (Leipzig 1853) 89, Petersen (1859) 56118, R. Schreiner, De genuinaCharacterum Theophrasteorum Forma Commentatio (Znaim 1879). Jebb (1870)2137 = (1909) 916 argues effectively against Petersen; with equal effect,against the whole theory of extracts, T. Gomperz, Ueber die CharaktereTheophrasts, SAWW117 (1889) x. Abh., 19. See also Gomperz, GriechischeDenker 3 (Leipzig 1909) 37583 = Greek Thinkers 4 (London 1912) 4809. Butthe theory has recently been revived: a Hellenistic compilation in whichTheophrastean material was redistributed under single headings (M. L.West, HSPh 73 (1969) 121 n. 29).41O. Immisch, Ueber Theophrasts Charaktere, Philologus 57 (1898) 193212.Others who see a rhetorical purpose are Furley (n. 39 above), S. Trenkner,The Greek Novella in the Classical Period (Cambridge 1958) 14754 (her claimthat T.s source was not real life so much as an existing tradition of narrativencc,ic, i.e. character-anecdotes, is not established by the detection ofparallel motifs in later Greek and Latin humorists), B. Stevanovi c, Contri-bution au probl` eme des mod` eles de quelques caract` eres de Th eophraste (IXet XXX), ZAnt 10 (1960) 7580, V. V. Valchenko, To what literary family dothe Characters of Theophrastus belong?, VDI 177 (1986) 162 (summary;article in Russian 15662), Fortenbaugh, Theophrastus, the Characters andRhetoric, in W. W. Fortenbaugh and D. C. Mirhady (edd.), Peripatetic Rhetoricafter Aristotle (New Brunswick and London 1994) 1535. Further documen-tation in E. Matelli, S&C 13 (1989) 32935, 37786. Pertinent criticismby Lane Fox 139; more ponderously (against Immisch) C. Hoffmann, DasZweckproblem von Theophrasts Charakteren (Breslau 1920) 928.42eine Motivsammlung, . . . ein Farbenkasten (Immisch 207). This argu-ment owes too much to their later history. They survive because, in theByzantine period, they were incorporated with the treatises of Hermogenesand Apthonius, whose discussions of nc and nctciic they were taken toillustrate. See below, p. 38.13I NTRODUCTI ONderive from Aristotles lost work on comedy.43They are a mereappendix at the end of a work on the theory of drama, anaid for the playwrights of contemporary drama, a handbook ofcharacterization for Menander . . . and his fellows.44Or the work is an otcuvnuc, wie das Skizzenbuch einesMalers zu seinen ausgef uhrten Gem alden, like a painterssketchbook to his nished paintings a preparatory sketch forthe His or ltpi ncv, to which it bears the same relation-ship as the various Aristotelian Constitutions to the Politics and theHomeric Problems to the Poetics.45Any attempt to interpret the work as a serious treatise comesupagainst anobjectionneatly formulatedby Jebb. The difcultyis, not that the descriptions are amusing, but that they are writtenas if their principal aim was to amuse.46Jebbs answer is that Theophrastus wrote the Characters for hisown amusement and that of his friends, who put them togetherafter his death and issued them in collections of various sizes43A. Rostagni, Sui Caratteri di Teofrasto, RFIC 48 (1920) 41743 = ScrittiMinori (Turin 1955) 32755, followed by P. van de Woestyne, Notes sur lanature des Caract` eres de Th eophraste, RBPh 8 (1929) 10991107, Ussher(1960) 56, 23, id. Old Comedy and Character, G&R 24 (1977) 719,A. Dosi, Sulle tracce della Poetica di Teofrasto, RIL 94 (1960) 599672(esp. 6356). For the Tractatus Coislinianus see R. Janko, Aristotle on Comedy(London 1984), Nessselrath (n. 26 above) 10262.44Ussher (1960) 23, (1977) 75. Much the same words in van de Woestyne 1107,Dosi 6356. Pertinent comment in Lane Fox 13940.45Gomperz, SAWW 117 (1889) x. Abh., 1013. The argument that the workis an empirische Materialsammlung zu seinem ethologischen Hauptwerkeltpi ncv (Hoffmann (n. 41 above) 32) is founded on the false assump-tion that the ethical dimension which the work now has was given to it byTheophrastus (see p. 12 above). I say nothing of the curious argument ofP. Steinmetz, Der Zweck der Charaktere Theophrasts, AUS 8 (1959) 20946 = Kleine Schriften (Stuttgart 2000) 11552, that T. is cocking a snook atDicaearchus, Zeno, and Epicurus.46Jebb (1870) 29 = (1909) 13. Comparable, in this respect, is the extractfrom T.s essay on Marriage, translated or paraphrased by Jerome (fr. 486Fortenbaugh; also Fortenbaugh, Quellen l 46, with commentary 20712).Casaubons often cited description of the Characters as aureolus libellus isan echo of Jeromes aureolus Theophrasti liber De Nuptiis.14THE NATURE AND PURPOSE OF THE CHARACTERSand shapes.47In evidence of this he adduces their lack of sym-metry, the capriciousness of their order, and the multiformityof the manuscript tradition. The manuscript tradition licensesno such inference.48With regard to symmetry, some sketchesare incomplete, and others may be.49As for order, accidentsof transmission may have disturbed a less capricious design; orwhat seems caprice may be designed to avoid the appearance ofa textbook.There is another possibility, which meets Jebbs objection, andgives at least as plausible an account of the origins of the sketches.Pasquali suggested that they were conceived as illustrative show-pieces for a course of lectures on ethics, a few moments lightentertainment amid more serious matter, and for that reasoncomposed in a simple style which suits oral delivery, and notdesigned for publication by Theophrastus himself.50According to a reputable source, Theophrastus was a livelylecturer:Lpuittc ot gni Otcgpc:cv tcpc,ivtci ti :cv ttpitc:cvsc cpcv cutpcv sci tnsnutvcv, t:c scicv:c oic:itci :cvc,cv cootuic ttycutvcv sivntc coot ynuc:c tvc. sci tc:tccgc,cv uiucutvcv ttipcv:c :nv ,ccv ttpitiytiv :c ytin(Ath. 21B = Hermipp. fr. 51 Wehrli = T. fr. 12).Hermippus [3rd cent. bc] says that Theophrastus would arrive at thePeripatos punctually, smart and well dressed, then sit down and deliverhis lecture, in the course of which he would use all kinds of movementsand gestures. Once, when he was imitating a gourmet, he stuck out histongue and licked his lips.47Jebb 1821, 3740 = 89, 1617. Lane Fox 141 detects much the samepurpose (see below, p. 37).48See the section on Transmission (pp. 3751).49V and XIX each consist of two parts, which come from separate sketches;in V both parts, in XIX one or both, are incomplete.50elaborazione dei punti salienti di un corso di lezioni di fenomenologia decostumi (RLC 1 (1918) 77 =Scritti Filologici (1986) 53), parte . . . di un corsodi etica descrittiva (ed. 1919, vi = 1956, x). See also (from his later reviewof Navarre) Gnomon 2 (1926) 868 = Scritti Filologici 8447.15I NTRODUCTI ONI can believe it. And I can picture him picking a speck of strawfrom anothers beard (II.3), stufng his cloak into his mouth tostop himself from laughing (II.4), ofciously arranging cushions(II.11), grabbing a dogs snout (IV.9), staggering forwardas if bur-dened by a jar, his hands plucking at documents which threatento elude his grasp (VI.8), dousing himself with a ladleful of water(IX.8), rummaging through the rubbish for a lost coin (X.6),wiping his nose on his hand while pretending to eat and scratch-ing himself while purporting to sacrice (XIX.5), sponging awound and swatting ies (XXV.5), and twisting his buttocks fora wrestling throw (XXVII.14), while reciting his sketches in thelecture hall.There was a famous Professor in Oxford who would introduceinto his seminars, as if on impulse, carefully designed sketchesof past scholars, one for each occasion. I heard him once: hesketched Pasquali.(vi) Authenticity and integrityDoubts have arisen from time to time that Theophrastus is theauthor of the Characters. Doubters include Victorius,51Valcke-naer,52Porson,53and Haupt.54The prooemium used to be a stumbling-block: its author isninety-nine years old, and Theophrastus, according to DiogenesLaertius, died at eighty-ve. Casaubon emended one or othernumber. But now we know better. The prooemium is spurious, avery late addition.When we have deleted the prooemium, what remains is not,as it stands, the work of Theophrastus. Several sketches (I, II,51Variae Lectiones 1 (Lyon1554) 302, 326, 2 (Florence 1569) 210 =ed. 2 (Florence1582) 196, 211, 434.52On Theoc. 15.33 (Leiden 1773, 333).53Putabat scilicet, nisi me uehementer fallit memoria, falso tribuiTheophrasto Characteras, antiquos tamen esse concedens, Dobree on Ar.Pl. 1021 (in P. P. Dobree (ed.), Ricardi Porsoni Notae ad Aristophanem, Cambridge1820).54Opuscula 3 (Leipzig 1876) 434, 498, 592.16THE NATURE AND PURPOSE OF THE CHARACTERSIII, VI, VIII, X, XXVI, XXVII, XXIX) have epilogues, whichbetray themselves as later (perhaps much later) additions by theirlanguage, style, and moralising tone.And there are the introductory denitions. Some reect thepseudo-Platonic Denitions,55others the phraseology of Aristotleor pseudo-Aristotle; some describe a form of behaviour whichhas little or nothing to do with the behaviour described in thesketch itself; even those which are unobjectionable are no betterthan banal. They were added before the time of Philodemus,who quotes def. II. They rst came under suspicion early inthe nineteenth century.56Nearly everyone continued to defendthem.57That they are spurious and must be deleted en bloc hasbeen established beyond all doubt by Markus Stein.58It may beobjected that Stein has proved only that some, not all, denitionsare spurious; and that there are some whose spuriousness cannotbe proved, nor does Stein claim to have proved it. In that spirit,a recent editor has deleted some but not all of them. This iswrong. We cannot pick and choose. The denitions have thesame stamp. They come from the same workshop. They standand fall together.When we have stripped the work of its prooemium, its epilogues,and its denitions, we still have not unwrapped the genuine arti-cle. Numerous further additions are embedded in the sketches,ranging inextent fromsingle words tobrief phrases (IV.4, VIII.7,XVIII.6, XIX.4, XX.9, XXI.11, XXII.7, XXX.10), whole55For which see H. G. Ingenkamp, Untersuchungen zu den pseudoplatonischen De-nitionen (Wiesbaden 1967); also Stein (n. 58 below) esp. 2835.56Priority is usually assigned to F. Hanow, De Theophrasti Characterum Libello(Leipzig 1858). He was anticipated by Bloch (1814), who stigmatised someor most (quaedam xii, xiii, 85, pleraeque 79) but explicitly con-demned only XIII and XXVIII, and by Darvaris (1815), who condemnedthem all.57Exceptions are Petersen (1859), Ussing (1868), and Gomperz (SAWW 117(1889) x. Abh., 24 , ibid. 139 (1898) i. Abh., 1113).58Denition und Schilderung in Theophrasts Charakteren (Stuttgart 1992). See below,p. 57. H. Escola, Le statut des d enitions dans les Caract`eres: de Th eophraste` a La Bruy` ere, Lalies 17 (1997) 17586, contributes nothing pertinent.17I NTRODUCTI ONsentences (II.9, VI.2, VII.5, VIII.5, XVI.13) and even a sen-tence of paragraph length (VI.7).Here is a simple proof that interpolation is a real phe-nomenon, not a ction designed to save Theophrastuss credit.In V.10 a show-off hires out his little wrestling-school to :cgiccgci :c cgi:c :c ctcuyci :c cpucvisc,for them to perform in. This quartet of philosophers, sophists,drill-sergeants, and music lecturers, listed in asyndeton, oughtto worry us. Theophrastus has several trios of nouns or verbsin asyndeton, but no quartets. Furthermore, philosophers andsophists are too much alike, when compared with the pair whichfollows, drill-sergeants and music lecturers. If we are to reducethe list to three, by getting rid of either the sophists or the philoso-phers, we must get rid of the philosophers, because sophists aremore likely than philosophers to wish to hire a place for publicdisplays. And the Herculaneumpapyrus omits the philosophers.There is an important lesson here. Anything that is anomalousshould be regarded with suspicion. Nothing is genuine merelybecause it is in the manuscripts and cannot be proved to bespurious.Much, then, has been added; and probably much has beenlost.59It has evenbeenarguedthat a whole secondbook, describ-ing virtuous characters, once existed.60This rests on three sup-positions, all false: (i) That the author of the prooemium, whenhe says that he will describe :cu ,cc as well as :cugccu, knew of a book of ,cci. The author makes sev-eral statements which show him to be a bungler and a fraud.(ii) That Diogenes Laertius, when he lists Xcpcs:npt twice,61refers to two separate books. His catalogue is made up of fouror ve different lists,62so that several titles appear more than59There are many lacunae. And there were once more than thirty sketches(n. 49 above).60For example, Rostagni (1920) 43940 = (1955) 3501, Edmonds (1929) 78,Ussher (1960) xi, 34, (1993) 3012, Torraca (1994a) xxxxxxii.61See p. 4. 62See n. 11 above.18THE NATURE AND PURPOSE OF THE CHARACTERSonce.63(iii) That Eustathius, when he says (in the passagequoted above, p. 6) that Homer created archetypal characters,as Theophrastus was later to do, cc utv c csiuc tv scipcicycu, cc ot c otic, ascribes to Theophrastus a descriptionof the csiuc as well as the otic. The words tv scipci cycushow that Eustathius is citing these characters from Homer, notfrom Theophrastus.64(vii) Integrity and styleAntiquity believed that Theophrastus was aptly named, becausehis speech was divine.65Quintilian praised its brightness (Inst.10.1.83 loquendi nitor ille diuinus), Cicero its sweetness (Ac. 1.33oratione suauis, Brut. 121 quis . . . Theophrasto dulcior?),66and he wasaccustomed to call Theophrastus his ioic :pugn, own specialdelight (Plu. Cic. 24.6).Some modern judges have looked in vain for sweetness andbrightness in the Characters. The Greek is not Greek at its mostlimpid;67sometimes obscure or inelegant . . . unvaried andabrupt, notes for lectures . . . they can hardly have been writtenfor separate publication as a literary work.68Let us take another lesson from the Herculaneum papyrus.The Greek for that little wrestling-school is, according to themanuscripts, coioicv tcci:piccv. The noun coioicv isattested once, as diminutive of coc, in the sense small tube.LSJ invents a sense for it to have here, place of athletic exer-cises, ring. The adjective tcci:picc is attested only here.63ltpi oiccn c (fr. 1.189, 252, 275), ltpi :cv tpcnu:cv guiscv c(fr. 1.227, 266), lpc:ptt:isc c (fr. 1.262, 284). Several other titles appearto be variants of each other.64For a further awed attempt to nd traces of a lost sketch in Eustathius seethe Introductory Note to II.65See p. 1.66For suauis and dulcis as terms of stylistic criticism see D. C. Innes in W. W.Fortenbaugh, P. M. Huby, A. A. Long (edd.), Theophrastus of Eresus: On his Lifeand Work (New Brunswick and Oxford 1985) 251.67R. G. Ussher (1960) 3. 68P. Vellacott (1967) 8.19I NTRODUCTI ONLSJ takes it to mean suited for a tcci:pc. Cobet replacedcoioicv tcci:piccv with tcci:pioicv. The papyrus con-rms his conjecture. But, if the papyrus did not exist, editorswould be as blind to its merits as LSJ. The lesson is the same asbefore. Anomalies ought to provoke suspicion. Nothing is rightmerely because it is in the manuscripts and cannot be proved tobe wrong.And the application of that lesson is this: we must not callTheophrastus obscure and inelegant and not limpid, simplybecause much of what we read in our printed texts is obscure andinelegant andunlimpid. Our printedtexts are nothing more thanthe best that editors have been able to make of what is probablythe corruptest manuscript tradition in all of Greek literature.Let us now see that Theophrastus can, and often does, writeGreek that is the reverse of obscure and inelegant and unlimpid.The A,pcisc is a countryman who comes to town and showshis country manners. Here is the rst sentence of the sketch:c ot c,pcisc :cic0:c :i cc sustcvc ticv ti tssnicvtcpttci, sci :c upcv gstiv cootv :c0 ucu noicv ctiv, sciutic :c0 tcoc :c otconuc:c gcptv, sci ut,ni :ni gcvnictv.The Country Bumpkin is the sort of man who drinks a bowl of gruelbefore going to the Assembly and claims that garlic smells as sweetlyas perfume, wears shoes too large for his feet and talks at the top of hisvoice (IV.2).What could be more limpid than that? The Greek is simpli-city itself, and conveys, in a very few words, a range of tellingimpressions, which develop logically the one from the other.First, he drinks for breakfast a sustcv, highly avoured broth orgruel. His breath will now be pungent. He goes to the Assem-bly, where he will meet townsmen, on whom he will pungentlybreathe. And he says that garlic smells as sweetly as perfume.There was (we infer) garlic in his gruel, and so there is garlic onhis breath. In the town they smell not of garlic but of perfume.But perfume and garlic are all one to him. And he clomps his20THE NATURE AND PURPOSE OF THE CHARACTERSway to town in boots too big for him, and talks too loud. Sound,sight, smell: a slovenly carefree inconsiderate yokel. All that intwenty-six words. Lecture notes, never intended for publication?Or loquendi nitor ille diuinus?Another illustration from the same sketch::ni pci (:nv pcv AB) otcsc0ci co:c, sci :cv svctpcsctutvc sci tticcutvc :c0 p,ycu tittv Oo:cgu::ti :c ycpicv sci :nv cisicv.He answers the door himself, calls his dog, grabs it by the snout, andsays This guards my estate and home (IV.9).First, he answers the door himself. Why? Normally, you wouldhave a slave to answer the door for you. Is he too poor to keep aslave for that purpose? On the contrary, he has an ample house-hold, as we learn elsewhere in the sketch. What follows suggestsa different answer. A knock at the door alarms him, and so heinvestigates for himself who his visitor is. Perhaps he does nothave many visitors, and anyone who knocks at his door is anobject of suspicion. Next, he muzzles the dog by taking holdof its snout. Again, why? Again, Theophrastus has prompted aquestion, and again we have to supply the answer. By muzzlingthe dog he shows his visitor that it can bark and bite, and willdo so if he lets go of its snout. If the visitor intends harm, hewill take the mans action to mean Beware of the dog. If heintends no harm, he may suppose that the dog has been muz-zled as a courtesy to him. Then the A,pcisc grandly describesthe dog as guardian of his estate and home. If the visitor is inno-cent, this is an expression of pride in the animal. Otherwise, itmeans This dog has got the measure of you. The words ycpicvand cisic, simple and prosaic on their own, when paired soundpompous and affected. There is something very similar in Petro-nius. Trimalchio summons his dog Scylax into the dining roomand calls him, with affectation and pomposity, praesidium domusfamiliaeque (64.7), the protection of my house and household.The lesson is this. By the simplicity and economy of his language21I NTRODUCTI ONTheophrastus can prompt us to think, to ask questions, to ll inthe details for ourselves and supply the thoughts at which heonly hints.Next, see how much he can hint at in the careful placing ofa single word. The Oiucn, The Late Learner, is a man whopursues activities for which he is too old:tpcv t:cipc sci spic (-cu V) tpccv :c pci tn,ctingc ot v:tpc:c0 oistci.He falls for a courtesan and rams her door, and when her other loverbeats him up he goes to court (XXVII.9).This is a masterly sentence, short and simple, with the mosttelling detail reserved for the nal word.69A man past his primehas fallen for a hetaira. He behaves like the typical infatuatedyoung lover from comedy, elegy, and mime: he tries to batterher door down. Along comes her other lover, a young man weassume, to claim not only the girl but also the role (as batterer)which the old man has usurped from him. So battery (but ofa different kind) follows: he beats the old man up. And nowcomes the real punch. Because we have not yet had an innitive,we know that the story is not over. What conclusion might weexpect? Any sensible man will now retire chastened, to lick hiswounds in silence and hush up his humiliation. But not our LateLearner. He takes the young man to court on a charge of assaultand battery. He steps out of comedy, elegy, and mime, and stepsback into real life, to become an ordinary litigious Athenian. Butat the same time he remains the man he was, insensitive to hisown absurdity, impervious to the ridicule of others: ridiculousthen as the elderly lover, now to be ridiculous again when hispast behaviour is exposed in court. What an ancient biographersaid of Sophocles could equally be said of Theophrastus, that69I leave for the commentary discussion of the conjecture spic, which addsyet more vigour to the picture.22THE NATURE AND PURPOSE OF THE CHARACTERShe can create a whole personality out of half a line or a singleword.70Now look at a couple of nouns. The Atcvtvcnutvc, TheMan Who Has Lost All Sense, comes into courttycv tyvcv tv :ci tpcsctici sci cpuccu ,pcuuc:tioicv tv :cytpivwith a boxful of evidence in his coat pocket and strings of little docu-ments in his hands (VI.8).This translation does not get the full avour of the nouns. Thetyvc is a sealed jar in which a plaintiff or defendant places allthe evidence relating to an impending court case. The tpcsc-ticv is a sort of pouch, such as kangaroos have. You make thispouch by pulling your yi:cv up through your belt and lettingit hang out in a capacious fold. Why he needs to carry the jarin this pouch is shown by the next phrase. His hands are full ofcpuccu ,pcuuc:tioicv, strings or chains of little documents.Some take this in a literal sense, to mean that the documents aretied together in a bundle. But a word exists for a bundle of doc-uments tied together. That word is not cpucc but otun. Thestrings or chains are probably metaphorical. And so the man,as he enters the courtroom, cuts a ridiculous and ungainly gureby carrying a bulky jar in the front fold of his cloak, while hishands are full of an endless chain of little documents. This isthe kind of picture that Dickens loves to draw, where farce andexaggeration teeter on the borders of the credible.Now see how a style of speech can characterise a man. TheMispcgic:iuc, The Man of Petty Ambition, while servingas a member of the Council, secures for himself the task ofannouncing in the Assembly the outcome of ofcial sacricesperformed by himself and his colleagues at the festival calledGalaxia.70TrGF 4 Test. a 1.901 ts uispc0 nui:iyicu n ttc uic ccv nctcitvtpcctcv. The same was said of Homer: 2d Il. 8.85 otivc t:iv Ounpcsci oic uic ttc ccv :cv cvopc nucivtiv.23I NTRODUCTI ONtcptstucutvc cutpcv u:icv sci t:tgcvcutvc tcptcvtittv 0 cvopt Anvcci, tcutv c tpu:vti [:c tpc] :niMn:pi :cv tcv :c Icic (,cp cic V), sci :c tpc sc, sciout otytt :c ,c.He steps forward wearing a smart white cloak, with a crown on hishead, and says Men of Athens, my colleagues and I celebrated theMilk-Feast with sacrices to the Mother of the Gods. The sacriceswere propitious. We beg you to accept your blessings (XXI.11).He asks for this task because it gives himhis brief moment of lime-light, a solo performance, garlanded and brightly robed, with asolemn and impressive script. It was not a demanding speech tomake, since it was composed entirely of traditional phrases, aswe can see from a similar announcement in Demosthenes:0 cvopt Anvcci . . . tcutv :ci Aii :ci c:npi sci :ni Anvcisci :ni Nisni, sci ,t,cvtv scc sci c:npic :c0 ouv :c tp.tcutv ot sci :ni ltic sci :ni Mn:pi :cv tcv sci :ci Atccvi,sci tscitpc0utv sci :c0:c. nv o ouv sci :c :c cci tc:utv tp gcn sci tcic sci scc sci c:npic. otyt covtcpc :cv tcv oiocv:cv :,c.Men of Athens . . . we sacriced to Zeus the Saviour and Athena andVictory, and these sacrices were propitious and salvatory for you. Andwe sacriced to Persuasion and the Mother of the Gods and Apollo,and we had propitious sacrices here too. And the sacrices made tothe other gods were safe and secure and propitious and salvatory foryou. Therefore we beg you to accept the blessings which the gods give(Prooem. 54).For all the community of phrases, the speeches are different instyle. The speaker in Demosthenes has sacriced to a multitudeof gods: to so many that he divides his list into three parts, whoselanguage and structure he varies. The Mispcgic:iuc has onlya single sacrice to report, and his report is accordingly barer.This sacrice was held for the Galaxia, which seems to havebeen a tranquil and somewhat unimportant affair.71We may71R. Parker, Athenian Religion: A History (Oxford 1996) 192.24THE NATURE AND PURPOSE OF THE CHARACTERSsuspect that the occasion which he chooses to report is not theone which would best have served his wish to be impressive,and that the mention of the Galaxia, which takes its name froma noun meaning a barley porridge cooked in milk, deates thesolemnity of the traditional phrases. The man himself, however,is satised with his performance. For the sketch has a wonderfullast sentence:sci :c0:c tc,,tic ttcv (ticv V) cisot oin,ncci (o-ci- V) :ni tcu:c0 ,uvcisi c sc ottpcnv nonutpti (tonutptv V).After making this report he goes home and tells his wife that he hadan extremely successful day (XXI.11).This brings to mind the deluded Harpagus in Herodotus: Hewent home . . . in his delight he told his wife what had happened(1.119.12). It was a stroke of genius on the part of each authorto bring in the wife to listen to her husbands naivet e.Here is the essence of the problem. We often nd that ourtext of Theophrastus exhibits qualities of language and stylevery different from those which he is capable of achieving, thatit really is obscure and inelegant, that it is not Greek at its mostlimpid. Let us concede that a writer may be inelegant at onemoment, elegant at another, at one moment obscure, at anotherlimpid. But I should not expect that a writer who is capableof writing with consummate elegance and limpidity will readilybe satised with inelegance and obscurity. And so, when ourtext exhibits these faults, we have a right to be dissatised andsuspicious.(viii) Literary inuenceThe Characters were imitated by Ariston of Keos in the late thirdcentury bc.72In the rst century bc Philodemus quotes V anddef. II, and a papyrus attests parts of VII and VIII.73Thereafter,until they reappear in the medieval manuscripts, the only trace72See pp. 910. 73See pp. 378, 50, and on def. II.25I NTRODUCTI ONof them is a papyrus of the third century ad, which attests anabbreviated version of parts of XXV and XXVI.74It has beenclaimed that they are imitated by Petronius75and Lucian.76These claims cannot be substantiated.77And when DiogenesLaertius lists themin the third century, he is merely reproducingan entry from a much earlier catalogue.78They are next men-tioned by Eustathius79and Tzetzes (Chil. 9.9345) in the twelfthcentury, after the date of our earliest manuscripts.80It is not until the seventeenth century, in England and France,that the name of Theophrastus becomes inseparable from thegenre of character writing. Some account of the impulse whichhe gave to the genre may be found in Jebb (Introduction ii),R. G. Ussher, Some Characters of Athens, Rome, and England,G&R 13 (1966) 6478, W. Anderson, Theophrastus, The CharacterSketches, translated, with Notes and Introductory Essays (Kent State1970) xxixxxii, 13353, Rusten 3441. For further study thefollowing are especially valuable: G. S. Gordon, Theophrastusand his imitators, in Gordon (ed.), English Literature and the Classics(Oxford 1912) 4986, R. Aldington, A Book of Characters, fromTheophrastus; Joseph Hall, Sir Thomas Overbury, Nicolas Breton, JohnEarle, Thomas Fuller, and other English Authors; Jean de La Bruy`ere,Vauvenargues, and other French Authors (London 1924), B. Boyce, The74See p. 50.75M. Rosenbl uth, Beitr age zur Quellenkunde von Petrons Satiren (Berlin 1909) 5662, O. Raith, Petronius ein Epikureer (Nuremberg 1963) 207, P. G. Walsh, TheRoman Novel (Cambridge 1970) 1334, D. F. Le ao, Trimalqui ao ` a luz dosCaracteres de Teofrasto, Humanitas 49 (1997) 14767. But J. P. Sullivan, TheSatyricon of Petronius: A Literary Study (London 1968) 1389, is suitably sceptical.76M. D. Macleod, Lucians knowledge of Theophrastus, Mnemosyne 27 (1974)756, B. Baldwin, Lucian and Theophrastus, Mnemosyne 30 (1977) 1746.77See on III.3, IV.9, VII.3, def. XXVII. Llera Fueyo (n. 23 above) prudentlystops short of concluding that Herodas was acquainted with them. F. Titch-ener, Plutarch, Aristotle and the Characters of Theophrastus, in A. P erezJim enez et al. (edd.), Plutarco, Plat on y Arist oteles (Madrid 1999) 67582, failsto establish that Plutarch was acquainted with them.78See n. 11 above. 79See pp. 6, 19.80Lane Fox 1278, in claiming that they were read by St. John Climacus(6th7th cent.), misunderstands (and misdates) Immisch (1923) 2.26DATETheophrastan Character in England to 1642 (Harvard 1947), J. W.Smeed, The Theophrastan Character: The History of a Literary Genre(Oxford 1985).III DATEThe main contributions: C. Cichorius in Bechert et al. (1897)lviilxii; F. R uhl, Die Abfassungszeit von Theophrasts Charak-teren, RhM 53 (1898) 3247; A. L. Boegehold, The date ofTheophrastus Characters, TAPhA 90 (1959) 1519; Stein 2145;R. J. Lane Fox, PCPhS 42 (1996) 1349. Three dates are in ques-tion: dramatic date, date of composition, date of publication.I begin with two sketches, VIII and XXIII, which allude tohistorical persons and events.In XXIII the Accv claims that he campaigned withAlexander (3), that he has received three invitations fromAntipater to visit him in Macedonia, and that he has declinedthe offer of permission to export Macedonian timber duty-freethrough fear of attack by sycophants (4). He also claims thathe made voluntary contributions to needy citizens in the grain-shortage (5).Antipater was appointed by Alexander as his military deputyin Macedonia in 334, and his appointment was conrmed afterAlexanders death ( June 323). His victory over the Greek statesin the Lamian war (autumn 323 to autumn 322) left him masterof Athens, on which he imposed Phocion, an oligarchic consti-tution, and a Macedonian garrison. He died in early autumn319.81Serious shortages of grain are attested in 330/29, 328/7,323/2, and there may have been others in the decade 33020.The shortage in 328/7 appears to have been particularly seri-ous.82The dramatic date therefore falls between 330 and 319.Cichorius asserted without argument that Alexander is dead.He then argued that the only occasion when Antipater stayed81For the date see R. M. Errington, Hermes 105 (1977) 488, A. B. Bosworth,Chiron 22 (1992) 59.82See the commentary for fuller discussion.27I NTRODUCTI ONlong enough in Macedonia to be imagined as issuing three invi-tations was between early 320, when he returned from Asia(where the dynastic intrigues of Perdiccas had called him), andhis death in 319.The Accv does not explicitly say that Alexander is dead.And Stein argued that, even while Alexander was alive, Antipa-ter, as his deputy in Europe, was a gure of such standing thatan invitation from him makes a suitable object of boasting.Stein suggested three possible dramatic dates: (i) between theend of the grain-shortage (he dated this 326) and the beginningof the Lamian war (autumn 323); (ii) between the end of theLamian war (autumn 322) and the beginning of the Aetolian war(he dated this summer 321)83; (iii) between his return from Asia(he dated this early 319)84and his death (early autumn 319).Lane Fox argued for a dramatic date in Alexanders lifetime,because friends of Macedon were politically safe from 322 to320, whereas, before that time, acceptance of them [sc. lettersof invitation from Antipater] risked attack by sycophants. Thisis mistaken. The Accv fears attack not for accepting lettersof invitation to visit Macedonia but for accepting a more com-promising invitation, to export Macedonian timber duty-free.The importation of goods from an enemy state was an offenceinviting prosecution.85If Macedonia were the enemy, the issuewould be clear-cut: he would be a legitimate object of attack.In 322319, when Macedonia is not an enemy but an ally, hewould be free to accept the invitation. He declines it ctc uno 83Stein 37, following R. M. Errington, JHS 90 (1970) 76; similarly J. D.Grainger, The League of the Aitolians (Leiden etc. 1999) 625. The conventionaldate is late 322: Hammond in N. G. L. Hammond and F. W. Walbank, AHistory of Macedonia 3 (Oxford 1988) 115, 120 n. 1, A. B. Bosworth, CQ 43(1993) 426 n. 34.84Stein 379, following B. Gullath and L. Schober in H. Kalcyk, B. Gullathand A. Graeber (edd.), Studien zur Alten Geschichte, Siegfried Lauffer zum 70.Geburtstag . . . dargebracht 1 (Rome 1986) 336. Alternative dates: autumn 320(Errington (1977) 487), spring 320 (Hammond 1289, 618, Bosworth (1992)5960, (1993) 255).85See the commentary on XXIII.4.28DATEog tvc uscgcv:nni, so that not even one person can bringa trumped up charge against him. It suits him to suppose thatthere is still the risk of a prosecution prompted by malice andjealousy. He is living in a fantasy world, and he has to nd somereason for declining an offer that was never made.Onthe most natural reading, Alexander is deadandAntipateris the most important man in the world. And this is what Antipa-ter was to become, when, with Perdiccas dead, he returned fromAsia in 320 or 319. Adramatic date of 319 is therefore more likelythan any other. And since familiarity with Antipater ceases to bea topical subject for boasting as soon as he is dead (early autumn319), the date of composition is unlikely to be much later than319.In VIII the Ac,ctcic claims that Polyperchon and the kinghave recently defeated Cassander, who has been captured.Antipater designated Polyperchon, a general of Alexander, tosucceed him as military commander in Greece, in preference tohis own son Cassander. The ensuing struggle between Polyper-chon and Cassander continued until 309. Polyperchon offeredthe Greek cities autonomy in return for their support. The Athe-nians executed Phocion and briey returned to democracy in318. Cassander captured Athens and placed it under the controlof Demetrius of Phaleron in 317. He invaded Macedonia, per-haps in 316, and defeated Polyperchon. Polyperchon invadedMacedonia in 309, but made peace with Cassander.There are three candidates for the title of king during thisperiod (319309):(i) Philip III Arrhidaeus,86mentally impaired half-brother ofAlexander, proclaimed Alexanders successor by the army atBabylon, a cipher in the hands successively of Perdiccas, Antipa-ter and Polyperchon, by whom he was murdered in 317.86H. Berve, Das Alexanderreich auf prosopographischer Grundlage 2 (Munich 1926)no. 781.29I NTRODUCTI ON(ii) Alexander IV,87posthumous son of Alexander andRoxane, elevated by Perdiccas to be joint ruler with Philip III,captured by Cassander in 317/16 or 316/1588and murdered byhim in 310 or 309.89(iii) Heracles,90bastard son of Alexander and Barsine, pro-claimed king by Polyperchon in 310 but murdered by him at theprompting of Cassander in 309.91The purported defeat of Cassander distresses the ruling partyat Athens (8). Therefore the ruling party are pro-Macedonian:the oligarchs either under Phocion or under Demetrius ofPhaleron. The outer chronological limits are therefore: (i)autumn/winter 319/18 (when Polyperchon opened hostilitieswith Cassander, by offering autonomy to the Greek cities)92tospring 318 (fall of Phocion);93(ii) early 317/summer 317 (begin-ning of the oligarchy of Demetrius of Phaleron)94to 309. Theplace of the battle is not specied. Since the news was broughtto the ruling party four days ago, but is not yet generally known(8), it must have taken place a good distance away; since themessenger came from Macedonia, it must have taken place inor near Macedonia.If the battle took place during the oligarchy of Phocion, theking may be either Philip (favoured by Cichorius) or Alexander.Both kings were in the charge of Polyperchon (D.S. 18.48.4, 49.4,55.1). Alexander, a mere infant, is less likely than Philip to bedescribed as sharing a military victory with him. But it maybe doubted whether Philip, any more than Alexander, would be87Kaerst, Alexandros (11), RE i.1 (1893) 14345. 88See n. 99.89For 309, Hammond 1657; for 310 (the traditional date), Stein, Prometheus19 (1993) 1503, Bosworth in A. B. Bosworth and E. J. Baynham (edd.),Alexander the Great in Fact and Fiction (Oxford 2000) 214 n. 32.90Berve no. 353. 91Stein (1993) 1503.92J. M. Williams, Hermes 112 (1984) 303, Gullath and Schober 33847,Bosworth (1992) 67.93Williams 3005, Gullath and Schober 33847, Bosworth (1992) 6870; notsummer/autumn 318 (Errington (1977) 48992).94Errington (1977) 494 (July/August), Gullath and Schober 376 (August),Bosworth (1992) 71 (early months of 317).30DATEdesignated as the king, at a time whenhe is only one king of two.The duality of the kings is widely and consistently recognised inboth inscriptions and literary texts.95Stein claimed that it wasin the name of Philip alone that Polyperchon offered autonomyto the Greek cities, on the evidence of D.S. 18.56.2 1iittc cnut:tpc tc:np, 7 1iittc . . . c tc:np, i.e. Philip II. This ismistaken. The offer was made in the name of both kings (18.55.4:cv citcv, 56.2 :n citic ti nuc scnscn). Thedesignation 1iittc c tc:np embraces not only the fatherof the one but also the grandfather of the other.96Lane Fox, aswell as emphasising the duality of the kings, questioned whetherCassander could have fought a battle in Macedonia during thisperiod. Soon after the death of Antipater (early autumn 319)Cassander left for the Hellespont, and people in Athens wouldknow that Cassander was no longer in Macedonia. If we needto circumvent this argument, we can simply locate the battle inThrace.If the battle took place during the oligarchy of Demetrius ofPhaleron, there are three options:(i) It is possible (but it has been disputed) that Cassanderinvaded Macedonia in 317.97If he did, and the battle took placeduring this invasion, we have the same difculty over the king,at least in the earlier part of the year. Philip was murdered byPolyperchon in autumn 317,98after his wife Eurydice, usurpinghis authority, aligned herself with Cassander. After his death,Alexander remains the sole candidate for king.95C. Habicht, Literarische und epigraphische Uberlieferung zur GeschichteAlexanders und seiner ersten Nachfolger, in Akten des VI. Internationalen Kon-gresses f ur Griechische und Lateinische Epigraphik, M unchen 1972 (Vestigia 17, Munich1973) 36777, Hammond 138 n. 2, Bosworth (1993) 4207, Lane Fox 137.96For a more sophisticated explanation see Habicht 3757.97In favour, Errington (1977) 483, 494 (late autumn 317), Hammond 1378,Bosworth (1992) 64, 713, (2000) 210 n. 12 (early 317); against, Gullathand Schober 35976; sceptical, Stein 2330. A dramatic date during thisinvasion is contemplated by Hammond 138 n. 1, Bosworth (1992) 72 n. 84.98Errington (1977) 402, Gullath and Schober 3368, Hammond 140,Bosworth (1992) 56.31I NTRODUCTI ON(ii) Cassander invadedMacedonia, defeatedPolyperchon, andcaptured the remaining king, Alexander IV, in either 317/16 or316/15.99During the one or other period, a victory by Polyper-chon and the king would be a plausible ction. The objectionof Cichorius that the king is too young, at the age of six or seven,to be linked with Polyperchon as winner of a military victory isvery weak. It is uncertain how strong is the objection that a boyof that age would not be referred to baldly as the king.100(iii) In 310/9 Polyperchon summoned the seventeen-yearold Heracles from Pergamum and proclaimed him king (D.S.20.20.12). He confronted Cassander in Macedonia, came toterms withhim, andmurderedHeracles (D.S. 20.28.13). Cicho-rius objected that in an Athens governed by Cassanders allyDemetrius of Phaleron the pretender would not be referred toas the king. But Cassander himself refers to him as the king(D.S. 20.28.2).101Of these options the third, advocated by Lane Fox, is the mostattractive.102But I do not rule out the second (317/16 or 316/15).Date of composition would be soon after the dramatic date, sinceinterest would fade as topicality faded. Against the earliest date,in the oligarchy of Phocion, the anomaly of a reference to theking, when there were two joint kings, is a serious obstacle.103By contrast with VIII, the dramatic date of XXVI (theOi,cpyisc) falls in a period of democracy. The theoretical99For 317/16, Hammond 1412, Bosworth(1992) 612; for 316/15, Errington(1977) 488, 495, Gullath and Schober 377, Stein 314.100See Stein 21. 101Cf. R uhl 325, Stein 21.102The argument which he builds on 9 (In 319 his era of strength was still inthe future, By 310/9, Cassander had indeed grown strong) is precarious,since the text is incurably corrupt at the vital point. See also Stein 22 n. 4.103The false report (D.S. 19.23) of the death of Cassander and the triumph ofPolyperchon issued by Eumenes, Cassanders adversary in Asia Minor, in317 or 316 (Errington (1977) 483, Hammond 141, Bosworth (1992) 624,(2000) 210, Stein (1993) 14650, Lane Fox 136), comparable though it is,has no bearing on the date of VIII.32DATEpossibilities are: (i) before 322 (advent of Phocion); (ii) 318/17(between Phocion and Demetrius of Phaleron); (iii) after 307(fall of Demetrius). The last of these is excluded by the mentionof liturgies in 5. These were abolished by Demetrius and neverreinstated.104In 2 the people are debating :ivc :ci cpycv:i tpcci-pncv:ci (tpc- V) :n tcutn :cu uvttiutncutvcu,whom they will appoint in addition to help the archon withthe procession. The eponymous archon organised the annualprocessionat the Great Dionysia withthe helpof tenttiutn:ci.According to [Arist.] Ath. 56.4 these were originally elected by ashow of hands in the Assembly and contributed to the expensesof the procession from their own pockets, but afterwards werechosen by lot, one from each tribe, and received an allowance.The change from election to lot occurred after 349/8, the dateof D. 21.15 sttcv tcu:cv ti Aicvic ytipc:cvtv ttiut-n:nv.105Rhodes has suggested that the change was a part ofthe reorganisation of Athens festivals in the Lycurgan periodand will have been very recent indeed when A.P. was written;106104W. S. Ferguson, Hellenistic Athens (London 1911) 558, 99, Pickard-Cambridge, DFA 913, Stein 40 n. 2, P. Wilson, The Athenian Institutionof the Khoregia (Cambridge 2000) 2702, S. V. Tracy in Fortenbaugh andSch utrumpf (n. 6 above) 342, H. B. Gottschalk, ibid. 371. See the commen-tary on XXIII.6.105MacDowell ad loc. suggests that the change occurred before 328/7, onthe evidence of IG ii2354.1516 c cycv:t ttiutn:[c]i :n toscuic:n ttp[i] :c tc:pcv. That these overseers of good order in the theatreare identical with the ofcials who are responsible for the procession (anassumption shared by Wilson 24, but not by Pickard-Cambridge, DFA 70)is unlikely. Perhaps the overseers of good order are the ttiutcutvci ofthe Dionysia mentioned by D. 4.35 (351 bc). These, who are described asappointed by lot, cannot be the ttiutn:ci of the procession, who were stillbeing elected at this time. We even hear of elected ttiutn:ci who wereresponsible for keeping dramatic choruses in order (Suda L 2466, DFA 91).Wilson 15960, again failing to distinguish these from the others, is wrongto accuse D. 21.17 of misrepresentation.106P. J. Rhodes, A Commentary on the Aristotelian Athenaion Politeia (Oxford21993) 628.33I NTRODUCTI ONthat it occurred perhaps in the mid 330s.107The date of Ath.56.4 is uncertain. Rhodes has suggested that an original versionof Ath. was composed in the late 330s, and that additions wereincorporated in the mid 320s (with 322 as the latest possibledate).108There is nothing to indicate that Ath. 56.4 was amongthe later additions.109Since Theophrastus species election, either (i) he refers toa time before the procedure changed, or (ii) he refers to a timewhen there had been a change back to the original procedure,or (iii) he ignores the change.Of these alternatives, (i) implies a date not later than c. 335,if Ath. 56.4 is dated in the late 330s; if Ath. 56.4 was addedin the mid 320s, and the change occurred in the early 320s,a date c. 330 becomes possible. For (ii), there is inscriptionalevidence that a change back to election did occur: certainly by186/5 bc (IG ii2896.345), possibly by 282/1 bc (IG ii2668= SIG3388.1315, 23).110Boegehold suggested that it occurredduring the oligarchy of 322318: lot is democratic, and oligarchsprefer election.111If this were right, the dramatic date would be318/17. Stein favoured (iii): Theophrastus ignores the changethrough oversight (historical accuracy was not crucial in a matterof this kind). In this case, the dramatic date might fall eitherbefore 322 or in 318/17.The treatment of Ath. 56.4 by Lane Fox is unconvincing. Heargues: (i) That the procession to which Theophrastus refersneed not be the Dionysiac procession. This requires us to believethat there was another procession, againorganisedby the archonwith ten ttiutn:ci, about which Ath. is silent. This is improb-able, since Ath. goes on to mention two further processionsfor which he was responsible, and appears to be giving usa complete list. (ii) That tpccipncv:ci means choose not107Rhodes 52. 108Rhodes 518. 109Rhodes 52, 628.110As Dittenberger observes (621 n. 3), not one per tribe, so possibly elected.111Cf. Stein 41 n. 3.34DATEelect, and the question is how many will be chosen, with-out specication of method, election or lot. This is impossi-ble, since :ivc indicates identity not quantity. (iii) That thereis no evidence that the procedure changed back. One, atleast, of the inscriptions cited above provides that evidence. (iv)That ten ttiutn:ci, the right number for the overseers of theDionysiac procession, reinforce respect for his eye for Athe-nian detail. This is a strange argument to use in support of theargument that Theophrastus is not referring to the Dionysiacprocession.Lane Fox argues more persuasively that the manner in whichthe Oi,cpyisc is depicted does not suit the two later periods.He is given no Macedonian connections and no words aboutrecent political upheavals. In 318/17 and after 307 oligarchshad just had power, could look back to Macedoniansupport andwould grumble at the harsh reprisals of a period when demo-cratic fervour ran extremely high. In any case, we may add,the period after 307 is ruled out by the reference to liturgies.112The manner of his depiction, Lane Fox observes, suits the ear-lier period. Our Oligarchic Man belongs in a stabler world, ina democracy against which the grumbles are those which mighthave been heard way back in the age of Alcibiades.A dramatic date before 322 is very plausible. Date of com-position is indeterminable. Lane Fox places date of composi-tion, no less than dramatic date, before 322. If Theophrastuswrote him up any later, he would have been characterizing hisman against a setting which had passed. Perhaps this is to takethe Oi,cpyisc too seriously. His vices are conventional andhis targets traditional. Even in the 320s he cuts a comic g-ure. Men such as he, upper-crust out-of-touch reactionaries, arematerial for caricature, whatever the current political climate.113Like Stein, I do not exclude the possibility of a later date of112See p. 33.113For further comment on his type see the Introductory Note to XXVI.35I NTRODUCTI ONcomposition, even during a period of oligarchy. I do not evenexclude composition in the 330s, for a reason which I shall giveat the conclusion to this section.A date before 322 has been suggested for other sketchestoo. Boegehold observed that Theophrastus regularly refers tojudicial activity as an ordinary feature of everyday life (I.2,VI.8, VII.8, XI.7, XII.4, 5, XIII.11, XIV.3, XVII.8, XXVI.4,XXVII.9, XXIX.2, 5, 6). During the oligarchy of Phocion thequalication for citizenship (and so for attendance at the Assem-bly and service on juries) was 2,000 drachmas, under Demetrius1,000. Boegehold inferred that these sketches were written dur-ing a period of stable democracy. By the same token one mightinfer that those sketches which casually refer to meetings ofthe Assembly (IV.2, VII.7, XIII.2, XXI.11, XXII.3, XXIV.5,XXVI.2, 4, XXIX.5) were also written before 322.114But cau-tion is needed. There were some 21,000 qualied citizens underDemetrius of Phaleron,115and the courts and the Assembly con-tinued to function.116We cannot say that the dramatic date ofany of these sketches is incompatible with this period. Muchless can we say that they could not have been written during it.Again, the allusions to liturgies (XXII.2, 5, XXIII.6, XXVI.5)set the dramatic date before their abolition by Demetrius.117Butthey say nothing about date of composition.118My conclusions are these. (i) There is no consistent dramaticdate. One sketch (VIII) is set during a period of oligarchy; manyof the others are set during a period of democracy. (ii) It is114Other passages which imply a democratic setting are XXVIII.6(onucspc:ic as a soubriquet for slander) and XXIX.5 (watchdog of theonuc), as R uhl observed.115Hammond 137.116A. L. Boegehold, The Athenian Agora, xxviii: The Lawcourts at Athens (Princeton1995) 41, S. V. Tracy in Fortenbaugh and Sch utrumpf (n. 6 above) 3389.M. Gagarin, ibid. 35961, arguing that there was a signicant decline inthe use of the courts under Demetrius, relies heavily on the unarguedassumption that Theophrastuss courts belong to the 320s.117See p. 33. 118See Stein 423.36TRANSMI SSI ONimpossible to assign a single date of composition to the wholecollection. (iii) Date of publication is indeterminable.The question when Theophrastus wrote the sketches andthe question when (if ever) he published them are inseparablefrom the question why he wrote them. If (as suggested above,pp. 1516) he wrote them as incidental material to illustrate hislectures, he may have written them over a long period, poten-tially throughout the whole of his career as teacher. Their unifor-mity of style and structure suggests that he may have reworkedthem for publication. Lane Fox (141) puts it well: Written forlike-minded readers, the sketches were meant to amuse, notteach. If they were rst shared with friends and pupils, they couldeasily grow up piecemeal, being increased as the years passed.We do not know what publication meant, but survival froma personal collection after Theophrastus death is an obviouspossibility.119IV TRANSMI SSI ON(i) PreliminariesTheophrastus composed the sketches in the later part of thefourth century. In what form and at what date they were pub-lished we do not know.120A century later they were imitated byAriston of Keos.121They were quoted by Philodemus in the rstcentury bc.122Before the time of Philodemus they had alreadysuffered interpolation: the denitions at least had been added.123They had also suffered serious corruption. For Theophrastuscannot have designed V.610 to follow V.15. Yet the papyrus119Others who have contemplated an extended period of composition areR uhl 327, H. Reich, Der Mimus (Berlin 1903) 309 n. 1, Regenbogen 151011,M. Brozek, De Theophrasti Characterum ueritate ac de obseruatiuncula,in K. F. Kumaniecki (ed.), Charisteria Thaddaeo Sinko . . . oblata (Warsaw 1951)6770.120See the section on Date (pp. 2737, esp. 367). 121See pp. 910.122See p. 50. 123See p. 17.37I NTRODUCTI ONof Philodemus (l1), like the medieval manuscripts, presentsV.110 as a continuous text.The general fabric of the text transmitted by the papyrus ofPhilodemus, and of the shorter portions of VII and VIII trans-mitted by another papyrus of the rst century bc (l2), is notessentially different from that of the medieval manuscripts. Theprooemium and the epilogues appended to nine sketches wereadded much later. But, those additions (and other interpola-tions) apart, our collection as its stands reects a version of thetext which had come into existence by the rst century bc.124Itis no longer possible to argue, as was argued before the papyriwere known, that it owes its form to large-scale editorial activityin the imperial or Byzantine period.125The archetype of the medieval manuscripts, containing 30sketches, was divided for copying, by chance or design, at adate unknown (not later than the eleventh century), into twohalves. One half (containing IXV) is represented by our oldestmanuscripts, AB (tenth or eleventh century); the other (XVIXXX), by V (thirteenth century). These manuscripts are cor-pora of rhetorical treatises. The text of Theophrastus will havebeen added to the prototype of the corpus in the early Byzan-tine period.126It may have become divided because an ancestorof AB lacked space for the whole, or because a half was feltsufcient; or through accident.127124For a misguided attempt to link the early history of the text to the allegedfate of the lost philosophical works of Theophrastus and Aristotle seen. 10 above.125So Diels, Theophrastea (1883) and his edition (1909) vviii.126Immisch (1897) xxviiixxxvi, id. Philologus 57 (1898) 2046, H. Rabe,Rhetoren-Corpora, RhM 67 (1912) 32157, E. Matelli, Libro e testo nellatradizione dei Caratteri di Teofrasto, S&C 13 (1989) 32986, Fortenbaughin W. W. Fortenbaugh and D. C. Mirhady (edd.), Peripatetic Rhetoric afterAristotle (New Brunswick and London 1994) 18. See n. 42 above.127Another rhetorical corpus, Par. gr. 1741 (10th cent.), is said by its index (14thcent.) to have once had the Characters, on pages now missing. How manyit had and when they were lost are questions which cannot be answered.38TRANSMI SSI ONIn addition to these three, 68 later manuscripts arerecorded.128The majority contain IXV; a few contain eitherIXXIII or IXXVIII.129Immisch (1897) classied these (orsuch as were known to him) into three groups, according tonumerical content: C = manuscripts with IXXVIII, D IXXIII, E IXV.Whether CDE preserve any trace of a tradition independentof ABV has long been debated. Cobet pronounced an uncom-promising verdict: omnem crisin Characterum Theophrastitribus tantum Codicibus niti: omne enim emendandi praesid-ium et fundamentum in capitibus XV prioribus esse in duobusvetustissimis Codicibus Parisinis [AB], in posterioribus capitibusXV crisin pendere totam a Codice Palatino-Vaticano [V]:reliquos autem libros ad unum omnes occi non esse facien-dos et criticam rem impedire tantum et quisquiliis nil profuturisonerare.130Diels argued vigorously in his support: true or plau-sible readings were lucky slips or medieval conjectures.131Manyhave remained unconvinced.132What scribe, protested Pasquali,For a conspectus of views on the former question see Matelli 367 n. 110,who suggests that it had room for all 30. The lost text has been claimedas a possible source of E (P. Wendland, Philologus 57 (1898) 1045), of CD(Immisch, ibid. 205 n. 26), of Marc. gr. 513 (no. 64 Wilson) (Matelli 364n. 101), and of V (Matelli 378). Appropriate caution is expressed by C.Landi, SIFC 8 (1900) 978, and Diels (1909) xxv (Sed ecce terret nos inABVsolis consos ex inferis citata umbra codicis celeberrimi et vetustissimiParisini, with splendid facetiousness).128N. G. Wilson, The Manuscripts of Theophrastus, Scriptorium 16 (1962)96102.129For brevity, here and in what follows, I stands for I plus prooemium.130Mnemosyne 8 (1859) 311. Similarly, Mnemosyne 2 (1874) 34.131Diels (1883) 1115, (1909) ixxiv. Similarly Wendland (1898) 10312.132For example, Immisch (1897) xlxlvii, (1923) iiiiv; Pasquali (1919) 1617 =(1986) 901, (1926) 912 = (1986) 8502, id. Storia della Tradizione e Crit-ica del Testo (Florence 21952) 2930; Edmonds (1929) 1130; Navarre(1931) 79, 301 (contra (1920) 12, (1924) xxxvxli); De Falco (1956)xviixxii; Steinmetz (1960) 2338 (arguing only for the independence ofCD from V in XVIXXVIII); Torraca (1974) 71, (1990) 202, (1994b)61416.39I NTRODUCTI ONwould have the wit to replace :iuit with the slave-name Tiitat IX.3, or an unexceptionable gcivtci with the more subtlysuggestive otcgcivtci at IV.4?In 1992 Markus Stein sketched a plausible picture of themedieval tradition, using only the piecemeal evidence alreadypublished.133Two years later I. E. Stefanis published his inves-tigation of the later manuscripts, which he had collated almostin their entirety.134His investigation conrms that the picturesketched by Stein is in all essentials right. Now that we can seethe relationships of the later manuscripts to each other and toABV, and the precise distribution of variants, we can establish(what Cobet and Diels inferred but could not prove) that nolater manuscript or group of manuscripts had access to a tradi-tion independent of ABV.The medieval tradition provides plentiful evidence of scribalinterference. For example, the version of XVIXXVIII in Candof XVIXXIII in D is an abridged version of what is in V, andthe abridgement did not happen by accident.135A reading likeotcgcivtci, if it is not an idle blunder, is an idle embellish-ment.136A