Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Claudius Aelianus’ Varia Historia and the tradition of the miscellany Johnson, Diane Louise 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1997-250733.pdf [ 13.07MB ]
JSON: 831-1.0099207.json
JSON-LD: 831-1.0099207-ld.json
RDF/XML (Pretty): 831-1.0099207-rdf.xml
RDF/JSON: 831-1.0099207-rdf.json
Turtle: 831-1.0099207-turtle.txt
N-Triples: 831-1.0099207-rdf-ntriples.txt
Original Record: 831-1.0099207-source.json
Full Text

Full Text

CLAUDIUS A E L I A N U S ' VARIA HISTORIA AND THE TRADITION OF THE M I S C E L L A N Y by DIANE LOUISE JOHNSON B.A. University of Washington, 1971 M.A. Western Washington University, 1983 M.A. University of Washington, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES Department of Classics We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A August 1997 © Diane Louise Johnson, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of, this thesis for financial gain shall not be allowed without my written permission. , Department of C L f l S S i C i The University of British Columbia Vancouver, Canada Date X l l / l M ? DE-6 (2/88) ii ABSTRACT Claudius Aelianus was recognized by Philostratus and the author of the Suda as a participant in the literary and intellectual movement of the Second Sophistic. Philostratus' biographical sketch in the Lives of the Sophists, however, makes it clear that Aelian did not perform publicly as did the other sophists whom Philostratus described; Aelian's retiring and scholarly nature is emphasized by Philostratus, who implies that Aelian's choice of literature over performance followed a pattern established by Demosthenes and Cicero. Most scholarship on the Varia Historia during the past 150 years addresses the question how Aelian made his collection, i.e. what sources he accessed. This directly reflects modern use of the Varia Historia as a quarry from which to mine information about the ancient world. Such scholarship must conclude that Aelian was not a modern research scholar with the goals, techniques, and readership of the modern "scientific" historian. What then were his goals, techniques, and readership? The Varia Historia cannot be fairly assessed without taking into account its membership in the genre of the miscellany. The Imperial miscellanist concerns himself with a specific subset of traditional literature: the material which supplements the standard literary education and may be termed polymathic. The miscellanist assumes a readership with whom he shares certain iii educative goals: specifically, further detailed education in literature beyond the primary level, including further work in the encyclic artes and a general increase in detailed information "for its own sake." Because the miscellanist adopts the stance of a mature amateur scholar gathering data for a younger reader, he reveals a patronizing tone in his collection. The data the miscellanist offers his reader is presented in a manner characterized by rroiKiXia or "variety"; as such it reflects the Imperial attitude toward the cultured person's correct use of leisure. An analysis of passages from the Varia Historia reveals that Aelian conceives his reader as a young person currently in the process of acquiring paideia. In his miscellany Aelian has provided this reader with material that conveys a moral message at the same time that it provides models of the correct way to respond to traditional literature. T A B L E OF CONTENTS Abstract ii / i Table of Contents iv Introduction 1 Chapter 1: Claudius Aelianus and the Varia Historia 6 The Life of Claudius Aelianus 6 The Text of the Varia Historia 21 Scholarship Addressing the Varia Historia, the Imperial Miscellany, and Paideia 32 Chapter 2: Compilator and Compilanda 43 Paideia and the Archive 52 Selection (ExXoyn) and Application (o7TouSf|) as Standards of Hellenistic Scholarship 76 noXuuaGia and Imperial Education 90 noXuuaOia and rpauucmKri 91 noXuuocOia and Post-Primary Education 101 The Miscellanist and Correct rioXouaGia 113 Chapter 3: The Miscellany' s Readership 118 Supplementary Reading and the Encyclic Arts 134 The oXcoc, jreTrcuSeuuevoc, 138 The Imperial Miscellanist and the Adult Learner 156 Chapter 4: Miscellany Structure and Style 170 Chapter 5: The Ai;ioo7Tou5aaTa in Aelian's Miscellany 192 Moral Anecdote in the Varia Historia 193 a) Minor Characters in Major Events 199 b) Paideia-Icons in Anecdotal Situations 200 c) Statesmen and Politicians 202 d) Philosophers and Poets 204 e) Musicians, Athletes, and Artists 208 f) Ethnic Anecdote 210 g) Characters not drawn from paideia 211 h) Intended Readership 212 i) Anecdote as Moral Modelling 214 Epideictic Biography 219 V Ecphraseis 227 Ethnography and Nouiucc 236 Natural History 242 Paradoxography in the Varia Historia 249 Lexicography 251 nivaKec, in the Varia Historia 254 Aelian's Use of the Progymnasmata 255 A<J)eXeia and the Varia Historia 265 Bibliography 268 1 Introduction The purpose of the following study is to establish Claudius Aelianus's Varia Historia within the tradition of the Imperial miscellany. Although the similarities between the Varia Historia and a number of other surviving collections of material compiled from earlier literature and scholarship during the Imperial period have long been recognized, few attempts have been made either to analyze the qualities which the Varia Historia shares with these other collections or to consider the various ways in which it diverges from them. Indeed, there has been little scholarly work done on the Varia Historia during the present century. Nor has Aelian attracted much attention from modern students of the Second Sophistic, who have tended to focus upon this period's more productive and flamboyant contributions in the fields of rhetoric and philosophy. Yet Aelian, too, forms a part of the intellectual culture of the second and third centuries AD. Consequently, in Chapter 1 I attempt both to place him within his social and intellectual context by considering the ancient witnesses to his life and influence and to review the recent scholarship which has addressed Aelian in terms of his own and his work's place within the Second Sophistic. Also in Chapter 1,1 consider recent scholarship on the Imperial miscellanists; as I shall attempt to demonstrate, this has generally avoided discussing the generic framework that has typed these authors. Chapters 2, 3, and 4 represent my efforts to establish and define this framework. Once understood, the generic conventions of the miscellany will serve as a means to analyze the content, structure, and style of Aelian's Varia Historia, this topic will be the focus of Chapter 5. 2 It is my purpose to demonstrate that this work was not intended to be a random and careless omnium-gatherum of amusing irrelevancies, in which terms it has been ridiculed and dismissed by recent scholars. Rather, the Varia Historia provides, as do all Imperial miscellanies, (1) material which the composer has found especially relevant to himself and to a reader with whom he identifies, and (2) models for the correct reception and utilization of the material. This compiled and miscellaneous material is drawn from paideia, the traditional literature, both curricular and secondary, which under the Empire forms the basis of all liberal culture. In a sense all Greek and Latin literature of this period provides its audience with both paideia-matter and models for its utilization. We can in fact construct a continuum of Imperial authors' utilization of paideia as matter and model, based upon the author's relation to his audience. The performing sophist, for example, reenacts paideia in a public venue; he himself becomes the model, while his ueXerai recreate personalities and events enshrined in the literary tradition. The lecturing philosopher provides models through his use of paideia-sanctioned means for seeking philosophical truth — eksyxoc,, dissertatio, 5iaTpi|3f| — as well as through his paideia-sanctioned garb of long matted hair and careless dress; his discourse will contain such paideia-matter as anecdote, chreia, and historical allusion. In the epistles and moral essays of Plutarch and Lucian, the writer and his audience are individuals, single voices in a private setting. Yet not only is the writer's thought supported, illustrated, and amplified through paideia-matter in the form of quotations, apophthegms, and allusions; it is also expressed through the imitatio of models drawn from the tradition, these mimetic reworkings thus providing further models 3 for the reception and recycling of the tradition. Thus all Imperial literature can be seen as sharing this twofold manner of incorporating paideia, differing essentially only in its conception of the audience as the community or as the private individual. But whereas other genres draw in paideia-content and paideia-form as subsidiary and ancillary to the author's purposes, the miscellany is paideia, in the form of data extracted more or less directly from the paideia-authors and scholars and recycled as a collection of compilations. What position does the creator of such a collection occupy upon our literary continuum of paideia-manipulation? Briefly, the miscellanist is not a paideia-manipulator so much as a paideia-purveyor. It is not the efficacy of rhetorical skill and of philosophical acumen in the recreation and modeling of paideia that determines the success and value of his undertaking so much as the quality of his selection of data from paideia —that is, their value to the reader. In selecting according to the reader's intellectual needs, the miscellanist shares some features with the creator of the pedagogic chrestomathy, textbook, and technical manual. But the miscellanist is not writing specifically for the classroom. His reader, like Plutarch's and Lucian's, has already acquired a basic liberal education. This reader has an adult relationship to paideia, with an adult's needs. The key to understanding the Imperial miscellanist's selective process lies, I believe, in determining his view of his reader's intellectual needs and requirements. As I shall attempt to demonstrate, the Imperial miscellanist believes that his reader needs access to paideia-extracts which are wide-ranging, detailed, true (or at least authorized by an acceptable paideia-figure), and omnivalent. In creating a miscellany 4 consisting of such data he envisages two stages in the compilation process: concentrated industry and mature selection. His industry is occasioned by his tacit acceptance of the positive status of polymathy and scholarly labor and commitment (Ttovoc,, 07rou6f|) as necessary activities of the polymath. But this polymathy has to be controlled and directed, for the miscellanist also believes that not all data contained in the literary tradition are equally valuable (ot^ ioc 07rou5fjg, a£ia Xoyov, a^iooTTouSaora). His selection of data to be included in his collection depends upon his views of the needs of his reader, whom the miscellanist considers as more or less identical with himself at an earlier stage in his intellectual development. This selection assures the value of the content of the collection. In the process of providing this reader with relevant extracts from paideia, the miscellanist has also provided, in his own activity of compilation, a model for the correct response to paideia: selective industry. The miscellanist may then go on to multiply his paideia-models by presenting his data within a dramatic frame, allowing the modelling characters to act out, as it were, further correct responses to paideia, polymathy, and scholarly selection. In discussing the generic framework of the Imperial miscellany I have depended upon the evidence provided by the surviving works of Aulus Gellius, Athenaeus, Macrobius, and Clement of Alexandria, as well as by fragments of other miscellanists such as Pamphila and Favorinus. Some of these writers provide direct and candid statements about purpose and readership. The works of others are described by ancient scholars as being accessed for purposes similar to those stated by the surviving miscellanists. It is from such statements and discussions that the framework of the miscellany tradition can be constructed. In the Varia Historia, however, Aelian provides neither a statement of purpose nor a discussion of how he selected and organized his materials. This collection lacks prologue, epilogue, and significant internal editorializing. But as is the case with the content of the other miscellanies, the chapters of the Varia Historia themselves provide evidence for Aelian's goals in creating his collection. By analyzing Aelian's subjects, his structuring of chapters, and his style, I have attempted to demonstrate that the Varia Historia fits into the Imperial miscellany tradition, providing a collection of relevant data and models for the correct reception of data, for a young adult reader needing guidelines to paideia's reception and relevance. 6 Chapter 1 Claudius Aelianus and the Varia Historia In the present chapter I address three topics: 1. Aelian's position within the Second Sophistic as a writer who exhibits the general archaizing qualities of that cultural movement; 2. the present state of the text of the Varia Historia, insofar as this state affects our interpretation of the work as exhibiting the generic qualities of the Imperial miscellany; 3. recent scholarship on Aelian, the Varia Historia, the Imperial miscellany, and the role of paideia in Imperial society. The Life of Claudius Aelianus Although frequently dismissed by nineteenth- and twentieth-century scholars as at best a marginal figure in the intellectual life of the second and third centuries A D , 1 Aelian was considered significant enough in antiquity to be included among the notable Imperial sophists in Philostratus' Vitae sophistarum, and to be provided with a biographical sketch by the author of the Snda. Philostratus' account of Aelian must have been written shortly after the latter's death. The Vitae sophistarum was dedicated to Gordian before he became emperor, and 1 Schmid. for example, found him a "winzige Persbnlichkeit (1893 vol. 5: 4), Dihle (1994) a "literary journalist." 7 must therefore have been written before A D 238. At this time Aspasius, who had been one of Aelian's fellow students, is described by Philostratus as still professionally active. In fact, Aelian is one of the latest figures to appear in Philostratus' work. Only the lives of Heliodorus and Aspasius follow the biographical sketch devoted to Aelian, and these three final figures were probably nearly exact contemporaries as members of the third generation of sophists after the seminal Herodes. Philostratus lived at Rome for some time, and it is likely that he met Aelian there (Swain 1991: 188). According to his account, Aelian was a curiously reticent member of the Second Sophistic movement. AiXiavog Se 'Piouaiog U E V x\v, f|rriKig*£ be, &OTiep oi EV xfj UEOOYEIO: AGnvaToi. ETTCUVOU UOI S O K E ! oti;iog 6 dvrip ouTog, TOCOTOV UEV, ETTEiSf) xaGapdv <j)iovfiv EgETrovnoE TOXIV okoov ETEPQC (jxovojc Xptouevnv, E7t£i6' on TrpoapnGsig ap<j>ioTf|g U T O TIOV xapiCouEvcov T & Toiaura O U K ETTIOTEIXTEV, OU5E E K O X & K E I X T E Tf)v EauroO yvwunv, OU6E E7rf|p6n UTTO roug ovouarog OUTOJ ueyaXov ovrog, dXX' ECCUTOV ev SiaaKEipauEvog tog UEXETTI O U K EmTf|6£iov Top cuvYPO«i>Eiv ETTEGETO xai sGauudoGn. IK T O U T O U . r\ UEV EjriTrav VSEO: T O U dv5pog afyekeia 7rpoo|3dXXouad TI Tfjg NiKoorpdrou copag, f) 8s EVIOTE, Trpog Aiiova 6pQ[ KCU TOV EKEIVOU TOVOV; EVTUXOJV 5E TTOTE auTio OiXoorparog 6 Aquviog PipXiov ETi rrpoxeipov EXOVTI xai dvayivvtbaKOVTi auro auv opyfj xai EmrdoEi TOO (J)GEyuaTog TIPETO GCUTOV, 6 TI orou8dCoi, xai 6c, "£KTO7Tovr|Tai uoi" e<\>r\ "KaTriyopia T O U IuwiSog, xaXio yap oi)Tto TOV otpri KaGriprifJEVov rupawov, errei5f| doEXyEia Tidou Td 'Pcouaicov rjoxuvE." xai 6 OiXoarparog "eytc O E " EVTTEV "sGauuaCov av, si CiovTog Ka-rriYopnoag," sivai yap 6f] TO UEV Ciovra rupawov ETUKOTTTEtV dv5p6g, TO 6e ETTEuPaivElV KElUEVCp TTOVTOg. E(j)C(OTC£ 5E 6 dvfip ourog unS' d7ro8£8r|unKEvai TTOI Tfjg yf\c, vnep TT]V 'IraXtov Xcopav, un8£ supfjvai vauv, un5£ yvtivai GdXarrav, 66EV xai Xoyou TiXEiovog xard TT)V 'Ptb|ur|v f|g"iouTO ibg TUIIOV rd fjGn. Ilauoaviou UEV ouv dxpoaiTig E Y E V E T O , EGauuagE 5E TOV cHptb8nv cog TTOiKiXcoraTog pnropcov. Epico 5E urop rd EgY|K0VTa ETT| xai ETEXEUTO: O U K ETri Traiaiv, 7rai5o7roiiav yap 7rapr|Tf|oaTO rap uf| ynuai TOTE, TOUTO 5E SITE Eu5aiuov EITE dGXiov ou T O U Trapovrog xaipoO opiXoaoopfjaai. Aelian was a Roman, but he spoke Attic Greek as authentically as a native from the heartland. He seems to me to deserve commendation for two reasons. 8 First, through his own great efforts he acquired [such] purity of language while inhabiting a city which used a different tongue. Secondly, although he was given the title "sophist" by those currying favor, he did not believe what they said, nor would he flatter himself and let himself grow conceited at so great a name. Instead, having carefully examined his own abilities and found that he was not cut out for declamation, he turned to writing, and thereby won much admiration. His style consistently displays that simple straightforwardness reminiscent of Nicostratus. He occasionally takes his cue from Dio and that author's tone. Once Philostratus of Lemnos came upon him with a text in his hand, reading it with passion and intensity. Philostratus asked . him what he was so involved in, and Aelian answered, "I have completed my 'Accusation of the Effeminate Man, 'for by that title I refer to the recently deposed tyrant who shamed the Roman state with his vice." Philostratus responded, "I would admire you if you were making [such] an accusation of him while he was still alive. " For anyone can trample a man when he is down and out; it takes a real man to aim a blow at a living tyrant. Aelian used to say that he had never gone abroad, had never even gone on board ship, and that he knew nothing of the sea. Hence he enjoyed a good reputation at Rome as a man who honored Roman customs. He was a pupil of Pausanias, and admired Herodes as the most versatile among the sophists. He lived past the age of sixty and died childless, having circumvented the acquisition of children by never taking a wife. Whether this is a blessed or a wretched state is a philosophical inquiry not suited to the present discussion. (VS 624) Although he specifies two aspects of Aelian's life which in his eyes are particularly worth noting, Philostratus in fact emphasizes three points about Aelian: 1. his peculiarly thorough mastery of Attic Greek, despite his Roman birth and residency; 2. his moralistic but timid response to social and political events; 3. his emotional bond to his native land. Among the sophists whom Philostratus discusses in the Vitae Sophistarum, Aelian is the only Italian-born Roman citizen to merit inclusion among the native Greek speakers. His grasp of the ancient Attic dialect is especially worthy of note, Philostratus insists, because of its surprising purity. Aelian's Greek was as untainted and as pristine as the language spoken by the people of the Attic interior. Elsewhere Philostratus describes this region as offering aya0oO SiSaoKaAeTov av5pi PouAouevw SiaXeyeaOai .... auuaoc; 9 Pappdpoig ouaa u y i a i v e i . . . . r\ ({xJavn K a i f| yXwrra rnv ctacpav 'AT9i5a ajroipaXXei a good classroom for one who seeks a style .... Unsullied by foreigners, ... the language there is sound, and intones as though upon a lyre the purest Attic strain (VS 2.553). Although deprived of the spiritual stimulation of the mesogeia, indeed in the very capital of the foreign empire, Aelian made up for these negative circumstances by effort and concentration. Philostratus' use of EKTTOVOO, "labor into completion," in reference to this achievement implies a positive assessment.2 Philostratus further maintains that Aelian was in a position to consider a career as sophist, but that he rejected this option through an awareness of his own incapacity for the peXeTX]. Philostratus does not specify the details of this incapacity, whether, that is, it arose from a physical or emotional inability to orate in public. However, Philostratus does make clear that this rejection of public performance in favor of research and composition3 did not prevent Aelian from reacting publicly to a politic event by composing with effort and care (eK7rejrovf|Tai) a formal accusation ( K a T n y o p i a ) of a deposed "tyrant," presumably Elagabalus (Schoener 1873: 4-5). Philostratus relates this incident in some detail, but the point of the anecdote is at first hard to grasp. Is Philostratus faulting Aelian as a cowardly recluse? Considering the political climate of the Severan era, if Philostratus' reference to this incident is meant to be critical of Aelian's behavior, the Lemnian Philostratus' attack seems hypocritical and "For a native Latin speaker to compose in Greek was not without parallel during the Imperial period. Suetonius wrote scholarly works in Greek. For L. Annaeus Cornutus, C. Musonius, and Marcus Aurelius, Greek was the medium of philosophical inquiry (Lesky 1966:876). Babrius composed Greek verse. Favorinus and Apuleius could declaim with ease in either language (Steinmetz 1982: 2). 3 £uyypa<i)£iv; the question of whether Aelian followed the sophistic practice of accepting students is not addressed. 10 petty. Such criticism could be leveled at any intellectual in the early third century A D irate enough to express his politically motivated indignation. One could hardly expect even the most intrepid public figure to attack an emperor as unstable as Elagabalus. The cases in which Philostratus relates a sophist's aggressive confrontation of a powerful public figure4 are not quite parallel, for the emperors involved in these accounts could be expected to display some magnanimity toward even the most presumptuous orator. A closer parallel to Aelian's KoerriYopio: in the Vitae sophistarum might be Philostratus' description of Dio's declamation against Domitian,-delivered after the latter's assassination (VS 488). Yet Aelian's peculiar position as a retiring Roman preferring literary composition to public declamation makes the comparison with Dio strained. Is Philostratus' intention to immortalize his kinsman's witticism? Anderson suggests as much (1986: 86). Again, the triviality of the Lemnian Philostratus' response hardly seems to justify this interpretation. It may be, however, that Philostratus is attempting to align Aelian with traditional paideia-icons who had been involved in similar situations. As a practicing sophist, Philostratus was familiar with the creative use of typology, as he makes clear from his definition of Second Sophistic subject matter: f) 8e per' eKeivnv, f)v ov>xi veav, apxaia yap, bevTipav 5e uaXXov TrpoaprjTeov, roue, 7rivr|Tac, UTTETUTTCOOCCTO KCU roue, TTXOUOIOUC, rai TOUC, apioreac; m i roue, rupawouc, Kai T&C, EC, ovoua vnoQeoexc,, act/ ag r) i a r a p i a Styei. . The [authors of the] so-called Second Sophistic sketched out the types of paupers and rich men, of aristocrats and tyrants, and took scenes and events in history as their plots and background. (VS 481) 4 Favorinus and Hadrian (VS 489), Polemon and Antoninus (539), Alexander Pelopolaton and Antoninus (570). Pliny the Younger (Ep. 9.1) discusses the moral issues involved in orating against the recently deceased. 11 That Philostratus occasionally used such typical relationships in his biographies can be seen from his alignment of Herodes' failed improvisational speech before the emperor with Demosthenes' similar failure before Philip (VS 2.565). He may be suggesting some connection here between the figure of Aelian and the type of the patriot faced with the pragmatic reality of tyrannical power. The abuse of strength was certainly a topos which interested Aelian; the Varia Historia, as I shall demonstrate, abounds in anecdotes in which a tyrant brutally exerts his power over a virtuous private individual. Although Aelian's vituperative K a T n y o p i a , like a Ciceronian Philippic, has no immediate political effect, it still suggests something about the ethics of its composer (cf. Swain 1991: 149): Philostratus may be attempting to characterize Aelian as a talented, responsible, but politically frustrated individual. If Philostratus is drawing upon types in this instance, he may have also done so in his discussion of Aelian's rejection of a career as a public speaker. The turning away from politics in favor of a quiet life of scholarship and composition is an act associated especially with Isocrates. Just as early in the Vitae sophistarum Philostratus describes Isocrates' decision to leave public life as influenced by an awareness of his own insufficient vocal power as well as by his fear of Athenian political (j)96voc, (VS 1.505), so Aelian sensibly retreats from the pressure and intensity of the sophist's public career as well as from direct exposure to a tyrant's vicious wrath. If Philostratus' description of Aelian's language, career choice, and political responses defines his virtues as traditional and conservative, the discussion of his 12 education contributes to this image. Philostratus states that Aelian "was a pupil of Pausanias, and admired Herodes as the most versatile of orators." As Philostratus has somewhat more to say about Herodes and Pausanias than he has about Aelian, a consideration of this material may shed some light upon Aelian's position within the paideia of the Second Sophistic. Herodes Atticus was indubitably a central figure in the intellectual world of the second century AD. For some, the entire Second Sophistic emanates from him (Anderson 1986: 108). Born into a wealthy Athenian family which had acquired Roman citizenship under Nero, Herodes cultivated the double persona of statesman and intellectual. He served as consul at Rome in A D 143 and made magnificent donations to a number of Greek cities, including Athens, Corinth, and Ilium (Graindor 1930: 10). More to our purpose, he was preeminent among the great virtuoso sophists of his day. His rhetorical ability attracted many pupils, a number of whom went on to become themselves orators and teachers in Herodes' tradition. What this tradition was in terms of Herodes' intellectual contribution is somewhat difficult to determine with precision. The one surviving work attributed to Herodes, the nepi TToXireiac;, is of very dubious authenticity; even assuming the work to be his, its style is disappointing in light of Philostratus' descriptions of the beauty of Herodes' compositions and style.5 Whatever the quality of the improvised performance and the finished document which resulted from Herodes' declamation, however, it was influenced by his purposive 5 These descriptions, although ornate and intriguing, prove to be frustratingly uncritical and imprecise; at one point, for example, Philostratus describes Herodes' style as xPuaou ipfjyua TTOTauw dpyupoSivri lOTOlUYOtCov gold dust gleaming at the bottom of a silver-eddying river (VS 564). 13 cultivation of TO dpxcuov. Herodes insisted upon grounding his discourse in the paideia of the canonic past, through the analysis and imitation of the works of ancient speakers and poets. The classical figure of Critias was especially associated with Herodes' teaching and research. According to Philostratus, Herodes was personally responsible for discovering and promoting this speaker's speeches as classical models of rhetoric (VS 564). In the next generation, Pausanias of Caesarea in Cappadocia, a member of Herodes' inner circle of special pupils, developed into a sophist of such prestige that he was appointed first to the Imperial chair of rhetoric at Athens, then to the analogous position at Rome. At the time of his installation at Rome in A D 192/3, Pausanias had reached the high point of his career (Avotins 1975: 324). Aelian, during these years probably in his late teens or early twenties, studied under Pausanias at Rome. Philostratus asserts that, while Pausanias exhibited many of Herodes' other excellent qualities, his acquisition of his teacher's ability to extemporize, avTocrxebia&w, was the most striking of these. Possession of this highly admired ability guaranteed Pausanias' success as a performer. But of more interest to an examination of Pausanias' contribution to Aelian's education is the archaic flavor attributed to Pausanias' work: oux duapTdvei TOO dpxaiou He does not fail to attain an antique flavor (VS 594), Philostratus maintains, and assures us that the statement is easily verifiable; toe, u T r d p x a TOUC, ueXeToac, quupaXeiv, TTOXXCU yap TOV nauaaviou Kara TT)V 'Ptounv One can get access to his declamations, for many of Pausanias' works are in circulation at Rome (ibid.) Aelian no doubt encountered his professor's copious declamations at Rome and profited from their emphasis upon the ancient Attic paideutic traditions promoted through Herodes' 14 instruction, an emphasis which Pausanias in turn passed on to his own pupils through his creative work and teaching. Aelian enrolled under Pausanias probably with the goal in mind of becoming a teacher of rhetoric and declaimer like his master. He would have already been fluent in Greek, and Pausanias' tradition-based program must have broadened Aelian's acquaintance with Hellenic literature. The profession of sophist required a mastery of manipulative and affective discourse, of techniques for the utilization and display of traditional poetry and prose committed to memory over a period of years, of the novelistic ability to sketch a person's character and life experiences in speech or tract (ethopoeia) and, most difficult of all to attain, the power of improvisation, to which all these techniques contributed. Pausanias' other pupils too presumably aimed at sophistic careers offering opportunities to display some if not all of these abilities. But there were other career options for them besides the performance circuit. Some with advanced rhetorical training will have sought positions in the civil service, governmental bureaucracy, and senate. They may have looked to imperial secretaryships ab epistulis. They may have sought an outlet for their talents and education in diplomacy, the life of the ambassador and public spokesman (Bowersock 1969: 43-58). A few, like Aspasius (VS 627), would follow in Pausanias' footsteps and become holders of chairs of rhetoric at Athens and Rome. Aelian rejected all of these options. Though ranked among the sophists in Philostratus' work, Aelian chose the path taken, Philostratus suggests, by Isocrates and by Cicero in retirement. He turned to literary composition and scholarship. Developing a style 15 striking in its directness and simplicity, he was admired, we are told, because of his contributions to literature. Yet through his education and attitude Philostratus' Aelian is representative of his period and can properly be considered a product of sophistic paideia and a member of "a group that shared a distinctive set of cultural, social, and political values" (Swain 1991: 149). The third point about Aelian which Philostratus makes in his biographical sketch is Aelian's physical and emotional bond to his native land. Not only did he love Rome and honor her traditions; he also never left the country. Indeed, we are told, he insisted that he had never left Italy, that he had no knowledge of sea travel. Some concern arises about the accuracy of this statement when we try to explain the contradiction between it and a remark Aelian makes in the De natura animalium about a five-hoofed calf which he claims to have seen at Alexandria.6 This problem has been addressed in several ways. Wellmann suggested that the claim to autopsy was part of Aelian's source, not a personal statement on Aelian's part (1893:486). Schmid assumed that Aelian's refusal to board ship and leave Italy was connected with his priesthood (see below p. 16), quoting other scholarly opinion (GGL. 786 note 6) which suggested that Aelian may have simply lied about the autopsy (ibid, note 7). Rudolph maintained that 6 NA 11.40: eydj 8e Koci JiEvxdnoSa pcuv iep6v e9eaad | ir |v, dvaSnua xcp 8ecp xcp5e ev xfj icolei xfj ' A^e^avSpecov xfj \x.zyakr\, ev xcp dSo|ievcp xov Gecu dlaei, ev6a itepaeoa auucjnrcoi a i a d v TtepiKaAAfj Koci y-u^iv direSeiKvuvxo. Kod f|v ubaxoc, fevxaijwa ir\v %pbav Knpcp TcpoaeiKaauevoc,, KOCI em xo\> &[iov 7166a d7tT)pTTip.evov elxe nepiepyov UEV baa E7ii(3f|vai., xeXeiov 5e baa kc, %\&a\v. K a i xavxa uev SOKE! xfj (jj-uaei buoXoyeiv o-b navv xi, feyco 5E baa EC, k[if\v b\\ixv xe K a i dKofjv dcfjlKexo E17COV. / myself saw a sacred five-footed ox, an offering to this god in the great city of Alexandria. It was in the celebrated precinct of the god; there the persea trees made a lovely shade and freshness. There was a calf there, the color of honeycomb, and it had an extra hoof on its shoulder, quite superfluous in terms of walking but all the same perfectly formed. All this was quite unnatural of course; but I report what I saw and heard. 16 Aelian made the journey to Alexandria later in life, after he had told Philostratus about never having left Italy (1884: 11). However, if we consider Philostratus's assessment of Aelian's Greek as being almost autochthonously pure, in combination with Aelian's position as a native Roman untainted by sea travel as a to/?os-symbol of the Iron Age, we might then interpret Philostratus' portrait as antiquarian with typological detail.7 In Philostratus' eyes, Aelian displayed ancestral Roman Republican virtues while discoursing in the language of Archaic and Classical Athens. Philostratus has thus created an image of a sober and sincere patriot of the old order, an image enhanced by the information that Aelian refused to listen to those who encouraged him to undertake a professional career as a sophist. He refused to flatter himself, to be so puffed up by a title that he did not trust his own judgment and careful self-assessment. Likewise, by emphasizing Aelian's refusal to leave his homeland, Philostratus reiterates in different terms Aelian's rejection of a sophist's career, a lifestyle which demanded much travel (Swain 1991:150). Philostratus maintains that by making this choice, Aelian enjoyed greater respect at Rome, tbc, TIULOV xd fj6q, as a patriotic antiquarian. Rather than being a failure as a performing sophist, Aelian was a success in the eyes of his contemporaries through self-knowledge and honesty. After Philostratus' insistence upon specifying Aelian's marginality in terms of the Second Sophistic, the .SWa-author's acceptance of Aelian as a sophist seems almost glib. AiXiccvoc, onto ripaiveoTou rfjc 'IraXiac, apxiepeuc <ai ao<J>iaTn.c;, 6 XPmianioac, KXauSioc. * 6c. ejteicXf|0r| ueXiyXwaooc f| |ueXi<j)0OYYoC ' «ai eoo<j>ioT£uo£v e v Ptoun airrfj em TIOV u e T d 'A8piav6v xpovwv. Aelian came from the Italian town of Praeneste. He was a pontiff and a sophist, and he was addressed as Claudius. They gave him the epithet "honey-Cf. SenecaMedea 301-379. 17 tongued" or "honey-voiced." He practiced sophistry at Rome in the period after Hadrian. (Suda alpha iota 178) Given his extensive use of Aelian's work, the Suda-mthofs sketch is also surprisingly brief. The Suda is the richest source of testimonia to the Varia Historia, and Aelian one of the authors most frequently cited by name in that work (GGL 788). The Suda quotes four passages from the Varia Historia specifically, and attaches Aelian's name (though without book title) to about 175 other quotations. There may be many more quotations from Aelian in the Suda which its author has failed to label (Dilts 1971: 6). The Suda's date for Aelian agrees with that of Philostratus. Hadrian of Tyre had preceded Pausanias the Cappadocian in the chair of rhetoric at Rome, and died in A D 193 (Gerth 1956: 753). Aelian's participation in sophistic culture as suggested by the Suda's term oo<t>ioTti<; must be interpreted with Philostratus' information in mind. The Suda's attribution of the office of priest to Aelian substantiates Philostratus' suggestion that Aelian took some responsible part in state business either at Rome or at Praeneste. But how are we to deal with the honey-tongued and honey-voiced epithets? A word search reveals that these terms, part of high lyric poetry's diction, were applied from the Archaic period of Greek literature to nightingales, Muses, Sirens, and to personified Song, but never to speakers or to writers of prose. Wellmann suggests that the Suda-author here repeats some term used by his sources in praising Aelian's command of Athenocentric Greek (1893: 486); this position receives some support from the ancient connection between innate language ability and honeycomb.8 Cf. Aelian VH 12.45 and West 1966: 183 on the connection made in antiquity between honey and persuasive speech. 18 The Sudd's identification of Aelian's birthplace, coming at the beginning of so succinct a biographical sketch, tempts us to make further inferences about the man from the city itself. Located in the Apennine foothills some twenty-three miles southeast of Rome, Praeneste was an old Republican foundation with a colorful local history. As a summer retreat for Rome's aristocracy, the city enjoyed a social season. The oracle at Praeneste's temple of Fortuna Primigenia for centuries attracted both a local and an international clientele (Wissowa 1902: 209-210); that the oracle was still functioning during Aelian's lifetime is shown by the record of a consultation made by Alexander Severus (Radke 1954: 1555). Among the public buildings connected with Fortuna's sanctuary was one containing the enormous Nile Mosaic, a mural-sized replica of a Hellenistic painting, constructed at some time during the second century B C and filled with scenes displaying the flora and fauna of Egypt (Boardman 1993: 180-181). Scholars have not been slow to make connections between Aelian's comprehensive fascination with animal life and this extraordinary work (Lukinovich and Morand 1991: 167). Do the sets of biographical data provided by Philostratus and the SWa-author supply any information about Aelian which could help determine his position within the Imperial miscellany tradition? As I shall attempt to demonstrate below, the Imperial miscellanist speaks in a private, nonrhetorical voice. He does not present himself as a professional teacher (ao(bioTf)<; or YpauuaTiKoc,), nor is his reader addressed as a pupil. Rather, the miscellanist represents himself as an ordinary person with social and professional responsibilities from which he has stolen precious moments to devote to the acquisition of paideia. He has in 19 this manner acquired considerable exposure to literature, he is oXioc, nenaibevuevoc,; and precisely because he has other responsibilities, he is especially qualified to select from paideia material which is pertinent and useful for a younger person in a similar position. If we assume that Philostratus even in part, and the .Swab-author completely, drew their conclusions about Aelian's life from his writings, then I believe it is possible to account for their image of Aelian—a literary man with social responsibilities and a respectable position in society—as one conveyed by Aelian himself. If Aelian retired from the limelight of sophistic performance, we need not assume that he rejected a public career in general, but only the career of the ueXeTat-performing orator. He may have justified the Suda's attribution of the title oo<bicrrf)c, by taking pupils, by writing (unperformed) U E X E T C U , or by composing material which suggested that he had interests similar to those of the performing sophists.9 Among Aelian's writings, which are currently accessible to us? Three of Aelian's compositions have survived in more or less complete form. These three are (1) the noixiXq ioTopia (Varia Historia), Aelian's miscellany; (2) the nepi £ibiov iSioTnToc, (De natura animalium), accounts of animal (in some cases plant) behavior as mirroring human qualities and virtues; (3) the AvpotKiKcri emoroXai, twenty brief and fictitious letters drawing u p o n scenes and characters from Old and New Attic comedy, self-consciously retailing Attic idiom and proverbs but likewise aligning purity and simplicity of thought with bucolic goodness.10 In addition to these we have a substantial body of fragments 9 Or the SWa-author may have simply interpreted Philostratus' inclusion of Aelian in the Vitae Sophistarum as reason enough for identifying him as a oo<|)ioTr|G. l0The final letter sums up Aelian's position: elusion EV roTg avpoT<;...ico(i 5ncaioouvn Kai aco<j)poauvr|, icai TauTa...6ev6pcov ra KaXAiara K a p n c o v ra xpnoi^ wTaTa....sori yap ng icon evTotuSa ao(j>ioc Righteousness and 20 connected with two further titles: the JJepi Trpovoiac, and the JJepi Beicov evapveicov. These titles may reflect two separate works, or two different titles applied to the same work; or one title may indicate the whole work of which the other title represents a subsection.11 The uncertainty arises because of the similarity in subject matter and tone in the fragments surviving under these titles. Both groups present anecdotes illustrative of deity's involvement in human life. In all the surviving work the same quality is apparent: Aelian works only in miniature. These documents consist of brief and independent units, each unit carefully structured and a self-contained whole. Of the surviving material which is not problematic (below, pp. 18-21) no topic's treatment exceeds seven Teubner pages of print. Most occupy less than half a page; some consist of a single sentence. The 'AypoixiKai emoroXai shares this miniature quality with the other works. Each little letter attempts to sketch a single sentiment. In some cases (e.g. Opora in Letters 7 and 8, Callipides and Cnemon in Letters 13-16) the speaker or situation can be linked with a figure from some surviving comedy text, and we see Aelian delicately developing a dramatic potential inherent in the model. From the letters alone we can understand the connection Philostratus draws between Aelian and the Attic mesogeia. But in terms of subject matter we must set the epistolography aside to consider the large collections, which among themselves share some further similarities (cf. Schoener 1873: 12). goodness grow in the country, and these are the fairest of trees and the most useful of harvests; and there is even here a kind of wisdom [20]. 11 Cf. Schoener 1873: 6-7 note 2, in which the opinions of earlier scholars are reviewed, and ibid:. 60 Thesis II. 21 These similarities concern Aelian's attitude toward his material and his manner of forming collections from it, if we may judge from the editorial statements contained in the prologue and epilogue to the De natura animalium (Tiepi £toiov ISIOTTITOC;). This work contains the only explicit statements which Aelian makes at any place in the corpus explaining his purposes and goals, and I have made use of these statements as arguments for Aelian's position in the miscellany tradition. Whether in fact the De natura animalium can be included in the discussion of Imperial miscellanies must be considered in further detail below. The Text of the Varia Historia The current state of the text of the Varia Historia complicates the attempt to assess its generic qualities. The work has neither prologue nor epilogue, unlike the De natura animalium. It also seems to have undergone some degree of scribal manipulation— primarily epitomization—at some point in its history. Lacking a direct authorial statement as to intent and readership, uncertain as to the complete contents of the work as it came from Aelian's pen, and with only a little biographical material through which we might construct a portrait allowing us to justify certain tastes and judgments on the part of the author, to what extent can we make any definite assertions about the genre of the Varia Historia'} This undertaking is further complicated by Aelian's own method of compilation, a technique observable in all his collections. Unlike Athenaeus and Gellius, for example, each of whom tended to copy out his compilation word for word, Aelian would either reword or condense his data. This situation threatens the effectiveness of analyzing 22 Aelian's sources, because when Aelian quotes an authority in the text we cannot be sure if in the process of compilation he has actually accessed the authority's text, or has quoted a secondary source which in turn quoted the authority. A third complication arises from the great range of detail among individual chapters in the Varia Historia, a situation which does not arise in the De natura animalium. Some chapters of the Varia Historia are written in so condensed and hasty a style that, lacking an authorial statement explaining the purpose of the collection or even justifying its intentionality, we cannot tell if this material was meant to be worked up later or to stand as it is in the text. Some chapters, on the other hand, exhibit considerable care on Aelian's part, both in the elaboration of detail and in the arrangement of topics. Because the majority of scholarship focused upon Aelian and the Varia Historia has been devoted to interpreting the state of the text and Aelian's relationship to his sources, at this point in the discussion we may consider how nineteenth-century scholars dealt with these textual and source problems. The transmission of the text of the Varia Historia has itself not been unusually problematic. According to its most recent Teubner editor, M.R. Dilts, codex V (Paris, suppl. gr. 352) and x establish the main branches of the manuscript tradition upon which the current text is based (1974: v-vii). V and x both originated in thirteenth-century Byzantium and were brought from there to Italy, where they served as the basis for some twenty-one apographs before x disappeared from the Vatican collection at some time around 1527. The text in Dilts' edition is based upon these two branches, V + x (as reconstructed from the apographs), with additional material from <J> (Vatican, gr. 96), a 23 thirteenth-century codex representing excerpts from the Varia Historia, the De natura animalium, and the Politiae of Heracleides Lembus, two texts which also form part of V and x and which "follow the TloiKiXn ioropia like a suffix," as Dilts says, throughout the tradition (Dilts 1965: 57 et passim). The stemma then is reasonably clear and in itself does not offer a great deal of room for controversy. The difficulties which the text presents arise when scholars, trying to assess Aelian's interests and purposes in writing this work, attempt to account for certain peculiarities present in all the exemplars of the Varia Historia. These peculiarities involve ( 1 ) the presence of doublets, that is, longer and shorter versions of some chapters involving primarily Books 12 and 14, (2) the use of the word on to introduce a number of chapters beginning in Book 3, and (3) a rather striking stylistic variation among chapters, some carefully and deliberately narrated and others succinct and condensed. We may consider the doublet chapters first. These occur in three "batches" in Books 12 and 14. The following is a list of these double occurrences. Batch One: 12.2 — 14.37 12.4 — 14.34 12.5 —- 14.35 12.6 - — 14.36 Batch Two: 12.12 — 14.46a 12.13 —- 14.46b 12.14 —- 14.46c 12.15 — 14.46d 12.16 — 14.47a 24 Batch Three: 12.22 —- 14.47b 12.29 — 14.48a In some cases the version of the material in the Book 12 chapters is shorter than the corresponding chapters in 14, and in some cases it is considerably longer. On three occasions (12.6, 12.12, 12.16) the earlier material is introduced by the word on ; on one occasion (12.5/14.35) both chapters begin with on. There are several doublets which are so close to each other as to be nearly identical (12.6/14.36; 12.13/14.46b). The fact that, with one exception (14.37), the sequence of the doublets is the same in Book 12 and in Book 14 adds to the problem of explaining this peculiar situation. The doublet chapters are not the only ones to challenge textual scholars with the presence of on. The Varia Historia contains about eighty chapters beginning with this particle, chapters which tend to be extremely succinct and condensed as opposed to fuller, more carefully written sections. But not all such paragraphs present a condensed style. Nevertheless, H. Liibbe undertook to explain this situation, summing it up in this manner: Diversa scribendi genera in [Varia Historia] esse perspicuum fit: alia enim capita scriptor luminibus rhetoricis largissime exornavit, alia vero sine ullo cultu sunt et mira exilitate laborant. (1886:1) Earlier editors of the Varia Historia had assumed that the presence of on and the condensed style, in conjunction with the situation with the doublets described above, indicated that the work had at some time been epitomized. The problem with this theory, first pointed out by F. Rudolph, was the fact that (1) on was used at the beginning of some chapters in which Aelian made mention of his own reactions to the material he was 25 compiling (e.g. 12.17; 12.48), thus including editorial comment which we would expect an epitomator to omit; (2) on was used to introduce chapters which, in terms of content and sequence of material described, showed close similarities with passages in Athenaeus' Deipnosophistae, i.e. these passages had not been epitomized in Aelian's work even though on was present, (3) some of the doublets begin with on and yet offer longer texts than their counterparts, or contain descriptive or synonymous words and particles which one would expect an epitomator to have eliminated; (4) there is no substantial stylistic difference between the portion of the Varia Historia from 1.1 through 3.13 and all material found after 3. 13, the point at which evidence for epitomization is claimed to begin (Rudolph 1884: 100-101). Rudolph and Liibbe tried to explain the situation by suggesting that Aelian himself had added on at the moment of compilation. According to Rudolph's theory, debetur illud on Aeliano ipsi fontium variae historiae epitomatori, qui eo praemisso argumenta inter legendum probata in codiciliis suis ut forte libido excerpendi et adumbrandi praevaluerat, ita on particula usus nudis rebus adscribendis continebatur; quo magis autem quaeque fabella ipsi arriserat, eo fusius earn tractare suoque iudicio augere malebat. (Rudolph 1884: 101-102) According to this interpretation on was part of Aelian's compiling process. As he went through the works he was excerpting and copied down passages, he began each compiled passage with on. When he came to an especially affective passage, he became so involved in the process of writing it up that he eliminated the particle. In connection with this theory Rudolph suggested that Aelian intended the Varia Historia for his personal use 26 only—a use which would suggest that the Varia Historia was meant to be a commonplace book—and that at his death the work was left unfinished. Liibbe accepted Rudolph's explanation of the on but felt that Aelian's more elaborately written chapters were not sufficiently accounted for by considering them to be simply notes. Liibbe suggested that the Varia Historia was substantially and purposively rewritten: Cum...ex iis quae breviter adnotaverat opus suum conficeret, quae maxime ei arriderent, iis larga manu fucum induxit eaque non solum ' omni ornatu rhetorico distinxit, sed etiam additamentis auxit; quae vero minus ei placerent, ad ea exornanda minus studii laborisque attulit, sed iisdem fere verbis ea in opus recepit quibus antea breviter consignaverat. ( Liibbe 1886: 6) However, those scholars such as R. Hercher (Aelian's first Teubner editor) who insisted upon the presence of epitomization in Aelian's text could argue for epitomization by citing evidence outside of the manuscript tradition of the Varia Historia. Citations in several sources from late antiquity reveal the existence of a text rather larger than the text represented in the manuscripts. Stobaeus, whose anthology contains the earliest testimonia to the Varia Historia, includes fifteen chapters from our text and five fragments attributed to the Varia Historia but not included in our text. Of the fifteen chapters in common, three are nearly identical. Stobaeus cut seven of the present chapters short at the end; but there are ten chapters which Stobaeus presented in a fuller form than occurs in the manuscript tradition. "He doubtless had a fuller text of noixiXq iaropia than is preserved in our manuscripts," Dilts concludes (1971: 4), which refuels the arguments for epitomization of the Varia Historia. More support for epitomization is offered by the Suda, which like Stobaeus expressly attributed to the Varia Historia four passages which 27 do not appear in our text. One further passage may reflect a fuller version of a chapter which our text contains (Dilts 1971: 5-6). Rudolph's and Lubbe's attempts to explain away these fuller texts as the results of misattribution and misquotation are not convincing. Proponents of the epitomization of the Varia Historia may add two further details to their argument: (1) Al l existing exemplars of the work lack any indication of a book division or a title for Book 6 (Perizonius 1701: Praefatio xxxi; 404), the current edition's division at this point reflecting the arbitrary division of Book 5 made by Peruscus in his printed edition of 1545 (Dilts 1974: xii). (2) What looks like the note of an epitomator appears in the middle of Varia Historia 6.8: 'ApTagepgriv T O V Kai 'Oxov emKXnOevTa, ore EnefiovXevaev auT(£> Baywag 6 euvouxog, og fjv Aiyt>7moc;, chaaiv avaipeOevra Kai KaraKOTrevra ToTg aiXoupoig 7rapapXr|0fivai • exdcbri 5e Tig aXXog & V T ' auTou Kai arreSoOr) raig PaaiXiKaig 0f|Kaig. [OeoouXiai uhv T O U "Ox o u K C ( i aXXai U E V XeyovTai Kai udXiora Kara rr\v AryuTrrov.] TW 5e Baycoa OUK T O djroKTelvai TOV ^Hxov K . T . X . When Bagoas the Egyptian eunuch plotted against him, they say that Artaxerxes surnamed Ochus was killed and chopped into little bits and thrown to the cats. Someone else was given a funeral and buried in his place in the royal tomb. [Ochus' sacrilegious deeds are discussed, especially those committed against Egypt.] It was not enough for Bagoas just to kill Ochus.... This puzzling addition was bracketed by Dilts in the current Teubner edition; it had been deleted in the previous Teubner by Hercher. Complicating the issue of textual manipulation is the fact that Aelian's title occurs in alternate forms. Stobaeus and the Swdla-autfior both add to the text's equivocal status in antiquity by quoting alternative titles for the Varia Historia. Stobaeus uses EuuuiKTog ioropia five times and ioropiai three times, while the Suda uses both LToiKiXri iaTopia and IIoiKiXn dcbfiYPOig (Dilts 1971: 5). 28 The study of Aelian's relationship to his sources began with Perizonius' edition of and commentary upon the text of the Varia Historia, a work first published in 1701. Perizonius drew his readers' attention to the close similarities between chapters of the Varia Historia and passages in Athenaeus' Deipnosophistae. Again it was Rudolph who began a detailed study of this relationship. In the 1884 study discussed above, Rudolph attempted to demonstrate that Aelian had compiled Athenaeus' material directly, pointing out that the order in which data appeared in the two texts was almost identical and that Aelian had changed only a word or two in his rendition. Rudolph then proceeded to propose other sources for the Varia Historia by aligning Aelian's data with those found in other authors such as Diogenes Laertius and Stephanus Byzantinus. By analyzing the shared data and linking them to sources quoted by these authors, Rudolph proposed a number of sources for the Varia Historia in addition to Athenaeus, including Favorinus and Pamphila (1884: 137). Several years later M . Wellmann challenged some of Rudolph's assumptions in a series of studies analyzing other possible sources for Aelian's material in both the Varia Historia and the De natura animalium (Wellmann 1890, 1891a, 1891b, 1892, 1895, 1896,1916). Unfortunately for our present inquiry, his findings were most applicable to the zoological collection, for which he suggested as sources the lost works of Pamphilus, Sostratus, Juba, and others. However, Wellmann concluded that Aelian had not drawn his material in either of his collections from Athenaeus, as most of Perizonius' successors had assumed, but rather that both Athenaeus and Aelian had accessed an unknown collection of material and that both had directly transcribed that author or authors (Wellmann 1893: 487). 29 In the end, an analysis of the sources of any of the miscellanists' data can only elucidate the extent of scholarship accessible by the miscellanist. Even when the miscellanist quotes his sources, we cannot determine whether he is referring to the source he holds in his hands or to the source which his immediate source held in his. The series of unattested acts of compilation could be extended back for generations. Aelian for example quotes Aristotle and Theophrastus as casually as he states, "It is said that... " Reconstructing the precise bibliography for his or anyone else's collection is an exercise in both patience and imagination. Yet this casual attitude on the miscellanist's part to documentation serves to emphasize his own peculiar attitude to the material he gathers into his collection. The miscellanist is primarily a collector. He does not compile with the intent of opening up and analyzing another's ideas, but of making them available to his reader. If he specifies his criteria for collection, they may be aesthetic, moral, recreative, all or none of these; but the material which he cuts and pastes must be, for one reason or another, worth remembering. Consequently, if Aelian attaches the name of Theophrastus or Aristotle to data he records, it is because the name adds value and authority to the data. The immediate source of the compilanda is hardly relevant to their value as collectibles. The problem of determining Aelian's sources for the Varia Historia does not, then, seriously affect our analysis of the work as an Imperial miscellany. From at least the fifth century AD, when Stobaeus anthologized passages from the Varia Historia, it has itself served as a source for later scholars. We must assume, with Rudolph, a series of 30 collections of compiled material available to the Imperial period, not only to miscellanists but to any reader interested in research and composition. Do the problems with the text of the Varia Historia affect the present discussion of the miscellany tradition? In this regard one must consider the implications of epitomization, at least insofar as they affect our interpretation of Aelian's assumptions in creating this work. In order to argue that in the Varia Historia Aelian was consciously working within the generic framework of the miscellany, it is a necessary assumption that he was writing for a reader. In his 1884 study, Felix Rudolph had argued that the Varia Historia represented not a finished work but rather the notebook, as it were, in which Aelian had been recording his compilations, itself neither ready nor necessarily intended for publication. There was no way to offer a counter-argument to epitomization for the state of the text, Rudolph maintained, unless we assumed that the Variam Historiam non esse opus perfectum, sed materiae collectionem futuris curis reservatam....Iam cur variis suis finem non imposuit [Aelianus]? Scilicet quia fato, antequam ea perficeret, abreptus est. Ergo hoc eius opus ultimum erat aut postumum et post historiam animalium exaratum est. (101-102). Rudolph could take this position because the individual chapters of the Varia Historia are in general much less structured, less stylistically homogeneous, and less detailed than those of the De natura animalium. Unlike the De natura animalium, the Varia Historia contains no internal cross-referencing to tie its data together and to show that Aelian was controlling the selection and placement of individual chapters. If the De 31 natura animalium is taken as an example of Aelian's writing style, then the Varia Historia is clearly the less carefully finished of the two. By suggesting that the on which previous scholars had accepted as a sign of epitomization originated instead in Aelian's own note-taking process, Rudolph relegated the Varia Historia to the level of a commonplace book. Is it possible to assert that Aelian had not purposely structured his collection in (more or less) the manner in which it has descended through the manuscript tradition? We have abundant enough references to notebooks and to the process of compilation from the Imperial period to suggest that the keeping of a notebook for one's private reference, created without the primary intention of publication, was a common enough practice among readers (cf. Steinmetz 1982: 278). Plutarch, for example, when on one occasion pressed for time and unable to create a polished essay, instead accessed his notebooks (uTrouvquaTo:) and sent off transcriptions from them.12 When faced with a problem in terminology while studying dialectics, Gellius accessed L. Aelius Stilo's Commentarium de proloquiis in the Bibliotheca Pacis. The work was so succinct and opaque, however, that Gellius had to assume that Aelius' text was a "reminder to himself rather than a teaching text (NA 16.8; cf. S. West 1970: 290). Again, at his death Pliny the Elder bequeathed his nephew 160 closely written notebooks, electorum commentarios opisthographos quidem et minutissime scriptos; qua ratione multiplicatur hie numerus uDe tranq. an. 464.E.7 46: (ifixe 5e xpovov k%wv, foe, 7rpor|po'()|a.r|v, y E v k o d a i npbq oTc, kpovXov [if\Q' imouevcov KEVOCTC, navxanaoi TOV dtvSpa x e P ° ^ v b4>8r|vat aoi nap tpcov dt^ vyp-evov, avEXetfx\i.r\v ... £ K TCDV i)7CO(i,VT|pdXCOV COV EUXXmCp 7C£ftOir)U.£VOC, £.ZVYXavov-Not having enough time to make a careful choice, and not being able to stand the thought of you beholding your man coming back with empty hands, I gathered together some things from my notebooks.. 32 notes consisting of compiled passages, written on the front and back of the page and in a very tiny hand-in fact, they were doubled in length that way. (Pliny Ep. 3.5). It is as such a notebook, containing material clearly considered valuable but in a condensed, sketchy, or outlined form, that Rudolph would have us interpret the Varia Historia. The lack of prologue and epilogue contributes to this interpretation of the work as a relic of Aelian's, left inchoate at the compilator's death. If the argument based upon the fuller texts of the Varia Historia in Stobaeus and the Suda is not accepted as proof of the epitomization of the text, then the only counter which can be made to Rudolph's thesis must be based upon internal evidence for intentionality. That is, does Aelian imply in his chapters the presence of a reader and a desire to communicate with him? As I shall attempt to demonstrate in Chapter 5 below, such intentionality can be traced both in the structuring of individual chapters of the Varia Historia as well as in the positioning of chapters within books. Despite the evidence for tampering with the text, which following Dilts I attribute to the hand of one or more epitomators, enough of the internal patterning of Aelian's work is in place to reveal that Aelian was compiling his material and addressing himself to a specific kind of reader. Scholarship Addressing the Varia Historia, the Imperial Miscellany Tradition, and Paideia As I have attempted to point out, scholarship devoted to Aelian and the Varia Historia has been focused upon the text and its sources. Studies of the literature of the Second Sophistic, on the other hand, have had to include Aelian and his collections in a survey of the second and third centuries AD. How then has modern scholarship assessed Aelian's literary achievement in relation to contemporary authors? 33 In this regard, two questions complicate an adequate assessment of Aelian's work. 1. Is every writer whose work consists primarily of "recycled" compiled material to be included in one general assessment? 2. Is such a work as Aelian's able to be assessed as a piece of literature? Making paideia accessible to others is the goal of the miscellany author, but it is a goal shared as well by many other writers of this period. Here we must make a broad distinction between the compilator whose material is process-oriented and him whose collection is data-based. The former category includes authors of artes, Texvou, and encyclopaedias, and here may be mentioned by way of example Philostratus' De arte gymnastica, [Plutarch]'s De musica, Martianus Capella's Nuptiae, and the fragments of Varro's Disciplinae. These technical exposes were intended to summarize the processes and the content of the arts which made up the eyKUKXioc; 7tai5eia or general liberal education, becoming increasingly standardized during this period (Hadot 1984:99-100). Although these works could contain various forms of information in their discussion of the origins and development of individual artes, their primary goal remained the exposition of procedure. The data-based collection may be termed "polymathic."13 They cover a wide area of intellectual interests. In attempting to distinguish among works so diverse as for example Lucian's Demonax, Diogenes Laertius' Vitae philosophorum, and the lexica of Phrynichus, Moeris, and Harpocration, the author's assumptions about his reader's needs provide the most efficient means of distinguishing among these paideia collections. 1 3 The importance and assessment of polymathia during the Imperial period will be discussed in Chapter 2 below. 34 Thus we may eliminate from the present discussion of the Imperial miscellany such works as the Atticizing lexica and commentaries on specific authors, as well as literary works which were primarily referential and exegetical, intended to be consulted as aids to the reading of canonic authors or as authorities in some area of traditional literature. But even after such a categorization, there still remain a considerable number of authors whose works consist primarily of compiled data. At this point, authors of modern assessments of Imperial literature have selected, from this still very broad spectrum of authors, those who for whatever reason strike the modern scholar as particularly worthy of note. In the Geschichte der griechischen Literatur, Schmid distinguished the literature of the Imperial period formally, between poetry and prose. Having among the prose authors separated out the epideictic orators (including Lucian and Philostratus), he then divided up the remaining authors among those whose subjects were broadly historical in the modern sense (including geographers, ethnographers, and paradoxographers), and philosophical. Schmid was compelled to create a special section for a discussion of Aelian's Varia Historia and Athenaeus' Deipnosophistae, a section which he entitled "The Bioitschriftstellerei of the Sophists." In this section of his study, Schmid attempted to analyze, categorize and understand the works of Aelian and Athenaeus. Schon in der hellenistischen Periode bemerkt man ein Bestreben, bunte, insbesondere auch abgelegenen Wissensstoff, der bei den Forschungen der Fachgelehrten abfiel, zu Unterhaltungszwecken zusammenzustellen Die Neusophistik hat sich dieses Gebiets bemachtigt, um der zwanglosen Stoffanreihung auch eine zwanglose oder zwanglos sein sollende sprachlich-stilistische 35 Einkleidung zu geben in [der] neumodischen dchEXeta. (1924: 785-. 786) Having offered a general introduction to this category of writers, Schmid went on to sketch Aelian's biography and then to assess his writings. Was wir von Aelianus besitzen, sind Auszuge teils geschichtlicher teils naturwissenschaftlich-paradoxographischer Art. An Kritik gegeniiber seinen Vorlagen denkt er nicht ernstlich. Ihm ist es bloss um pikanten Inhalt, um Stilkiinste im Sinn der modernen otcbeXeia um eine gewisse erbaulich moralistisch-mystische Tiinche zu tun (786-787) Schmid then summarized Wellman's work, referred to above, on Aelian's sources. Several peculiarities arise in the course of Schmid's summary of "Buntschriftstellerei." We are, for example, told of the connections between paradoxography, mythography (as represented by Ptolemy Chennus), Xuaeicj-collections (Plutarch's Ouaestiones conviviales), and the miscellany, although these categories of scholarship have quite distinct traditions of their own. We are reminded that all of these works consist of data uncritically gathered and artlessly put together, yet told of the "artificially informal" style which the miscellanist purposely assumes. All of the main points about Aelian's work which Schmid makes in the Geschichte der griechischen Literatur are touched upon in the section devoted to "applied" rhetoric in Reardon's Courants litteraires grecs des lie et Hie siecles apres JC. We find here the same insistence upon Aelian's carelessness and artificiality, supplemented by a seemingly gratuitous desire to ridicule what Reardon believes is Aelian's purpose in writing. 36 Reardon focused his discussion upon the De natura animalium. In describing the De natura animalium he states that Cet...ouvrage est muni lui aussi d'un theme (on ne peut pas parler d'une structure) cense, selon l'auteur, justifier l'assemblage de ces curiosites saugrenues .... Nous avons droit a bien des bizarreries....On est heureux de constater qu'il ne s'efforce pas de tirer trop de conclusions, mais se contente, la plupart du temps, du role de conteur. Dans l'histoire variee, il n'y a meme pas d'excuse [i.e. presumably Aelian does not have a "theme" which justifies the quality of the compiled data] .... En somme, Elien ne tient pas trop, malgre ses pretensions, a developper un theme; les faits divers font son affair...II veut plaire et (a sa facon) instruire; bref, il veut amuser. Notons qu'il croit le faire en employant un "style simple," mais en fait son apheleia est tres artificielle et plutot facheuse. A condition de le lire par petites quantites, il est assez attachant. Mais il n'est pas serieux ... sauf... a ses propres yeux. (1971: 225-226) Although he does not discuss his reasons for doing so, Reardon appears to dismiss the value of a miscellaneous collection of data ("il n'y a pas d'excuse"). Basing his general conclusions, as Schmid had done, upon the work on animal ethnology and applying them indiscriminately, Reardon repeats Schmid's pejorative dismissal of Aelian's work as "entertainment." Yet he does not specify the manner in which these data, which are admitted to be "absurd," "oddities," and "curiosities," can be expected to entertain. Reardon types Aelian as a "storyteller," yet does not specify the difference (if indeed he recognizes one) between an anecdote (which is a true narrative) and a story (which need not be "true" in the historical sense). 1 4 Indeed, in his two-page discussion of Aelian's work he quoted only once from one chapter of the Varia History — the first chapter of the first book - which raises the question of the extent to which he had examined the collection. 37 Given such assessments by Schmid and Reardon, it is not surprising that Anderson, in his 1993 survey of the literature of the Second Sophistic, should dismiss Aelian and his work in a very succinct manner. Of the miscellany he states, Manuals of Variae Historiae and Mirabilia were at best the scrapbooks, and at worst the scrapheaps, of the educated. But pedantic trivia could acquire an entertainment value of their own (193). Again, the categories of scholarship are mixed, is a paideia collection here seen as a technical "manual" to be used in a process-oriented educational experience? Is the paradox collection in fact the same as a miscellany? And how can "pedantic trivia" be entertaining? Anderson suggests that Aelian is incapable of understanding the value of his compiled data. The least engaging writers are those who...often seem unaware of the real potential of the basic ingredients [drawn from paideia] and combine sophistic materials with naive moral platitude ... as in the case of Aelian (ibid. ASS). Anderson repeats Reardon's attribution of naivete to Aelian, but does not suggest the possibility that this is an affected simplicity rather than a basic simple-mindedness. Is it possible to account for the three elements of disparity of treatment and content, entertainment potential, and naivete, which Schmid, Reardon, and Anderson have isolated in their critiques of "Buntschriftstellerei"- authors and of Aelian in particular, by seeing these qualities as genre-bound rather than as faults of taste, judgment, or intellect? In fact, studies of individual figures have proved more fruitful in this regard than have the surveys of literature. Peter Steinmetz' 1982 analysis of the Latin authors of the second century AD, for example, contains a sensitive analysis of Gellius' Nodes Atticae. 3 8 Although Steinmetz reviews the Greek miscellanists in what are essentially Schmid's terms,15 when he comes to consider Gellius' achievement in the Nodes Atticae he finds more specific and positive qualities to emphasize: a sense of focus and direction in terms of the reader and the reader's needs (279), and a concern with providing means for the correct use of the reader's leisure (290), in a careful structuring of material within chapters and of chapters within books (281-287). Steinmetz considered Gellius' achievement in the Nodes Atticae to be substantially different from that of the Greek miscellanists. In der Auswahl des Stoffes, in der Auswahl der Exzerpte und Notizen mochte sich Gellius aber von den Gepflogenheit der Buntschriftsteller unterscheiden (280). In adopting this position, Steinmetz was taking his cue from Gellius himself, who in the prologue to the Nodes Atticae complains of the lack of discrimination in selection of data shown by early miscellanists, both Greek and Roman. As I shall attempt to demonstrate in Chapter 2, however, Gellius did not differ qualitatively from the authors he rejected, but rather codified in his prologue the generic features of the miscellany as it was known to readers of the Imperial period. In specifying Gellius' goal as one of providing a "Wissenschaftspropaedeutic" (279), Steinmetz identified a quality apparent in all polymathic Imperial collections (cf. for example Pausanias 3.18.10), yet which is especially pertinent to the miscellany because of the miscellanist's peculiar relationship with his reader. As Steinmetz suggested in the case 15"Vorstufen dieser Gattung finden den Sammelwerken...des aristotelischen und theophrastischen Peripatos...lehrreiche Unterhaltungsbucher...wird das Sensationelle (das paradoxon) zur Unterhaltung des Lesers dieses Unsystematische und dieses durch immer neuen Themen aus den verschiedensten Bereichen den Leser Verlockende also wichtigste Merkmale der Buntschriftstellerei ansah" (275-276). 39 of Gellius, and as I shall attempt to demonstrate in the case of Aelian, the miscellanist conveys relevant information at the same time that he, and sometimes the characters in the material he conveys, model the appropriate way in which the data are to be received and applied to life situations. Relevancy and application are the two major concerns, then, of the miscellanist, and both depend upon the miscellanist's right relationship to paideia. In this regard, recent scholarship analyzing the attitudes toward and transmission of paideia during the Imperial period throws light upon the miscellanist's position in the cultural tradition. Whether they concern individual authors, genres, or general cultural movements, all such studies must address the primacy of paideia in all areas of Imperial culture, especially the conveyance of paideia and the manner in which it provided the "cultural ecology" of the Imperial period (Anderson 1993: 242). It is certain that the archaizing environment of second- and third-century literature and culture lies at the base of all interpretations of the period's cultural achievement. II n'y a pas de siecle qui soit plus conscient de la tradition que ne Test le deuxieme, et Ton ne devrait Petudier autrement qu'en fonction de cette conscience (Reardon 1971: 5). Yet in the process of emphasizing this pervasive importance of tradition-based paideia, it is possible to overlook the fact that literature did have a connection with daily existence, that the educated reader and writer during the early Empire had lives in the everyday world, to which they were expected to give their attention and energy. And although it is indeed true that the Imperial authors do not consider rhetoric-inspired literature the proper venue for a discussion or analysis of contemporary events, concerns, and circumstances (Schmid 1924: 666-667; Reardon 1971: 3-4), this convention does not 40 prevent literature from reflecting, however indirectly, the individual author's and reader's bonds with the quotidian (cf. Bompaire 1958: 477; Steinmetz 1982: 119, 289; Raster 1989: 13-14). It is in this area of interface between the paideia conveyed by a polymathic collection and the individual's incorporation of paideia into his life in the present that the miscellany must be encountered. Therefore those recent studies of the Imperial period which have focused upon the place of education in Imperial society are especially pertinent here. Two works in particular may be mentioned. H.I. Marrou's 1938 study of depictions of paideia on Imperial sepulchral monuments directly addresses the role of education in the life of the individual. In MOUOTKOC , dvf|p, Marrou describes a social elite of educated and privileged individuals held together by a common culture acquired exclusively through the study of literature and the artes, and dominated by archaism of language and intellectual stance. Recent studies by R. Raster, building upon this conception of paideia's social and spiritual primacy, emphasize the ideal formative function of a literary education in the moral life of the individual during the Imperial period. In his 1980 study of Macrobius' dramatization of the grammarian Servius in the Saturnalia, for example, Kaster analyzed the way in which Servius' attitude toward paideia (verecundia) mirrored the attitudes of the creators of paideia (e.g. Vergil) and of the current social order itself as reflected in the participants of the Saturnalian discussions. In addition, the Saturnalia became as a whole a kind of model of the ideal working of paideia. The values and behavior elaborated in the dialogue become the well-spring of the dialogue's substance. Macrobius chose to make a . virtue out of a fact of life: the fragmentation of 41 redeemed here, not because knowledge is coordinated and redirected toward some new synthesis, but because it is endowed with the unity of the social order. The behavior of the participants goes beyond the polished good manners of urbanitas, to become inseparable from, and as important as, the information conveyed (248). In Guardians of Language: The Grammarian and Society in Late Antiquity, Kaster further developed his interpretation of the social and moral qualities imparted by and through paideia. Analyzing the role played in the establishment of a political and social elite by a traditional literary education, his work provides what I believe to be solid support for my interpretation of the Imperial miscellany as a vademecum for the proper utilization of that education. My task in the present study is to position Aelian's Varia Historia within the spectrum of demands made upon paideia and its transmission. As I have attempted to show in this review of recent scholarship, students of Imperial culture have demonstrated that paideia is not only a conveyance of data or of process, but entails a moral and ethical formation as well—in short, the creation of a human soul able to react correctly to the demands made by society Aelian and the Imperial miscellany play a role in this transmission, but have been marginalized and, I believe, partly misunderstood in their relationship to paideia and the social demands made upon it. Yet this inability correctly to assess the miscellany occasions difficulties with the further assessment of the entire literary experience of the culture. This is especially observable when the literature is seen as either escapist or even illusory (cf. Bowie 42 1970:36-41). Anderson's puzzlement in the face of one Imperial genre—the declamation —helps to explain his out-of-hand rejection of Aelian. 1 6 It is difficult to arbitrate about the success of it all, since the criteria themselves are so elusive. But if the goal was to pretend to be in the fifth century BC, however contrived or perverse such an ideal might seem to us, then the Second Sophistic was well on its way to achieving it. If the aim was to invest present literature with a sense of continuity with the classical past, then again the illusion was largely successful (237). As I shall attempt to demonstrate below, one of the means to an understanding of the interconnectedness of this literature, polymathic and otherwise, lies in determining the relationship not only between the composer and paideia, but between the composer and reader and between the reader and his reception of paideia. The following chapters discuss these relationships as they appear in the generic framework of the Imperial miscellany. 1 6 "It seem misleading to try to pull all the surviving material together, call it literature, and impugn its quality on that account. For a start we ought to be able to put aside material that seems purely preparatory, such as Aelian's Varia Historia, or even for that matter Marcus Aurelius' Meditations, which are in the nature of an informal commonplace book" (ibid. 242). 43 Chapter 2 Compilator and Compilanda Before Aelian's Varia Historia can be analyzed as an Imperial miscellany, the generic qualities of the miscellany must be examined and a framework established within which to view Aelian's collection. This framework has three aspects which may be classified as: the compilator's attitude toward his subject matter (paideia); the compilator's relationship with his reader, including his conception of the reader's needs in the acquisition of paideia; the relationship between paideia and entertainment, including the role of style and structure in providing entertainment. The present chapter deals with the first of these aspects. Here it is my purpose to demonstrate the following: 1. For Aelian and other Imperial compilators, paideia has two components, both of which must be addressed by the system of education: a. paideia-sanctioned skills, the teyyax or artes. b. data, which we may summarize as TroXupaOia. 2. The acquisition of paideia involves two successive processes, each demanding self-conscious zeal and dedication (o7rou5f| or diligentia) on the part of the reader/scholar: a. a formal literary education under a YPauucmK6c, and perhaps a OO(JHOTT|C,; b. the self-education of the educated adult, the oXcoc, TTCTrouSeuuevoc,. 3. The miscellany is created by a oXioc, rorrouSeuuevoc, who, through his successful completion of formal education and his present correct approach to self-education, is in a 44 position to provide his reader with two valuable things: relevant jroXuuaGia, and a good model for paideia acquisition. Paideia as Education Up to this point I have been using the term "paideia" in the sense of "liberal education," "cultural tradition," and "literature-based education." This Greek concept now requires some unraveling. No single word in English covers this range of meanings. "Culture," for example, places an emphasis upon the contents at the expense of the process, while "education" shifts the weight in the other direction. And neither "culture" nor "education" implies the indoctrination in right values and social forms which "paideia" contains. To position the miscellanist's activity within the literary production of the Imperial period, however, all three aspects of paideia must be borne in mind: the process of acquisition, the contents of this process, and the results of the process through the acquisition of content. The extent to which the miscellanist's selections reflect the value system imparted by paideia, as well as the manner in which the miscellany furthers the process of education itself, will be dealt with in later sections of this chapter. The question which concerns me here is the content of paideia during the Imperial period. How is the material which the miscellanist compiles and preserves related to the subject matter of education? 45 The question is an important one in determining the generic framework of the Imperial miscellany. The scholarly assessments of the miscellany examined in the previous chapter emphasize the miscellany's instructive or at least scholarly aspect.1 Statements by the miscellanists themselves emphasize the pedagogic value of their material as well. Clement speaks of the dxbeXeux (Strom. 1.1.2) which he hopes men will derive from his undertaking. In the prologue to the De natura animalium Aelian prides himself on the 6tc;iov constituting his collection. Gellius' assertion of educative value is the most explicit. Ea ... sola accepi, quae aut ingenia prompta expeditaque ad honestam eruditionis cupidinem utiliumque artium contemplationem ... ducerent aut homines ... a turpi agrestique rerum atque verborum imperitia vindicarent. (NA Prol. 12) / included only those items which would direct ready and prepared minds to the proper desire for learning and a study of the useful arts, or would protect men from a shameful and boorish ignorance of language and information. Gellius had already identified this material as anything which he had found to be memoratu dignum (Prol. 2); later in his prologue he adds to this general kind of material notice of another sort which he has included and which he requests his reader to treat with indulgence as it is a little more difficult to read and understand, pauca quaedam scrupulosa et anxia vel ex grammatica vel ex dialectica vel ex geometrica, quodque erunt item paucula remotiora super auguris iure et pontificis. Some few items rather thorny and troublesome, from grammar, dialectic or geometry, some things — even more widely ranging — drawn, for example, from the law of the augur or pontiff. (NA Prol. 3) 'e.g. Reardon: "[Aelian] veut plaire et ... instruire"; Schmid: "abgelegenen Wissensstoff, der bei den empirischen Forschungen der Fachgelehrten abfiel..."; Steinmetz: "lehrreiche Unterhaltungsbucher"; cf p. 33-38 above. 46 Gellius thus divides his paideia excerpts into two categories: those which are related to the artes, which may be a little challenging; and those which are in general worth remembering. In regard to the latter, more diverse kinds of excerpts, Gellius distinguishes his selection as represented in the Nodes Atticae from the miscellanies created by less discriminating compilators, quoting in judgment Heraclitus' adage TroXupaGiq voov ou 8i8daKei (Prol. 12). Gellius implies that vouc, results from his collection, while the collections of other miscellanists foster only jroAuuaGia. Thus in terms of data which are pedagogically beneficial we are given a choice between those related to technical subjects and providing instruction in the artes, and those which, while not being related to the artes, yet run the risk of contributing to a mere jroAuuaGia. Is Gellius dismissing TroXupaGta as a desideratum for the miscellany? How does the material which Gellius relates, in its use of anecdote, chreia, and thematic list having many points in common with Aelian's Varia Historia,2 avoid being labelled as polymathic? Does noXvuaQia have positive aspects, which Gellius has chosen to overlook here? Or is TroXuuaOia a general risk run by all miscellanists in gathering relevant data, a quality or state to be avoided at all cost? In fact, the term jroXuuaGia was modified somewhat in connotation over the centuries. Gellius had tapped into its earliest occurrence when he cited Heraclitus, but Stobaeus quotes similar usages in Pythagoras3 and Democritus,4 suggesting that among 2e.g. W 8 . 12 and AM 8.9; W 9.20 and AM 19.1. 3 T O §6 7TE7Tou5e0a9ou O U K ev TioXuuaOeia Xovcov avaXtiy>Ei (Stob. 2.31.96). According to Diogenes Laertius 8.6, however, Heraclitus criticized Pythagoras' polymathic scholarship: nuOayoprn; Mvnaapxou ioropinv £7Toir]0O(TO eauTou ao(|)inv, TToAuuaOrinv, KOtKOTEXvinv. 47 philosophers there was a need to distinguish data (derived from research or from formal/sophistic education?) from wisdom (derived from one's aristocratic nature? from experience? from age and virtue?); for the sixth and fifth centuries, the scrutiny of the wellsprings of excellence is a frequent literary topos (cf Jaeger 1944: 5-14). Plato and Platonic dialogues continue the negative assessment of 7roXuua0ia. At Laws 811, for example, occurs a discussion of the role of literature in the classroom, and whether a lot of time devoted to learning a number of poets fosters a child's social and moral development, e i UEX XEI TIC, dya06g r|utv K a i aocboc, E K TroXuTTEipiaq K a i 7roXuua0iacj YEVEaOai whether one might become good and wise from learning many skills and many things. The conclusion is that total ignorance is a better alternative than a polymathy associated with the overlearning of a great number and variety of poets and the incorrect learning of mathematics (819a). In the Platonic Amatores, Socrates undertakes to demonstrate that cbiXooocbia is not the same thing as 7roXuua0ia. In the process he demonstrates that the polymath relegates himself to a position of inferiority by trying to do too many branches of knowledge. oux OUTCOCJ, w (JHXE, EXWOI , un5' f| T O C T O cbiXoaocbEiv, jrepi TOCC, TEXvac, EOTTOuSaKEvai, OV36E jroXuTrpayuovouvTa KU7rTdCovra £fjv O U S E 7roXuua0ovjvTa, dXX' dXXo TI, ETTE! Eyw wunv K a i OVEISOC, Eivai T O U T O K a i Bavauaooc, KaXEio0ai TOUC; ixepi TCHC, TEXvac; EorrouSaKOTac,. They are not philosophers, nor is this pursuit of the arts and crafts, this life of prying and peering and learning many data, philosophy. No, rather it is something blameworthy, and those who have devoted themselves to the arts are called low technicians. 'VoXXoi 7ToXu|ua0ee<; vouv oik E X O U O I V (Stob. 3.14.8). 48 The author of Alcibiades II likewise links together skills or processes (jroXurexvia) and TroXuuaGia (147a). These passages do not suggest that TroXupaBia means learning quantities of data, the meaning which Gellius seems to convey in his use of TroXuuaGia. Much learning in Academic terms appears rather to be the learning of many skills, emaTfjuat and jiyyau Hippias, the polymath of the Hippias Major, is famed for his acquisition not so much of information as of xiyyav. arithmetic and geometry, "grammar" (rhetoric and poetry), music, astronomy (285b); we lack only dialectic to have the full battery of the seven liberal arts canonized in late antiquity. But these were not the only techniques which Hippias professed. According to the Hippias Minor, he was also an adept at jewelry-making, pottery, woodworking, cobbling, weaving, and macrame (368b). Granted that the author of these dialogues heaps together such sophistic accomplishments as rhetoric with the banausic crafts with some polemical intent, we can still see the technical, process-oriented side of traditional Greek education emphasized here.5 There is, however, one topic Hippias professes in this dialogue which does not represent a technical skill. This is his dpxaioXoyia, information on genealogies and the foundations of colonies (Hipp. maj. 285b). Hippias insists that these data are real crowd-pleasers: ouXXf||38nv Trdcmc, rfjc, dpxccioXoyiac, fj8iaTa dxpoiovTca, loar EVIOVE 8i ' auTouc, nvdyKaouou EKueuaGnKEvat TE Kai EKPEUEXETTIKEVCU 7TavTa rd Totaura. What they altogether most dearly love to listen to is every kind of discussion about the past. They're the reason why Vve taken such pains thoroughly to research and master all such matters. ' Apuleius (Flor. 9.36) praises Hippias as homo multiscius, but carefully segregates his rhetorical skil from his skill in crafts: quin ipse Hippian laudo, sed ingenii eius fecunditatem malo doctrinae quam supellectilis multiformi instrumento aemulari; fateorque me sellularias quidem artes minus callere. 49 Such information, though it draws closer to Gellius' assessment of 7roXuua0ia as content-oriented, is yet linked here with process. Its acquisition is a matter of learning by heart, and it is associated with the skill of public speaking. It is precisely this thorough learning of poetry (oXouc, TTomTdc, eKuav0dveiv) that, Plato insists, exposes children to the dangers of become polymaths (Leg. 81 la.l). Xenophon too presents Hippias as the model polymath. At Mem 4.4.6 Hippias' polymathy is again a kind of appendage to his rhetoric. And here again occurs the suggestion that 7roXuua0ia is linked, like dpxaioXoyia, with content or data. Socrates is the speaker: ob Troria, ou uovov del raurd Xeyco, dXXd Kai ropi TCOV aurwv • cru 5' iatocj 6td TO 7roXuua0f|c, rival Trepi TWV auTiov ouSejroTe Ta airrd Xeyeic,. dueXei, ecbri, jreiptopai Kaivov TI Xeyeiv dei. "Not only am I always saying the same things, but I'm always discussing the same subjects as well. But you, perhaps because you are a polymath, never say the same things about the same subjects. " "As a matter of fact, I really make an effort, " he said, "always to have something new to say. " Yet this jroXuua0ia too has negative connotations. Hippias always has something new to say because he knows a lot of things; Socrates always says the same thing because, it is implied, he knows the truth. There is an ambivalence about the assessment of 7roXuua0ia in a quotation attributed to Aristotle by Plutarch (Quaest. conv. 734d5): TTIV TroXuudOeiav rroXXdg Tapaxdc, TTOIEIV. Polymathy causes much confusion. As I shall discuss below, the transition from negative to positive roXuuaOia occurs during the Hellenistic period, when scholarship is deeply influenced by Peripatetic activity (see below p. 84 footnote 31). Does Aristotle 50 view the "confusion" arising from TroXuuaGia as a necessary preliminary to further positive research and synonymous with ocropia, or does he interpret it as a negative, troubling state which prevents one from reaching clarity of vision? Either interpretation suggests that 7roAuua6ia is more "information" than "versatility," especially given the empirical orientation of Aristotle's own research. Less ambiguous than Aristotle's assessment of 7roXuua0ia but still expressing its author's awareness of polymathy's ambivalent qualities is a statement which Stobaeus attributes to Anaxarchus: TroXuuaGvn xdpra uev axbeXeei, rapra 5e pXdTrrei T O V EXOVTOC. dxheXeei uev TOV 8e£iov av5pa, pXarrrei 5e TOV pniSkoc, (hcoveuvra jrdv OTOC, xai ev 7rdvTi Sqpcp Polymathy both benefits and harms the polymath. It benefits the clever man, but harms the man who speaks indiscriminately. (Flor. 3.34.19) Polymathy is a two-edged weapon, one which requires respect and careful handling. Positive assessments of TroXuua9ia begin as early as Isocrates. eav fie, (J)iXoua9f)c,, eoei jroXuua9f|c;. a uev emoTaoai, rauTa 8ia(j)i3XaTTe Taiq peXeTaig, a 8e pq pepd9qKag, 7rpooXdpPave Talc, emoTTipaig If you love knowledge, you will be a polymath. Preserve what you already understand through practice, and acquire through study what you have not learned. {AdDem. 18.1) In a context of rhetorical efficacy, the emphasis still seems focused upon skills acquisition. But with Strabo we find jroXuuaGia as skills definitely giving way to TroXuuaGioc as data. Information in the form of factual details is needed by the man undertaking to study geography, and 7roXuua9ia is the knowledge associated with the empirical researcher. 7ToXuud9eia, 5i' fjc, M O V R I C £<JHKEO9OU Tou5e T O U epyou 5uvar6v, O U K dXXou Ttvoc, eanv q T O U T d Geia KCQ T d dv9pcoTreia empXerrovTog (1.1.1). 51 Polymathy is the only means of accomplishing this work, and is the function of none but the man whose view encompasses the whole range of heaven and earth. Strabo would increase this treasury of knowledge by adding, basically, all the information in the world. cbepe 6r) rfj ToaauTfi rroXuuaOeia 7rpoaQ<Jouev TOV eTriyeiov ioropiav, oiov Coocov Kai cbuTwv Kai TCOV aXXwv, ooa xPnOTua fl Suaxpnara cbe pei yfj TC Kai OdXaaaa. To such a quantity of knowledge let us add research into terrestrial phenomena such as animals, plants, and the like, all the flora and fauna which the land and sea produce for good or ill. Although Strabo traces such comprehensive factual knowledge back to Homer (1.2.20), we are dealing here with a concept of knowlege more empirical, more open-ended, more data-based than that discussed by Plato and Xenophon. noXupaOia is more inclusive than the variety of poems and poets in Plato's Laws, more empirical than the learning of skills in Xenophon, the [Plato] of Alcibiades and Amatores, and Isocrates. It is this positive assessment of data which determines Imperial uses of the term jroXuuaOia. Unless an Imperial author is arguing for a specific philosophical or theological position,6 the word denotes a valuable and respectable trait, applicable both to scholars and paideia-icons alike. So Aristotle, Posidonius, Varro, Cicero, and a series of others are labelled TroXupaOflg or TroXuuaOeoraToc, by Plutarch and Athenaeus.7 But Athenaeus also extends the epithet to literary figures. At 1.24.28 he types Odysseus as the ideal polymath. ai Leipfjvec; 5e qc5ouai TO) 'OSuaaei TO udXiora aurov Tepipovra Kai TroXuuaOeia Xeyouaai. lauev yap, <bacn, T & T ' dXXa Kai oaaa yEvnrai ev xOovi 7roXu|3oTeipr|. 6e.g. Philo De congressu 20.4, De somn. 1.206.1; Aristides 12.82.30; at Sacra parallela 96.93 Joannes Damascenus compares the pious individuals who are dAiYOuaOelc to the hypocritical T T O A U U C X O E I C . 7e.g. Athenaeus 9.58.35, 11.112.41; Plutarch Luc. 22.2, Crass. 3.6, Comp. Dem. etCic.1.3. 52 The Sirens sing to Odysseus those things which especially delight him, communicating with him through polymathy. "We know," they say, "many things, all the information in the world. " For later writers such as Eustathius, Photius, and the Suda-auihov, TroXuuaGia and its related adjectives and adverbs are used consistently as terms of praise. Although Gellius had insisted upon rejecting TroXuuaGia as not fostering vouc, in his miscellany, elsewhere TroXuuaGia is found as a positive term in reference to miscellanies, sometimes used by miscellanists of their own work. Photius, for example, praises Sopater's twelve-volume 'ExXoyai 5ia4>opai as a work both learned and convenient, dc, TroXuuaGiav ex T O U eroiuou (Bib. cod. 161.105a). Clement describes the Stromateis as Tfj TroXuuaGia oiouaTOjrotouuevoi (Strom. 1.2.20). When we examine the kind of material which miscellanists consider valuable for inclusion in their collections, we find that in fact it is all — the contents of the Noctes Atticae as well —able to be considered TroXuuaGia, as material over and above what the reader is expected to have acquired in the course of a more or less standard literary education. As a data-base, TroXuuaGia represents material which builds upon a foundation in the artes, but includes, as well, information drawn from the paideia tradition and contributing, in the miscellanist's view, to the reader's intellectual improvement. It is TroXuuaGia in this sense which is conveyed by the Imperial miscellany. How then is it seen as relevant and improving to the reader? What is the relationship between TroXuuaGia and Imperial paideia? An analysis of the relationship between this positive, relevant TroXuuaGia and paideia during the Imperial period has two components: 53 1. The relationship between paideia and the text as an archive for data. The miscellanist creates an archival text through the selective manipulation of earlier texts which themselves stand in varying relationship to the literary tradition. What relevant data can such an archival text convey? Does such an archive mirror, supplement, or interpret paideia? 2. The purpose which text-based TToXuuaGia serves in the enculturation process. What purpose does a data-base serve in paideia transmission? How is the miscellanist's reader expected to utilize relevant TroXuuaGia? Paideia and the Archive Greek paideia had always represented a blend of instruction in process and content, with literature, including literature set to music and performed in a community venue, as the vehicle of both. In his 1963 study of the effects of literacy upon classical Greek thought, Havelock coined the term "tribal encyclopedia" to denote this pedagogic quality of archaic Greek poetry. Poetry was encyclopedic in the sense that it transmitted all the cultural lore which this society deemed indispensable. Poetic passages could enculturate an audience in data such as that conveyed by the Iliad s Catalogue of Ships and the Odysseus' Pageant of Heroines in the Underworld. M L . West refers to such poetic material as "elementary brute facts" (1985:7), a term representing data which we, with differing concepts of literary decorum, prefer to store away as "history" rather than as literature. But peoples for whom written records play a smaller part or no part at all, and with whom the scientific study of history is underdeveloped, often think very differently. They delight in factual knowledge for its own sake, especially where it relates to people 54 and places beyond their own limits of time and space (West 1985: It is this knowledge "for its own sake" which will in the Imperial period find a partial counterpart in the 7roAuua0ia of the miscellanists. In these early centuries there exists another kind of knowledge codified in the "tribal encyclopedia" of oral literature and passed down through the generations. This consists of specific instructions on how to do things: how to sacrifice to the Olympian gods, how to win a chariot race, how to build a proper plow, when and whom to marry. And not only does the poetic literature provide a vehicle for these empirical tasks; it also provides instructions and proper models for behavior. Here I refer not only to the wisdom of proverb and saw contained in Hesiod, but to the moral content of elegiac and lyric verse and to the kind of character models provided by epic and lyric poetry as well (cf. Dover 1993: 14-15). So long as community venues exist for the transmission of paideia, literacy does not substantially affect this balance of fact, process, and value judgment in the content of the cultural tradition. When, that is, public performances of poetry regularly instantiate paideia on the community level, all three aspects are experienced by and transmitted to the community as cultural consumer. These are the conditions prevailing during the sixth and fifth centuries BC, at a time when the poet can be considered 5i6aoKaXoc;, speaking directly to the Greek community at all levels of sophistication. It is this quality of classical literature which Reardon labels "la largeur de vue des siecles de la polis" (1971:3). We form the impression that the authors of this literature communicate directly with the 55 community, that the community responds to the author's words directly and homogeneously. During the centuries of the Imperial period audiences are still capable of responding under certain circumstances to ancient paideia. From Prusa to Carthage, people gather in great numbers to hear lecturing philosophers reiterate the values and declaiming sophists recreate the contents and processes enshrined in the epic and tragic poets. Judging from reports of such performances of paideia, half a millenium of wars and social change have made no substantial alteration in the way Greeks respond to their tradition. By the second half of the first century A D declamation seems to have moved into the first rank of cultural activities and acquired an unprecedented and almost unintelligible popularity .... Its practitioners ... displayed their skill to enraptured or critical audiences, not only in their native places, but throughout the Greek world .... The favored themes of the sophists harked back constantly to the classical period. (Bowie 1970: 5-6) Yet Plutarch witnesses to the importance of literacy in acquiring paideia, and to the fundamentality of the written text: opyctvov TT\C, naibeiac, ij xpfloig TIOV pipAiwv. Books are the tool-box of paideia. (De lib. educ. 8b). Even more explicitly does Diodorus emphasize literacy not only as the sine qua non of all paideia, but as absolutely essential to civilized living. ric; yap av ag"iov evKtouiov 5id0oiTO Tfjg TIOV ypauudriov ua0tjaetog; Sid yap TOUTCOV UOVCOV oi uev TeTeXeuTt]KOTeg ToTg gwai StauvnuoveuovTcu, oi Se uaicpdv ToTg Tonoig SieaTWTeg ToTg TtXeToTov cuiexouaiv tog TrXnoiov TtapeoTwai 8id TWV yeypauuevtov ouiAouar raig TC Kara jroXeuov auv6nicaig ev e6veoiv fj PaaiXeuoi npog 5iauovnv TWV ouoXoyiwv rj docbaXeia PepaioTctTnv exei jrioTiv • Ka66Xou 5e rag xapteoTarag TIOV cbpoviucov dv8p(bv owrocbdoeig Kai 0etbv xpnououg, CTI 5e (biXooocbiav Kai jTctoav 7Tai5etav uovr] TripeT Kai roig eTnyivouevoig dei napa5i5cooiv eig djravra TOV aiwva. 5i6 Kai TOU uev t,f\v rqv (buatv atTiav u7ioXn7TTeov, TOU 8e KaXtog Cflv TTIV EK TCOV ypauuaTwv ouyKeiuevriv 7rai5eiav. 56 Who could compose an encomium truly worthy of literacy training? Through literacy alone the dead are preserved in the memory of the living, and those who are in very distant places communicate through written documents with those who are far away as though they were very near. When it comes to compacts made between nations or rulers in time of war, that security which arises from a written treaty carries the greatest confidence in terms of the continuation of the agreements. Literacy alone preserves the most pleasing pronouncements of wise men and of divine oracles. Moreover, it hands on uninterrupted to the next generation philosophy and all of culture. Therefore it also follows that while nature must be understood to be the cause of life ,that paideia which consists in literacy must be assumed to be the cause of living well. (12.13) Although paideia for the Imperial period had a performance aspect, its acquisition and transmission are now bound to the written text. Scholars investigating the growth of literacy in archaic and classical Greece, in comparing the Greek phenomenon with that of other cultures, debate the validity of an "autonomous" as opposed to an "ideological" model of literacy in Greek society (Thomas 1989: 6-26). In basic terms, the autonomous model of a society's adoption of literacy attempts to demonstrate that a society's mental capacities are shaped as a result of its acquisition of literacy. According to this position, literacy calls forth and fosters rationalism, science, logic, and systematic scholarship. Suggested here is a kind of philosophical and philological Darwinism, with the fittest being those most responsive to the mental requirements of literacy acquisition. The ideological model of literacy, on the other hand, demonstrates how literacy allows a society to develop and foster certain tendencies innate within that society. The skills and concepts that accompany literacy acquisition, in whatever form, do not stem in some automatic way from the inherent qualities of literacy ... but are aspects of a specific ideology. (Street 1984: 1) 57 The alphabetization of Greek culture was, according to this view, simply a tool which the Greeks, especially the Athenians, used to pursue certain cultural goals. Thomas focuses upon the opportunities literacy offered the developing Athenian democracy for creating for itself a new "tradition" able to compete with the oral and aristocratic one challenging its legitimacy (cf Thomas 1989: 7Iff, 88ff, 108ff). With the possibility of recording, comparing and preserving cultural material comes the additional mandate of getting recorded data right. Oral traditions can exist in several exclusive versions, all with claims to accuracy; they can all be "right" in the eyes of a society limited to an oral tradition. A society like the fifth- and fourth-century Athenian, which demands accuracy in terms of current social relationships (specifically in terms of establishing Athenian citizenship), may come as well to demand accuracy in the cultural material drawn from the past. This demand for accuracy based upon written evidence becomes apparent in the Attic orators about the middle of the fourth century. References to archives, to inscriptions, to revelatory documents of various kinds appear in their orations, suggesting that the Athenian citizens as jurists have come to recognize written evidence as conclusive in determining the truth behind oral testimony. In regard to figures and events from the past, the same respect for written evidence can now be seen. It is no longer enough simply to refer to the achievement of the ancestors, remembered in the old oral traditions. It was more impressive if their achievements could be documented with the written word, in fact by the precise texts of their decrees (Thomas 1989:88).8 8 An interesting result of this developing demand for written documentation is the fabrication of documents allegedly from the period of the Persian Wars (Habicht 1961: 13-15). So quickly can a new expectation be catered to. 58 When we consider the role of the text in paideia transmission, the ideological model of literacy in ancient Greece amounts to this: a scholarly element in the traditional literary culture can be developed to a much more elaborate and detailed extent, for society now sanctions and may even demand the use of written records to validate an oral statement. The text functions as an archive for the storage of a work of literature which before had existed potentially in the memory of the paideia transmitter, actively in the voice of the performer. The potential likewise exists for the manipulation of texts through compilation, epitomization, and interpolation, as well as for the unprogrammatic acquisition of written information for its own sake, i.e. detached from a specific work of literature. Alphabetic texts in inscriptional form exist in Greece from the end of the eighth century. Scholars refer to the useof writing during the seventh and sixth centuries as "craft literacy," however. Artisans, perhaps poets, learned to record information for specific purposes. But we are seeking a bond between paideia and a written text. When does paideia archived in a written text begin to enter the Greek community at large? In a sense, the ability to read and write never permeates to a level of general literacy such as we know it in industrialized countries in the twentieth century. Indeed, under the Roman Empire illiteracy throughout the Greek and Roman worlds may have exceeded 70% (Duncan-Jones ap. Kaster 1989:36). However, evidence, primarily from Athens, for various kinds of writing on papyrus (piPAia) and for their use in schools in which reading and writing were taught, including the physical representations of the acts of reading from papyrus rolls and writing upon 59 them, indicates that texts are at least beginning to impact the Greek community by the end of the fifth century BC. Birt's collection of depictions of books and readers in art provides a number of graphic scenes with pedagogic backgrounds. The schoolroom shown on the Duris Cup, for example, depicts a teacher holding up before the viewer a book containing a line of verse, while a young boy stands before him; the child is presumably reciting from memory (Birt 1907: 138 fig. 76). Are such books compilations of traditional poetry? "Originally no doubt they were," Robb suggests, "prepared by, and were a part of, the professional equipment of the paid teachers of the rich" (1994: 186; cf. Turner 1952: 13). The fifth-century Athenian "Sappho Vase" also depicts a figure with a book in hand, this time the poetess (she is identified by the painter) concentrating intently upon a closely written text which she holds propped up in her lap (Birt 1907: 147 fig. 83; cf Turner 1952: 14). Another fifth-century Athenian grave-stele (Birt ibid. 157 fig 90; cf Turner ibid.: 15) shows in deep relief the figure of a fine-looking young man reading a thick book; he sits comfortably on what looks like a low bench, his legs crossed at the ankle, slowly unrolling his text in deep concentration. A small dog lies quietly beneath his chair. Evidence in fifth-century literature leads to the conclusion that the concept of the textual transmission of relevant data, as well as the importance of reading as a form of communication, is becoming familiar to the community as a whole. Significantly these references occur in a poetic format designed for public performance. Thus Pindar declares at Olympian 10.1, "Read out to me the name of the Olympic victor, Archestratus' son, there in my thoughts where it has been written," while dramatic poetry from the middle of 60 the fifth century contains references to books and the data stored in them. Aeschylus (Cho. 450, Eum. 275) and Sophocles {Track. 682, Phil. 1325) refer to an idea written down in one's thoughts, or to a thought written down on a tablet. The author of the Prometheus Bound refers to the "retentive tablets of thought" at 789, and at 460 presents Prometheus discussing his invention of letters. References to texts in Aristophanes are more frequent and explicit. This author assumes that the use of books is so obvious to his audience that it has a humorous aspect. Both the Sausage Seller in the Knights (188/9) and Labes the dog in the Wasps (959/60) can read (though admittedly not very well). In the Frogs, Dionysus reminisces about reading Euripides' Andromeda while serving aboard ship (52/3); and Euripides takes his texts with him into the scales-pan in order to have his literary worth weighed (1409). At Frogs 1113/4 the chorus states that the members of the audience are holding texts during the performance, EaTpaTeupevoi yap eiai, BiBAiov T ' ex^v eicaoToc; uavOdvei rd Secjid. "They have all seen active duty, each has his book in hand and has exquisite taste" (cf. Turner 1952: 22; Dover 1993: 34-35). From the closing years of the fifth century come references to a book trade. References to the marketing of texts occur in Plato and Xenophon in connection with accounts of Socrates. In Apology 26D, Socrates, accused by Meletus of believing that the sun is a rock and the moon made of earth, deals with the charge by stating that everyone recognizes such ideas to be those of Anaxagoras, whose works can be bought in the Orchestra for a drachma. At Anabasis 7.5.14, in recounting his army's march through the coastal territory of the "Millet-Eating Thracians," Xenophon describes the scavenging 61 habits of these people. Treacherous shoals cause merchant ships to run aground and capsize in the area, and the Thracians pillage them and market their booty. Xenophon describes seeing "much furniture, many chests, many written manuscripts" acquired through such activity. Presumably some of the ships wrecked in this part of the Aegean were supplying the local book trade.9 But what information from this period throws light upon the teaching of literacy? We are trying to connect the literary tradition with the text itself, and with the act of accessing paideia in textual form. In this regard we may consider a statement made at Memorabilia 4.2. Here Xenophon recalls a conversation Socrates had with Euthydemus. The young man was known to be collecting as many texts of learned men as he could find. Socrates mentions that he knows Euthydemus possesses all the poems of Homer, then goes on to inquire what his other books deal with: medicine? architecture? geometry? astronomy? The reference suggests that such subjects were being dealt with in the form of technical manuals or at least anthologies of matter relevant to these skills and available through the book trade (cf Harris 1989: 82), and that Euthydemus was acquiring them with the intent of using them as learning texts. He was looking at books qua texts; not texts as the voices of Homer, Pindar, or Hippocrates, but texts as objects which one can store up for one's personal consumption at some future date. 9 Although Flory has argued against the use of this passage for evidence for the book trade (1980: 20 note 33), one may counter by asking why the Thracians would have been willing to fight and die for such booty, as Xenophon claims they were, if it consisted only of "business documents, letters, and state papers" of questionable resale value. 62 In this regard is Thucydides' statement that he has created in his history a xxfjua ec, aie t (1.22), a possession for always, not a Preislied for the moment; "in der Form des geschriebenen und veroffentlichten Buches beabsichtigt er sein Publikum zu erreichen, das Publikum der Mitwelt und einer spaten Nachwelt" (Kleberg 1967: 5). The text, that is, is not so much a recorded voice as an archive containing important matter able to be stored for later consumption.10 Xenophon emphasizes this aspect of a written text at Memorabilia 1.6.14. Here Socrates is discussing the difference between his relationship with his friends and the sophists' with their students. Socrates and his associates learn together as friends: TOUC; Onaaupouc, TCOV rraXai aocbtov avSpcbv, ouc, exeivoi KareXiTrov ev pipXioig Ypaipavrec,, aveXiircov Koivfj auv TOTC; (biXoic, Srepxouca, K a i av TI opwuev dyaGov eKXeyoueGa. Along with my friends I read the treasuries of the wise men of the past, the ones they bequeathed by writing them down in books; and if we find something good, we select it out. Socrates uses the verb eKXeyouai to select out for oneself. This word can also be translated excerpt; its nominal form will be used as the title of, among other authors' works, the Eclogae of Stobaeus and Sopater. The text is viewed as a mother-lode of rich ore at the disposal of any reader with the industry to access it. Socrates' circle is engaged in the perusal of texts for the purpose of extracting this useful material, precisely the activity which such a later author as Macrobius will describe himself pursuing as he collects the data written up in the Saturnalia (Praef. 2-3).11 1 0 Gellius NA Prol. 21: quasi quoddam litter arum penus. 11 The context of friendly exchange and mutual benefit is important to keep in mind. It will become part of the background to the Imperial miscellanist's program. 63 Xenophon provides one last reference to this early textual manipulation. At Symposium 4.27 Charmides is reminding Socrates that he too is subject to strong emotional responses when he is around young men like Critobulus: aurov 8E oe, Ecbn, eyco et8ov vai ud TOV AroXXio, ore Trapd Tto Y P A M u a T i a T n &v T V « U T t p PipXitp dp(J)OTepoi EPOCOTEUETE fi , TT|V KEchaXqv jrpoc, rfj KEcbaXfj xai TOV copov yupvov rrpoc, yupvtp TCO KpiToPouXou topic Exovra. / saw you myself when you were at the schoolteacher 's,and the two of you were looking something up in a book, your head next to his head and your bare shoulder next to his. Plato's, Xenophon's, and Thucydides' readers are literate adults. They associate texts with recorded data, and are comfortable with such a form of access and manipulation. But what connection does the text as such a data-vehicle have with the process of childhood education? The description of early childhood education in Plato's Protagoras is the earliest clear discussion of the subject. The account is focused upon indoctrination in skills: specifically, the acquisition of the techniques of reading and writing, of singing and of playing a musical instrument, and of moving the body rhythmically and effectively. Content in terms of empirical facts disassociated from the activity itself12 is not even mentioned as an education desideratum. oi be 8i5dcncaXoi ... £7TEi5dv av YP«UUOCTOC udGtoorv Kcd PEAXCOOTV auvqaEiv Td yeypamxeva ... TrapaTuSEaorv auToic, em TOJV pdGptov dvayiYvtooKeiv jrotqTcov dyaGcov jrovfipaTa xai EKuavGdveiv dvavKdCouaiv .... oi T ' av Ki0apurred ... £7T£i5dv Ki9apiC£iv udGiooiv, aXAiov av Ttoinriov dYaGtov rroifipaTa 8i8daKouoi UEXOTTOIIOV . . . . ETI Toivuv 7rpdc, T O U T O I ^ eic, raxiSoTpipou 7T£p7rouaiv .... xai T a u r a TToiouonv oi pdXioTa 8uvdp£vot pdXtora (udXtora 8E 8uvavTai oi 1 2 E.g. the historical development of the lyre, a description of the stringed instruments used in Persia or India, or a discussion of Orpheus or Musaeus as early performers upon the lyre, as opposed to how to string and tune the instrument. 64 TrAoucncoTOTot) K a i oi TOUTCOV veiq, Trpcoa i raTa rig 8i8aaKaXcov Tfjg riXiKiag dpg"duEvoi cbovrdv, o i p i a i r a r a dTraXXdrrovTai. When [the children] are learning to read and at the point of understanding written works, their teachers set before them on benches the poems of good poets to read and to memorize .... When [the children] are learning to play the lyre, the lyre-teachers teach them the poems of other good poets, the lyricists .... In addition they send [the children] to the trainer .... And the parents with the most means do this to the greatest extent — the wealthiest being those with the most means — and their sons start out their education the earliest and are done with it the latest. (Prot. 325e-326a) Plato's discussion, which reflects a system in effect by the middle of the fourth century if not during Protagoras' own time, clarifies two points: a written document as learning aid is a fixed part of the classroom setting; and such an education is capable of being extended, the wealthier parents giving their children a longer exposure to this formative process. Plato's speaker, Protagoras, does not specify here whether a longer exposure to the educative process will introduce material differing in quality from what has gone before. The dialogue simply states that the material children read and sing is morally sound (TTOinriov dyaScov Troif juaTa, 5iEgo8oi Kai ETraivoi K a i evKcbuia 7raXaiwv dvSpcov dyaGcov). As discussed above (p. 47), Plato in the Laws is aware of a risk in making children "too learned" — TroXuuaOac; — if too great a variety of poetic material by too many poets is introduced into their curriculum. Early childhood education is not a matter of data-transfer. It is a training in skills and traditional (oral) literature which is considered to be most effective in transmitting social values. The extended education available to children of wealthy parents to which Plato's speaker in the Protagoras refers is associated from the fifth century with training offered by the sophists. This secondary education implies excellence in doing something; in fact it 65 amounts to training in logical thinking and public speaking, mastery of processes rather than transmission of data. In what way relevant to paideia and the Imperial miscellanist's activity did sophistic education utilize the written text? How, that is, was the sophist's text related to paideia and the transmission of data? It is the fifth- and fourth-century sophists who provide the Imperial miscellanists with their earliest models for paideia-compilation in an educational context. From the surviving information concerning the texts created by sophists for educational purposes, two kinds can be distinguished, both related to the archival use of literacy suggested by Xenophon's and Thucydides' statements discussed above: those which convey models of discourse, and those which convey matter — uXn or copia ~ to be manipulated in composition.1'1 Some of the texts produced by fifth- and fourth-century sophists were referred to in antiquity as TEXVCU . Despite the fact that the later systematic digests of the rules and methods applicable to a given craft were also called TEXVCU (cf Fuhrmann 1960 ; LSJ 5. v. Ill), these earlier works were collections of excerpts meant to illustrate rather than to summarize proper procedure. In considering the work of Alcidamas, for example, O'Sullivan states that the earliest rhetorical TEXVCU may not have contained abstract rules and theories, but rather have given expression to these in concrete examples (1992: 64 note 12) .... Such speeches were themselves called TEXVCU by the ancient critics, and it has been argued that all early rhetorical instruction was carried on through this medium; the handbooks with their abstract rules and organized presentation of material were perhaps the invention of a later age (ibid 104). 1 3 The purposes of sophistic education, whether it is em rexvn or em naiSeia (Prot 312b), will be considered in Chapter 3 below. 66 Another example of such a collection may be Polus' Mouoeta Xoytov, referred to at Phaedrus 267M0 (cf Nestle 1952: 1425-6). Such collections of models may have provided Aristotle not only with the examples of sophistic discourse used as illustrations in his Rhetoric, but also with the material which he compiled in his EuvavtoYn rexvcov (Diogenes Laertius 5.24). When Isocrates refers to the popularity of his own published speeches, the implication is that they are being studied at least in part as models of discourse (Panath. 251; Dionysius Halicarnassensis hoc. 18,; cf Stemplinger 1927: 6, Turner 1951: 19-20).14 Isocrates also suggests, through the manipulation of documents and of the works of earlier authors in his own discourse, that texts are being treated as archives of data (cf. for example Paneg. 72-121; self-compilation at Antidosis 194). Although he does not seem to have treated data-acquisition in any systematic way in the school he instituted at Athens in the 390's, it is very likely that in his teaching Isocrates encouraged the incorporation of specific information into his students' work. According to Richard Johnson, clearly there is no safe way of deciding exactly what subjects Isocrates taught besides formal rhetoric .... His "curriculum" was almost certainly ill-defined, within certain broad limits: he taught no mathematics ... still less the other sciences; he taught the technique of oratory; and his pupils learned the matter necessary to form their political, social, and ethical judgments and to provide content for their speeches (1959: 25-26). From what source will Isocrates' students have acquired this content? Johnson suggests several possibilities. 1 4 It is likely that Isocrates used his own and his students' oratory as the basis of classroom instruction (Johnson 1959: 28-29). At In sophistas 19 Isocrates warns students away from "the so-called artes" of the earlier sophists. 67 The historiae of Thucydides and of Hellanicus were both books requiring no specialist knowledge beyond the ability to read ... yet from one book the pupils learnt history, from the other geography .... It seems likely that they were not so much taught to the pupil as read by him, and the knowledge employed in his composition .... [Isocrates] had no objection to his pupils' learning material from books. There is certainly no evidence that he preferred the lecture method. Therefore it seems likely that as the pupils' compositions came to require political knowledge or history, geography or an ethical message, Isocrates recommended the appropriate reading to them and supplemented this with his own knowledge or opinions {ibid. 30). This suggests that students applied on an ad hoc basis to texts viewed as archives of information, to acquire data to be used for the corroboration of theses or as illustrative exempla. The second type of sophistic work appears to be the data-archive suggested by Isocrates. Hippias' Euvayooyrj may have been such a work. A fragment quoted by Clement (Strom. 6.2.15) may be the introduction to the Euvaywyri: T O U T W V i'aooc; ei'pnTou TCX uev 'Opcbel, rd 8e Mouoaiw, Kara Bpaxu aXXcp cxXXaxou, rd 6e 'Hcn66op, rd 8e 'Oufjpco, Td Se TOIC, dXXotc, TCOV 7Toir|Ttov, To. 8e ev cruyypatbaTc; TO pev "EXXncn, rd 5e fkxpBdpoic/ eycb 5e £ K Trdvroov TOUTCOV Td ueyiaTa Kai ouocbuAa ouvOeic; T O U T O V Kaivov Kai jroXueiSfj T O V \6yov rroiriaouai. Some of these things may perhaps have been said by Orpheus, some briefly here and there by Musaeus, some by Hesiod, some by Homer, some by others among the poets, some in prose writings whether by Greeks or by barbarians. But I will put together the most important and interrelated passages from these sources, and will thus make this present piece both new and varied in kind. (Trans. Kerferd 1981: 48) "This suggests," Kerferd maintains, "that the Luvaycoyri was a collection of various passages, stories, and pieces of information concerned with the history of religion and similar matters" (ibid)15 Snell insists upon the form because he is attempting to Snell considered Hippias' EuvaYcovii to be a forerunner of Peripatetic-style doxography, with information about the opinions of early philosophers; he acknowledges, however, that it contained as well "was wir zur 68 demonstrate that Hippias' work was a doxography, a collection of philosophical opinions compared and contrasted; the genre will be one frequently met with in the literature of the second and third centuries AD. It is tempting to view Hippias' compilation proleptically as scholarly research in the Peripatetic tradition (below, footnotes 22 and 32). This is not likely, as Pfeiffer has indicated All the sophists, in his view, are working toward their rhetorical goals in the tradition of the Homeric rhapsodes (1968: 16ff). Just as Gorgias and Hippias dressed in purple robes and performed at PanHellenic festivals in rhapsodic manner (cf. Aelian VH 12.32), so their studies were still in a more immediately organic relation with the oral literary tradition. The old poets were for the sophists the authorities and repositories of language. Conversely Protagoras can claim that Hesiod, Homer, and all of the early cultural figures were in fact sophists incogniti (Plato Prot. 316d; [cf. O'Sullivan 1992: 67]). Similarly, Alcidamas refers to himself and Homer as ioropiKoi (cf. Richardson 1981: 6). Thus traditional literature becomes of a piece with the sophists' own productions, and analysis of the poet's diction, prosody, and persuasive techniques is seen as providing valuable direction for teaching rhetoric in the present. To summarize: by the fourth century, Greek society has accepted the text as a vehicle of paideia. This acceptance has taken two forms: 1. The text is seen as an archive for works of literature, allowing them to be accessed, read, and manipulated during the educational process in lieu of experiencing Literaturgeschichte rechnen wurden ~ vielleicht auch mancherlei Grammatisches" (1144: 181 ff). The varied quality of some of this other material is shown by Diels-Kranz Hipp. B.4, a quote from Athenaeus dealing with Thargelia of Miletus, a woman so beautiful that she married fourteen husbands. The later tradition will make her a powerful Ionian hetaera — Plutarch sees her as a forerunner of Aspasia — who manipulated Greek aristocrats into medizing. 69 them through oral performance. Such archives would include not only socially mainstreamed texts (e.g., Protagoras' jroinTwv dyaQcov TTOifipara) but also new works designed as paideia models (e.g., sophistic oration; cf. also Pfeiffer 1968: 16). 2. The text can also archive useful data; that is, records that may prove useful to the student or to the educated man applying the learning he has acquired through literacy. Such texts would include, for example, the 'E6viov ovopaoicu and the 'OXupmavuxoov avaypacfofi of Hippias, works associates with his apxaioXoyia (Pfeiffer 1968: 52, Kerferd 1981: 46-48), as well as his LuvavoJYfi; the works, for example, of Thucydides and Hellanicus, referred to above in relation to Isocrates' school, would also function as such pedagogically effective archives of relevant data. This material, which Greek society's valorization of literacy has now made both storeable and physically portable (cf Robb 1994: 253; Goody and Watt 1968: 34), may mirror or supplement data which the community as a whole finds useful, the sort of data, for example, accessed and preserved by the Atthidographers. But they may also counter or challenge the community's concept of the relevancy of paideia. Here, of course, arise charges such as Heraclitus' TroXupaBin voov ou SiSdoxei • TToio8ov yap av e5i5a£e xai nu0av6pnv aung re Hevo(j)dved re xai 'ExaTaTov (Diels-Kranz 40D). It is significant that Heraclitus singles out Hesiod, with his scholarly catalogues and his "elementary brute facts," to head the list; likewise that the early prose-writer Hecataeus should conclude it. Aristophanes' portrayal of Socrates' Phrontisterion (Clouds 92ff) speaks to the same issue: the pathetic irrelevancy of Socrates' investigations into the activity of flea and gnat, the pointless gaping at the orbiting moon and at the opaque soil underfoot, the silly 70 attempts at eristics. When in other plays the same author brings on stage characters with books in hand, they carry texts full of empty nonsense: the Paphlagonian's and Sausage-Seller's archived oracles need confidence-men to interpret them (Knights 997-1097), the oracle-monger-to-the-birds manipulates his potted prophecies to his own more or less sinister advantage (Birds 959-991). The texts are dumb and meaningless until a speaker can valorize them. They are consequently ambivalent documents subject to manipulation for good or ill. In such terms Plato addresses the value o f the written record in the Phaedrus. In so doing he introduces a discussion o f another paideutic application o f literacy: the creation o f u7rouvf|ucxTa, the individual's personal notes or responses preserved in textual form. The double issue of the potential for both irrelevancy and ambivalence in a written record arises at Phaedrus 275d-e. Here written discourse is treated by Socrates with patronizing contempt. The voice conveyed in a written document is defenseless and even naive, Socrates asserts. OTOCV 5e cmcdi, ypacbfj, KuXiv8eiTai uev jravTaxou irdc; Xoyocj ouoiioc; TTapd TOIC, eTraiouorv, wc, 6' auToocj 7rap' olc, ou5ev 7rpoof |Kei, Kai O U K eTrioraTat Xeyeiv ok; 5ei ye Kai uf|. jrXr|uueXouuevog 8e Kai O U K ev 8iKr| Xoi5opr|0eic, T O U Traxpoq dei 5errat (3or|0ou' aurdcj yap O U T duuvaaOai O U T E BorjOfjaai 5uvarog aurco. When discourse is once written down, it is tossed about in every direction and encounters both those who understand it and those for whom it is not fit. It does not know how to discriminate between those with whom it must speak, and those whom it must avoid. Wronged and unjustly abused, it needs its father's help, for it is incapable of defending or helping itself. The written discourse is identified at 276a9 as the eiStoXov of the spoken, the latter being the true document "written in the learner's soul" and therefore more real and meaningful 71 to him. True knowledge is transmitted through the verbal exchange o f dialectic, between two parties one o f whom has the truth within and who "implants" it like a seed into the other's soul. According to such a concept o f the transmission o f wisdom there is simply no use for a written record. The written document at best can contain "reminders" — Socrates calls these uTtopvfipaTo: at 276d3 ~ but not truths. To record data in written form is, Socrates insists, a TOUSICX, a pastime, analogous to the planting o f seeds in a "Garden o f Adonis" which will germinate quickly and just as quickly wither away and die (276b). The terms Plato opposes in this section o f the Phaedrus are TTOU5I& and OTtouSn,, amusement and serious attentiveness.16 Written words cannot be taken seriously, for they are incapable o f defending their own semantic content (abvvaxa ... auToig Xoyto pon0e!v) and incompetent adequately to teach the truth (ocSuvaTa ... kavtog TdXq0fj bxbatpx). Al l they can do is to entertain. Toug pev E V YpdMM a a i xf|7Toug, tog E O I K E , jraiSidg yapxv o7TEpei T E xai ypcupEi, orav 5e ypdthn., Eaurop re UTropwipara SqaaupigopEvog, sic, T O Xf|6n,g yfjpag EOCV ixqiax, K G U 7ravri rto raurov fyvog penovn, fia6f|a£Tai T E amove, Getoptov (j)uou£voug a7raXoug • OTCCV S E OCXXOI 7Tou5ia!g aXXcag xP^vrai, aupTroaioig T E ap5ovT£g atrroug Exspoig T E ooa TOUTtov d5EX<bd, T O T ' EKEivog, tog E O I K E V , dvr i T O U T O J V olg X E V C O . When he writes he will sow in the garden of literature, it seems, for the sake of amusement, treasuring up notes (v7ro/jv?j/jaTa) for himself when old age makes him forget, and for everyone who follows after him. He will enjoy watching their tender growth. And when others are amusing themselves in their various ways, at symposia and the like, he will probably just keep on enjoying himself as I have described. (276d 1 -8) 1 6 {1LQ..} b vow excov y E c o p y o c , , &v c r a E p p a x c o v Kf]8oixo KCCI e y K a p u a povXovxo y e v E a 8 a i , robxepa crao-uSfj dv Gepouc, E I C , ' AScbviSoc, Kr\novq dpd>v xcdpoi Qecopcov KaXovq fev T|pepaiaiv O K x d ) y i y v o i i E v o - o c , , f\ x a u x a p e v 8f| raxiSidc, X E K a i Eopxfjc, %apw Spopr) dv, 6xe K a i icoior kfy' olq 5e 'EOKOV&OLKEV, xfj y E G o p y i i c f i xpojpevoq dv xexvn, arcEipac, eic, xo JtpoafJKOv, dyarccpri dv ev by86cp pnvl baa eajreipev X E X O C , Xapovxa; {4>AI.) Oiixco nov, & EcbicpaxEc,, xd psv ajtauSfi, xd 8e (be, E T E P C O C , dv fj Xfeysiq T C O I O T . {Phdr. 276b2-c3) 72 Plato does not distinguish subcategories of these orcouvriuaTa, but rather treats them as though they were all on the level of random notes to oneself. He allows Phaedrus to confess that the act of composing them is, at least in comparison to participation in symposia, a fine way to amuse oneself (TrayKdXnv AEVEIC, 7iapd (j>au\nv rrcaSiav) but insists that Socrates stress the superiority (KaXXitov crjrou8r|) of oral dialectic.17 The ujrouvriuaTa which Plato brings into the discussion of the impact of literacy on paideia introduce a subjective element into the text as paideia-archive. They represent an individual's personal thoughts, reactions and assessment of what deserves recording. Insofar as ujropvriuaTa affect the interpretation of the role of 7ToAuuoc0ia in paideia, they require some analysis here.18 The term 07r6uvqua is drawn from the verb UTrouiuvqaKto, meaning "to remind, to mention," and in its earliest usages reflects the act of calling to mind, from which the word can concretize into "reminder" (eg Thucydides 4.126.1; Xenophon Ana. 1.6.3). "Memorial" is a rather more concrete use of the word found especially in inscriptions (e.g. K&AAIOTOV ujTouvnua amov ec, TOV aTravra xpovov IG 112. 677 from the early second century BC). Plato is the first to emphasize t>7rouvr|uaTa in the sense of written notes or memoranda, a usage which in the fourth century BC takes two different directions. The 1' Intellectual activity as earnest nouSid, and the importance of ortouSri in approaching literature, are concepts which are addressed by the miscellanists. This passage from the Phaedrus is imitated by Clement Strom 1.1.14. 1 8 How do Plato's own dialogues stand in relation to the value of the written record? Although Plato refuses to grant serious educational value to the written word, he clearly recognizes the value of paideia models. The dialogues are to be seen as dramatic presentations of a philosopher's legitimate activity in dialectic. They are not registers of facts, but pictures of the process of truth's acquisition (Hackforth 1952: 163-4; Lynch 1972: 58; Robb 1994: 236). As such, as written texts they are analogous to the early sophistic Tsxvai, the collections of model discourses meant not as sources of data but as models to be imitated. As models of right philosophizing and of right orating the dialogues entered paideia, and became part of the system of Imperial secondary education (cf Quintilian 10.1.81, Gellius 17.20). 73 first of these is associated with the modern concept of personal memoirs, one's own reaction to events in daily life. cY7r6uvr|ucx in this sense is first attributable to the journals of Aratus of Sicyon (Bomer 1953: 222). Polybius' documentary sources for the composition of history are termed UTrouvriuaTa (12.25a),19 while Strabo labels his own geographical study UTrouvriuaTa (18.1.36) and his activity in creating it urouvnuaTi£ea6cxt (2.1.9). The emphasis seems to be upon personal experience. As Bomer's study shows in detail, this usage issues in the Latin commentarius, applied both to Caesar's accounts of the Gallic and Civil Wars, and to the chapters of Gellius' Nodes Atticae. The second usage of uTTouvrjuocTcx continues to convey a meaning which suggests a subjective recording of information, but here the material eliciting the response is related to the u7touvr|uaTC< as a primary text or object of study is related to a work of secondary scholarship. The Peripatos develops this usage into a more or less specific "research notes," applied now to compositions attributable to Peripatetic authors; in other words, to the more or less amorphous "school literature" associated with the Aristotelian library and the Corpus Aristotelicum (cf During 1950: 58; Grayeff 1974: 80-81). Yjrouvf|uaTa in this sense become part of what Bomer describes as the "Schultradition" (ibid 218), from the l oToptKa t>7Touvr|uaTa attributed to the Peripatetic researchers Aristoxenus, Theophrastus, 1 9 In his attack upon the methodology of Timaeus (12.25el), Polybius establishes a three-part program for the writing of rrpayiuaTiKri ioropia: an analytic study of source materials, including both archival documentation and scholarly research (noXuTtpayuoauvn E V T O I C ; uTtouvriuaai, TtapdBeaic rfjc £K T O U T I O V uXnc), autopsy (8ea, which seems to include the examination of expert witnesses [cf Schepens 1974: 281-282]), and personal experience (jrpdgeic). Although for what may be purely personal reasons Polybius ranks the first of these as the least important [Walbank 1962: 10-11], the examination of written sources is still a valuable part of the process of compilatory scholarship. The memoirs of individuals who participated in the events subjected to the process of icrropia provide the raw material — uXq ~ for the finished account. 74 and Hieronymus of Rhodes through to the Alexandrian scholars and beyond (cf. footnote 32 below). The word is first applied to a work of secondary scholarship explicating a work of "primary" paideia in the case of a study of Aristophanes' Plutus written by Euphronius, the teacher of Aristophanes of Byzantium (Pfeiffer 1968: 161 note 1). 'Ynouvr]u<xTa in this sense, i.e.of research scholarship characteristic of Alexandrian scholarship, acquired a recognizable form.20 Specifically, the U7r6pvr|pa was expected to follow the order of lines in the primary text being studied, to which it formed a companion volume (Leo 1904: 391). Critical signs in the margins of a literary text were matched in the, first by a quotation of part or all of the relevant text — the lemma — and then by an explanation of why the text had been marked with a critical sign at this point. The formula marking the transition between the lemma and the scholarly explanation was generally "the sign is placed because ... " : crnueTov on. This formula became in time shortened to on (Turner 1968: 115-116). The range of material which a researcher could incorporate into such a work of secondary scholarship reveals the extent to which, by the third century BC, the text had become the recognized vehicle of paideia. A scholar working at Alexandria had, first, the work of the author he was commenting upon, represented it may be by a number of manuscripts acquired through Ptolemaic agents and perhaps already edited by other scholars (cf van Groningen 1963: 15-16; for the 6iop0cbrj£ic, of the great Alexandrian philologists, cf Pfeiffer 1968: 87-231, Reynolds and Wilson 1968: 9-15; Turner 1968: 112-2 0 S. West (1970) questions the validity of applying to these early works of research a term such as "commentary" or "monograph," more descriptive of modern works of scholarship. 75 113; Blum 1991. 110-111). From registers or pinakes of various sorts the scholar constructing the ujrouvnua could acquire information about other works by his author, their titles and lengths, performance information if they were scenic, and some basic biographical data (Schmidt 1922: 66-70, Blum 1991: 150-157). From the research of the glossographers the commentary could be supplemented by various explanations of words whose meanings were obscure either because they were old, dialectal, technically specific, or foreign. Obscurities in content could be researched through the Xvoeiq and 7Tpo0XrjuaTa collections which were abundant at Alexandria (Gudeman 1927: 2520; cf. Deas 1931: 6-7). Material referring to unexpected, surprising, or puzzling phenomena in the natural world might be compiled from the paradoxographers. Manuals to the xexvat and other studies of the arts were also available to a scholar working at Alexandria. Euclid was one of the first of the foreign scholars at the Library. Aristarchus wrote on astronomy. Bolus of Mendes, "Hermes Trismegistus," and a number of others produced many works on astrology during the Hellenistic period. The tradition of medical research was especially rich during these centuries at Alexandria and extended out to other Hellenistic states (Fraser 1972: 339-440). Historical research would have been assisted by such compilations as Istrus' 'Armed, a auvayioyfj of excerpts compiled from the various Atthides available at the library ("ein.Buch, in dem das von den verschiedenen alteren Atthiden gebotene Material zu Arbeitszwecken bequem vereinigt war," Jacoby 1914: 2273). An example of such a UTrouvnua found as a papyrus text of the first century BC illustrates a commentator's use of compilation. The text being elucidated in POx 1086 76 (Pack 914) is the second book of the Iliad, the commentary to lines 751-826 is the portion preserved. Since the Iliad passage occurs at the end of the Catalogue of Ships and the beginning of the Catalogue of Trojan Allies, the commentator has geographic material to elucidate. Lines 49-51 establish the location of the "land of the Arimoi" (77. 2.783), with a supportive quotation from Pindar (frag. 93). Lines 63-73 of the commentary explain why Aristarchus athetized Iliad 2.791-795. Line 100 of the commentary glosses the Homeric usage of KoAu>vr|. At line 109 the commentator, seeking to pinpoint the meaning of KopuOorioXoc, at Iliad 2.816, offers several suggestions and a brief illustrative quotation from Alcaeus. In the same vein is POx 1087 (Pack 926) of the same approximate date as 1086, revealing further research possibilities for commentators. The author of this work quotes profusely: Pindar, the tragedians, Xenophanes and Archilochus, Alcaeus, Hesiod, and Stesichorus. His interests seem to be primarily lexicographical, for he shows a real interest in peculiar formations such as the jrapcbvoua (Hunt 1911: 100). At line 65 the commentator begins a note on burial practices, unfortunately lost. Presumably he would be accessing ethnographic collections, Homeric Xuaeic,, and Xecjeic,, for supporting material. The Hellenistic commentary required, then, a variety of sources from which to compile its material. Its value could be assessed according to the standards applied to the use of those sources. Selection (EKXoyr|l and Application ((movbr\) as Standards of Hellenistic Scholarship Two standards are referred to by Hellenistic writers as important for accurate and effective scholarship: thoroughness and detailed application in gathering source material, 77 and the use of good judgment in its selection. Imperial authors may describe these two acts as 07rou5f| or TTOVOCJ, and xpicncj, exXoyfj or their compounds and synonyms. As I shall attempt to demonstrate below, when paideia is preserved in textual format, these are seen as the correct steps in approaching paideia. They are applicable to the acquisition of primary and secondary education as well as to the adult reader's leisure contact with culture.21 Hellenistic scholars, able to access large and disparate collections of paideia texts and secondary scholarship on those texts, are among the first to enunciate the importance of selection and application in carrying out effective research and in composing relevant scholarship.22 The potential for 7roXoua6ia has radically increased with royally funded library collections and the patronage of professional scholars, as the preceding examination of surviving fragments of v>7rouvf|uaTa suggests.23 More important, the kind of knowledge which TroXuuaSicx fosters is becoming valorized for a culture which has transferred to the written text the respect it had once held for its oral paideia. Two examples of Hellenistic scholarship will suffice to show the necessity now felt for 07rou5fi and xpioicj in paideia. 21 The Imperial miscellanist's success in his compilations from primary and secondary paideia will likewise be judged in terms of his OTrou5r) and Kpiaiq: ajtouSii assures TtoXuuaOia, while Kpiaic determines that the polymathic data are afya uvr)ur|c . 2 2 According to Strabo (13.1.54), Aristotle was the first to create a library, TTPIOTOC TIOV iouev auvavaY^ v (3i(3Xia, and influenced the formation of the library at the Alexandrian Museum, 5i5&i;ac roue EV AiyuTTra (3aaiXsac (3i|3Xio9iiicr|c owrai;iv. Aristotle himself was well aware of the value of secondary scholarship, as he states at Topica 105b. 2 3 The books housed in the Alexandrian Library alone amounted to several hundred thousands, in "simple" and "mixed" formats (Aristeas ap. Eusebius Praep. Ev. 8.2.1-4). Ptolemy's own grandiose orrouSti in undertaking this project is suggested by Aristeas' statement of the commission Ptolemy gave Demetrius of Phalerum, T O auvaYotYsTv ei 8 U V C X T O V oaravTa ra Kara T H V o i K O u p s v n v (3i|3Xia. 78 Selection as a Standard o f Hellenistic Scholarship If the anecdote tradition reflects even dimly the quality of Ptolemaic text acquisition, the government-funded scholars at the Alexandrian Museum must have been working among veritable book dumps. Thus if we are to accept Bing's assessment of the Library's collection as "a kind of microcosm of Greece on Egyptian soil," we shall have seriously to reinterpret the implications of a cosmos (1988:14). The sources give little indication of discrimination or selectivity in acquiring documents; anything written as a volumen seems to have been fair game for Ptolemaic agents and customs officials charged with amassing texts for the Library. Pliny's story of the papyrus embargo against Pergamum (NH 13.70) and Galen's account of the confiscation of books found in ships' cargoes (In Hipp. lib. 17a606.8) suggest some degree of avarice and rapacity, with quantity alone as a goal, in the assembling of the collection. Galen mentions that as a result of their (biAoTtpia in acquiring book rolls, the Ptolemies were easy targets for forgers (In Hipp, de nat.\5.\09). The term PtpXioGfJKCu itself implies no more than "box of book rolls" and could include the sense "archives of nonliterary documents," a sense it retained in the Imperial period (Wendel 1940: 3). 2 4 No guidelines for or principles behind the collection, that is, seem to be in effect. Specific bibliographical studies, works explaining or describing the acquisition of texts, will not appear until the first century BC, with Artemon's Hepi ouvayioYfjc, PtpXitov and ncpi PtpXitov xpnaecoc, (Athenaeus 12.11, 15.49) and Telephus' PipXiaxfi eproipia (Suda tau 495).2 5 " Polybius at 12.27.4 seems to provide the earliest use of the term fhpXioOfjKou to suggest the contents of the book boxes, that is, a collection of texts. 2 5 It is not until the second century AD, with the titles of Herennius Philo's twelve-volume ilepi K T i i a E u g Kai E K X o v n c ; (3i(3Xiwv and Demophilus of Bithynia's Ilepi a £ i o K T r ) T c o v (3i(3Xuov and in discourses such as 79 One wonders how the earliest scholars working at the Library were able to find any given text, let alone submit it to a process of ordering or arrangement. Galen has stated that the only labelling done to new acquisitions at the library was to indicate the manuscript's immediate source; that is, whether it had come from a ship's cargo, from a previous individual owner, or from a city. Fraser suggests that there was editorial method behind this practice.26 It may be, however, that the early library organizers felt the need for some mark of identification of a manuscript and recorded the only information at their disposal, information easily acquired at the moment of the purchase or confiscation of a text It is therefore puzzling to see rather focused bibliographical purposes ascribed to the organizers of the Alexandrian Museum. When Fraser, for example, asserts that "the early organizers [of the Museum] aimed at a complete corpus of Greek literature" (ibid. 329) he implies that there was some concept of an established corpus or, for that matter, of "Greek" literature as opposed to the poems of Homer or Sappho or Alcman. Imperial authors like Aelian could indeed look back upon Greek literature as a more or less closed corpus of "Classics," but only after centuries of pedagogic practice had established a canon of school texts. That literature at the end of the Classical period was viewed as a more local and diverse produce is suggested by the anecdote related by Proclus, according to which Plato, desiring to read the poems of Antimachus, sent a student to Antimachus' hometown of Colophon to obtain a manuscript copy (In Timaeum Comm. I.90d). Pfeiffer Lucian's IIpoc T O V & 7 T O U 5 E U T O V Kai rroXXd pMpXia cbvouuevov that we find a suggestion that discrimination should be exercised in the making of a library collection (Suda phi 447, delta 52; cf Callmer 1944: 145). 2 6 "In the task of establishing the history of a text, or of determining the superiority of one exemplar over another, the provenance would provide the only satisfactory form of description" (1972:327). 80 insists that at the beginning of the third century the Greeks "became conscious of a definitive break between the mighty past and a still uncertain present" (1968:87), and "the whole literary past, the heritage of centuries, was in danger of slipping away .... The first task was to collect and to store the literary treasures in order to save them forever" (ibid. 102). Perhaps we can more accurately rephrase this in terms of an awareness on the part of society in general that there was a literary heritage in written form and, for that matter, that there were reasons why any kind of written transaction might be stored. The new uses for written material discussed above will have become familiar to the wider public as the boundaries of a Greek-speaking world expanded dramatically during the last quarter of the fourth century. Greeks moving out into this new world will have now developed expectations both concerning the availability of texts and about methods for using those texts. They will also have become aware that the literature with which they were familiar could be packed up and taken along to any new settlement as part of the baggage. The cultural heritage which Pfeiffer describes as in danger of "slipping away" is in fact now capable of slipping into any milieu. Literature in duplicable text form was now being transported to the classrooms on the frontiers. Presumably the ships' cargoes, to which the Ptolemies helped themselves so liberally in building the Alexandrian Library, were supplying the needs of readers and pedagogues in these newly founded Greek communities. For a scholar working among the texts at third-century Alexandria, the potential for disorder, miscalculation, false attribution, multiple or divergent copies of the same text, and the general confusion inherent in such a large and expanding book collection 81 presented a serious problem. The selection was given immediate attention, for we have information which suggests that, once these texts reached the hands of the Alexandrian librarians, once they had been labelled according to their place or person of origin, they underwent a process of classification. Information is scarce concerning this major preliminary step of imposing taxonomy on documents; it is drawn from two sources only. The first is Tzetzes' statement at the opening of the Prolegomena to Aristophanes. ioreov O T I 'AXexav5poc, 6 A I T I O X O C , Kai AuKOcbpcov 6 XaXKi8euc, urro ITroXeuaiou TOV) <PiXa8eX<bou TrpoTpajrevTec, Tac, oxTivucdc, 5icbp9coaav pipXouc,. AuKOcbpcov uev rdc, Tfjc, Kioutp5iac,, 'AXecJav5poc; 5e TCXC, Tfjc; Tpayto8iacj. dXXd 8fj Kai rac, aaTupiKdc, . . . T a g 5e Troir|TiKoi)c; ZT|V68OTO<; TrpcbTOV Kai uorepov 'Apiorapxoc, 8itop6tbaavTo. Alexander of Aetolia and Lycophron of Chalcidice were appointed by Phi lade Iphus to "straighten out" the dramatic texts, Lycophron the comedies and Alexander the tragedies and satyr plays .... Zenodotus first and then Aristarchus "straightened out" the poetic texts. This passage suggests a classification of texts into "poetic" and "scenic," the latter falling into three divisions: comedy, tragedy, and satyr-play.27 The one hundred and twenty volumes of Callimachus' JTivaKec, T I O V ev Trdatp jrai5eia 5iaXauipdvTtov imply both a massive bibliographical research project and a set of preconceived standards of relevancy in approaching the Alexandrian book-collection (cf. Blum 1991: 136-160).28 The Pinakes was not a catalogue of the books contained in the Alexandrian Library, although it presupposes that such a listing did exist which Callimachus could take 2 7 This system of classification does not reflect Aristotle's in the Didascaliae. He had distinguished the plays according to festival, and included the satyr plays with the tragic trilogies (cf. Regenbogen 1950: 1415-1416). 2 8 The complete title of the work, assuming that this represents its original title and not simply a description of its contents, is known only from the Suda kappa 227. Other information about the work and its composition can be derived from references in Dionysius of Halicarnassus, Diogenes Laertius, Athenaeus, Harpocration, Tzetzes, and various scholia (Schmidt 1922: 22-25). 82 as a starting point for his more ambitious project (Regenbogen 1950: 1420). It appears rather to have been a digest of information about writers and their works. The authors treated in the Pinakes were divided first into writers of prose and poetry. The latter were then divided according to their distinctive metrical genre: authors of epic, elegiac, iambic, melic meters, and finally the tragedians and comic poets. Prose authors were arranged according to subject matter, historians (and presumably all "researchers" in the Peripatetic sense of ioropia), orators, philosophers, doctors, authors of vopo i 2 9 , and writers on miscellaneous subjects (TravToSarrd; works dealing with such topics as cookery and fishing, in short all those works which could not be classified under the preceding rubrics). Once Callimachus had distinguished authors in this very general way, he alphabetized them in each category and then under each name listed their works in alphabetical order of title. He indicated for each work both its title and its opening words, its genre, and sometimes its dialect. Finally, he stated the number of lines each work contained, taking the dactylic hexameter as a standard in this regard (Herter 1973: 399). For each author in the Pinakes Callimachus provided a biography. A work of this form and size would have required several stages of research. A comparison of various kinds of book lists both in terms of acquisition listings and of library catalogues probably gave Callimachus an idea of what was immediately available at the Library. He would have been able to access more data by consulting such Peripatetic works as Aristotle's Didascaliae. A study of this sort could inform him about the works 2 9 It is not clear whether these were compilers of vouoi collections or themselves ancient legislators. Callimachus may have had both possibilities in mind. 83 which the Alexandrian Library did not contain. Callimachus probably acquired his biographical data on authors from various sources including the earlier compilations of the Peripatos, historians, local antiquarians and chronographers, and the authors' own texts (Fairweather 1973: 236-237). The importance of the Pinakes for the present discussion of the relationship between paideia and 7roXupa0ia lies in Callimachus' attempt to categorize the masses of texts accumulated at Alexandria. His one hundred and twenty volumes are not a tidy encyclopedia of standard school authors but rather a farrago of documents originating in many different social contexts and having served a variety of different purposes in their original contexts. Yet they all came together as paideia in Callimachus' pinacography, all put on a par and arranged, not by content but by medium (poetry or prose), and in alphabetical order/1 Only after this formal categorization had taken place were content and circumstances of composition taken into consideration, assuming that such data existed. In the case of Demosthenes' speeches, for example, Callimachus was able to make 3 0 That there was information available about the books which the Library did not contain is made clear by Athenaeus, who reports that, although he had himself made an ExXovn of the eight hundred Middle Comedies he had read, he still could not find a copy of Alexis' 'AotoToSiSdcncaXoc either at Alexandria or in the library register at Pergamum (8.15; cf Fraser 1972 vol 2: 486 note 179); yet Athenaeus knew that there was such a work in existence. 31 Callimachus was aware of other principles of arrangement and chose a different structure for the nival; Kai dvavpac|>r| T W V KaTa xpovouc; Kai cm dpxng ysvo^ivwv SiSaoKdXuv. This register contained the names of the Athenian dramatists and the titles of their plays, including the lost ones, from the earliest period of the scenic fesetivals. (A series of three fragments from a list of New Comedy authors found at Rome and probably drawn from Callimachus' register reveals the structure [Korte 1905: 443-446].) Callimachus listed each author chronologically under the date of his first victory. The poet's name was followed by a list of all his known plays, and these were arranged chronologically with an indication of each play's ranking by the respective judges in the Dionysiac or Lenaean festivals. Callimachus' arrangement is interesting because in it he reveals how he has drawn upon and then manipulated the material in Aristotle's Didascaliae. Aristotle had arranged his records of authors and titles by festivals. Callimachus, wanting to provide a register which would be more accessible to scholars who would have conceived of "Attic Drama" as a specialized body of literature, transformed the original data into a chronology of poets' names (Herter 1973: 401). 84 a finer classification. He could arrange the corpus into those speeches addressed to the assembly, those delivered in court, and purely epideictic works, the first two groups being large enough to allow a thematic division specifying, for example, those speeches which concerned Philip of Macedon (Regenbogen 1950: 1422). But the very existence of this work and its creation at this time reveal the need felt for selectivity and judgment in approaching texts. First, the Pinakes exerted a selective control over the thousands of books in the Ptolemies' collection. In Callimachus' archival compilation there is focus upon authentication, standardization, and critical evaluation. The title by which antiquity referred to this work specifies this focus as oi tv naar\ 7Tca8ria 5iaXapipdvTeg. Paideia, conceived now as a body of texts, contains its greater and its lesser lights deserving mention; presumably those figures who emitted no light at all were left out of the selection. Callimachus' listing of titles may itself have been a creative or selective act in the case of those works which had no title, or several. The quotation of opening (and sometimes closing) lines and the stichometry and volume count of individual works may also have had a prescriptive purpose. Zeal and Application as Standards of Hellenistic Scholarship As represented by the collections of texts at the Hellenistic libraries, Greek paideia had expanded radically with the recording and proliferation of cultural material in textual format. Callimachus' pinacography highlights the need for selectivity in dealing with an abundance of texts, each with a different provenance and each with a specific communal venue. In a work such as the Mirabilia of Antigonus of Carystus we can see the growing 85 awareness of the importance of scholarly application and diligence in approaching paideia, specifically through the self-conscious and purposive manipulation of sources. Representing the scholarly community centered at Pergamum during the second half of the third century BC, Antigonus wanted to make clear that his compilation of 9aupao"ica was a collection of earlier collections; that is, he was concerned that the reader understand that Antigonus did not consult a primary authority in the process of compilation. In taking this approach, Antigonus is aligning himself with Peripatetic empirical research.32 It is not therefore surprising to find Antigonus compiling the work of an author whom he believes to be Aristotle. Antigonus very clearly states the sources he accessed for his compilation. His work falls into five sections according to the manner of citation. The surviving work begins with a group of brief excerpts from a series of authors (Mir. l-26a). Then follows a lengthy 3 2 Aristotle's contributions to Hellenistic bibliographical methodology were mentioned above, footnote 22. Aristotle had responded to Plato's developed theory of ideal forms by developing in his own terms a concept of form pervading nature on the level of the individual entity, a concept which, while universally applicable, depended for its further development upon an inductive approach to material reality (cf Brink 1940: 915; Jaeger 1948:337). Working under the conviction that 6 8e 0s6<; KCX\ f] ( J H X T I C ouSsv udcrnv TToioucnv (De caelo 17la33), Aristotle and his immediate pupils encouraged the study of phenomena in nature and in the areas of human activity to demonstrate the validity of this basic Peripatetic concept. Material acquired as a result of this research entered the Lyceum's library collection to serve as data for further research. Rather than systematically and routinely archiving texts in the manner of the Metroon at fourth-century Athens, however, the Peripatos actively participated in the creation of new texts which drew upon those older works which it had produced or acquired and subsequently preserved. Secondary compilation of early work, as well as the editing, study, and storage of various kinds of documents and records, will have formed part of this activity. This concept of a community of scholars focusing their work upon the collection and storage of written documents consisting of both "primary" texts (e.g. a poet's or philosopher's book, or a transcription of archival material) and secondary studies and compilations of these (e.g. Theophrastus' digest of philosophical positions in the form of doxographical collections, or Aristotle's and Callisthenes' work on the victors at the Pythian Games) will have formed the basis of the system which, according to Strabo, Aristotle "taught" the Ptolemies (cf Blum 1991: 62-63; Blum here discusses Aristotle's own bibliographical principles as well). Alexandrian scholars working in this manner were sometimes labelled "peripatetic" in antiquity. "Hier wandert," Brink states, "der Name mit der Methode und dem Stoff. Aus dem Ausdruck Schulzugehbrigkeit ist eine Gattungsbezeichnung geworden" (1940: 904; we must however limit the connections between the Peripatos and Alexandria to these "exoteric" circumstances, rejecting a close connection between Alexandrian and Peripatetic philosophy [Brink 1946: 26]). 86 section drawn from an author whom Antigonus identifies as Aristotle. In fact this is the author of Historia animalium 9, selections from whose work are quoted chiefly in the sequence in which they occur in Historia animalium 9 (Mir. 26b-60a). Antigonus then adds another selection (Mir. 60b - 114) compiled from various other books of the Historia animalium, in general following the order of material in the original books but not the sequence of books, the existence of this ancient sequence of books as used by Antigonus probably reflecting the instability of the tradition of the Historia animalium (During 1950: 47-50). Following this double compilation from Peripatetic texts, material which Antigonus took to be Aristotle's, is another section of excerpts from assorted authors (Mir. 115-128). The fifth and concluding section consists of excerpts from Callimachus' paradoxographic collection, preserved solely through its inclusion in Antigonus' compilation (Ziegler 1949: 1146-1147). Of special interest to us are the second, third, and fifth sections of the Mirabilia, which represent texts which Antigonus prefaces with brief references to his compilations of another author's text. At section 26, for example, Antigonus notifies the reader that he is about to make selections from Historia animalium 9: Kai ufjv rac, re Xomdcj evrpexeiac, TOJV COJOJV, oiov ev udxatcj, ev Oeparreiaic; rpauudrcov, ev TrapaaKeuaicj TOJV Ttpoc, TOV Biov dvavKaicov, ev (biXooropyiaic;, ev uvtjuaig, dKpipeaTaT ' av TIC, CK Tfjc; TOU 'ApiororeXoug auvavioviic, KarapdOoi, ecj rjcj fijaetg TrpcoTov jroir|d6ue0a TT)V eKAoyf)v. One could very precisely learn about the remaining instincts of animals, in battle, care of wounds, acquisition of life's necessities, affections, memory, from Aristotle's collection, from which we shall first of all make a selection. The r iepi 9au(aacncov cxKouaudTcov and Book 9 of the Historia animalium represent early Peripatetic scholarship. Though generally assumed by modern scholars not to have been written by Aristotle, large parts of them can be ascribed to Theophrastus (During 1968: 315), and they are certainly fairly early products of the Peripatos (Fraser 1972: 771). Both Historia animalium 9 and ilspi Oauuocoicov cxKOuopdrcov compile material found in the Aristotelian portions of the Historia animalium (i.e. Books 1 through 6 and 8; cf. During 1950: 48). 87 Because the individual sections of Historia animalium 9 have a detached and succinct quality, making them easily transferable to another compilation, sections from this work show up both in the Pseudo-Aristotelian LTepi Occuuacricov aKouauaTOJV and in Callimachus' paradoxography. Antigonus likewise values the detail and accuracy of his source, which he emphasizes in introducing the compilation here. Antigonus makes a similar brief statement about provenance and quality when he compiles Callimachus' collection: jreTroinTou 5e Ttva Kai 6 KuprivaToc, KaXXiuaxoc, eKXoynv T O J V 7rapa86c;ojv, rjc, avaypacbouev oaa TOTC flurv ecbaiveTO elvat aKofjc, acjia. Callimachus ofCyrene has made a paradox collection, of which we write up all those matters which seem to us worth hearing about. Again, Antigonus' selection must have been fairly simple to put together, as the material had already been compiled once by Callimachus. But Antigonus must explain his procedure in rather more detail when it comes to scanning the multivolume text of the genuinely Aristotelian portion of the Historia animalium. Because Antigonus presumably was not accessing this work in an epitomized or pre-compiled form, he had to exert more effort in acquiring relevant data. 6 ye 'ApioroTeAnc, ... Kai Toiaura Tiva 8tec;epxeTai, jravu 7roXXf|v emueXeiav 7re7roir|uevo<; ev TOIC ; TrXeiaroic, auTwv Kai oiov epv°P ou jrapepvop xP^uevoc, rfj jrepi T O U T O J V ei;riyf|aei. TO youv TrdvTO axe86v e(35ouf|Kovra jrepi aurcov KaTaPepXr|TOi PipXia, Kai jrejreipaTai e^ riyri'nKOJTepov f\ ioropiKtoTepov ev eKaoroic; dvaorpe(bea9ai. Trpoc; Tfjv fipeTepav eKXoyf|v eKTroiei T O J V Trporipripevojv auTW T O cjevov Kai 7rapd5ocjov eK Te T O U T O J V Kai T O J V dXXtov em5paueTv. Aristotle also goes through such topics in detail, having expended upon most of them a very great deal of care and giving his undivided attention to his explanation. Consequently he has composed nearly seventy [better "nine"; cf Ziegler 1949: 1146.61-63] books in all on this subject, and has attempted to conduct himself in each more like one delivering a continuous account than 88 like one writing up research. As far as my collection is concerned, of that which he has deliberately selected it is sufficient to summarize the unusual and unexpected material. (60b) The point of Antigonus' brief introduction lies in the phrase e£Y|yT|TiKibT£'pov r\ ioTopiKcoTepov. If I am right, Antigonus found some initial resistance in removing the data which Aristotle had marshalled as evidence for the thesis he was developing at any given point in the original work; that is, his doctrine hung together like a continuous exposition, e^nyriTiKtbTepov, and the illustrative material required some teasing of the connective fibers before it could be presented as 7rapd8o^a. Antigonus enters his text as an editorial voice in introducing these sections of the collection. He also on occasion comments upon and evaluates his sources, a gesture which Aelian too will make when presenting data not able to inspire confidence. While excerpting Callimachus, for example, Antigonus intrudes into the collection long enough to complain that he has had to add a bit of relevant detail which his source had overlooked, . T O U T O 6e Kai Eu5o£oc. xai KaXAipdxoc, 7rapaXei7Touoiv (Mir. 161.2). Aristotle, on the other hand, is praised for the diligent professionalism he expends upon his iaropia (rcavu TTOXXqv empeXeiav TTETrompEVOc, ... oiov epytp ou jrapepyco xpwuevoc, xfj ... e<f>qyr|aei [60b.1.8]). In the sections in which he compiles short passages from a variety of authors; Antigonus is willing to apply to poetry for evidence or support, and in these cases the poet may be evaluated for his precision, accuracy, or attention to detail. So Homer, quoted as an authority on the behavior of dogs, is introduced as being ixavtog ... empeXqc, Kai TToXuTTpdypiov (24.1). Philetas too, in giving evidence for the spontaneous generation of bees, is iKavtoc, tov Trepiepyoc, (19.2). Antigonus is less accepting of paradoxical material 89 which takes the form of uv30og, that is, a story pattern with a beginning, middle, and end. A critical term such as TEpaTioSric, or UUG IKOOC , is generally applied to such material; thus sanitized, it can be compiled. When, for example, Amelesagoras' Atthis is drawn upon for the information that a crow will not fly up to the Athenian Acropolis, Antigonus compiles the story of the birth of Erichthonius, to which is appended an arnov about a crow that becomes an avis non grata upon announcing to Athena the birth of the infant Erichthonius and is thus denied access to her precinct. ' AueAnoayopac, ... ajroSiScoaiv 5e T p v ahicxv UUG IKCOC , , Antigonus insists (12.2). The paradoxographer can thus make his pretension to sober accuracy and zeal at the same time that he adds variety to his material. But not all such data were admissible into the collection. Antigonus informs us that he might have compiled some material about ravens from Ctesias, but was compelled by his own standards to reject it: KTnoiac, iaropeT TrapajrXfjoiov T I T O U T O I C J [an excerpt from Theopompus] 5id be T O C X U T O V jroXXa ipeu5eo0ai TrapeXriTrouev Tf|v eKXoytjv. Kai yap ecbaiveTO TepaTibSr|c,. Ctesias gives information about something very similar to this. But because he is given to telling many lies, we left it out of our compilation. For it seemed fantastical and strange. (15b 1) As a responsible scholar working with a great range of written texts, Antigonus sees his contribution to this process in the diligence and accuracy with which he carries out his compilation. He makes similar demands of his source authors. Thus Antigonus praises Aristotle's EmueXria, and points out that the poets he quotes are emueXfjc;, TToXuTrpdyptov, and Trepiepyoc,.34 3 4 jToXujTpaYUoouvn. ropiepYia, and their derivatives and synonyms, terms which are primarily political and ethical for the fifth- and fourth-century Greeks, for Hellenistic and Imperial scholars describe ideal research practices; cf Ehrenberg 1947: 46-62. Antigonus' near-contemporary Polybius, for example, draws upon similar terminology to describe Timaeus' literary research: £K T I O V (3u(3Xicov 8 U V O C T O U 90 Antigonus also shows a concern for selectivity. Material must be a£\a dxofjc; (Mir. 129). This formula and its variants — d£ia 07rou5fjc; and Xoyou, atya pvf|pn.c, and GaupocToc,, a£ia uaGetv and avaypdtheiv — are used frequently by later authors in presenting material acquired through selective compilation. Aware that he reveals his own scholarly ability through the quality of the material he selects for his collection, the compilator tends to attach the name of an authority to his information. Likewise he will quote his source's authority when compiling a secondary compilation; Antigonus, for example, quoting Callimachus, also quotes Callimachus' sources. A stage will arrive, clearly apparent in Aelian's work, when the secondary source's name has been dropped, the compilator being satisfied with citing the name of the primary source alone (Fraser 1972: 772). rioXuuaGia and Imperial Education Literacy had its first impact upon Greek education in providing archives for school authors. Peripatetic and Hellenistic activity created an archival system of primary texts of paideia and of secondary scholarship explicating it, both supplemented by a diverse body of empirical data. All of these texts formed paideia for the Imperial period. To which aspects of this system can the term TroXupaGia be applied? That is, to which areas of paideia does the miscellanist address himself? Briefly, for Imperial authors TroXupaGia represents that area of the traditional body of paideia which forms a significant and relevant supplement to the texts which have now 7ToAu7TpaYuov£Ta6ai x^P'C K I V 5 U V O U Kai KaKOTtaOsiag sav ng auio T O U T O 7Tpovor|8n uovov wore Xa(3siv f\ rtoXiv sxouaav ojrouvnucVrcov 7iXf]0og r\ (3u(3Aio0r)Kr|v T T O U vEiTvuSaav One is able to carry out intensive research in books quite free from danger and discomfort if one only takes care beforehand to have access to a city with an abundance of texts or a neighboring library. (12.17.4-5) 91 become standard school-authors. JJoXouaGia still applies to the educational process, but now more specifically to the area of self-directed adult reading. Unlike Plato, Socrates' Hippias, and Isocrates, who used TroXuuaGia in a context of primary and secondary skills-acquisition and the memorization of traditional poetry, Imperial authors do not apply the word to children or adolescents still in the process of acquiring a basic literary education. They use it rather in reference to the educated adult who has chosen to extend his paideia-related data-base beyond the ordinary to the remarkable. How does the polymathic data-base function as a supplement to Imperial education? Here we must consider separately both the primary education of the child and adolescent, and the more advanced studies of the young adult. noXouaGia and rpauuaTucf) Since the end of the fifth century BC, a child's basic education depended upon literacy acquisition; but the basis of education continued to be, as it had been from the beginnings of Greek culture, traditional literature. The skills of reading and writing constituted ypauuaTucri, instruction in which began about the age of seven.36 The quality Gellius' negative reaction to 7toXuua0ioc is specifically occasioned by irelevancy. The data he criticizes suffer from taedium et senium. (Prol. 11) 3 6 Most analyses of Imperial education (eg Marrou 1948, Clark 1957, Christes 1975: 228ff, Beck 1970) make fairly specific statements about children's ages and "grades." Steinmetz, for example, citing Marrou, Ziebarth, and others, assigns the seven- to twelve-year-olds to a YPaMMoriGTTig, the thirteen- to eighteen-year-olds to a YPotuuomicoc.. But cf. Kaster (1983), who demonstrates that this division was very fluid, sometimes nonexistent, between YPauucmarrig and YPotuuomKoc.. Marrou recognized a varying degree of applicability of these terms, according to time period and location (1948: 223-224). By "primary education" I mean in the present study all formal education in YPapuomKf], probably normally limited to the under-eighteens, and to whom Dionysius Thrax directs the content of his manual. The scholiast to Dionysius Thrax (Gramm. Graec. 3.164) had specified the two parts of grammar as jraXai and vEtorepa, the former having as its primary goal ev ypafyew (not our expository writing but rather orthography [Quintilian 1.7, Dio 18.18]), the latter ev dvaYiYvwaxsiv or explication, a younger science. Cousin may be correct in assessing these as two components reflecting a change in the way the grammarian views himself (1975: 9); from being the pedagogue concerned with the teaching of 92 and extent of training which the child received in ypauuaTucrj was determined by the parent. We may accept these early years of training as the formative period of literacy acquisition and of an introduction to the basic paideia-authors. Secondary education in rhetoric, medicine, law, the more amorphous "philosophy," in fact any organized technical discipline — "das Hochschulstudium im eigentlichen Sinn" (Steinmetz 1982: 80) — began after the individual attained his majority (though the parent may still be financing his education), and involved the training not of children but of young adults. Although YpauucxTucfi is the primary Texvri, during the Imperial period it demands a data-base. The Texvri of Dionysius Thrax, composed at the end of the Hellenistic period, defines the province of the art in such a way that the need for secondary scholarship is clear: ypauuceriKri e a n v eu;reipia T W V rrapd rroiriTaic; T E Kai auvypatbeuorv OJCJ em T O TTOAI) Xeyouevojv. Grammar is the skill which deals with the things said for the most part by poets and prose writers. (1.1) Reading and writing have as their primary field of activity the works of traditional paideia. This represents little change from the school curriculum of Plato. But when Dionysius proceeds to define grammar's parts we see clearly the point of access for philology and all the fields into which ioropia has been introduced. TrpojTov dvayvojoTCj evTpipficj K a r a 7rpoaop8iav, 8euTepov e^riynaic, K a r a T O U C , evujrdpxovrac, TTOITITIKOUCJ T P O T O U C J , TpiTOv yXojaaojv T E Kai ioropiojv jrpoxeipoc, d7r68oaic„ TerapTov eTuuoXoyiac, eupecnc,, jreunrov dvaXoyiac; eKXoyiauoc,, E K T O V Kpiaic, TromudTOJV, 6 8rj KCtXXiorov e o n jrdvrojv TOJV ev rfj Texvn. 1. Skillful reading in meter 2. Detailed explanation in terms of the predominant poetic devices reading and writing to very young children, he has expanded his horizons to include the teaching of all materials which elucidate text-based paideia. 93 3. Ability to explain difficult expressions and to give background information 4. Tracing etymologies 5. Ability to explain morphological analogy 6. An evaluation of creative literature; this is the finest of all the techniques in the art of grammar. (1.1.5-1.1.6) Dionysius' grammarian directly addresses a text in the first and fifth items on this agenda. The student must be able to understand and correctly to enunciate his poets, an increasingly difficult task as each generation's colloquial Greek grows farther removed from the archaic and classical poetic languages of his texts (cfiBrowning 1969:44-52). But for an exegesis of the poetic tropes, difficult vocabulary items and phraseology, and obscure references ( ioropiai) in the text, secondary information becomes necessary. The uTrouvrjuaxa, as well as various other scholarly works produced by Hellenistic researchers, supply such a data-base for YP«MM«TtK'1- m these introductory lines Dionysius has formulated the purpose for which grammar, a socially relevant skill, can and must access the paideia (primary texts and secondary studies) archived and created by the Hellenistic scholars associated with Alexandria, Pergamum, and other library facilities. Grammar both defines these scholars' work and becomes its reason for being. Dionysius' definition of the province of ypcxuuomKri remained standard in Imperial schools (Cicero De or. 8.187; Quintilian 1.4.2-3). Building upon Dionysius' theoretical base, Quintilian in the early books of the Institutio oratorio provides a manual of what must and must not be actually taught in the classroom on a daily basis (Cousin 1935: 53, 73; von Fritz 1949: 337; Cousin 1975: 6-7). The education in grammar which Quintilian describes begins at the nearly rudimentary stage. Although he assumes that the pupil has learned to recognize the letters 94 of the alphabet and to handle a stylus, he offers advice on the basics of reading and writing. For Quintilian, the grammaticus is both the elementary-school's language-arts instructor and the more philologically sophisticated interpreter of literature (cf Kaster 1988: 447-452). Quintilian's grammaticus places at the center of his professional activity the written text, not only expounding that text's contents by first "emending" the written document itself but standardizing the pupil's language performances both oral and written by judging them against the central text as a model of performance. For such a system of education, the text has become not, as Plutarch claimed, an opyavov T i a i b d a c , , but in fact paideia itself. Once the child had acquired the ability to read and to write with some fluency, the grammaticus could devote classroom time to the enarratio poetarum, the study of literature. Here Quintilian demonstrates that, during the years of childhood and early-adolescent education, it was the grammarian who channelled TroXupaGia into the learning process. The grammarian's ability to discriminate the relevant from the pedantic, otiose and morally offensive determined his excellence in imparting paideia. According to Quintilian, the classroom analysis of a literary passage involved two processes, one involving systematic presentation of the rules of language and one focusing upon literary content: pe0o5n<f) and iaTopiKf|. Quintilian details the range of the data-base required of the grammarian in describing the "methodic" process of praelectio: deprendat quae barbara, quae impropria, quae contra legem loquendi sint .... id quoque inter prima rudimenta non inutile demonstrare, quot quaeque verba modis intelligenda sint. circa glossemata etiam, id est voces minus usitatas, non ultima eius professionis diligentia est. enimvero iam maiore cura doceat tropos omnes ... praecipue vero ilia infigat animis, quae in oeconomia virtus, quae in decore rerum, quid personae cuique convenerit, quid 95 in sensibus laudandum, quid in verbis, ubi copia probabilis, ubi modus. He must point out those terms which are wrong, improper, and contrary to language rules .... It is very useful at the beginning to point out in how many ways certain words are to be understood. Peculiar expressions should receive plenty of attention right away. He must be very careful with all tropes. He should especially emphasize to -the children the excellence in brevity, in ornament, in propriety, and what is praiseworthy in ideas and diction; and where abundance, where restraint of illustrative material is worthy of note. (1.8.14-17) The grammarian has to be able to deal with metrics, lexicographical detail, the different demands of poetic and prose usage and syntax, and some general concepts of the critical evaluation of literature (cf Degenhardt 1909: 7-60, Marrou 1948: 375). Though much learning is required to meet these demands, it consists of the sort of standard material found, condensed and abstracted, in the various manuals written by grammarians and becoming increasingly available during the Imperial period. This professional knowledge is readily compiled and generously shared, viewed not as original intellectual property but as the revered contents of a tradition. Charisius, for example, describes his own Ars grammatica as sollertia doctissimorum virorum polita et a me digesta (GL 1.1), taking credit chiefly, it would seem, for the arrangement of material already "polished" by earlier scholars. Phocas too explains and defends his utilization of earlier works: [in hoc opere] nihil mihi sumam nec a me novi quicquam repertum adfirmabo. multa namque ex multorum libris decerpta concinna brevitate conclusi, ut nec ieiuna parum instruat condensata nec verbosa prolixitas fastidium legentibus moveat. In this work I shall neither claim anything as my own nor make any new additions. I have included herein much matter excerpted from many authors' books with seemly concision, in such a way that neither a bare and succinct content gives too little instruction nor a wordy overgrowth dismays my readers. 96 Phocas values his compiled sources because they have allowed him to excerpt effectively (concitma brevitate) only the really relevant and clearly stated matter. He feels no personal need to reformulate what they have already stated clearly. Cledonius judges his sources similarly: De diversis veteribus aptos huic operi sumpsi tractatos et his mea quoque ut potui et quae potui ... copulavi ablatisque limitibus campo piano dispersi, ut inoffensibili cursu fructus sibi lector colligat maturatos usu. / have accessed from various old authors passages which befit this present work and have added what I could of my own material to these as I was able. I have smoothed out the divisions and have arranged it all in an open field, as it were, in order that the reader might harvest for himself with unhindered passage these fruits which have ripened through use. (GL 5.9) The TroXupaGia which concerns the Imperial miscellanist is of a different order. This kind of learning is more like that described by Quintilian as necessary for the second, "historical," portion of the enarratio poetarum. Now the grammarian must provide background material to the characters, situations, mythic allusion, geographical points, social and chronological relationships, authorial biography, in short to the imaginative context of the reading selection. At this point the grammarian must be able to access a wide variety of materials, in a range which allows us to describe these data as TroXupaGia. It is in this "historical" portion of classroom work that the need for discrimination in selection becomes clear. Quintilian specifies that the grammarian's enarratio historiarum be diligens quidem ... non tamen usque ad supervacuum laborem occupata; nam receptas aut certe Claris auctoribus memoratas exposuisse satis est. Carefully detailed, certainly, but still not busy to the point of being filled with superfluous detail. It is enough to explain the "histories" related by the famous authors. (1.8.18) 97 A few lines later we are sententiously informed that mihi inter virtutes grammatici habebitur aliqua nescire it seems to me that not knowing some things belongs among a grammarian's excellent qualities. No such restriction had been placed upon the methodic prae lectio Essentially, the Imperial grammarian must beware of fleshing out literature's imaginative universe into one filled too full of distracting detail — that is, into a world too like the quotidian. Although the mature amateur scholar (cf Ch. 3 below) has the leisure, distance, and resources to approach a poetic text diachronically, the grammarian and his young pupils experience the text's imaginative world immediately, synchronically.37 In the cases of Homer and Vergil, the temptation, or the burden, of metonyinically "filling in" details will have always been present, requiring careful if selective erudition on the part of the diligens grammaticus but also restraint, balance, and taste.38 Like the Imperial artes of grammar, handbooks appear during this period which answer the need for relevant historiae. The Genealogiae of Hyginus, for example, presented categorized lists of mythological and legendary figures and events, while [Apollodorus]' Bibliothecae of mythological data arranged a wealth of legends and folklore into a chronological narrative. Other texts provided a mixture of material suitable for classroom explication. Indeed, one advantage of a commentary like Servius' on the Aeneid was that it could balance for the grammarian lexical, metrical, and purely Especially considering that Antiquity did not provide literary texts geared to the schoolchild's cognitive development; one is reminded of Rousseau's stressful encounter with the imaginative world of La Fontaine. 3 8 We can perhaps attribute at least part of Menander's and Terence's popularity in the classroom to the fact that, as realistic New Comedy authors, they do not require much historical, geographical, and mythical funding on the young reader's part. 98 grammatical material with the potentially overwhelming historiae, providing a premixed "formula" of erudition for young students. By selecting and compiling only those excerpts from scholars and clari auctores which experience had shown were most useful and relevant to a given reading passage, a good commentary could help the grammarian avoid supervacuus labor and could thus function as a real teaching text. Of these supplementary commentaries, of which the Hellenistic u7Touvf|uaTa formed precursors and sources, the most useful parts would slip into the margins of codices and, like the Homeric scholia, be replicated along with the primary text. That the Hellenistic and Imperial commentaries were honed down over the centuries to the form of marginal scholia reveals the limitations put upon 7roXuua0ia by an elementary education in ypauuaTncrj. The Imperial grammar school which taught the child language, literacy, literature and its encoded value system did not expect him to become a professional scribe or, for that matter, a grammarian.39 Hence Seneca emphasizes that the . study of literature must be propaedeutic only: non discere debemus ista, sed didicisse these are things which we ought to have learned [as children], not to spend time now in learning (Ep. 88.2). Plutarch reiterates the grammar school's lack of connection with the hard daily realities of pMocj, 7rpdc;vcj, and TroXireia at De audiendo 42. The Imperial grammarian, as Kaster has pointed out (1988: 205) limits himself to this propaedeutic though seductively manifold world of literature. His professional The distinction between schools teaching literature as the basic art to children of the educated elite and those teaching craft literacy to the lower classes, has been given considerable discussion in recent scholarship. Booth, for example, has maintained that such schools existed in first-century Rome, the Ypauuomorric being the teacher of the lower social classes and the Ypauuomicoc the teacher of the children of the elite; while Kaster insists that a similar arrangement can be assumed for later centuries as well (Kaster and Booth ap. Kaster 1983: 339-346). 99 attention is focused upon the presentation and explication of standard texts, year after year, to children who may well outshine him in their acquisition of a broader education and in their wider horizons. Within the schoolroom, the grammarian's competence is, probably rightly, unchallenged. Outside the school, exposed to the real world, he is more vulnerable. Gellius' miscellany abounds in anecdotes in which the grammarian is seriously nonplused in questions of language and literature which demand that he function beyond the canon, by men whose opportunities have allowed them greater intellectual scope. The grammarian does not have even the comfort of being accepted as one whose craft is applicable to society as a whole; Sextus Empiricus (Math. 1.97-98) maliciously reminds him that his subject matter consists entirely of lies. Little wonder that the intelligent grammarian funds that tissue of lies with detail where he can, and by the acquisition of systematized information about it — TroXupaGia — gains intellectual power through the control of that detail. In the eyes of the wider public this activity will indeed seem trivial and supervacua. Seneca depicts the Alexandrian scholar Didymus as a sort of arch-grammarian: Quattuor milia librorum Didymus grammaticus scripsit. miserer, si tarn multi supervacua legisset. In his libris de patria Homeri quaeritur, in his de Aenaee matre vera, in his libidinosior Anacreon an ebriosior vixerit, in his an Sappho publica fuerit, et alia, quae erant didiscenda, si scires. Didymus the grammaticus wrote four thousand books. I would feel sorry for him if he had had to read so many superfluous works. In these books he researchs the fatherland of Homer, Aeneas' real mother, the question as to whether Anacreon was more lustful or more intemperate, whether or not Sappho was a prostitute,, and other matters which would have to be unlearned if you knew them already.(Ep. 88.37) These subjects are to be rejected because they are irrelevant to life in the real world, and a waste of time; as far as Seneca is concerned neither pupil nor teacher is improved by 100 discussing them. The grammarian thus must focus upon relevancy with morally paedeutic material in his text. He must select. Suetonius mentions several Imperial grammarians who committed their TroXuuocOioc to writing. Aurelius, for example, composed many volumina variae eruditionis (Gramm. 6), while C. Julius Hyginus' learning earned him not only the cognomen Polyhistor but also an appointment as director of the Palatine Library (Gramm. 20). L. Ateius Philologus left eight hundred books of compiled data. Gaius Melissus' collection of assorted compilations amounted to one hundred and fifty papyrus rolls. If these men published their polymathic collections with schoolchildren in mind, we can well imagine the need Quintilian and Seneca felt to speak out against such pedantry. But were these works, some of the titles of which sound very much like miscellanies (Ateius' works was entitled "YXn, Melissus' Ineptiae), in fact addressed to the classroom? The possibility that Suetonius' grammarians were directing their collections as paideia-supplements to a different audience is suggested by his description of the grammarian Valerius Probus. Compared with the works of Hyginus, Ateius, and Melissus, Probus' published scholarship is not impressive: a few short studies, carefully researched.40 Kaster describes Probus as a "literary guru" who self-consciously set himself apart from the ordinary professional grammarian: Probus had followers (rather than pupils), three or four of whom he would admit to his home of an afternoon (not meet in larger groups, in a classroom, in the morning), where he would recline (not sit in a teacher's cathedra) and hold conversations (not deliver lectures). It is the picture of an intimate Probus seems to have focused most of his writing upon textual criticism: multaque exemplaria contracta emendare ac distinguere at annotare curavit, soli huic nec ulli praeterea grammatices parti deditus (Suetonius Gramm. 24). 101 and elite coterie, gentlemen meeting in an aristocracy of letters (1984. 54). Probus purposely removed himself from the enarratio poetarum and instead pursued the study of literature as an adult's leisure activity, yet Suetonius insists all the same that Probus was in fact a grammarian. Implicit in Probus' activity is a view of a social role of paideia extending beyond the classroom but yet more erudite (i.e. polymathic) than the public paideia-performance associated with drama and declamation. Suetonius' description of Probus' associates as sectatores prevents their being seen as friends or colleagues; they are in fact his graduate students (cf. Gellius NA 2.2.2; 3.1.5; 13.5.2; 19.51).41 - . It is implied here that the elementary study of literature is meant not only to have an impact upon adulthood, but to extend in some form into adulthood as well. How is this impact assessed? If the grammarian has avoided overwhelming the child with too much detailed erudition during the early years of reading the canonic authors, how does such a literary data-base concern the educated adult? Polymathic scholarship has been accumulating in textual form, as we have seen, since the library foundations of the fourth and third centuries BC. How does it relate to the educated adult's experiences with literary culture? noXuuaBia and Post-Primary Education Grammatical education was not standardized in antiquity. There was no evaluative testing to determine the completion of a preformulated curriculum. Dionysius Thrax set as the goal of ypappariKri the correct xpiatc; of literature, the ability to judge the worth of 41 "The bond is different from and more intimate than the exchange of cash for learning" (Kaster 1988: 59). 102 a piece of literature, to see its relevance to paideia. Presumably the person fully educated in YpauucrnKf] could demonstrate satisfactorily this ability to assess literature's paedeutic value, and to integrate its relevant qualities into his daily life. In determining the manner in which the educated adult used his primary grammatical studies in his mature approach to paideia, two aspects of this elementary education in literature must still be considered: the extent to which other artes were incorporated into the study of literature, and the moralistic component of the elementary study of literature. I shall argue that society's expectation of certain basic skills and of standardized moral responses in the adult with a literary education encouraged a sense of group identity, of a cultural community with new demands made upon its members. rioAuuaGia will be among those new demands. An education in ypauuaTiKf) ideally imparted all the skills detailed by Dionysius Thrax: the ability correctly to read, understand, explicate, and assess the traditional curriculum-authors. The repetition of grammar's province by successive authors from Cicero to Martianus Capella indicates that-these were at least conceptualized as the component skills of grammar. From the earliest references to its content, Greek education had also included gymnastic and music; and during the Hellenistic period, training in other skills became increasingly available (Marrou 1948: 244). These came to be known as the E V K U K A I O C J jroaSria, the "common" or "standard" education. Elle demeura toujours une notion aux contours assez vagues: l'usage qu'on eu fait hesite entre deux conceptions: c'est tantot la culture generate qui convient a l'honnete homme, sans reference explicite a l'enseignement et qui reunit l'apport de toute l'education secondaire et superieure scolaire et personelle; c'est d'autres fois la 103 culture de base, la propedeutique: les 7tpo7rai5ei3uaTa, qui doivent preparer l'esprit a recevoir les formes superieures de 1'enseignement et de la culture (Marrou ibid.)42 Quintilian suggested that other artes might be necessary for the correct teaching of YpauucxTucfi. neque citra musicen grammatice potest esse perfecta ... nec si rationem siderum ignoret poetas intellegat ... nec ignara philosophiae. Grammar cannot reach completion without music; it can only make sense of the poets if it possesses a knowledge of astronomy; it cannot be ignorant of philosophy. (1.4.4) To the conclusion of his grammatical expose in Book 1, Quintilian added a discussion an oratori futuro necessaria sit plurima artium scientia.(\.\0). He answered affirmatively, that the perfect orator must be familiar with the encyclic paideia, and was to begin the study of the artes during these early years. Quintilian did not however give detailed instructions on how or when they were to be incorporated into the basic grammatico-rhetorical curriculum. They are certainly not to be pursued during the time consecrated to grammar, but to tempora velut sitbseciva (1.12.13; cf Marrou 1948: 378). Not to have this basic, "encyclic" acquaintance with the mathematical artes43 drew critical comment. Theon, with a passing nod to Aristotle (cf note 41 above), complains of ill-prepared rhetoricians who begin public speaking careers " Grammar's propaedeutic status is emphasized by theorists. It provides both a finite skill and a means of acquiring the other arts: Ille ... per quam pueris elementa traduntur, non docet artes sed mox percipiendas locum parat. That art through which the alphabet is taught to ch ildren does not teach the liberal arts but instead prepares the place for the arts which are soon to be acquired. (Seneca Ep. 8.20) The concept of literacy as propaedeutic goes back at least to Aristotle: en 8e rat T U > V xpnciuwv o n 5ei nvot J T O U S E U O E O O O U T O U C 7iai8ac ov> uovov 5ta T O xpiioiuov, otov niv T C O V Ypauu&Tcov uaOnatv, aXXa K a i 5ia T O noXXdc ev5exea9ai yiveaOai 61' auTiov uaOrjaetc ETepac T O 5e Cnretv navTaxou T O xpiicnuov f )Kiora a p u o r r s t T O I C usyaXoipuxoic K a i T O I C sXeuOepioic. We are also entitled to say that the reason why some of the useful subjects ought to be taught to children — for example, reading and writing — is not only the fact of their being useful: it is also the fact that they make it possible to acquire many other branches of knowledge .... to aim at utility everywhere is utterly unbecoming to high-minded and liberal spirits. (Pol. 1338a-1338b: translated Barker 1946: 337) 104 rcpiv duto<Tye7Tcog aipao6at (JnXooocpiac, Kai Tfjc, E K E I G E V Ep7rXqo"0fivai peyaXovoiac, • vuv 8E o i TrXsiouc, T O O O U T O V 8 EOUOI T I O V T O I O U T I O V Xoyiov EjraiEiv, I O O T E ou5e TCOV E Y K U K X I C O V xaXoupeviov pa6n.pdTiov O T I O C V uETaXaupdvovxeg ctTTouatv em T O Xeyeiv. Before undertaking in some manner or other the study of philosophy and imbibing the magnanimity to be derived from that source. Nowadays the majority are so ignorant of these basic concepts that they rush into oratory without even a basic grasp of the so-called encyclic subjects. Plutarch, in discussing the raising of children, brings up encyclic education after he has given his views on proper rhetorical training. 5 E I Toivuv T O V 7ral5a T O V E X E I 3 6 E P O V pn5Ev6c, Ttov aXXtov Ttov KaXoupEviov E V K U K X I I O V 7rai5EupdTiov pf |T dvf)KOOv pnr' dGsaTov edv eivai. TauTa pev C K 7Tapa8popfjg pa0e!v tboTrepei yeupaTOC, C V C K E V . The freeborn child must in no way be allowed to be ignorant of unfamiliar with, and uninstructed in the so-called encyclic arts; he should learn them superficially, as though getting just a taste. (De lib. educ. 7d) The lack of a controlled curriculum and the extent to which the individual or his parent could personally select a program of study from a range of desirable artes are clear. Care, forethought, and selectivity must be exercised in such a situation, as the finished product, the fully educated adult, will prove the value of his curricular choice in his daily life as an adult. If the first of society's expectations of an education in YpappaTucf) was a thorough grounding in the canonic authors, some level of acquaintance with the other artes of the E V K U K X I O C , jrai5£ia was the second. The third expectation was of a different order: a standardized, moralistic response to ethical issues contained in the curricular texts. As literature had always been the vehicle for the transmission of Greek cultural values, it is scarcely surprising that Imperial grammarians would devote considerable time I.e. music, geometry, arithmetic, astronomy. During the Imperial period the list varied with the author discussing it; cf Bompaire 1958: 95. 105 to the moral and ethical models which their texts conveyed.44 Recognizing such models formed part of the Kpioic, of literature which the educated person should have mastered (Marrou 1948: 234-235; Raster 1988: 27-28). Curricular authors provided models for language and thought (Bompaire 1958: passim). But they also conveyed to the student patterns of behavior which contributed to his moral development. Such models could function on the level of correct linguistic usage, with the text providing effective (i.e. correct, legitimate, proper) words and expressions. Since education in grammar dealt with language in its written mode, the text would be the primary source for models to imitate, in this case models of effective words and expressions. Quintilian sorted out these functions thus: Sermo constat ratione, vetustate, auctoriate, consuetudine, rationem praestat praecipue analogia, nonumquam et etymologia, vetera maiestas quaedam .... auctoritas ab oratoribus vel historicis peti solet... consuetudinem sermonis vocabo consensum eruditorum. Language is determined by reason, convention, authority, usage. Analogy especially and etymology frequently explain the rational portion of language. A certain archaic elegance, an authority, is usually sought from the orators and the historians (1.6.1-2). I shall determine usage through the consensus of the learned (1.6.45). Quintilian is discussing primarily diction in this passage. Of correct diction's four sources, three are located in the canonic authors. Analogy and etymology are simply ways of defending or criticizing diction found within a- text and analyzing it according to a norm which the child can understand (von Fritz 1949: 349-351). Vetustas implies the charming dignity of archaic words, the maiestatem ... non sine delectatione (1.6.39) which old words can contribute to discourse. We learn of this attraction from Gellius and Fronto too "L'etude grammatical ... sa finalite serait... d'ordre moral, et en cela le grammarien ... reste bien dans la ligne de la vieille tradition, a la recherche, dans ces annales du passe, d'exemples heroiques de ape-ni." Marrou 1948: 234. 106 (e.g. NA 13.21. 13-14; 19.7.12). Again, the text is both source and model for such terms. Consuetudo, usage, alone rests upon imitation of the spoken language, nam fuerit paene ridiculum malle sermonem quo locuti sint homines quam quo loquantur it would be almost risible to prefer outdated to current language (Inst, or at. 1.6.43). But here Quintilian curiously hedges his assertion: not everyone' usage is to be imitated, but only that of the eruditi. The many's linguistic models will not do, any more than will its moral examples, a periculosissimumpraeceptum indeed (ibid). So in the end consuetudo too depends upon the imitation of the written text, for upon this the eruditi have modeled their usage. Quintilian goes on to conjecture a second way in which the grammar school's core texts can provide models for imitation. Not only do they provide correct language, but an inspiring moral tone as well: et sublimitate heroi carminis animus adsurgat et ex magnitude rerum spiritum ducat. Let the heart swell with the sublimity of heroic song, and draw its breath from the greatness of the enterprise. (1.8.5). Yet the grammarian must also be warned to be careful of a text's possible danger to the students' moral development. Although the canonic school authors had undergone centuries of selection for educative purposes, they still represented high adult art. Antiquity did not "write down" to its children's level of comprehension. Quintilian expresses the need to censor some texts to make them fit for the classroom, while other Greats must simply be omitted from the curriculum or at least postponed ad firmius aetatis robur (1.8.6) because of an indecent or lascivious component unsuited to the pre-adolescent. Plutarch worries that the texts themselves corrupt: poetry, the main diet of the schoolchild, is a fabric of lies; in fact, its charm lies in being ev nevckeyuevr] SiaGeotc, pu0oXoyiag, O U K i'apev 8' apu0ov ov>8' dipeuSfj 7roincav. 107 a finely woven tissue of fable; we do not know of a poem which is not fabulous, which is not false. (Quomodo poet. aud. 16b) Consequently, it must be handled with intelligence and respect in the classroom: 7Toir|TiKfj rroAu uev f)5u Kai rpocbiuov veou ipuxfjcj eveortv, O U K eAarrov 5e T O TapaKTucov Kai irapdcbpovov, av uf| Tuyxdvri TraiSaycoyiac; opGfjc, f| ocKpoacnc,. There is in poetry much sweet and nourishing for the soul of the child, but there is also no less an element of the upsetting and the seductive if instruction is not attended by the right focus. (Quomodo poet. aud. 15c) Quintilian, more pragmatic, states simply non modo quae diserta sed vel magis quae honesta sunt discant let them learn not only what is eloquently expressed but even more what is morally sound. (1.8.5) If the Imperial grammarian needed more precise instruction on how and when to incorporate moral relevance into his enarratio, he could find it in various commentaries and u7TouvtjuaTa to his texts. Degenhardt has collected a number of scholia containing brief moralistic summaries of a variety of passages from Homer, Hesiod, Pindar, Sophocles, and Terence. In the Quomodo poetae audiendi, besides the cautionary comments quoted above, Plutarch offers some concrete positive advice about scouting out opportunities in a text for introducing moral discussions.45 When reading the Iliad, for example, the child's attention should be drawn to the fact that Achilles, though wrongly impelled to attack and kill Agamemnon (oure rrpocj T O KaXov 6p0ooc, O U T E jrpocj T O auucbepov), yet rightly restrained his anger (opGcocj raxXtv Kai KaXwc, 26d). Diomedes, a 4 5 28d: ETisi 8' worrep E V O C U T T E X O U c])uXXoic Kai KXrjuaoiv euOaXoucn T T O X X & K I C 6 Kapjroc a7TOKpu7TTETai Kai XavOdvei KaTaaKiaCouEvoc, O U T C O C E V jroiriTiKn Xel;ei Kai uuOsuuaai T T E P I K E X U U E V O I C noXXd Siac^EuyEi T O V V E O V ib(f>sXiua Kai xpnoiM°<. $ E I 5S T O O T O ur) jrdaxeiv un5 ' aTTOTrXavdoOai T C O V TTpaY d^rcov, dXX ' eu(|>u£a0ai uaXiora T O I C Tipoc dpeTiiv (J)£poucn Kai Suvausvoig TtXdTTEtv T O f|6oc. As when fruit lies hidden, overshadowed by the leaves and branches of the vine, even so many beneficial and useful points escape the child as he reads poetry because they are overshadowed by the words and the fantasies. He must not miss them, and he must not wander from the subject, but must cleave especially close to those elements which are conducive to excellence and which shape his character. 108 model of discreet discourse, refrained from using free speech to his commanding officer in public, but spoke out manfully to him in private: x] yap ToiauTn 5iacj)opd pf] Trapopcopevri 8i8d£ei T O V veov doretov riyeloQai TT)V dTU(}>iav Kai peTpioTnTa, T T | V 8e p e y a X a u x i a v K c a 7repiauToXoyiav ibc, (haOXov euXaPeioOai. If such a detail is brought to the child's attention, it will teach him to consider moderation and freedom from arrogance to be proper, but to avoid conceit and boastfulness as base. (29b) When ambiguity arises in a text, the child's response must be correctly directed toward the proper interpretation: Nausicaa's assertive manner of speaking to Odysseus, for example, must be interpreted in such a way as to avoid the conclusion that she is modelling behavior unacceptable in a young girl (27a; and cf. Basil of Caesarea De legendis gentdium libris 7-10). Thus literature which the twentieth century sees as too aesthetically complex to warrant such Aesop-like moralizing, for the Imperial period channels values and behavioral models to succeeding generation of grammar-school students. During the Imperial period, then, the young person who had completed his education in ypappaTncri was expected not only to have studied the canonic authors, but to have picked up information from a number of supplementary fields of study as well, and to have developed a conventional, moralistic way of reacting to his school texts. Imperial authors may view such a literary education not as a preparatory training period in the processes necessary for carrying on adult activities, but rather as an initiation into a state of enculturation. It opened a door to an imagined, idealized past, viewed both as the emotional heartland from which the present had moved away, and as a source from which the reader was encouraged to draw moral foils and exempla relevant to his own life. Paideia thus viewed offered the educated person not an escape from the realities of his 109 daily life but an identity with the past, an identity which allowed him to reinterpret the present and thus more effectively to participate in it. In Quintilian, an education in ypappaTucfi is applicable to all stages of life, and in itself immensely satisfying: necessaria pueris, iucunda senibus, dulcis secretorum comes a requirement for children, a pleasure for seniors, a sweet companion in retirement. (1.4.5).46 Paideia was a valuable, lifelong possession, man's best treasure and the element which marked him as truly human: euyeveia KOCXOV pev, aXka Trpoyoviov dyaGov. T T A O U T O C , 5e riptov pev, aXka ruxnc, Krfjpa ... S6ca ye pqv oepvov pev, dXX' dpe|3aiov. KaXkoq be TrepipdxnTov pev, aXk' oXvyoxpoviov. uyieia 6e ripiov pev, oikX eupeT&aTCCTOv. ioxuc, 8e CnXiordv pev, aXka vooip eudXcoTov xai yr]pa .... ;rai5eia be TCOV ev fiplv povov eoriv dBdvarov Kai Gevov. High birth is a fine and good thing, but depends upon our ancestors. Wealth is valued, but it is in the hands of fortune. Reputation is fine but insecure. Beauty is admired by many but lasts only a short time. Good health is precious but transitory. Strength is longed for but easily removed by sickness and age. Paideia alone of our possessions is' a thing both immortal and divine. (Plutarch De lib. educ. 5d) But the study is not to be tainted with worldly concerns: ne velim quidem lectorem dari mihi, Quintilian insists, quid studia refer ant, computaturum. I would surely not want as my reader the kind of person who would reckon up what an education is worth (1.12.16; and cf Aristotle Pol. 1338M). Quintilian is echoing an earlier statement of Cicero: -Haec studia adulescentiam agunt, senectutem oblectant, secundas res ornant, adversis perfugium ac solacium praebent. Delectant domi, non impediunt foris, pemOCtant nobiscum, perigrinantur, niSticantUr. These studies shape us in youth, delight us in age, are ornaments at the dinner table, offer refuge and solace in times of trouble, please us at home, are handy to travel with; they share our insomnia, our absences from home, our days in the country. (Pro Archia 16) 110 Thus conceived, paideia is a shibboleth, a union ticket, and a lifelong commitment.47 Idealized paideia on the adult level is regularly imaged in one of two ways: the ascesis of the athlete and the ineffable state of one initiated into the Mysteries. The use of imagery drawn from athletic training to illustrate the purposive focus of paideia acquisition occurs frequently in Imperial authors' discussions of education. Galen, for example, conceived of the arts of the E V K U K I O C , raxiSsia as exercise for the soul, a means of strengthening its various functions as though they were so many muscles and sinews. yuuvat'oueBa yap irptora pev UTTO TOIC, ypappaTncoTc, E T I ncfibec, O V T E C ; , EI6' ECTJC, jrapa T E TOXC. pnTopncovc, SiSaoKaXoic, vEtopETptKoTc; T E xa i dpiBpnTiKoTc, Kai A O V I O T I K O I C . . We receive such exercise first under the grammarians while still small children, then afterwards with the teachers of rhetoric, of the mathematical arts, and of dialectic. (De consuet. 125) That such imagery was taken seriously by the educated elite is revealed by a number of funeral inscriptions in which a literary education is figured as doTcqorg, the result of dcnceiv (Raster 1989: 17 note 11). The shared connotations of athletic training and intellectual culture emphasize the effort required to attain paideia. But it is the transfigured state of the paideia graduate which is expressed by images suggestive of religious initiation. Marrou's 1938 study of sarcophagi of learned individuals of the Imperial period graphically illustrates the transformation of the soul brought about by education. The educated individual as initiate is identified by Gellius with his reader and himself, while those unable to appreciate the value of his compilations are turned away from the Nodes Atticae as profane: ea ne attingat neve adeat profestum et profanum volgus a ludo musico diversum. Let that 4 7 Cf Gellius NA. Prol. 20; on civil employment for the literarily educated, cf Bowersock 1969:43-58, Bowie 1970: 6 I l l uninitiated and impious mob, uncultivated in the activities of the Muse, neither touch nor approach this work. (NA Prol. 19-21). Terminology drawn from the Mystery religions is frequently used by Imperial authors to refer to literature as well as to a literary education. So the works of Homer and other poets — TJuf|pou ... Kai auuTravTocj T O U rapi T O V TJuf|pbu xopov ~ are referred to by Libanius, for example, as d7t6ppr|Ta (1.6.45); advanced students are o i TeAouuEvoi (Or. 15.27), and Photius quotes Olympiodorus on the "rites of passage" at a school of rhetoric at Athens (Bib. 80.b0b.14 ff; cf Kaster 1988: 16 footnote 7). Macrobius refers to the penetralia of Vergil's poetry (Sat. 1.7.5; cf. Kaster ibid). Marrou refers to "le culte des classiques," which "autant et plus que le neoplatonisme ... constitue le dernier bastion ou la vieille religion se defend contre l'envahissement du christianisme" (1948: 411).4 8 Such imagery emphasizes not so much the potential religiosity of the educated elite, however, as their acquisition of a special status. Raster's assessment of such an educational experience as "a transfiguring revelation, a passive experience, an irreversible change " (1989: 16 footnote 7) is particularly appropriate when taken in conjunction with the image of the educative processes as ao-Knoic,. After the sweat and labor of the childhood process comes an inexplicable and ineffable alteration of viewpoint which can only be acquired through paideia. The initiate acquires a trustworthy and accurate Kpioic; which allows him to interpret and to utilize paideia correctly. The initiate's status is made manifest by his successful performance of a new paideia-activity, now no longer the child's classroom performance but rather a daily and ongoing life in paideia. 4 8 The roots of such a religious stance in relation to the arts and to learning in general may perhaps extend to the Muse cults associated with the fourth-century schools of philosophy at Athens as well as with the Alexandrian Museum (cf Fraser 1972: 305). 112 Such a performance has two aspects: the application of learning to daily life, and the acquisition of more and relevant jroXuucxSia. To both of these the miscellany responds. In providing both a data-base of relevant 7roXuucc0ia and models for its correct application to daily life, the Imperial miscellanist sees himself as supplying a real need arising from this peculiar concept of the role of education in adult life. The supplementary data-base allows the educated adult to interpret paideia by intensifying his command over literature and language. It further allows the educated adult to identify more fully with the cultural elite of past and present, thus fostering the archaizing, exclusionary, and highly conservative qualities of Imperial paideia itself. Such 7roXuua6ia cannot avoid a moralistic bent, because in fact a literary education is seen in the Imperial period not just as fostering but as the very foundation of the moral life. 4 9 Knowledge was pursued not for its own sake but as a predominantly social phenomenon, as an appanage of personal relations and a token of accepted virtues .... The centrifugal force of [polymathic] learning ... was balanced by the centripetal force of mores, urging conformity to established values and behavior .... Good learning and good mores are assumed to be inseparable .... The union of qualities is part of the line of continuity in the classical tradition from the early to the late empire and between literary and social conventions. (Kaster 1989: 64-65) The one who has aquired such an education ~ who is jre7rai5euuevoc; ~ increases his moral excellence with his polymathic fund (cf. Marrou 1948: 234). A chief attribute of the professional [grammarian] ... is taken over by the amateur literary tradition and regarded as a moral trait, one of the attributes of the good man — his scrupulous attention to the details of his cultural tradition (what impatient modern readers of Holford-Strevens attempts to separate moral response from aesthetic judgment in Gellius: "The overriding concern for morality ... is true only of his philosophical chapters .... It is less important to him than literary taste... Gellius discusses moral problems ... but easily slides into antiquarianism" (1989: 28-32). For cultural initiates, however, morality cannot be separated from any intellectual activity. 113 Macrobius and Gellius commonly call their "pedantry"). (Kaster 1989: 66) The Miscellanist and Correct noXuuaSicc The modern application of "pedantry" to TroXuuaSia represents our negative reaction to what we consider, to be incorrect or irrelevant learning. For the miscellanist concerned with collecting relevant data from the entire cultural tradition, selection is as necessary as it was for the Hellenistic scholar ordering and evaluating his paideia archives. What determines the miscellanist's assessment of data as relevant? Basically, his selection rests upon his concept of his reader's needs. The miscellanist knows these needs intimately because, as I shall attempt to demonstrate in Chapter 3, he can identify his reader with himself at an earlier stage of his own intellectual development. And he is able to provide relevant 7ToXuucx0ia to satisfy these needs because he has himself adopted the correct approach to paideia: diligentia or 07rou8tj with TTOVOCJ in the acquisiton of a polymathic data-base. The miscellanists may themselves describe in several ways the zeal and labor with which they have compiled their collections. Gellius describes his notes as quasi quoddam litterarum penus (Prol. 2) which he has "squirreled away" (recondebam). ego...ipse quidem volvendis transeundisque multis admodum voluminibus per omnia negotiorum intervalla in quibus furari otium potui, exercitus defessusque sum. / wore myself out in leafing through and perusing a good many volumes in those spaces of time which I could steal from my duties to devote to leisure pursuits. (Prol. 12) 114 Clement applies to his own activity, now the Biblical parable of the talents properly and zealously invested {Strom. 1.1.3), now the image of a hunter tracking down relevancies in the course of his compilation. Kcx0a7rep ouv 6 Tfjc; aypac, epcoTiKoc; c/|Tfjaac, epeovrjaag dvrxveuaac; KUvoSpoufjaac, aipe! T O 0r|piov, OUTGO Kai TdXr|0ec; Y X U K U T T | T I cbaivETai Cr|Tr|0ev Kai Trovop 7ropia0ev. Just as one who is enamored of the hunt catches his beast after having searched, trailed, tracked and run with the hounds, so too it is clear that the truth is both hunted out with pleasure and conveyed through labor. (Strom. 1.2.20) Aelian too, in the epilogue to the De natura animalium, utilizes hunting imagery to define his pursuit of meaningful facts. ri 5e TOTC, ©npariKoTcj Kai ev £6pov eupeiv 5oKeI TTCOCJ euepuia, dXXd T O ye T W V T O O O U T O O V or) Td Txvn., ou5e Td ueXn. cruXXa|3erv eycb (jmui yevvatov, dXX' bnbcsa f| cbuaic; eSooKe Te auToic, Kai oacov fjcjicoaev dvixveuaai. Hunters consider it a stroke of good luck when they have tracked down just one beast. I see nothing particularly fine in picking up the trail and in bagging an abundance of animals. I prefer to track down all those qualities with which nature endowed them. Pliny the Elder details the extent of his scholarship with an almost avaricious glee. X X rerum dignarum cura (quoniam, ut ait Domitius Piso, thesauros oportet esse, non libros) lectione voluminum circiter II, quorum pauca admodum studiosi attingunt propter secretum materiae, ex exquisitis auctoribus centum inclusimus X X X V I voluminibus, adiectis rebus plurimis, quas aut ignoraverant priores aut postea invenerat vita. Nec dubitamus multa esse quae et nos praeterierint. Into the compass of thirty-six volumes we ha\>e compiled twenty thousand noteworthy items — as Domitius Piso says, one needs treasure-rooms, not books, to hold them all — resulting from the reading of about two thousand books (some of them so recondite that only dedicated scholars read them) from one hundred carefully selected authors; and we have added a number of things which earlier scholars have either been ignorant of or which came later in time. I don't doubt that there are also many data which have escaped our notice. In Pliny's case we have additional information about his diligentia. According to the younger Pliny, his uncle was obsessed with the need to fill every free moment with 115 scholarly labor. In a letter to an admirer of his uncle's, he draws the portrait of a man who literally immersed himself in books; the picture is consistent with the scholar's own description of his research considered above. The nephew reports that in autumn and winter Pliny would begin his day as early as midnight, in order to give himself extra time for study. He would work steadily through the small hours; then, after attending to his administrative duties after daybreak, reversus domum, quod reliquum temporis, studiis reddebat. post cibum saepe ... aestate ... iacebat in sole, liber legebatur, adnotabat excerpebatque. nihil enim legit, quod non excerperet; .... frigida lavabatur, deinde gustabatur dormiebatque minimum; mox quasi alio die studebat in cenae tempus. super hanc liber legebatur, adnotabatur, et quidem cursim. Once returned home, he would give any remaining time to his studies. After lunch in the summer he would often lie in the sun and have a book read to him; he would take notes and make excerpts. He compiled constantly ... He would have a wash, a snack, a little nap, then study until dinner, as though it were a new day. During dinner a book would be read, and he would even make quick notes while dining. (Ep. 3.5.9-11) This was the regular daily schedule. The younger Pliny goes on to describe how, during vacations, his uncle would give his mornings too to study, and would even be read to while at the baths (15) and while riding in his sedan chair (16). Pliny the Younger found his uncle's diligentia particularly admirable because of the way in which the Elder had managed to balance the responsibilities of his public career with his scholarship, all within a relatively brief life span. Miraris, quod tot volumina multaque in his tarn scrupulosa homo occupatus absolvent, magis miraberis, si scieris ilium aliquandiu causas actitasse, decessisse anno sexto et quinquagensimo, medium tempus distentum impeditumque qua officiis maximis qua amicitia principum egisse. If you wonder how a busy man can have composed so many works, many of them carefully researched studies, you '11 really be surprised to find out that for some time he acted as an advocate, that he died at the age of fifty-six, and that during the time in between these two events he was preoccupied with very high offices and with the friendship of the Imperial family. 116 In the epilogue to the De natura animalium, another thematically controlled compilation collection, Aelian prides himself on having resisted the temptations to a lucrative career to devote even more orrou8r| to research: O U K dyvoco 5e o n dpa Kai TCOV ec, x P n u a T a opibvTcov 6£u Kai TeGqypevcov ec. npdc, re Kai 8uvdpeic, nvec, Kai Ttdv T O ( JHXOSOQOV 81 ' airiac, e^oucnv, ei TTIV epaurou oxoXqv KareGepnv ic, rauxa, e£6v Kai (b^puiooGai Kai ev ravg auXaig e£eTd£eaGai Kai em peya Trpor|Keiv T T X O U T O U .... dXXd ou poi cjnXov ouv TOia8e TOIC . 7rXouoioig dpi0peia9ai Kai rrpog EKeivoug i&Ta&oQai .... PouXoipnv yap dv pdGnpa ev youv jrerraiSeupevov TrepiyeveoGai poi f\ r d a86peva TCOV Trdvu TrXouaicov xpripaTa re dpa Kai KTripara. I am well aware that among those who are eager for material advancement and public office, power, and reputation, there are those who will find fault with me . because I have devoted my leisure to the present work, when I could have exerted myself and proven my abilities in the public forum and made a lot of money .... I, however, had no desire to be numbered and assessed among the wealthy ....I would prefer to have one lesson well learnt than all the celebrated property and possessions of those wealthy individuals. The Imperial miscellanist can provide his reader with a relevant polymathic data-base. Can he also provide him with the vouc, or Kpioic, which will allow the reader in turn to approach jroXopaGia correctly? The miscellanist's data-base itself will not do so. The miscellanist simply offers up relevancies; it is up to the reader to apply them as needed. But the miscellanist is in a position to offer the reader correct models for paideia acquisition and for the incorporation of TroXupaGia in daily life. He may do so in his own person, by discussing his own experiences with TroXupaGia; or he may do so indirectly, by narrating circumstances and anecdotes which illustrate paideia acquisition and the application of TroXupaGia. Only models can point to real relevancy by showing paideia-initiates using their TroXupaGia in real-life circumstances. 117 These models, and the reader for whom they were presented, will be discussed in the following chapter. 118 Chapter 3 The Miscellany's Readership To appreciate both the compilatory activity and the data collection of the Imperial miscellanist, it is necessary to analyze his relationship with his reader. The miscellanist may himself express his awareness of the reader and his conception of the reader's needs. Here Gellius and his compilator Macrobius are the most specific in describing this relationship. In making his selection of data for inclusion in the Nodes Atticae, Gellius states that he had a fairly specific program in mind. volvendis transeundisque multis admodum voluminibus ... modica ex his eaque sola accepi, quae aut ingenia prompta expeditaque ad honestae eruditionis cupidinem utiliumque artium contemplatione celeri facilique compendio ducerent, aut homines aliis jam vitae negotiis occupatos a turpi certe agrestique rerum atque verborum imperitia vindicarent. In perusing very many books, I only admitted those data which either would inspire quick minds to a desire for proper erudition through a rapid overview and handy digest of the useful arts, or which would rescue men preoccupied with daily business from a shameful and low ignorance of fact and language usage. {Prol. 12) Macrobius, borrowing from Gellius, insists that nihil enim huic operi insertum puto aut cognitu inutile aut difficile perceptu, sed omnia quibus sit ingenium tuum vegetius, memoria adminiculatior, oratio sollertior, sermo incorruptior. / have included nothing in this work which is of no intellectual use, nothing hard to understand. I have instead compiled all that material through which your memory may be better equipped, your oratory more flexible, your speech more pure. {Praef. 11). Clement opens the Stromateis with a long discussion, echoing Plato's Phaedrus (cf. p 70 above), on the utility of committing his TroAuuaOia to writing. He shows considerable concern for his reader and the risks the reader runs in being exposed to Hellenic (non-Christian) paideia. 119 TavJTa 5e dvaCojjropGov uTrouvf|uaoi rd uev eKobv jraparreuTrouai eKXeyoov emaTnuovcoc;, (boBouuevoc, ypdqSeiv a Kai Xeyeiv ecbuXac;dunv, ou Ti nov cbGovojv (ou ydp Geutc,), 8e8tojc, 8e a p a 7tepi TOJV evTuyxavovTOJV, uf| TTT) eTepojc, ocbaXetev Kai jraiSi udxoupav, f\ cbaorv oi 7rapoiuiaC6uevoi, opeyovrec; eupeGwuev. In rekindling these [compiled data] in the form of vnofiVTJfiata, I willingly omit some material from my selection, for I hesitate to record that which I guarded against even reading, not through ill will — that is immoral — but because I feared lest my readers might happen upon this material and be misled by it; thus I would be found to have offered "a sharp knife to a child, " as the proverb runs. (1.1.14) Aelian in the De natura animalium is aware that readers have a variety of uses for jroXuuaGia. ei Se TOJ Kai dXXoj cbaverrai TauTa XuaiTeXfj, xpflcrGoj auToig • OTOJ 5e or) (bavetrai, COTOJ TOJ TraTpi GdXjreiv Te Kai TrepieTreiv. ou yap TrdvTa Trdoi KaXd, ou8e a£\a 8oKeT a7rou5daar Traor TrdvTa. If this material appears useful to someone, he can go ahead and use it. If he doesn't think it's beneficial, let him give it to his father to cherish and study. People hold different opinions about what is fine and worthy of study. (Prol. 1) The present chapter continues the analysis of the miscellany tradition by considering the manner in which the miscellanist's selection of data responds to his reader's needs. Here I shall attempt to demonstrate the following: 1. The miscellanist provides relevant data for a reader who must navigate a flood of available texts. Abundance and availability of literary resources must be dealt with by diligent study and by selection, the same activities seen as significant by Hellenistic scholars. 2. The reader is mildly patronized by the miscellanist. His interests in and uses for jroXuuaGia are identical with those of the miscellanist, but he is viewed as a younger person who has not experienced polymathic paideia to the same extent as the miscellanist. 120 3. The reader approaches paideia in a nonprofessional, elitist manner, with the expectations of an educated adult. 4. The miscellanist, in the process of providing the reader with a polymathic data-base, demonstrates the correct way for an educated person to use his leisure (oxoArj, otium). Bompaire has described Imperial paideia as based "sur les livres lus (dvayvcooig) ou dits (ocKpoaaic,). La lecture est ... Pintrument essentiel de culture .... II faut avoir beaucoup lu, beaucoup entendu, et chaque jour" Throughout later antiquity we constantly encounter "la persistance d'un gout profond pour les livres, independant de toute doctrine" (1958: 33-41). One senses that this constant close contact with the written text was not only a pleasant luxury for the leisured literary amateur, but an absolute necessity both for him and for the creative writer. A free access to paideia was of course limited during the Imperial period to those fortunate enough to have acquired both basic literacy and an education in ypappaTiKT). One entered the ranks of the culturally initiated elite through such a preliminary education; but a reputation for excellence and a preeminence within that elite could only be attained through relevant 7ToXupa9ia acquired through further study of written texts. From the first century BC, references are increasingly made in literature to the necessity of a written text for all forms of intellectual activity, including both original composition and secondary scholarship. The library is viewed as a source both for inspiration and for matter — uXn, silva, doctrina (Cicero De or. 3.8. 103, 125) — which funds intellectual creativity, a kind of stockpile of resources. The poet cannot create 121 without his books: ignosces igitur si ...haec tibi non tribuo munera .... nam, quod scriptorum non magna est copia apud me,hoc fit ....hue una ex multis capsula me sequitur. Forgive me if I don 't supply you with the gifts I owe. I hcr\>en't a lot of documents with me, for when I came here I brought only one of my many book-boxes (68: 33-40)1 Drawing upon Catullus' frustration over his capsulae, Ovid describes a similar balk placed upon his own creativity by the lack of texts in exile: non hie librorum per quos inviter alar que / copia. Here there is no supply, of books by which I may be enticed, on which I may be fed. {Jr. 3. 14. 37-38) During the early years of the Principate, those authors involved in both amateur and professional scholarship and in paideia and its transmission also enjoyed an intimacy with written documents. " A considerer Pactivite litteraire du premier siecle," Salles states, nous pouvons dire que e'est Pepoque qui a integre le mieux la litterature dans les activites quotidiennes et lui a donne une veritable fonction sociale .... Petit a petit la noblesse au premier siecle avait de la creation litteraire P equivalent d'une carriere politique avec ses charges et ses grades" (1992: 47). The literary individual was not typed by his genre but by his use of literature to create literature: Pliny, for example, saw poetry, oratory, and history all as possible venues for his talent (Ep. 5.8), while Juvenal in his seventh satire decried society's financial neglect of paideia by describing the mutual plights of the poet, historian, forensic orator, and teacher of rhetoric. All such practitioners shared one common feature: the written documentum, which in its etymological sense was the repository of doctrina both drawn from and concerning the past. Paideia was "le lien entre une formation litteraire telle que la donne Penseignement, et la culture generate qui resume la civilization" (Bompaire 1958: 94). 1 This might have been part of the burden of the doctus poeta, a development from the scholarly element in Hellenistic poetry. Callimachus had also insisted that auaprupov O U 8 E V deiSeo / sing nothing that has not its witness (fr 612 Pf; cf. Quinn 1973: 380; Marshall 1976: 251, 255). 122 The production, distribution, and acquisition of written texts in the first century BC appear to be not functions of commerce and systematic manufacture so much as a social gesture reflecting bonds of friendship and a sense of decorum. There are no significant data which make it possible to consider Atticus, for example, a publisher in our sense of the term (Starr 1987: 220-221 note 54).2 If a person in late Republican Rome wanted a text, he acquired an associate's copy and had it replicated either by his own hand or that of a scribe. Cicero never mentions going to a bookshop (Starr 1987: 225); he does, however, have quite a lot to say about private library collections, both his own and those of friends and associates. Quintus is building a book collection and needs advice (OFr. 3.4.5); given a complete library by someone in Greece, Cicero fusses about its transport to Italy (Att. 2.1.12). Friends sold books to friends. Marshall describes the manner in which Cicero bought up Sulla's library (1976: 259). Ego ... pascor bibliotheca Fausti I graze upon Faustus' book collection, Cicero wrote, graphically expressing his satisfaction at getting access to this collection (Att. 4.10). Faustus' texts represented personal wealth, some of them doubtless having formed part of the plunder Faustus' father brought to Rome from the sack of Athens in the preceding generation. Lucullus' library too consisted in part of such booty, in this case the texts which had formed Mithridates' royal library. It was all to be shared by friends, as Cicero indicates when describing how Lucullus generously opened his text collection to personal friends (Acad. Prior. 1.1). " Cf.. Zetzel's assessment that "private enterprise and private interest are more significant than commerce in ... respect [to getting possession of a desired text]" (1981: 235). 123 Plutarch refers to Lucullus' library as a kind of hostel at Rome for visiting Greek scholars, Mouooov TI KaTaytoyiov {Luc. 41; Birt 1881: 563-564; Callmer 1944: 154-156).3 One's personal book collection marked the extent of one's learning, forming at times a kind of alter ego. Mark Antony's pillage of Varro's library must have been viewed as the ironic tragedy of this scholarly man's career (Gellius NA 3.10.17). For the late Republican author, collections of books as physical objects were a mark of wealth, rank, status, and personal identity (Starr 1987: 223; Salles 1992: 197). Like exotic food, fish ponds, and country villas, libraries could become outward signs of an indulgence in luxuria, moral decline through material possessions. Plutarch had purposely to eliminate this connotation in Lucullus' case when he insisted that, as far as Lucullus was concerned, rj XPhOX h v (blXoTlUOTEpa Tfjc; KTrjaecoc; He took more pride in the use [of his library] than in its possession (Luc 41).4 Texts and their acquisition are frequently referred to in the literature of the early Empire. Martial's numerous casual references to bookshops and bookdealers allow us to build up a rather clearer picture of the Roman book trade than we could do for the Republican period. Martial identifies his poetic voice with the physical volumina containing it, and consequently insists upon the personal quality of his relationship with booksellers, naming their shops specifically as the places where "Martial" could be found. We hear of the booksellers Tryphon and Atrectus, Secundus and Polius Valerianus (e.g. 3 This element of individual sharing of paideia may be traced back to the second century BC. Aemilius Paullus kept only the booty from Perseus' library, with the intention of sharing it out among his sons (Plutarch A em. 28.6). Marshall points out that the friendship between Polybius and Scipio began with the loaning of texts (1976: 258). 4It..was as patrons of the city of Rome that Pollio and Augustus opened their libraries to public use (Marshall 1976: 261). 124 1.2, 113, 117; 4. 72; 13.3), functioning as Martial's editors, publishers, and distributors. For Martial seems simply to have entrusted these men with copies of his poems and allowed them to replicate the manuscripts according to their own judgments (Birt 1887: 357-359; van Groningen 1963: 3-4; Starr 1987: 219-221; Salles 1992: 156-170). Seneca mentions the shop of Dorus, who retailed Cicero and Livy {Ben. 7.6.1). Pliny refers familiarly to bibliopolae as his usual means of publication (Ep. 1.1.6). During the second century A D the bookseller's shop was occasionally depicted as a spot frequented by intellectuals, including authors. Gellius set three of his longer chapters in bookshops, locales where assertions made in the course of a discussion on literature or philosophy could be ratified or refuted by consulting an available text (NA 5.4; 13.31; 18.4). Athenaeus' lexical scholar, Ulpianus, is described as a man who acquired his reputation 5id rac, ovvexeic, Cnrfiaeig ac, dvd jraaav wpav Troievrai ev rale; dyuiaig, TOpiircxTOic;, BipXiircoXeioig, PaXaveioic, on account of the continual examination he would make at all seasons in the streets, porticoes, bookshops and baths. (1.2) For Athenaeus this list must have been an exhaustive one in terms of the places where books and scholars might regularly be found. Does it also imply that booksellers at the end of the second century A D dealt in such lexical reference works as Athenaeus himself must have consulted in the course of composing the Deipnosophistae? Here the information is not so abundant as for contemporary belles-lettres and the canonic authors of paideia. Of the books Gellius refers to in the Nodes Atticae as being for sale in bookshops , we have only one reference to compiled texts. These were bundles, fasces librorum of raggedy used volumina which Gellius bought at a very low price at the 125 market in Brundisium (9.4), and which seem to have contained paradoxographical and geographical mirabilia from Aristeas, Isigonus, Ctesias, Onesicritus, Polystephanus, and Hegesias. Gellius does not indicate whether he was dealing with compiled excerpts or whole works. That he goes on to describe his own cautious use of the texts suggests that they were already auvaycoyai of compilations which had been streamlined into easily excerpted units — that is, polymathic compendia. Although Imperial authors do not make frequent references to their sources for acquiring scholarly texts, they are clearly accessing the texts by some means. In a letter to a fan of his uncle's works who had asked for more bibliographical information, Pliny the Younger prepared an index of titles for him, remarking tarn diligenter libros avunculi met lectitas ut habere omnes velis (Ep. 3.1). Though Pliny did not indicate a source for his uncle's texts, these may have been available through the Pliny family, or copies which friends had earlier made from a family manuscript could be replicated in turn. This practice of copying from the author's autograph lent out to friends was, as van Groningen has shown, the ancient equivalent of our publication (1963: 3; he applies the term 5id8ooi<; to this process; cf. Zetzel 1981. 233-237). However Gellius acquired access to the miscellaneous compilations he lists by title in his preface to the Nodes Atticae (6-9), he was familiar enough with them to dismiss them with some contempt, and felt that his reader was familiar enough with the genre to recognize their quality by title alone. Diodorus learned to his own regret just how readily available a work of scholarly compilation could be at Rome. Parts of his own universal history, he relates at 40.8, had been pirated before the whole work could be published, despite the overview of the entire 126 project which he included in his preface with the stated purpose of roug 8e 8iaoKeudCeiv ricoOorag Tag BiBXoug cxTTOTpeipai T O U XuuaiveaOai rag aXXorpiag Trpayuarriag discouraging those in the habit of making compilations from spoiling other people's work. Of course friends and associates could freely offer their own works to others. A friend offered Gellius a manuscript of his own compilations for possible inclusion in the Nodes Atticae (14.6). Judging from the size of some recorded private libraries, there were individuals during the Imperial period who could afford to be generous with their texts. Persius left a library of seven hundred volumes (Suetonius Persius). Silius Italicus spent so much money on his library and books (ad emacitatis reprehensionem) that even Pliny was shocked (Ep. 3.7.8). The Suda attributed to the grammarian Epaphroditus a library of thirty thousand book-rolls. Aelian's contemporary Serenus Salmonicus bequeathed to Gordian sixty-two thousand rolls (Wendel 1940: 38). Even Symmachus, late in the fourth century, had enough material in his private collection to re-edit Livy (Ep. 4.18.5). Gellius mentions an occasion when Antonius Julianus rented an old copy of Ennius to check a manuscript reading (18.5.11); the owner of this text had evidently found a lucrative way to utilize his personal library collection. When writers of the Principate and early Empire mention a source for scholarly texts, they refer frequently to copies in public libraries. Although we may question just how public such institutions were (Marshall 1976: 261, Starr 1987: 216 note 23), the social class which found in paideia the sole means of entry into the cultural elite must have seen in the public library an extension of the opulent personal collections of the wealthy statesmen of the Republic. The first public library at Rome had been established in 39 BC, 127 and several large institutions followed in the next few centuries. By the fourth century Rome had twenty-eight public libraries. Italian and provincial cities took Rome's libraries as models and instituted their own, with the result that the existence of a public library was a primary sign of urban status (Wendel 1940: 45-55; Callmer 1944: 156-183). Imperial public libraries were more than book repositories, considerably more central to a city's political and social life than their Hellenistic counterparts had been. Gellius describes animated discussions among authors and the social elite in the library of the Domus Tiberiana (NA 13.20) and in the hall of the Aedes Palatinae, the latter discussion taking place in the midst of omnium fere ordinum multitudo opperientes scdutationem Caesar is an enormous mob of clients of all ranks and classes waiting to greet Caesar (4.1.1). Tacitus describes a meeting of the Senate at this library cum temple summoned by Tiberius, (Ann. 2.37) where, according to Suetonius, Augustus himself had been in the habit of convening the Senate (Aug. 29). Authors frequently refer to the elaborate accoutrements of the public libraries at Rome, especially to the busts and statues of both contemporary and ancient heroes of paideia (e.g. Pliny HN 7.115). The furnishings sometimes, it appears, attracted more attention than the books. Pausanias, for example, describing the public library built at Athens by Hadrian, mentions the text collection itself as an afterthought. KCfTEOKeixxt/xTo P E V teed dAXa AOnvaioic; ... td 8e im§avio~TO.Ta exaTov eioi KiovEC, Opuyiou XiOou • TTETroinvToa 8e Kai Talc, O T O O U C , xard rd a u r a oi T O T X O I . Kai o k f i p a T a EvrauSd E O T I V 6p6<j)ip T E emxpooto xai dXa|3doTptp XiGop, npoc, 8E dydXpaai KEKOopqpeva Kai ypa(J)ai<; • KaraKeiTai 5E be, ama pipTia. Among the constructions he made for the Athenians, the most impressive are the [stoa consisting ofJ7 one hundred columns of Phrygian marble. The walls have been made just like these walkways. There are rooms here with gilded ceilings and with alabaster, fitted out with paintings and statues. And there are books deposited in these rooms. (1.18.9) 128 The busts and paintings may in fact have been one means of locating texts, besides decorating the libraries and encouraging a canonic approach to authors and their works. But libraries could contain other fixtures as well — specimen collections, for example, inscribed tablets, and assorted antiquities — with the result that, as Salles describes it, the library "devient une sorte de 'decor' qui ... suffit a creer un univers imaginaire .... L'abondance d'ouvrage d'erudition de tout ordre, de curiosites diverses tant philologiques qu' historiques ou mythologiques rassembles dans les bibliotheques publiques ou privees a favorise la mode des oeuvres de compilation, des abreges et des anthologies." (1992: 185) One might add to Salles' list of connotations the simple concept of uXn or silver, an abundance of paideia's resources ready to be worked up into scholarship and art. Despite the clutter of the decor and the crowds of people, study and research did take place at the public libraries, and it is significant that the first living author to be commemorated with a library bust was the polymath Varro (Pliny HN 7.115). All individuals involved with the creation of literature, whether belles-lettres or scholarship, were active in the public book collections; in other words, the library was as appropriate a milieu for the compilating research scholar as for the docti poetae. Quintilian describes in passing the opportunity for excerpting offered in a library setting. In responding to protests to his reading list for young orators, Quintilian invited negative critics to construct a list of their own favorite authors by compiling one from pinakes of names and titles located in a library collection. nec sane quisquam est tarn procul a cognitione eorum remotus ut non indicem certe ex bibliotheca sumptum transferre in libros suos possit. 129 Surely there is no one so unfamiliar with these [available authors] that he ' could not easily copy down into his books a list taken from a library. (Inst. 10.1.57) We have more indications of the availability of texts of scholarly research for the library than for the book trade. Philological, historical, and philosophical studies are all referred to as available in the private library collections of the late Republic. Cicero represents himself in the De finibus looking up an Aristotelian commentarium in Lucullus' library (3.3.10); elsewhere he refers to studies by Dicaearchus and Varro which he would like to have in his own collection (Att. 13.31; 4.14). During the early Empire we find evidence for such works in public libraries as well. Gellius refers to a number of scholarly texts which he has gotten access to in public libraries. Among these he names an archival collection of praetorian edicts (NA 11.17), a work which we know as the pseudo-Aristotelian Problemata (19.5), and a handbook of the principles of logic by Aelius Stilo, the latter a collection of Stilo's personal notes (Gellius types it as a commentarium) sui magis admonendi quam aliorum docendi gratia more for the purpose of reminding himself than for teaching others (16.8.3).5 Less precise but more provocative is a passing reference to some sort of scholarly work which Apuleius makes in the course of his defense against a charge of sorcery. Having listed the names of some famous magi, Apuleius states that these names are easily found in any public library, haec et multo plura alia nomina in bibliothecis publicis apud clarissimos scriptores me legisse these and many other names I have read at the library in the works of the most well-known authors (Apol. 91). "Apuleius can hardly refer to magical treatises," his commentators maintain, "which would assuredly not have 5 S. West has referred to such works of scholarship as being mostly privately owned documents made by individuals "for their own use" (1970: 290). If that is what Stilo's work was, the text had by Gellius' time managed to find its way into the scrinia of the Bibliotheca Pacis. 130 been kept in the public libraries. Such works were publicly banned in the third century BC .... He must refer to learned works such as Pliny's Natural History" (Butler and Owen 1914: 164); in other words, to compilatory scholarship. Authors undertaking scholarship craved a quiet, well-stocked library and the opportunity to use it. To these Imperial authors Cicero's description of Cato in Lucullus' library seems particularly applicable: in summo otio maximaque copia quasi helluari libris, si hoc verbo in tarn clara re utendum est, videbatur. There he was, at his ease and surrounded with abundant resources, having a binge of books — if I may use that term to refer to so noble a pursuit (Fin. 3.2.8). Frequently forced by his public responsibilities to deprive himself of texts, Cato becomes a paideia glutton once he has free time and unlimited access to a library. Plutarch echoes this sense of abundance and resources offered by a large library in the introductory lines to De E apud Delphos. Here Plutarch writes to his more fortunate friend, the learned Serapion: eyib youv Trpoc, oi Kai 8td aou TOIC, auToGt chiXoic, TCOV H U G I K C O V Xoycov eviouc, COOTTEP oarapxdc; OCTTOOTEXXCOV opoXoyco 7rpoo5oKdv ETEpouc, Kai TrXEiovac, Kai PEXfiovac; Trap' upiov, orre 8f] Kai TTOXEI XpcopEvtov pEydXr) Kai axoXfjc, paXXov E V PtpXioic, TTOXXOIC, Kai 7ravTo5a7rai"c, 8taTptPaic, E U T T O P O U V T C O V . In sending to you and, through you, to others there who are interested in information about the oracle these first fruits, as it were, I admit that I expect longer and better accounts from you. After all, you do live in a big city, and you have more free time for textual research and a variety of studies. (384d) Living in Antioch, Serapion had access to the Seleucid library foundation in that city. Plutarch elsewhere describes in greater detail his personal frustration in trying t o carry o n research in a small town. Plutarch feels himself deprived at Chaeronea of the texts necessary for research. Tip P E V T O I ouvTa^iv UTropEpXripEvio Kai ioropiav, E £ O U rrpoxeipcov ou5 ' OIKEICOV , dXXd £EVIOV T E T IOV JTOXX IOV Kai 8iE07Tappeviov iv 131 ETEpotg ovvxovoav avayvcoopdTcov, TO) O V T I xp f i v TOCOTOV imdpxEtv Kai pdAiora TCXV roXtv EuSoxtpov Kai cbiXoraXov Kai rroXudvGpcojrov, cog PipXicov T E 7ravTo5a7rcov dcf)6oviav sxcov, K C ( i ° a a roug Ypdcbovrag Siacbuyovra acoTnpia pvr|png emtbaveaTepav eiXn(j)e monv, uTToXapPdvcov aKofj Kai StaTruvSavopevog pn6evog TCOV dvavKaicov bvoekc, aTroStSoin. T O epyov. ripetg Se piKpdv pev okouvreg JTOX IV . The person who has undertaken a work with its attendant research depending upon passages from documents which one does not have on hand but many of which are at a distance and scattered in various locations, really needs first of all and especially a city which is glorious, discriminating, and populous. Here he has an abundance of all sorts of books and oral information, which though it has escaped written records still has been preserved in memory and is dependable, and which he could investigate. These things he needs to produce a work which is lacking none of the necessary ingredients. But as for me, I live in a small town. (Dem. 2-. 1-2) Although he recognizes the researcher's need for reliable witnesses (in conformity with the ioTopia-program established by Herodotus), Plutarch cannot properly begin his work without the library.6 Diodorus likewise values the library resources of a large city. He attributes the success of his history to ri ev 'Pibpn. x°P n Yi a T (J°V roog if|v UTOKeipevnv OTToBeaiV dvnKOVXCOV the abundance of materials answering to the needs of my project, and to the ease with which he could access information E K TCOV jrap' EKeivoig ujropvripdTcov E K TTOXXCOV XPOVCOV TETTipTipEVtOV from the notes and memoirs preserved there from the distant past (1.4.2-4). Imperial authors developed a cluster of expectations and conventionalized responses to their work and its relationship to the library. In an oration delivered at Carthage, Apuleius stated si erudita [mea verba] fuerint, [habetote] ut si in bibliotheca legantur. If my words seem to you learned, consider them as though you were reading them in the library (Flor. 18.85). The standard which measures Apuleius' erudition is physically stored 6 Since the Hellenistic period there has been an interesting change of emphasis in favor of the role of the library collection in research. Polybius had faulted Timaeus for limiting his research to library collections (12.25e). Yet Cicero's much more positive estimation of Timaeus as longe eruditissimus et rerum copia et sententiarum varietate abundantissimus (De Or. 2.58) suggests that this author is to be valued precisely because of that work in the library. 132 on library shelves. The public library has in this sense become a reference institution. From earlier associations with luxuria and the booty acquired in war (the two libraries which Trajan included in the Forum Traiani were inscribed E X MANUBIIS [Gellius NA 13.25.1; Callmer 1944: 162-164]), the library has come instead to represent the storehouse of authoritative learning. When the general contents of libraries are referred to by Imperial authors, as often as not the antiquity of the texts is the important point. So Athenaeus' Larensis, a paragon of polymathic learning, surpassed even Polycrates, Nicocrates, Euripides and Aristotle in his collection of ancient Greek texts (1.4). Ammianus Marcellinus draws upon such an association in lamenting the decline of paideia in his own day, a time when bibliothecae sepulchrorum ritu in perpetuum clausae libraries sealed up forever, like tombs (14.6.18) figure metonymically the falling off of a society's bonds to the past and its irresponsible preoccupation with the ephemeral pleasures of the present moment. Some authors even extend the connotation of the library as a storehouse of ancient learning and apply the term to especially erudite individuals. Pliny's Titus Aristo, for example, is described as non unus homo sed litterae ipsae omnesque bonae artes ... quantum rerum, quantum exemplorum, quantum antiquitatis tenet ... mihi certe quotiens aliquid abditum quaero, ille thesauros est. He's not an individual man but the incarnation of literature and all the liberal arts. So much matter, so many exempla, so much antiquity he possesses! Every time I have a problem, he is a veritable treasure house of information. (Ep. 1.22.1-3) Eunapius' description of Porphyry's learned teacher Longinus is even more mannered and condensed. He was, Eunapius asserts, PipXto9f)Kr| TIC; ... eutpuxoc; Kai TrepmaTobv uouaeiov A library incarnate, a walking shrine of the Muses (4.1.3). 133 Books composed primarily through the compilation of earlier texts can themselves be entitled "libraries." Diodorus' BtpXio6fJKr| icrropiKf|, for example, boasts in its title the comprehensive manipulation of earlier scholarship. Diodorus' introductory remarks refer both to the extensive labor and time —thirty years — which went into compiling the work, and to the advantage to subsequent research offered by a work which drew together under one roof, as it were, many earlier treatises. TOIC, uev ydp empaXXouevoic, 5ie£;ievai TCKCJ TOJV TOOOUTOJV auyypacbeojv ioropiac; jrpojTov uev ou pd5iov euTropfjaai TOJV eicj Tfjv Xpeiav TnTrrouoojv PipXitov, e j re iTa Sid Tf]v dvoouaXiav K a i TO rrXfjGoc; TOJV a u v T a y p d T t o v 6uoKaTdXr|7rTocj y i v e T a i TeXeojcj K a i 5uoe(biKTOc, f| TOJV TreTTpaypevojv dvdXrupic,. Those undertaking to work through the histories of so many authors find first that it is not easy to get access to the necessary books; then, because of the diversity and the quantity of works involved, a comprehensive understanding of the events under consideration becomes, in the end, difficult and hard to arrive at. (1.3.8) Diodorus' compilation, however, shares with a library's book collection its ease of access and comprehensive selection of topics, Trjv uev dvdyvojaiv eToiunv rrapexeTat rfiv 6' dvdXrivpiv exei TravTeXoJC, eu7rapaKoXou6r|TOV. It makes reading easy and facilitates comprehension (1.3.8). In the case of [Apollodorus]'s Bibliotheca we have not only the work itself but also an assessment of it by Photius.7 Photius' description attributes to [Apollodorus] the antiquarian emphasis of the public library, for the book rrepieixe Se T d jraXaiTaTa TOJV 'EXXfp/ojv oaa Te rrepi 9ewv Kai npojojv 6 xpovoc; auToig Socja^eiv CSOJKCV. It included the most ancient statements of the Greeks, all those which they formulated over time about the gods and heroes. (186.142a) 7 Photius' own work is entitled Bibliotheca. 134 In analyzing the primacy of the text in cultural activity we have moved from the concept of library collection as booty to library collection as storehouse of paideia, through to the learned text or learned individual's memory as like a library collection in its retention and storage of authoritative data. The common feature in all four of these Imperial-age formulations of the book collection is the image of matter, of stuff, assorted objects, miscellaneous materials acquired in the first instance through the physical act of the pillage of book collections, then through the scholarly practice of compilation. Libraries, learned texts, and polymathic minds are created by the amassing of relevant data, in a process involving two steps: acquisition and selection. Documents thus viewed are receptacles of literature seen as a kind of polymathic raw material, the UAn, silva, or copia of paideia. The miscellanist's reader must navigate this flood of texts, however, because he has paedeutic needs which must be met. We may begin with his need for further contact with the encyclic arts. Supplementary Reading and the Encyclic Arts Up to this point in the discussion I have been using the term "educated adult" to apply to those individuals who had completed their formal education in ypauuaTiKf|. For most people this point probably arrived at about age sixteen (Beck 1970: 372, Marrou 1948: 223-225); but in terms of the miscellanist's reader, the status which such an education gave lasted a lifetime; his reader may therefore be inclusively termed an adult, 135 though one expected to apply to paideia for both self-improvement and amusement over the years.8 The miscellanist's reader is assumed, then, to be an-educated person in need of a general rounding out in various areas, specifically in the encyclic arts other than those specifically devoted to language acquisition and manipulation. These he has already acquired in the form of grammar and rhetoric. Having gained control over language, he has acquired access to the manuals of artes which are becoming increasingly available to the general reader from the first century BC. Varro's Disciplinae had included architecture and medicine among the artes. Vitruvius (De Arch. Praef. 3) and Galen ( n p o T p e j T T t K o g ETT ' iaTpixfj 9) had also insisted that these respective pursuits be considered liberal. Celsus added agriculture and military science to the list. These authors are in a position to treat the artes in a summary way in their collections of technical manuals because the first century BC witnessed the composition of a number of such works using dialectic structure and definition to formulate the vocabulary, processes, and aims of the arts peBoSiKtoc,, systematically (Fuhrmann 1960: 156ff; cf. Cicero De Or. 1. 187-188). However, this codification of the artes had not guaranteed their inclusion in the child's standardized curriculum. Both Greek and Roman schoolchildren of the Imperial 8 During the Imperial period, termination of grammatical studies marked the point at which one sought further training with a rhetor in a more or less formal school setting (cf. Raster 1983: 323-324). There were other educational options for the older adolescent, however. The four chairs of philosophy established by Marcus Aurelius provided state-sanctioned studies in the major Greek philosophical sects. Gellius (NA 17.20.4; 18.10) reveals that young people studying abroad could seek instruction in both rhetoric and philosophy. Vespasian encouraged teachers of medicine at Rome (Singer and Wasserstein 1970: 662). There were well-established schools of law at Rome by the second century AD (Marrou 1948: 387-388). An edict of Valentinian I of 370 AD put the age limit at twenty for students from the provinces studying at Rome; in the time of Justinian, law students at Beyrut were required to finish their studies by age 25 (C. Theod. 1491; C. Just. 10.50; Marrou 1948: 403). 136 period learned some arithmetic, at least enough to allow them effectively to rem servare suam, as Horace describes the process (Ars P. 329; Marrou 1948: 366; cf. Christes 1975: 170). But the competency thus acquired was minimal. As for the other artes viewed as liberal (that is, the performance of which did not produce money-making goods and services), they played apparently no fixed part in the young person's indoctrination in ypauuaTiKf|. Quintilian assumed that the grammarian himself had some acquaintance with those artes such as astronomy or music, references to which might arise in a canonic author (1.4.4); and in a discussion an oratori futuro necessaria sit plurima artium scientia (1.10). Quintilian affirms that the perfect orator must be familiar with the encyclic paideia, and is to begin the study of the artes during these early years. But as we have seen, Quintilian does not give details as to how or when these subjects are to be addressed in the basic grammatico-rhetorical curriculum. In rounding out and supplementing the orator's education with the encyclic arts, Quintilian draws upon Cicero's concept of the doctus orator. In the dialogue in which he formulates his image of the ideal statesman as the educated speaker, however, Cicero himself traces his concept of complete education back beyond Isocrates to the archaic Greek statesmen and their concept of sapientia. The ancient Greek statesmen fulfilled their civic duties and, when time permitted, amused themselves with artes, circumstances or personal inclination determined the amount of time they could devote to them (De Or. 3. 56-58). So Isocrates, so Gorgias and Thrasymachus, so Socrates. And Socrates, through a disinclination to appear as a political figure, effected a discidium, Cicero insists, between the public figure and the private lover of learning, thus precipitating the 137 regrettable divorce of rhetoric from philosophy (cf. Schulte 1935: 37-46). Cicero essays to recreate the ancient model. The statesman needs a complete grounding in the art of rhetoric. But he needs much more than this, namely, a thorough familiarity with has artes quibus liberates doctrinae atque ingenuae continerentur, geometriam, musicam, litterarum cognitionem et poetarum those arts through which liberal learning is preserved: geometry, music, literature (De Or. 3.127). But even the artes are not enough in a society in which literacy has ensured information's storage in textual format. Cicero's ideal statesman now needs to know facts, data, antiquarian lore, et ilia quae de naturis rerum, quae de homimtm moribus, quae de rebuspublicis dicerentur (ibid.); in short, relevant polymathic material from all the matter now collected into texts through a process of ioTopia (Schulte ibid. 60). From this larger, less structured field Cicero's orator acquires the matter which feeds his discourse, a wide and varied knowledge of a world cuius cognitio magnam orationis suppeditat copiam a knowledge of which supplies a great fund of material for discourse (Orat. 16). How and when is the orator to learn all these things? Cicero does not specify. If Cicero's historic paradigm is consistent, the orator will acquire them in the moments he can spare from his public duties and responsibilities. Cicero's description of his own education in Brutus 300-324 mentions training in rhetoric and dialectic only, while Crassus in De oratore admits that he himself had not had the time for a truly full education: Fateor neque hodie ... nec ... ullum habuisse sepositum tempus ad discendum ac tantum tribuisse doctrinae temporis quantum mini puerilis aetas, forensis feriae concesserint. / must admit that neither today [nor in the past] have I had any time especially set aside for learning, and have allowed to the acquisition of learning only so much time as my childhood years and public holidays have permitted. (3.85; and cf. Rawson 1972: 35-37) 138 Vitruvius states at the opening of his handbook of architecture a similar conviction about the importance of an education in the artes. Ut litteratus sit, peritus graphidos, eruditus geometria, historias complures noverit, philosophos diligenter audierit, musicam scierit, medicinae non sit ignarus, responsa iurisconsultorum noverit, astrologiam caelique rationes cognitos habeat. [The ideal architect] should be learned in literature, drawing, geometry; he should know a number of historical accounts, should have studied the philosophers with care, have learned music and a little medicine, be familiar with legal matters, and should understand celestial phenomena. (De Arch. 1.3) The architect without this background will necessarily lack auctoritas (1.2), we are told, and we are given some situations in which the cultured architect will find his liberal training of great benefit. His polymathic knowledge of literary historiae, for example, will permit him to explain antiquities of design such as the origin of the Caryatid columns in an ancient war between Athens and medizing Caria. Detailed knowledge of the natural world will help the architect avoid such problems as air pockets in plumbing systems (1.5-7). But again, Vitruvius, does not explain how this 7TOXUUCX6ICX is to be incorporated into the future architect's program of study. The oXcoq TrejTaiSeuuEvoq The historian and philosopher Nicolaus of Damascus was, according to the Suda, a prodigy of the EVKUKXIOC, TtaiSeia, mastering grammar, rhetoric and music, attracting attention through his original dramatic compositions before turning as an adult to Peripatetic philosophy. The Suda, drawing presumably upon Nicolaus' autobiography, attributes this thorough early education to Nicolaus' father, a connoisseur of the artes (ropi TCXUTCX U&XIOTCX ajrouSdaai). But Nicolaus' own description of the liberal arts 139 suggests that most people acquired them in other ways, not in childhood as a result of broad educational curricula but irregularly and later in life. According to the Suda's account, eqbr| 6e NixoXaoc, opoiav eivai T n v oXnv 7rai5eiav d7ro5npia. obc, yap ev TaoTTi jrpoaaupPaivEi TOIC, djro6qpoi3ai Kai p a K p d v 666v 5IEC;IO0CTIV 07100 pev evKaTdyeaOai TE Kai evauXig"ea9ai povov, OTTOU 6 ' Evapiarav, ojrou 8e TrXeiouc. EvSnpeTv qpEpac,, EVIOUC, 8E TOTTOUC, EK 7rap68ou SEwpsTv, £7ravEX96vTag PEVTOI rale, sauxcov EVOIKEIV Eoriaig, OUTOO Kai 5id Tfjcj SXnc; ncaoeiac, SiEpxopevou 5eiv ev okj pev E m T n 5 E u p a o T V ETTI TTXEOV Ev8iarpi|3Eiv, tv ok; 5' ETT' EXOTTOV • Kai TO PEV 6Xa, r d 5E EK uipovc,, r d 5E dxpi aroixEuooEcog TrapaXapPdvEiv Kai TI eKeivcov xP^uov K a r a a x o v r a g era rriv cog dXnBox; 7iaTp6pav EOTiav eXSovrag (j)iXoao<J)eTv. Nicolaus said that the encyclic arts were like going abroad. When a person goes abroad and journeys far, he puts up for several days now at this place as in an inn, now at that place as though only stopping for breakfast. Some places he sees only as he travels by them. But in the end he comes hack home and lives in his own house. In the same way, those who make their way through the liberal arts must spend more time with some of them than with others, studying some thoroughly and others only in part, of some acquiring just the major points and really practical parts; then, returning to their true home, they must practice philosophy. (Suda nu 393:28) The sightseeing image is appropriate. One must learn as much as one can as opportunity permits and interests and time allow. The best that a realistic person can expect is the systematic overview, the casual acquaintance in some areas and more intensive study in others.9 Cicero's orator doctus was skilled in the technique of oratory and funded his rhetoric with material from the encyclic arts as well as the polymathic copia provided by history, philosophy, and Hellenistic scholarship. The educated adult of the Imperial period 9 Plutarch De lib. educ. 70 adopted the image of travel to figure an education in the artes in a similar discussion: T O O T C X U E V E K TrapaSpoufjc uaOelv .... T T ) V 5E <J)iAoaoci>iav 7TpsaPeu£iv ... coarrep Y«P T T E P I T T A S U C T O U MEV 7T0XX(XC TToXsic; KaXov, EVOlKqOOU §S Tfj KpaTioTf l X P H 0 1 " 0 ^ [The student of the encyclic arts] should learn these things in cursory fashion ... but give primary attention to philosophy .... just as it is fine to sail around among many cities, but in the end to take up residency in the best. 140 inherits that model as well as the limitations which daily reality imposed upon it. Although Cicero still considered political effectiveness to be part of the ideal statesman's persona, he was himself aware of the transitoriness of power, most of his own philosophical and rhetorical works having been composed in more or less forced retirement. If he took Demetrius Phalereus and Licinius Crassus as model statesmen, it was not because he was ignorant of the personal disasters they had suffered in the course of their political careers. Imperial authors were aware of Cicero's, as they were of Plato's and Xenophon's, discussions of the relationship between political power and education. But they also knew that the system had changed significantly with the establishment of the Augustan Principate. They did not imagine that the truly educated adult would affect politics in any appreciable way, on the level, that is, of a Cicero, a Demosthenes, or a Scipio. Indeed, when an emperor such as Hadrian or Marcus showed any interest at all in paideia, Imperial authors reacted with a delight that must have originated in surprised relief. These authors were nevertheless affected by the past's model of the educated man as politically responsible, to the extent that they incorporated political responsibility, be it only a knowledge of jurisprudence or of the antiquities of public religious cult, into their concept of the ideal educated man. The Imperial oXcoc, TreTrcaSeuuevocj highlights his social responsibility in a twofold manner: by downplaying his actual political power while upgrading his relationship with the state's authoritative past. For the Roman, the fully educated and socially responsible individual is learned in his country's legal and religious antiquities; for the Greek, he focuses his learning upon local (i.e. Hellenic) genealogical 141 and topical antiquities. Although he does not wield power, the educated man is yet publicly responsible as a guardian of the state's authoritative past. In addition to this political stance, the truly educated adult of the Imperial period must also display a personal responsibility to paideia as a whole. Here too he is seen as an example to others, since as a public figure he. cannot avoid the public eye. A time will come when the public insignia fall away and only the paideia is left to mark the responsible statesman. In the fifth century, Sidonius will maintain, iam remotis gradibus dignitatum per quas solebat ultimo a quoque summus quisque discerni, solum erit posthac nobilitatis indicium litteras nosse. Now that there are no ranks of office through which each highest individual used to be distinguished from the lowest, henceforth the sole mark of nobility will be a literary education. (Ep. 8.2) As a consumer of paideia, the educated and socially responsible individual must model the integration of paideia into daily life in a moral way. For the Imperial period shares with its classical past a profound desire to believe that right education has a direct and positive effect upon the individual soul.1 0 During the Imperial period, the image of the young adult seeking further exposure to the encyclic arts and philosophy, and that of the young orator seeking further polymathic copia to fund his rhetoric, merge into the image of the mature and politically Plutarch De lib. educ. 8a: Tpicov yap O V T C O V (iicov cbv 6 uev eon TipaKTiKocj 6 8e BecopriTiKOc 6 5e djToXotuGTiKog, 6 uev, E K X U T O C K a i SouXocj T I S V fj6ovtov cov, Ccpcb5r|c K a i uiKpoTrpsTrrig eonv, 6 be SetopnTtKoc, T O O TTpaKTiKou 8iauapTdvcov, dvcoc|)£Xr|c;, 6 5e npaKTiKOc;, duoiprjaac (tnXoaoclnac, cxuouaoc; Kai rrXnuueXricj. TTEiparsov ouv eicj Suvauiv K a i r d Koiva TTpdrrav K a i rfjcj ctnXoaocjnac; dvnXau(3dv£cs0ai Ka ra T O TtapetKOv T U V Kaipcov. O U T C O C J enoXiTeuaaxo riepiKXfjg, ouxcoc. 'Apxurac; 6 Tapavrivoc, O U T C O Aicov 6 LupaKoaiog, otiTcog 'ETtauEivcovSag 6 ©nPafoc;. There are three basic lifestyle: the life of action, the life of contemplation, and the life of pleasure. The latter is dissipated and enslaved to pleasure, vulgar and bestial; the contemplative life, lacking an active component, brings no benefit; but the life of action without philosophy is bereft of culture and without aesthetic grace. One must therefore try one's best both to function within the community and to participate in philosophy, as opportunity allows. This is how Pericles carried on his public career, as did Archytas ofTarentum. Dion of Syracuse, and Epaminondas of Thebes. 142 active adult at leisure. The oXioc, nenaibevixevoc, knows how rightly to take advantage of these leisure moments.11 If opportunity allows, he may participate in the public display of paideia offered by, for example; lecturing philosophers and declaiming sophists. Philostratus' descriptions of some of the successful sophistic showpieces reveal that these performances had a very dramatic quality about them, dealing with human issues from a given culture-hero's perspective after the manner of ancient tragedy (cf. VS 520, 589-590). The public image of the sophist as declaiming teacher melds into the image of the lecturing philosopher; these lectures, too, were paedeutic and often not seen as substantially different from a declamation. Favorinus and Dio are given equivocal treatment in Philostratus: are they philosophers or are they rhetoricians? Apuleius, whether defending himself in court or declaiming before the city of Carthage, insists that he is a philosopher. Even Lucian insists 1 1 The concept of the correct use of leisure (cr/oXr)) as opposed to recreation (dvaTtaucng) can be traced to Aristotle's discussion of education in Book 8 of the Politica (1337b30-1338al3): Tf]v ())uaiv O C U T T I V CnTeiv ... \xr\ uovov doxoXetv opGcjg dXXd K a i oxoXd^ eiv 8uvao6ai K O X W C . aurn yap dpxn j r d v T t o v uia • ei 5' auc|>co uev 5ET , udXXov 8e aipeTOv T O cr/oXdCsiv Trig dcxoXiac K a i xeXoc, CnTnTeov 6 i i 8ei Troiouvrag oxoXdCeiv. ou yap 5r| jraiCovTag • T E X O C yap dvavKafov sivai T O U |3iou T T ] V TtaiSsidv riuiv. si Ss T O U T O dSuvaTov, Kai udXXov ev raig dcr/oXiaig x P n 0 T £ ° v T(XK TtaiSiaTg (6 yap T T O V G J V S E T T O I rf\c, dvarrauoEcog, f\ 5s natSid xdpiv dvajrauaeiog eonv • T O 5' doxoXetv auu|3aiv£i usrd J T O V O U K a i auvroviac).... ro 8e cr/oXdCsiv exeiv auro 8 O K E T T T I V iqSovriv Kai T T ) V euSaiuoviav K a i T O Cqv uaKap icoc . T O U T O 8' ou T O I C doxoXouaiv urrdpxEi dXXd T O I C ; oxoXdCouorv .... O J O T E 4>avepov o n 5eT Kai npdc, TT\V E V xf\ Siaywyrj oxoXriv uavOdveiv d i r a K a i rraiSeueoGai, Kai Taura pev r d naiSsupaTa K a i naurac. Tat; ua6r|0£i<; sauTwv sivai XdpiV, Tag 8e rrpog TTIV aOXoXiav tOC dvayKaiag K a i xdplV dXXwv. Our very nature has a tendency seek of itselffor ways and means which will enable us to use leisure rightly, as well as to find some right occupation; indeed, it is the power to use leisure rightly... which is the basis of our life. It is true that both occupation and leisure are necessary; but it is also true that leisure is higher than occupation, and is the end to which occupation is directed. Our problem, therefore, is to find modes of activity which will fill our leisure. We can hardly fill our leisure with play. To do so would be to make play the be-all and end-all of life. That is an impossibility. Play is a thing to be chiefly used in connexion with one side of life: the side of occupation .... Occupation is the companion of work and exertion: the worker needs relaxation: play is intended to provide relaxation .... The feelings which play produces in the mind are feelings of relieffrom exertion; and the pleasure it gives provides relaxation. Leisure is a different matter: We think of it as having in itself intrinsic pleasure, intrinsic happiness, intrinsic felicity. Happiness of that order does not belong to those who are engaged in occupation : it belongs to those who have leisure .... It is clear therefore that there are some branches of learning and education which ought to be studied with a view to the proper use of leisure in the cultivation of the mind. It is clear, loo. that these studies should be regarded as ends in themselves, while studies pursued with a view to an occupation should be regarded merely as means and matters of necessity (trans. Barker 1946: 335-336/ The pleasure derived from the correct use of oxoXrj lies at the heart of the miscellanist's stylistic choices; cf. Chapter 4 below. That Nicolaus of Damascus made philosophy the ultimate goal of a broad encyclic education probably derived from this Aristotelian position as well; cf. above n. 8. 143 upon his personal contact with both rhetoric and philosophy (Hermotimus 13). The audiences who gathered for the educational experience of listening to any public speaker do not seem to be absolutely certain whether they are being entertained or improved — or both. Seneca describes the audiences which came to hear Atticus as ranging from the intensely sincere (himself) to the appallingly nonchalant, quos ego non discipulos philosophorum sed inquilinos voco. quidam veniunt ut audiant, non ut discant .... Magnam hanc auditorum partem videbis cui philosophi schola diversorium otii sit. These I would call not the philosophers' students but their tenants. Some come to listen but not to learn. You will see many who think the philosopher's class is a lounge. (Ep. 108.6) Some attend such a lecture only to improve language skills: aliqui tamen et cum pugillaribus veniunt, non ut res excipiant sed ut verba. Some come with notebook in hand, not to write down concepts but just vocabulary items. Gellius' Calvenus Taurus is likewise annoyed by students pursuing philosophical studies only to improve their rhetorical abilities (NA 10.19). Plutarch seems to be describing specialist philosophers when he advises the young autodidact to fit his questions to the lecturer's capabilities, not for example challenging the ethics expert with questions on natural history, and vice versa (De aud. 43 b-c). Perhaps we may see in the Platonic, moralizing Maximus Tyrius, and in Apuleius with his Peripatetic zoological interests, examples of such diverging types of popular philosophers. As Plutarch and Seneca stress, oral teaching is a group phenomenon, and as such brings with it all the disadvantages associated with such a setting: distractions, fixed format, restless audiences, acoustical problems, delays, and unexpected or last-minute changes of location or topic. At Florida 16 Apuleius describes a lecture interrupted 144 because of the rain, then further delayed when Apuleius (the featured speaker) sprained his ankle in the gymnasium . But for the adult learner such drawbacks were all eliminated from the learning process when it was simply a question of the learner addressing his text. For one thing, the student could control his own time, not fit himself into a prearranged schedule. The eccentric study schedule of Pliny the Elder (being read to in the baths or while traveling, rising before dawn or staying awake late at night) was only possible because he. had his own texts with which to work. Plutarch offers a description of Brutus epitomizing Polybius on the eve of Pharsalus (Brut. 4). Text-based learning was probably the only option most of the time for the adult involved in an active public career. Gellius imagines his Noctes Atticae being read interstitione aliqua negotiorum (NA Prol. 1). Indeed, Pliny the Younger even incorporated studies into more physically demanding athletic exercise; while out hunting ad retia sedebam, Pliny recalls, erat in proximo non venabulum aut lancea, sed stilus et pugillares. I was sitting there watching the nets, and nearby had neither my hunting spear nor my lance, but my pen and notebook (Ep. 1.6). The adult learner of the Imperial period could not and would not subordinate himself to an instructor (although he would readily be instructed by a friend; cf.. Dio Or. 18). He must, as an adult, judge what he needed to learn and the amount of leisure time he could reasonably allot each subject, given the realities of his career and social responsibilities. His texts in hand, he had more freedom to arrange his study schedule during his free moments, and the situation was the same in the case of the adult involved in scholarly or creative work as well: Caesar composing the De analogia on campaign (Suetonius ltd. 56.2-5; cf. Dahlmann 1970: 53), Horace packing up Greek comic texts to 145 use in the country (Sat. 2.3.11-12), Pliny the Younger on vacation and reveling in his freedom from urban distraction, drawing inspiration for his writing from the silence and solitude. mecum tantum et cum libellis loquor .... O mare, verum secretumque uouaeTov, quam multa invenitis, quam multa dictatis. I speak to myself and my notes alone .... O sea, O shore, true and private Muse-shrine, how many ideas and words you inspire! (Ep. 1.9.6). The processes of study and of composition are not viewed as different operations, nor is there a point at which the individual "shifts gears," stops researching and studying, and begins to write. Both processes require texts, both require judgment on the part of the busy autodidact, and both require that he select with care the subjects he writes on as well as the subjects he studies. The results of the autodidact's judgment and taste in selecting texts and topics appear in the public persona both of himself and of any scholarship which he writes up and makes public. As I shall attempt to demonstrate below, it is to such a reader, ranging from the young student seeking further indoctrination in the liberal arts to the mature adult with an established public career, that the Imperial miscellanist addresses his work. Once one is beyond the control of the grammarian and the paideutic curriculum, once one has finished a practical rhetorical education, he must acquire such learning as he can by seizing the opportunities which come his way. Artes and toTopia in text form increase the number of encounters which the seeker after knowledge may have with paideia. As we have seen, under the Empire the literate adult could access information through the book trade and the public and private library. Varro and Celsus, and later Augustine, Martianus Capella, and Boethius, made collections of artes. Quodsi [quis] ... 146 minus instructus erit magnarum artium disciplinis, Cicero had maintained in the protreptic preface to Orator (1.4), teneat tamen eum cur sum quern poterit If a person be less instructed in the fields of the liberal arts, still, let him make what progress he may. The self-learner may progress straight to the library, where he will find a variety of learned texts. The person who completes his education through focused reading will probably not concern himself with poetry. He has presumably exhausted the canonic authors in grammar school, or at least drawn from them all he can practically use. During the Imperial period, the classical poetry of both Greece and Rome is peculiarly associated with juveniles and early childhood education (Colson 1914: 46). Canonic poets have become more authority-icons than sources of direct inspiration for creative poetry, and while for example Dio or Philostratus may react to Homer, he does so by rewriting Homer's account as a personal response to a challenge as in Philostratus' Heroicus or Dio's Troicus. Perhaps this association contributed to the relative neglect of poetry by creative writers during this period. The studious adult must turn his attention to other sources more significant in terms of his status as an adult learner. Strabo for example recognizes in poetry the attraction of uDOoc, (tale) and TO Gauuaorov (wonder) — the element of ipuxaytoyia, (amusement) which makes poets attractive to children and iSiojTnc, 5e TTCXC; m i aTraiSeuToc; every uncultivated individual — and, influenced by the Peripatetic doctrine, sees in this attraction proof that man is by nature (biAetSfjuojv, a lover of learning. But he also states that grown-ups need more substantial fare: &vdyKr| ... Trpoorouoric; .. Tfjc; fjXiKiag em Tfjv TOJV OVTOJV ud9r|0"lV ayeiv as one gets older one must turn to learning about the real world (1.28). 147 F o r 6 7rpdTTcov, the man o f affairs functioning in the real world, the kind o f knowledge contained in Homeric verse may be evSocjov xai r]5v, aXk' OUK em jroAu • udXXov y a p aTTOuSdc^oUCnv, ibc; eiKOCj, TTEpi Td xpfiorua respectable and delightful, but not to be overdone; for [adults] are, properly, wore interested in useful lessons. For Strabo, it is the geographer who is especially concerned with xpflotua, but 6 Trpdrriov too is expected to be concerned with utility drawn from i a r a p i a and ua0f|uocToc, Kai yap TOUTCOV T d x p n a i u a d d uaXAov XTITTTEOV Kai TO TnOTOTepov in fact one ought always to seek out the useful and more plausible elements contained in these. (1.1.19) In his discourse Hxpi Xoyou daKfjaeioc, (Or. 18), Dio undertakes to direct the private education o f a wealthy, mature man who wants to participate in his city's government. To be taken seriously as a statesman, Dio recognizes, his friend must have technical training (2). But the friend has already acquired a respectable, level o f paideia (4), and it is not as a forensic orator (OUTE xPflCe i c. cHKavncfjc, Suvduecocj TE Kai 6EIV6TT|TOCJ you aren't looking for forensic capacity and force) that he wants to SUCeed but as a TToAlTlKOC, (5). This man has, in other words, already received a standard education in literature and in rhetorical technique, and wants to round out his background to become the kind o f cultured politician whom Cicero praised. With such a goal in mind, Dio recommends that the only poets to be experienced are Homer, Euripides, and Menander. In the case o f Euripides and Menander, they are to be listened to, absorbed, that is, as rhetorical phenomena, not read (6-7); other poets must be cut from the list: UEATI 5E Kai EXsyEia Kai I'auPoi Kai 5i9upauPoi TW UEV oxoX.f|v dyovn TTOXXOU a & a • TOO 8E TrpdrrEiv TE Kai dua Tdcj Trpd^Eic; Kai TOUC; Xoyouc; aucjeiv 5iavoouuev(p OUK av eir| Trpoc; auTd axoXf). 148 Lyric, elegiac, iambic and dithyrambic poetry are fine for the man of leisure. For the man of affairs who also wants to increase the range of his activities and his power of public speaking, there is simply no time for them. (8) Dio's list of helpful authors consists of orators and historians, in particular Xenophon, whose matter, style, and embedded oration cannot help but aid the self-taught statesman (14-17). Drawing upon earlier representations of the perfect statesman, the Imperial period is attempting to form a concept of the fully educated and therefore socially and morally responsible individual. This concept as we have so far examined it consists of a person who has acquired literacy and a foundation in his culture's paideia through a more or less standard training in the grammar school. He may have already built upon this primary education in the school of the orator, or is at least in the process of doing so. He now must use his own taste and good judgment in selecting further sources of paideia. Insofar as this paideia goes beyond the standard, it represents 7roXuua9ia. Such a person must likewise find the time, opportunity, and energy to spend upon this additional education. He may, for example, choose to listen to the lectures of the philosophers as part of his adult education, as Seneca describes his own mature relationship with Attalus in his one hundred and eighth letter. This attendance must, however, be an act of personal choice and commitment; no longer a child under the control of a master, he still needs to commit himself to self-education with attention, energy, and drive. Dio calls this approach to self-improvement chiAoKaAelv (18.1), Seneca studere (108.4). Others apply the term diligentia, diligens, the same term used above in the 149 discussion of the miscellanist's approach to TroXuuaGia. Ammianus uses the term to describe the emperor Julian furthering his education under the adverse conditions of an army camp in Gaul: Incredibile quo quantoque ardore principalium rerum notitiam celsam indagans per omnia philosophiae membra ... currebat ... poeticam ... et rhetoricam Graecam diligentius tractans ... et nostrarum externarumque rerum historiam, multiformem. It was almost unbelievable with what great passion he would track out the lofty learning of foremost subjects, how he would work his way through all branches of philosophy, with even greater industry studying Greek rhetoric and the various histories of our own people and of others. (XVI. 5.6; cf. Ensslin 1923: 37) For Nicolaus of Damascus the adult pursuit of learning and the arts is above all euSidycoyov, "amusing" but also "educative" (cf. LSJ Siaycoyfi 1.2). As we have seen above, (pp. 110-111), authors may liken the person who has received a literary enculturation to one initiated in the Mysteries. Diligence in further learning is the recognizable mark of such an initiate, and Gellius drives away from his Nodes Atticae all those unfortunate enough to fail to respond to its call. Erit ... optimum, ut qui in lectitando, percontando, scribendo, commentando numquam voluptates, numquam labores ceperunt, nullas hoc genus vigilias vigilarunt neque ullis inter eiusdem Musae aemulos certationibus disceptationibusque elimati sunt, sed intemperiarum negotiorumque pleni sunt, abeant a Noctibus his procul atque alia sibi oblectamenta quaerant. It will be best that those stay far away from this work and seek out other amusements for themselves, who have never derived any pleasure from study, research, writing, note-taking, those who have never made any efforts, have never stayed up late at night in such vigils, who have not been filed smooth by struggles and disputes with those passionately involved with the same Muse, but who instead are filled full ofpetty anxieties and busy-ness. (Praef. 19) What activities mark the Muse-initiate? All those, presumably, which serve to drive away the profane: reading and rereading, research, writing and note-taking, all 150 pursued with determined energy and at impracticable hours. Is Gellius describing himself or his reader? Is the person who can respond to Gellius' work a consumer of paideia, or a creator of it in his hours of study? In fact he is both. The individual seeking knowledge as a learner is also in a position to extend that knowledge to others, to share, with his associates in paideia, the fruits of his labors. This communal relationship among self-learners is at the base of the miscellany, as I shall attempt to demonstrate. Because he is working without a master, however, the adult learner's text-based and self-selected paideia can have its dark side. The autodidact, unlike the child at grammar school, lacks the supervision of a competent teacher. There is a variety of errors, of false directions, and of misconceptions to which he might fall victim. The grossest misconception involves the relationship between the learner and his text. Any fool with money, as Lucian mercilessly points out in his essay LTpocj TOV d7rai5euTov Kai iroXXd BiBAia WVOUUEVOV, can buy a whole library of beautiful and valuable texts. But Ti ocbeAoc, ... TOU KirjuaToc, OUTE EI86TI TO KOCXXOCJ auTiov OUTE xpnaousvto TTOTE; What good is their possession when you can neither appreciate their beauty nor ever be able to use them? (2) Lucian's victim seems to have lacked even a grammar-school education in literature (TauTa uf| uaGtov f|piv). Like Gellius, Lucian responds to this lack by adopting a tone of outraged religiosity at the man's pretense: aoi Kai uEuvfjaGai Mouatov dvoaiov ! For you even to mention the Muses is sacrilege! (3) A man who could assume that the mere ownership of texts brings paideia is of subhuman mental capacity, a monkey, a jackass, a dog (3.5). The adult autodidact may, however, in all good faith be led astray by a corrupt society's concept of paideia, of ^ EuSorouSEia as the figure is allegorized by Cebes early in 151 the Imperial period (Ross 1970: 218). In the dialogue known as The Pinax, Cebes' speaker describes a painting allegorizing modern society. Included is the figure of False Learning, a seemingly chaste and courteous female (5OKET rrdvu raGapioc, Kai euraKTog) whom those courting True Learning must first encounter before getting access to their true bride. Pseudopaideia is presented surrounded by her many deluded lovers: oi pev rroinTai ... oi SE pf)Topec., oi 5e 8 i a X s K T i K 0 t , oi 8E pouaucoi, oi 5e dpiGpniTKoi, oi be yewpETpiai, oi 8E darpoXoyoi, oi 5E KpmKoi, oi be fi5oviKoi, oi 5E TTEpiTraTnTiKoi, Kai oaoi dXXoi TOUTOIC, Eiai rraparrXf|aioi. Poets, orators, dialecticians, musicians, arithmeticians, geometers, astronomers, critics, Epicureans, Peripatetics, and all those others like them. (13.1-2) Not only has False Learning ensnared practitioners of the liberal arts, but she has captivated proponents of several of the more popular Hellenistic philosophies as well. (Cebes' omission of the Platonists and Cynics suggests his own philosophical position.) The Christian Tertullian will echo Cebes' frustration with a stupid system which turns a blind eye to what these authors see as the world's most obvious truths (Labhardt 1960: 216-218). It is significant that, as a Christian and a miscellanist, Clement had to specify the quality of his compilations from Hellenic paideia (e.g. Strom., and cf. Mahat 1966: 333). But even when society's general conception of paideia goes unchallenged, there are still plenty of more subtle ways in which the adult supplementing his education can go astray. Either the individual reading and studying for self-improvement focuses too closely upon subjects and skills which are trivial and without meaningful application, or he allows himself to be warped from a balanced temperament by an overindulgence or wrongheaded application to his subject. These involve basic imbalances: the learner selects 152 the wrongs things to learn — irrelevant rroXuua0ta — or he uses bad taste and judgment in incorporating his learning into his daily life. Cicero had been aware of the dangers of being distracted by sheer curiosity about things which did not really deserve the busy and intelligent adult's attention. In the De finibus he refers to this cupiditas discendi as the song of the Sirens: scientiam pollicentur quam non erat mirum sapientiae cupido patria esse cariorem They promise knowledge, in the eyes of one desiring wisdom dearer (no surprise!) than the fatherland (5.49). Seneca with greater panache dismissed all inane studium supervacua discendi meaningless zeal for unnecessary learning (De brev. vit. 13.3). All of the standard liberal arts come under attack by Seneca at some point in his work (Stuckelberger 1965: 21). He found grammar, as we have seen, especially easy to attack, though none was immune (cf. especially Ep. 48.6-9 on dialectics; Ep. 88.9-17 on music, geometry, and astronomy); that it was not the inquiry so much as the triviality of the arts' content to which Seneca objected is clear from his own essay into the natural sciences. However, the Quaestiones naturales, attempting to give rational explanations for nature's more startling phenomena — lightning bolts, rainbows, earthquakes — rest upon the moralistic purpose of freeing people from the wrong explanation to these questions supplied by the superstitious and the misinformed. Cum timendi sit causa nescire, non est tanti scire ne timeas? quanto satius est causas inquirere et quidem toto in hoc intentum anima. Since the cause of fear lies in ignorance, is knowledge to prevent fear not a valuable thing? How much better to seek out causes with a total commitment (QNat. 6.3.4). Seneca's eager approach to research may be compared to the orrou5r| of the miscellanist. Labhardt points to a similar dichotomy of response to 153 empirical learning in Apuleius' work. Although in the Metamorphoses Apuleius' Lucius is a character who gets himself into trouble because of his curiositas about the magic arts, Apuleius himself in Apologia refers with pride to his research into species of fish, faulting his accusers for being so stupid as to mistake scientific analysis for sorcery and sacrilege (Apol. 27.1-15; Labhardt 1960: 216). Elsewhere in the Apologia, Apuleius insists upon his status as philosopher because of his scholarly contributions to taxonomy (38) and experiments in optics (15-16). The self-learner could thus go astray if he were attracted to the wrong areas or subjects of study. But he could also fall short of his goal of a complete education by falling victim to certain imbalances in his own emotional response to the pursuit of learning. By definition the student — the zeal in Latin studere is echoed in Greek 07rou5aCeiv — pursues his goal with energy, diligence, and conviction. Qui ingenuis studiis atque artibus delectantur, Cicero states in De finibus, nonne videmus eos nec valetudinis nec rei familiaris habere rationem omniaque perpeti ista cognitione et scientia captos et cum maximis curis et laboribus compensare earn quam ex discendo capiant voluptatem. Just look how those who take delight in the liberal arts have no regard for health and property and in their passion for learning, suffer all things and compensate all that effort and labor with the pleasure they receive from learning. (Fin. 5.48) Voluptas can render the activity truly addictive. Imperial authors are aware of the difference between balanced and obsessed students, and the miscellanists present them most effectively through the dramatized 154 portions of dialogues and through vignettes. Gellius, for example, draws a portrait of the scholarly neurosis of otpiucxOeia.12 The opsimath suffers from vitium ... serae eruditionis ... ut quod numquam didiceris, diu ignoraveris, cum id scire aliquando coeperis, magni facias quo in loco cumque et quacumque in re dicere. A fault of late-born erudition, which makes you speak out as a thing of great importance and on absolutely any occasion what you were long ignorant of and Just lately became familiar with. (NA 11.7.3) The opsimath, delighted with the new words and ideas he has learned but not having the restraint and the patience to wait until he has learned their correct context, violates usage by displaying his new acquisitions in all the wrong places. The humorous side of oipiuaOeia is displayed later in Gellius' chapter, but a less pleasant encounter occurs at NA 15.30, where an over-confident opsimath actually lies to Gellius to defend his version of a word's etymology. Gellius later checks the word in Varro and discovers the cheat. A similarly neurotic bent is showcased in Lucian's Lexiphanes, the portrait of a man who has learned many impressive Attic words but uses them with unintentionally hilarious results (Lex. 2.1 - 15.7). Athenaeus' Cynulcus teases Ulpian with accusations of similar misbehavior (3. 52-53; cf. 9.29). Another kind of scholarly imbalance is described at Nodes Atticae 1.10. Here Favorinus criticizes an adolescent who insists upon making colloquial use of words which he has come across in his reading but which no longer represent current usage. The speaker claims to have made a moral choice in selecting such diction, for the words carry with them antiquity's aura of righteousness and authority. Favorinus insists that clarity is 1 2 In the Characteres Theophrastus presented the first detailed picture of the opsimath (27). It is significant for the change in the concept of the contents of paideia that Theophrastus' opsimath learns primarily processes, i.e. how to sing the latest songs and execute newly-acquired gymnastic techniques. 155 to be preferred to age, advising the young man vive moribus praeteritis, loquere verbis praesentibus Base your life upon the morals of antiquity, but speak with the words of today. The sentiment struck Macrobius as particularly appropriate; he compiled it arid worked it into the first book of the Saturnalia, where he has the abrupt Avienus use this phrase to mock the grammarian Servius' discussion of archaic terminology (1.5). These authors emphasize that the self-directed learner must exercise care over the kinds of subjects he studies and the extent of detail in which he pursues them. No author sets precise guidelines, but insists only that the autodidact pursue a balanced and socially responsible relevancy. Obviously such relevancy depends upon personal judgment; the same society that read Seneca's attacks upon supervacua studia also read the abundantly detailed and diverse Historia naturalis of Pliny the Elder, and both authors were absolutely committed to relevancy (cf. Plin. NH 2.1). As we have seen, the term TroAuucxGia as used by authors under the Empire mirrors a flexibility of attitude toward knowing a lot of detailed information which appeared in Greek culture as early as Heraclitus. If the earlier centuries had distrusted TtoAuucxOicx as implying the learning of many skills and ways of explaining the world which were distracting and mutually exclusive or contradictory, by the first century BC authors like Strabo could praise TroAuucxGia as a necessary component of convincing scholarship (e.g. 1.1.1.; 1.1.12; 1.2.20; 16.2.10), while for Plutarch it had become a generally attractive quality and a term of mild praise.13 From Athenaeus through to the Byzantine period, TroAouaOia has become 1 3 On one occasion Plutarch found a mildly undesirable habit associated with TtoAuuaGia. At De garrulitate 519c he notes that Tfj UEV ouv TToAuuaBEia xr|v TToAuAoviav STOOGOU auu(3aiv£i (816 Kai IIu0aY6pag sTat;s TOIC VEOIC; TrEVTaETfj atioTrrjv, ExsuuBiav n p o a a y o p e u o a g ) .... a yap r)5£coc d K o u o u a t v r)5ecog AaAouai, K a i a Trap' aAAcov o7Tou5fj auAAsyoucn npdc, srspoug usra x«P«g £K(j)epoucnv. Garrulity is a more or less natural consequence of polymathy. That's why Pythagoras required of his young followers a five-year silence .... For 156 a very positive assessment among people concerned with ancient Hellenic paideia, implying a thorough grounding in the classical authors and in sound scholarship, and the good character which Imperial society associated with paideia.14 From what source is the adult learner to find guidance in acquiring relevant TroXupaGia? The Imperial Miscellanist and the Adult Learner The Imperial miscellany conveys a polymathic data-base relevant to such needs. Whether this data-base funds further education in the liberal arts, provides copia for public discourse, or simply feeds the learner's passion for information, it is relevant insofar as it bolsters and legitimizes his status, in the public forum as well as in private circles, as an educated, enculturated man. The miscellanist is himself such an educated adult; his diligentia and Kpimc, in amassing his polymathic collection substantiate his position. In undertaking to provide polymathic data for another such adult reader, the miscellanist assumes the patronizing tone of an older person addressing a younger. This implied personal bond between compilator and reader is one of the basic qualities of the Imperial miscellany, and more what they delight in hearing, they delight in talking about, and what they gather together from some with industry, they exhibit in the presence of others with delight. At De garrulitate 514c, the learner's own eagerness may hinder him in gathering more information, especially if he is given to TroXuXoviot: E V ioropiaig 6 dvaYvwcmKOC, E V TEXvoXoyiaig 6 ypauuanKog, E V 5inYnuo:ai gsviKotg 6 rroXXriv xwpav E7i£XnXu6cbg K a i 7T£7rXavnuEvog ... av U E V ng suraon Xoyog, EE, ou uaGsTv n 5uvaxai Kai TiuGsaGai Ttov dvvoouuEvcov, T O U T O V sgtoOEi K a i E K K P O U E I , uioSov O U T I O (3paXUV 8ouvai T O Oltonfioai un 8uvdu£V0g. These individuals [are particularly subject to noXvXoyia]: the great reader when the conversation turns to historical research, the grammarian when details of his art are being discussed, and the world traveler when the topic is foreign affairs....If some discussion arises from which he might learn something of which he is ignorant, his inability to keep his mouth shut makes further learning impossible. 1 4 Athenaeus' assessment of Chrysippus is a clear example of this association. At 13.18.6, Athenaeus prefaces a quotation from Chrysippus' Ilspi TOU K O X O U K a i Tfjg r)5ovfig with the words, O U K araipwg 5', ubg suauTov 7rei0to, UEUvrioouai Tfjg XsgEcog- xaipw yap n&vu Tto dv5pi 5id T E ifiv rcoXuuaBiav K a i T T I V T O U fjGoug ETTlElKEiav. / am quite certain that his words are appropriate here. I admire the man very much both for his polymathic knowledge and for the probity of his character. 157 than any other sets it apart from the other polymathic collections reviewed above as well as from the handbooks of artes. Gellius and Macrobius dedicated their works to their sons, and in so doing took their places in a long and broad tradition, putting their accumulated wealth of learning and wisdom at their sons' disposal as part of their patrimony. The literary convention ... is a compound of actual practice and normative pressure: it reflects both the fact that a father supervised his son's education and the belief that such was the father's proper role .... No professional grammarian we know in later antiquity dedicated a work to his own son .... The professional's distinguishing mark was his stepping aside from his role as father. (Raster 1988: 67-68) Although the prologue to the Stromateis is missing, Clement too expresses a solicitude for his reader's correct reception of his polymathic collection (e.g.;; As I shall attempt to demonstrate below, Aelian in the Varia Historia (a work also lacking a prologue) adopts toward his reader both a patronizing attitude and an ingenuous style. The miscellanist's adoption of a paternal attitude as opposed to that of a professional educator directly influences the polymathic content of the miscellany, because it aligns the data-base with the reader's position within the cultural elite. In the miscellanist's implied view, his compilation is as much a part of his estate as is his real property (referred to as a KeipqAiov by Aelian [NA Epilogue], a storehouse [penus] by Gellius [NA Prologue 21], a scientiae supellex by Macrobius [Sat. Praef. 2]); it is, figuratively speaking, as effective as a physical legacy in assuring the heir's social status. The value of this intellectual legacy demonstrates not only the scholarly industry of the "father," but his discrimination, good judgment, and sense of responsibility. 158 The miscellanist's paternal stance as opposed to that of the professional's can be taken one step farther. The professional grammarian or rhetorician is necessarily limited in scope, for his contact with his reader is in one area of paideia alone. The father, however, is concerned with the entire range of his son's paideia-acquisition, and especially with the son's successful adoption of paideia into his adult life. Hence arises the tone of intimacy and direction in many chapters of Gellius and Aelian. The miscellanist's solicitude for the reader's correct use of paideia in adult life leads to the presentation of his compiled data-base in two basic forms: as directly compiled copia or uXn per se, and as copia presented with a dramatic modeling of paideia-acquisition and paideia-utilization. The amount of attention the miscellanist gives to each, the stylistic treatment of individual chapters and discussions, and the extent to which directly compiled material is reworded or rewritten, will depend upon the miscellanist's own taste and judgment. Gellius, for example, intersperses chapters consisting variously of anecdotes, lexicographical discussion, or direct compilation from philosophical, historical, or technical writers, with vignettes representing, he claims, his personal experiences in acquiring paideia. In these latter vignettes, Gellius may write himself into a scene in which people are constructively or incorrectly incorporating paideia into daily life (e.g. NA 9.9, 13.20, 13.25, 16.119.7), or into a scene in which a well-known paideia encounter is reenacted in "real" life to make its message more vivid and effective (cf. Holfred-Strevens 1989: 47-51, Steinmetz 1982: 283-285). At NA 19.1, for example, Gellius claims to have relived an experience related of Aristippus in a familiar chreia: the philosopher who 159 asserted that he was indifferent to danger and death had blanched during a rough sea-crossing, and needed to defend himself against a charge of hypocrisy (cf. Aelian VH 9.20). In the case of Aelian's Varia Historia, as I shall discuss in Chapter 5, this presentation of anecdotes and chreiae directed toward modeling correct paideia-responses is uppermost; Aelian, however, chooses to lecture the reader directly rather than to allow him to draw his own conclusions about the material's message.15 For Clement, on the other hand, paideia-models tend to be limited to lists and registers; for this miscellanist, Biblical figures carry more authority than Hellenic ones do (cf. Mahat 1966: 184-187). Athenaeus and Macrobius (perhaps Pamphila too, cf. Chapter 4 below) preferred to construct a narrative framework of a banquet, symposium, or informal holiday 1 5 The function of paideia-modeling is discussed by other Imperial authors. Plutarch, for example, in De profectibus in virtute, refers to people who seek excellence through study: svioi Ss xP£iag K a i ioropiac dvaXeYOuevoi Trspuaaiv .... Kai Trpoxsipd ye 8EI K a i o"uxv<* T C O V sm^ avcov K a i dyaGcov dvSpcov E X E W d7TO(|)0EYMaTa K 0 ( i U V H U O V E U E I V npdc, T O U C ; Suacoirouvrac; .... K a i Troirjuaoiv ouiXcov Kai ioropia 7rapacj)uXaTTE aeauTOv ei ur|§£v O E Sia^ suysi T C O V rrpog STravopGcoaiv fj0ouc M T T & O O U C ; Koucfuauov suueXcog X E Y O U S V C O V . Some people occupy themselves with gathering anecdotes and chreiae .... One must have many of the sayings of men recognized for their probity at hand, and mention them when confronted by those who would discourage us .... When you study poetry and history be careful that nothing escapes you which could benefit your character or lighten suffering. (79c) Plato and Xenophon as particularly profitable sources of moral models: T O U C ; M E V yap nXdcTcovi Kai H E V O < ] > C O V T I xpwuevouc 8id T T J V X E 5 V , E T E P O V 8E M H S E V dXX' T O K a 0 a p 6 v T E Kai ' A T T I K O V coarrsp Spoaov Kai xvouv dTroSpETtouEvouc; T I av aXXo 4>otir)Cj fj cfiapudKcov T O L I S V sucoSsg K a i d v 0 n p o v dYaTTdv, T O 5' dvcoSuvov K a i KaBapriKov ur) 7tpoai£O0ai HHOE 8iaYiYvwaK£iv; dXX' oi YE uaXXov E T I T T P O K O T T T O V T E C ; O U K O C T T O . Xoycov uovov dXXd Kai 0£audTcov K a i H'paYpdTCOV rrdvTCOV C0(J)£XEra9ai 5 u v a V T a i Kai CnjvdYElV TO 0iK£l0V Kai XP'IO'MOV- How would you describe those who make use of the writings of Plato and Xenophon for style alone, culling as it were nothing other than the pure Attic diction as though it were so much dew and blossom? Aren't they like those who would use a drug because it tastes or smells good and not because it relieves pain and removes impurities? So those who progress in virtue can benefit not only from all the descriptive and lexical material [in Plato and Xenophon], but can also gather that which is relevant and useful to self-improvement. (79d) Moral models can function as charms and spells: T O I Q T O I O U T O I C ; TtapETtETai T O . . . Ti0£o0ai npo 6(])0aXucov roug o v T a g dYaQouc; fj yEvousvoug, K a i 8iavo£io0ai -ri 8' av £7Tpat>v E V T O U T C O tlXdrcov, T I 5' av S I T T E V 'ErraiuEivcbvSag, jrotog 8' av coc}>0r| AuKoOpyog f| . 'AYnoiAaog, oiov T I rrpog E o o T i T p a Koopouvrag sauroug rj UETappuOniCovTag fj cj)covfjg dY£W£OT£pag aurcov £mXau|3avou£voug fj rrpog T I jrdGog dvn(3aivovTag .... fj 8 E T W V dyaOcov dvSpcov enivoia K a i uvfjun raxu napiora)ievr| Kai 8iaXau(3dvouaa r o u g TrpoKOJTTOvTac], ev nacn nd0EOl Kai djTOpiaiC dTldaaig 6p0OU<; TE Kai dnTCOTaC 8iacj)uXdTTei. After this one should set before the eyes examples of men who are virtuous or who are becoming virtuous, and one should consider carefully what Plato did, or what Epaminondas said, what kind of man Lycurgus or Agesilaus was seen to be, as the dressing or practicing before a mirror, reprehending some unworthy utterance or confronting some unpleasant experience .... The recollection of the models of good men, coming to the aid of and inspiring those advancing in virtue, keeps them on their feet and stable in all their sufferings and quandaries. (85a-b) 160 gathering for the presentation of their miscellaneous collections.16 The potential which such scenarios have for modeling good and bad behavior was known at least since Xenophon (Flamant 1968: 303-319). At the relaxed social event, men could candidly demonstrate how they succeeded or failed at incorporating their expressed views or doctrines into their daily lives. For Athenaeus the desire to demonstrate the impact of polymathic learning upon man's life as displayed in social interaction occasions the frame. In the Introduction to Book 5 we read what may well stand as Athenaeus' goal for the Deipnosophistae (cf. During 1936: 249): TO pev Trdpepyov epyov obc, rcoioupeOa, TO 8' epyov cbc, rcdpepyov eKrrovoupeOa. We hold our avocation to be our main concern, we labor at our work as though it were our avocation. The orrou5r| / diligentia with which Athenaeus made his collection — he was probably compiling at the Alexandrian Library (During ibid.: 238; Schmid 1924: 788) is figured in the spontaneous and candid polymathy of his characters.17 And as Raster has demonstrated, Macrobius, too, gives much attention to the ethical implications of polymathy in constructing his miscellany's narrative frame (cf. Raster 1980: 238-239; 1988: 60-62). Here we may consider in more detail three models of the oXwc, rcerraiSeupevoc; dramatized by Imperial miscellanists, characters who are presented as having made good paideia selections and enjoying the results of their educational choices in their current ,6Other kinds of polymathic collections during the Imperial period may use such a format, e.g. Plutarch's Quaestiones conviviales (a Xooeic -collection), [Plutarch]'s De arte musica (a technical manual). 17 Athenaeus apparently created the frame of the Deiphnosophistae by first amalgamating several earlier polymathic dialogues and then supplementing these with liberal additions from his notebooks; cf. During ibid.: 237-241. 161 social positions. In all three cases we see these men in the context of relaxed conversation among friends, when they can behave most naturally, least self-consciously. According to Gellius' characterization, Cornelius Fronto wears his education, as it were, effectively and well. Although the chapters of the Nodes Atticae are, like those of Aelian's Varia Historia, each distinct and independent of the others, Gellius yet manages to weave this character into five similarly structured chapters of the work. Each time that he appears, Fronto directs a conversational inquiry or quaestio into the precise meaning of a Latin word. In the first chapter in which he is featured (NA 2.26), Fronto is described as consularis, his prestigious position as a responsible political figure thus typing him as a learned amateur in opposition to figures such as Favorinus, specified in this chapter as a philosophus and therefore in career terms a public teacher-lecturer, or Sulpicius Apollinaris (NA 19.13), a professional grammarian. Fronto is, then, a learned but nonprofessional connoisseur of paideia. (Gellius makes no reference here to Fronto's historical relationship as tutor at the Antonine court). But Fronto does have a political career, and as a statesman is aware of the demands upon one's time which public responsibilities can make. At NA 19.8.14, for example, he dismisses his listeners gracefully: Quin his quoque ipsis, quae iam dixi, demoratos vos esse video alicui, opinor, destinatos But I see that you have other business, and have been delayed by my discussion, urging them to pursue their inquiry independently. Twice Gellius explains why Fronto himself does have the opportunity for such discussion: he is forced by an attack of gout to stay at home (NA 2.26; 19.10). Unlike Herodes' moralistic dismissal of a posturing student (NA 1.2), Fronto's approach to his inquiries consists not in extracting an ex cathedra textual 162 statement from a written authority, but in amassing instances of usage recorded in text form. (On the one occasion on which he consults the authority of Caesar's De analogia [NA 19.80] he challenges his listeners to find exceptions to Caesar's statement.) Fronto's discussions of usage require a profound memory and an energetic commitment to exhaustive research. For he is engaged in these chapters not in a Formalist-style discussion of "levels of ambiguity" in Latin diction, but in something more exclusive and exact and in fact quite opposed to the ambiguous. Fronto wants not a range of connotations but a word's precise denotation, honed by ancient usage and fixed irrevocably in canonic texts. The discussion of Latin color words (NA 2.26) is an exercise in precision, and no word is too trite or dull to escape a word-search in archaic usage (NA 19.10). In some chapters Gellius collects the lexical windfalls which result; the color words are, for us, a lucky acquisition. Elsewhere — praeterpropter, the etymology of nanus, the correct number of harenae — the results of the quaestiones are not so valuable as the example set by Fronto is protreptic. While Fronto's conversations are themselves purissimi and bonarum doctrinarum pleni, the man in action is an inspiration to others. Unlike Aelian and Gellius, Athenaeus constructs a dramatic setting for his entire work, and allows his characters continually to interact with one another. Among Athenaeus' twenty-three deipnosophistae, only Larensis, the host of the dinners recounted in this text, can be considered an amateur of paideia. All the rest can be typed as poets, philosophers, grammarians, musicians, physicians. (Although Ulpian too is introduced as a statesman, his obsessive Atticism makes it difficult for us to see in him the individual who 163 has integrated his learning with a responsible public career.) Larensis is described by Athenaeus' epitomizer in this way: vno (fuXoTiuiac JTOXXOUCJ xtov and TtaiSeiac. ouva0pi£tov ou uovov xoTc. dXXoic dXXd Kai Xoyoic eiaxia, rd pev rtpopdXXtov xwv dcjitov £nxr|0Etoc., xd 8e dveupioKtov, OUK dpaaaviaxioc ouS ' EK TOV jrapaxuxovxoc xdc t/|xfjo£ic TTOIOUUEVOC, dXX ' toe evi pdXiaxa pexd Kpixucfjc xivoc Ka i EtoKpaxiKfjc. emoTfipng, toe navrac, Gaupd^Eiv xtov £nxn.o£tov xfjv xtipnotv. Xeyei 8' auxov Kai KaSeaxapevov erri xtov iepwv eivai Kai Guaitov UTTO xou Tidvxa dpioxou PaoiXetoc MdpKou Kai iir] eXaxxov xtov rtaxpitov xd xtov 'EXXr|viov pexaxeipiCeoGai. raXe! Se auxov Kai ' Aoxeporcaiov xiva, ere' tone, dpc|)oxeptov xtov (jxovtov rtpoioxdpevov. Xeyei 5 ' auxov Kai eprceipov eivai iepoupvuov xtov vopioGeiawv UTTO xe xou xfjc rtoXetoc. eraovupou 'PtopuXou Kai riopmXiou Noupd Kai emoxripova voptov JTOXIXIKWV. rtdvxa 8e xauxa povov e^ eupeiv EK naXaiwv ti;r|<biapdx<ov K a i Sovpdxtov xnpnaewc, exi 8e voptov auvaytOYfjg, oi3c exi SiSaoKouoiv ... Kaxaaeoiyaapeva UTTO xfjc xwv rtoXXwv dtJuXoraXiac. r|v 8e ... Kai PipXitov Kxfjaic auxw dpxaitov 'EXXnviKi&v xooauxn toe uTtepPaXXeiv rcavxac xouc em auvaytovfi xsGaupaopevoucj.... Siorcep ... -dyXaiCexai 8e Kai pouaiKdc ev dtoxip • oia rcaiCopev cfnXav avSpec dpcjn 0apd xpdne^ av, Kaxd xov ©nPaTov peXonotov. He eagerly gathered together many cultured men and gave them both a real feast and a feast of discourse, proposing some points of discussion that deserved examination, also finding solutions himself. He set the subject of these examinations not in an extemporaneous or unexamined manner but with as much judgment and Socratic spirit as possible, so that all wondered at the way he studied these inquiries. [Athenaeus] says that he had been appointed to certain religious offices by the emperor Marcus, and administered Greek no less than Roman public cult. He calls him Asteropaeus because he had an equal facility in handling both languages. He says that he was both expert in the priestly rites established by that Romulus who gave his name to the city and by Pompilius Numa, and that he was also expert in jurisprudence. He had learned alt these things by himself by studying ancient decrees and documents, and further by making a collection of laws which they no longer teach ... but which have been silenced by the apathetic bad taste of the majority. His collection of ancient Greek texts was so great that it surpassed the collections of all those who have ever been admired for their libraries. Hence, He glories even in the flower of culture, in such amusements as we frequently delight ourselves with when we men gather around the hospitable table. As the Theban poet sings. Larensis has all the social graces, kept in perfect balance. Athenaeus' quote from Pindar is meant to link Larensis with Hieron, the king cultured enough to appreciate true paideutic merit (cf. Aelian's depiction of Hieron at VH 4.15 and 9.1). As host Larensis welcomes 164 into his home both Greeks and Romans. He is csuunocriapxog (4.50); in three of the twelve contributions he makes to the discussion, his statements conclude a book. His education in both Greek and Latin is flawless, putting him in the position of artistic critic as well, able to judge with discrimination and yet displaying no compelling tendency in any one art. Larensis knows enough about all the arts to direct learned discussion without appearing the expert in any one field of study. Athenaeus has noted his lavish library holdings with approval. These texts will have been the source of his rich fund of quotations displayed throughout the Deipnosophistae. Larensis has made a collection of obsolete religious and community laws, preserved through his Tfjpr|cncj : careful study and observation. The same word is applied to Larensis' care in formulating conversation topics, TOJV £,r\TT)OEU>v Tf)v Tfjpr|cnv. Because of this interest, Larensis is the spokesman for Roman historians and Latin terminology throughout the body of the text, but is as capable of quoting long passages of Greek poetry as is any of the guests. He wears his learning thoughtfully, socially, with good taste and discrimination (cf. Baldwin 1976: 37-38). Like Fronto and Larensis, Macrobius' Praetextatus is an historical reality (cf. Ammianus 22.7.6) written into the Saturnalia. Drawing upon Praetextatus' position as a statesman to model the fully educated man, Macrobius dramatizes his social graces as an example of the learned amateur's behavior in society. As in the Deipnosophistae, the contents of this miscellany are woven into the dramatic framework provided by the leisure of the Saturnalia. It is precisely because it is a public holiday that Praetextatus and the others can have these conversations. Macrobius' work displays the educated individuals at a time when they can relax and candidly display their personalities as products of their 165 education. Like Athenaeus' host and guests, the celebrants of the Saturnalia represent various practitioners of the liberal arts. But of the twelve participants in Macrobius' text, half are prominent career-statesmen at Rome (Davies 1969: 4). The entire cast is referred to as Romanae nobilitatis proceres doctique alii (Sat. 1.1.1), and as viri et docti et praeclarissimi (1.1.4). As the first and most prominent host of the various gathering, Praetextatus not only throws his dining room and library open to his guests, but directs, as Larensis directs, the topics of conversation; indeed, even when the group gathers at the homes of others, Praetextatus still functions as master of ceremonies (e.g. Sat. 7.1.1; 7.4.1-2). Unlike Gellius' Fronto and Athenaeus' Larensis, however, Praetextatus does not so much initiate quaestiones as "access" people as information sources, as though they were themselves compilable texts. At Sat. 1.2.20, for example, he calls upon Caecina for information verum quia te quidquid in libris latet investigare notius est quam ut per verecundiam negare possis because it is too well known that you track down whatever information is found in books for you modestly to deny that fact, and Caecina then answers by enunciating within the framework of the conversation the information from Varro found in Gellius' NA 3.2. As Macrobius had already excused this direct compilation of entire data-sources (Sat. Praef. 4-9), we must view Praetextatus' rather artificial manner of eliciting information, as well as the equally artificial response, as part of Macrobius' program. Macrobius is aware of the artificiality; at Sat. 1.11.1., for example, after Praetextatus has spoken at great length on the antiquities related to the Saturnalia celebration and drawn upon various earlier sources to do so, a guest accuses him of flaunting ingenii sui pompam et ostentationem loquendi. The other guests shudder in distress at such an unscholarly (and metafictional) 166 attack; Praetextatus, elegant and urbane, simply smiles and calmly takes a new tack in his discussion of antiquities (on the smile as a gesture of scholarly superiority cf. Kaster 1980: 238-239; Pliny the Younger likewise refers to accessing information from a learned friend as from a thesaurus, Ep. 1.22.2-3). His flexibility and charisma carry him across the interruption. Like Larensis, Praetextatus is a master of Roman antiquities because he has made a careful study of the sources for early Roman religion: sacrorum ... omnium ... unice conscius (Sat. 1.7.17), he is the princeps religiosorum (1.11.1). The object of his private studies reflects his personal piety and religious convictions, which are in turn funded by the material he has studied. Another guest describes Praetextatus as nec in moribus Socrate minor et in re publica philosopho efficacior not less than Socrates in morality, more effective than that philosopher in politics. (Sat. 2.1.3) The three examples of the oAiocj 7re7rai5EuuEvoc, which we have here examined display the fruits of their focused and consistent scholarship while enjoying the social status which their active political careers have won them. They have developed both the private (intellectual) and the public (moral and responsible) sides of their personalities, and all three hold important positions in the Roman state. Though these three men are Romans by birth, this circumstance is not a requirement to status as a fully educated individual. We could, for example, include Gellius' Herodes Atticus in their number, vir facundia et consulari honore praeditus, whose approach to textual authority has been referred to above. The important qualities they share, qualities which are being modeled in their approaches to scholarship and the educational process in relation to their own careers and personal responsibilities, are a serious and directed commitment to paideia, prodigious 167 memories stocked with data worth remembering, a taste and judgment which allow them to recognize the difference between trivial and meaningful scholarship, and an admirable personality influenced by the moral content of paideia. These are, interestingly, qualities to which both Gellius and Macrobius lay some claim in their prefaces. Although the Varia Historia is lacking a preface, Aelian's prologue and epilogue to the De natura animalium likewise claim similar qualities for this author, to be discussed in more detail below. The dramatic scenes at the symposium, at the dinner party, in conversation with friends and acquaintances allow the educated individual to display his learning in a spontaneous manner. In the give and take of discussion, the speaker cannot have a dissertation prepared, and must access his learning in an ad hoc way. As readers we build up, metonymically, some concept of the vast range of reading and study carried out by the 6Xwc nenmbeviievoq . In a sense, his learning is more sincere, more selfless than that of the public teacher, sophist, or philosopher, who has a performance in script or sketch, or has at least a few moments to prepare an impromptu discourse. Philostratus speaks at length of the sophist's admirable powers of avroaxebia^ew, improvisation. Quintilian insists that this ability is the final goal of an oratorical education (10.7.1; cf. Cousin 1935: 602). The educated men we have examined here display a similar ability in auTooxe5idCeiv , but on the private, text-bound level. They can mentally reproduce data to respond to any given scholarly need, whether as fuel for a quaestio or as solicited background information in a discussion of antiquities or natural history. That this ability is limited to the oAooc, TCTrai5eupevoc, is displayed on the occasions when others fail to reach his level; the semidocti grammatici in Gellius' NA 4.1, 19.10, for example, become tongue-tied and 168 embarrassed through their inability to provide the necessary information to support their positions in an argument. That Athenaeus at least is playing with this concept of auTooxeoidCeiv in a semiprivate context is suggested by his title; although most of his characters are not sophists in the professional sense, they all do get to perform at dinner like the great public virtuosi on stage. How does this display of polymathic paideia on the level of the intimate paideia-gathering relate to its text-based status and the goals of the miscellanist to provide truly useful polymathic data? Being bound to a textual vehicle and representing a culture now nearly a millennium old, Greek and Roman literature were acquired, as we have seen, first by the child under a teacher's directions, a necessarily involuntary and public experience. The busy adult acquired further learning in literature, the other liberal arts, and the world in general, according to his own energies, drive, and commitment. The information and skills he acquired gave him pleasure and improved him as a person, but the latter growth did not become apparent until he interacted with others; material he learned would lie dormant until it took on a moral quality in social interaction. Just as archaic and classical paideia as performed on the community level had educated the public through poetry, drama, and music, so the scholarly conversation, quaestio, and symposium were a more intimate educational experience for their participants. Polymathic data are derived through study by individuals and presented to the social gathering as a piece of text-based information, an epavoc, Tfjc; roAuuaSeiac,.18 By becoming a clearing house, as it were, for each individual scholar's contribution to 18Dionysius Hal. De imit. 31.1.27 Us.; Athenaeus refers to Larensis' banquets as supplying party-favors consisting of Aovdpia (2.1). 169 knowledge, the discussion becomes a counterpart on the level of the personal and friendly gathering to the miscellanist's own solitary study session, better because representing the best of each participant's compilations from written texts. The element of selection determined by taste and good judgment is still operative, and the 7roAuua0ia conveyed to the reader of such a miscellany can therefore be only of the beneficial kind. 170 Chapter 4 Miscellany Structure and Style In analyzing the generic qualities of the Imperial miscellany, the miscellanist's approach to content and its selection was discussed in Chapter 2. In Chapter 3 the miscellany's reader was discussed, in particular the needs which the reader's social position as a member of an educated class imposed upon the miscellanist's selections and tone. In the present chapter my purpose is to examine the third element in the miscellanist's program: the miscellany's structure and style. To what extent do the surviving miscellanies display a recognizable structure? To what extent do these features contribute to the readability of the text? I shall attempt to demonstrate the following: (1) The miscellany shares, with a number of other works of Imperial scholarship, an origin in the process of compilation. (2) The miscellanist approaches the task of composing his compiled notes into a finished work by applying models of blending and cosmetic arrangement, drawn from nature and the arts. (3) The miscellanist purposely retains a quality of disarrangement and disorganization in his arrangement. It is in this quality of the finished work that the "entertainment" value of the miscellany is viewed as lying, and not in a trivialization of content. Paideia in the form of canonic literary texts used in schools, works of scholarship contributing to the understanding of those texts, and the writings of generations of philosophers, scientists, historians, and other intellectuals made the Imperial miscellanist's 171 task of selecting relevant polymathic data especially demanding of diligence. As we have seen, the flood of texts which the Imperial reader must navigate was daunting. To adopt a different image, there was much chaff among that crop of literary wheat, and the miscellanist needed to work hard to thresh out only relevant data for a reader with whose intellectual development he was really concerned. Compilation had been a legitimate means of manipulating textual material since the appearance of the written text in classical culture. In the Laws, for example, Plato refers to a text of anthologized poetry (81 le. 1-5). Papyrus fragments of such reading primers have been found in the Fayum from the third century BC (cf. Gueraud and Jouguet: 1938). Aristotelian "school literature" as represented by significant portions of the current Corpus Aristotelicum was built up in part through a directed program of compilation of a variety of written documents (cf. During 1950: 58).1 We have examined examples of Hellenistic compilations and some of their authors' responses to their sources. In some cases these sources are previous compilations; thus Antigonus of Carystus, for example, made selections from Callimachus' paradoxographic collection of over a dozen identified authors (Giannini 1964: 108 note 45). Scholars of the Imperial period did not hesitate to compile previous collections; indeed, much of the limited scholarship expended upon Aelian in the last century has been occupied with determining his sources. However, Imperial compilators treated these compiled authors each as a voice with its own authority irrespective of its context. The secondary vehicle of the voice may well be overlooked; Aelian quoted Aristotle frequently 1 Epitomization, a process of textual manipulation analogous to compilation, is first recorded in the fourth century BC, of Theopompus' epitome of Herodotus (Suda theta 172; Bott 1920: 6-7, 11). 172 from secondary or tertiary compilations because Aristotle's was the voice whose authority he wanted to access, not that of Pamphilus or Juba. Pausanias offers an interesting example of an Imperial scholar analyzing, manipulating, even wrestling with precompiled sources. In the course of his research into the antiquities of Helicon, Pausanias drew information from a poetic Atthis attributed to Hegesinus and compiled by Callias of Corinth. TCU3TT|V TOO 'Hyqoivou TTJV Tcoifjaiv OUK ETTEXEgapnv, Pausanias rather wistfully admits as he records in his own text the material he can use, dXXd rrpoTepov dpa EKXeXoimna qv rrpiv q ep.e yivEoGai • KdXXirrog be KopivGiog EV xfj Eg 'OpxopEvioug auvypa^fj paprupia HOiEirai Tup Xoyto rd 'Hyeaivou eirr\. oboauTcog 8E Kai qpEig TTETroiqpEGa reap' auTou KaXXirrou 5i5ax6evTeg. / didn't read for myself this poem by Hegesinus, for it had been lost before I was born. But in his study of the people of Orchomenus Callipus quotes Hegesinus' lines as corroborative evidence, and thus instructed by Callipus I quote them in turn. (9.29.1) The previous chapter's discussion of the educated individual of the Imperial period focused upon this individual's uses of rroXupa6ia in a social context. In the miscellanies of Macrobius, Gellius, and Athenaeus, we see educated adults furthering each other's intellectual development by funding discussions with TroXupaGia. Paideia is presented as an important element of social intercourse, and one which contributes to the spiritual well-being of the peer group. But the educated adult's most concentrated and meaningful encounters with paideia occur in private, faced with his texts and prepared to make the best use of the time at his disposal. Here the image we have of the Imperial oXtog rreTraiOEupEvog is that of the scholar with notebook in hand, extracting from his texts the material which in his own 173 judgment is most useful and relevant. His compilation reveals his judgment, for not all readers will find the same data equally relevant. Seneca refers to this process in Ep. 108, describing the manner in which people with different interests approach the same authoritative text: non est quod mireris ex eadem materia suis quemque studiis apta colligere; in eodem prato bos herbam quaerit, canis leporem, ciconia lacertam. You shouldn't be surprised at the fact that each person gathers from the same material what suits his own interests. In the same meadow the cow heads straight for the grass, the dog for the rabbit, the stork for the lizard. (29) Plutarch uses a similar image: as the bee, goat, and pig hunt out different foodstuffs in the same locale, so among reader of literature 6 uev djravOiCeTca TT]V ioropiav, 6 8' epchberai Top KaXXei Kai Tfj KaTaoKeofj TCOV ovouctTtov. This person plucks the flowers of history, that one is attracted by the beauty and the arrangement of words. (Quomodo poet. aud. 30c) The reader's notes thus become a kind of barometer of the reader's responsible scholarship and taste, and rejection of offensive data may be as indicative of the reader's intelligence as acceptance of relevant material. This situation arises particularly among Christian scholars faced with the compilation of Hellenic texts, as St. Basil emphasizes, using the same imagery as his pagan counterparts: Kara Traaav br\ ovv TOJV ueXirnov TTJV ekova TOJV Xoyoov f|uTv ueOeKTeov. eKefvai Te yap OUTC ajraoT TOTCJ dvOeai jrapaTrXr|criojcj errepxovTai, oure uf)v ok; av emTTTcoorv SXa cbepeiv emxeipouorv, aXX' oaov auTwv emTfj8eiov Ttpdc; TT|V epyaaiav Xa(3ouaai, TO XOITTOV xaipeiv dchfjKav • f|uel<; Te fjv atoct)povojuev, oaov okelov f|ufv Kai auyyevec; Tfj dXr|0eia Trap' auxcov Kopiadpevoi, ujrepPriaopeOa TO Xemopevov. In our intercourse with literature we must use as our image the activity of the bee. Bees neither approach all flowers indiscriminately, nor do they take everything they happen to find. Rather, they take only what suits them and is appropriate to the task at hand, leaving the rest behind. And we, if we use good sense, shall draw from literature what befits us and is consonant with our beliefs, and shall pass over all the rest. (De leg. gent. lib. 4.45) 174 Intent upon extracting from his reading all that is beneficial and meaningful, the adult learner of the Imperial period does so with notebook in hand, compilation being his primary means of drfav6i^ea6ai. Gellius refers to his adnotatiuncula (NA 19.7.12), his notes ad subsidium memoriae quasi quoddam litterarum perms (NA Praef. 2). Herodes Atticus, Philostratus records, at his death left eyxeipiSia re K a i K a i p i a if|v dpxaiav rroXupaOeiav ev (3paxeT dmivetopeva (VS 565). Pliny the Younger refers to one hundred and sixty electorum commentarii... opisthographi quidem et minutissimis scripti, qua ratione multiplicatur hie mimerus notebooks full of selected data .... They were written in a tiny hand, and on the backs of the papyrus, which means that their total number was multiplied (Ep. 3.5.17); these notes represented the "leftovers" from the Elder's seven completed works. Sidonius deserves special commendation for his energy and diligent commitment. Having gone in hot pursuit, literally, of a man who he discovered was carrying books in his baggage, Sidonius finally caught up with him: Capti hospitis genua complector, iumenta sisto, frena ligo, sarcinas solvo, quaesitum volumen invenio produco lectito excerpi. / grasped and embraced my guest's knees, I made the baggage train stop, I fastened up the reins, opened the pack,-found the book I was looking for, drew it out, perused it, made excerpts. (Ep. 9.9.6-8; discussed at Starr 1987: 218) He can, if he chooses, publish his notes in the sense that he can put them at the disposal of his friends. The educated individual has the Xenophontic and Platonic symposium as a model for this nonprofessional sharing of paideia among friends (cf. Marrou ,1938: 210-214). He can relate to the image of the educated as initiates in a 175 mystery cult of the Muses, his personal research being a visible token of his membership among the elect (cf. Marrou ibid. 231-267). He may have a sincere desire to help a personal friend or relative with his private study, offering this person short cuts to polymathic knowledge which he himself required much more effort to attain. We have been describing here miscellaneous, unspecified notes which the amateur has gathered from his study. If he has focused his reading with the purpose of acquiring material dealing with some specific topic, we would have to specify his work as i o T o p i a : directed and focused research. The terms ioropia and historia have come to have a number of rather specific denotations in the Imperial period, from a statement of what is real as opposed to fabulous, through narrative accounts of past events, to anecdotes explaining the basis in reality out of which literature has created an imaginative construct (Dietz 1995: 95). Whether the object of research be the natural world, the "stories" behind epic and tragic art, or a given social and political entity, the common term is historia as research. As Fornara defines the concept, when method designates a class of literary works it is obvious that the activity described is the sine qua non of the genre ...The method [of ancient historiography] consisted of piecing out the record in detail on the basis of a search for information from knowledgeable sources, and the resultant works attested to the diligence of the seeker. (1983: 47; cf. Press 1982: 70-72). "Diligence" is a quality of the amateur adult directing his own education, and "knowledgeable sources" are the only kind such a responsible reader will consider worth compiling if he has taste and discrimination ((biXoxaXia). Likewise he expects to be judged by these qualities when his notes are written up and made public. As a process, ioropia in 176 the Imperial period explains a phenomenon by accessing information about it and ordering that information in such a way that it will form a unified whole, a auvypcxcbf), with unity imposed by subject matter, chronology, geographical area, or any number of other things. For subjects which the researcher cannot know by autopsy, because they belong to the distant past, are geographically remote, or demand more skill or wisdom than the researcher commands, he will compile from a trustworthy source. Originality of view would be nonsensical under such circumstances, and the point of the research is not to be original but to give an accurate account of the subject being studied. Pliny the Younger, faced with the option of writing what we would call ancient history, hesitates to undertake such a project because of the vast amount of compilation it would require: vetera et scripta aliis? parata inquisitio, sed onerosa collatio You suggest that I write ancient history, subjects dealt with by others? My work is cut out for me, but the compilation is a burdensome task (Ep. 5.8). The scholar's readiness to compile in the course of creating a work of research can be seen in the preface to Dionysius' first book of the Antiquitates Romanae. Having insisted upon the importance of diligence and judgment in gathering material for an historical account (perd TroXXfjcj empeXeiac te Kai (J)iXo7roviac ... OTrouSfjg o£,ia 1.1.2-3), having defended the value of and need for a balanced presentation of the political and social development of Rome for a Greek audience (oi35epia yap aKptPqg egeXf|Xu0e rcepi aurcov TiXXnvig ioTopia pexpi TWV KOCG' qpag X P ° V W V on pf) Ke<J)aXaicb8eig E m r o p a i navv ppaxeiou [1.5.4]), Dionysius states the sources from which he has collected his information: rd pev jrapd TWV XoYitoTdrwv dvSpwv, ok, eic opiXiav rjXBov, SiSaxfi rtapaXaPtov, rd 8' BK TWV iaropicov dvaXe^dpevoc ac oi rcpoc aurwv 177 OTaivouuevoi 'Ptouaitov ouvsypaipov ... an EKEIVCOV opurjuevog TCOV T r p a y L i a T E i t o v ... TOTE ETTEXEipnaa Tfj ypacbf j . Having been instructed in some information by the most trustworthy men whom I conversed with and having collected other data from the histories written by the authors respected by the Romans themselves, basing myself upon these sources I undertook the writing (1.7.3). The program is conservative, based upon the guidelines established by Herodotus; but considering that Dionysius is writing the history of Rome down to the First Punic War, all of his data must come ultimately either from written texts or from Roman scholars — oXioc; 7r£7rai5EuuEvoi one would suppose — who have shared with Dionysius in the communal associations of scholarship which we have been considering here, material a£,ia a7rou5fjc; which they in turn have derived from texts. The compiled notes of a scholar can thus be accessed and written up into a auyypactri by a person involved in researching any given subject. Compiled information thus used as raw material for more polished and focused written accounts might be referred to as iiXn, silva, copia, "stuff' or "supplies"; such polymathic data could fund oratory as well as written scholarship (cf. Cicero De Or. 8.103, 125; Oral 16; Quintilian Inst. 8 Praef. 28). Photius can thus account for the rather disorganized Xoyoi i a T o p i K o i of Olympiodorus: auToc; iacoc; auviStcv ou auYypa(j>f)v auTop TauTa KaTaaKEuaaOfjvai aXXa uXr|V auyypacbfjc; EKTTOpioOfjvai 5ia0£PaiOUTC(l. He himself has maintained that it was not with the intention of writing a complete work of historical research that he prepared this material, but of providing the raw material for such a study. (Bibl. 80.56b). Such OXn does not necessarily have to result in an historical account of the past. Photius uses the same term for the Atticist lexicon of Phrynichus: EOTI 5E 6 auyypactjEuc;, EI TIC; jroXupaOEoraToc;, dXXtoc; 8e XaXoc; Kai 7repiTT6c;...Kai KaXov Kai copaiov Xoyov uXnv aXXoic; auvaOpoiCtov. Although this author is 178 extremely learned, he is wordy and prolix; all the same, he collects a fine account to be used as a resource by other scholars (Bibl. 158.101b). Plutarch uses these concepts rather suggestively in describing the activity of the Spartan Cleombrotus who, he relates, having acquired a comfortable income, used his leisure for research. He sailed beyond the Red Sea, ouvfjyev lOTOpiav OlOV iJXnv (blXoaotJHCXC, He collected [the results of] his research as raw material for philosophy (De def. or. 410b; cf. Babut 1975: 207-208). A oXcoc, 7rE7rai5eupevo<; may simply write up his notes, then, and make them public, allowing the reader in turn to use or reject them for his own research as he sees fit. Aelian, for example, seems to have done just this in the De natura animalium2 And although we cannot know for certain, Aelian seems to have done likewise for the Varia Historia as well. But its Greek title, noixiXq ioropia, suggests that in some way Aelian's historia must result in a "piecing out of the record," as Fornara's definition above demands. Perhaps we should refer to Aelian's polymathic activity in Varia Historia as a "piecing in," i.e. a supplementing, of the "record" rather than a comprehensive survey of areas of knowledge from which he compiled. The concept of a miscellaneous supplement is clearly typing Aelian's research as something different from, for example, the universal history of a Cassius Dio or the apxaioXoyioc of a Dionysius of Halicarnassus. The concept of jroiKiXia or disparilitas is important for the Imperial miscellany. It is primarily this quality, sought by the miscellanist in arranging his notes, which distinguishes the miscellany from other polymathic collections of the Imperial period. Other polymathic compilations share with the miscellany an awareness of the adult 2 OTio 8e ou (JjavsiTai [ T O U T C X XuoiTsXfj] , E C X T U ) T U J Troup! OaXnsiv T E K a i Ttspisjrsiv. (NA Prol.) 179 reader's paideia-needs. Hesychius, for example, a professional grammarian working at Alexandria probably in the sixth century A D (Kaster 1988: 292), in the luvaytoyrj naacov Xecjecov Kara OTOIXETOV provided his reader with synonyms for hard words found in literary sources. Hesychius prided himself on having made a comprehensive compilation of specialist lexica from all the canonic genres. He mentions in particular the works of Aristarchus, Apion, Apollonius, Heliodorus, and Herodian, but especially the text of Diogenianus. Hesychius greatly admired Diogenianus, as one both G7iou6aToc. and (fnAoKctAoc., the author of a work entitled nepiepYOjTevnTec., "Poor Men but Scholars." The description which Hesychius gives of this work, which he seems to have taken as the core of his own lexicon and to whose alphabetical structure he added further compiled material, interests us here. Diogenianus had worked under Hadrian, and seems to have compiled not only the epitomized Pamphilus and specialist lexica, but material illustrative of medicine, the historical and antiquarian writers, and oratory as well: TCXI3TT| xpncropEvoc, Tfj 8icxvoia fiysfro yap, oiuai, ur) uovoicj irXouaioicj aXAd Kai TOTC, 7revr|0i TCOV dvBpcbjrcov xpr|OTUEvJaEiv TE K a i dvri 8i5aaK&Xcov apxEOEiv aura, EI uovov TCpiEpyaa&UEvoi TTCXVTOXOGEV dvEupElv raCra 5uvn0Ei£v Kai eyKpaTElc, auroov YEveaSav. ETtaivco uev EytoyE TOV av5pa Kai Tfjc; (biXoKaXiacj Kai Tfjc; o"7Tou5fjc;, OTI xpnoTpcoTdTnv rrpayuaTEiav Kai TOIC; cmou5aioicj TCOV cbiXoXoycov oocbeXipcoTdTr|v x ° P n Y i a v Trpoc; aTraaav TratSsiav rrpoEiXETO TrapsxEiv. / think that his intention was to provide useful material for rich and poor alike, material which could take the place of teachers if they could just get access to it in the course of comprehensive research. I must commend the man's good judgment and scholarly industry. He elected to provide for serious scholars a most useful study and a body of material of immense benefit for every manner of paideia. (Praef.) It is Diogenianus' concern with getting polymathic resources into the hands of students working without teachers that places him beyond the range of the professional 180 grammarian and within the world of the adult self-learner. The Imperial period was aware of the needs of the individual who was autodidact from financial necessity as well as from the demands of a career and public service.. Collections of compiled passages speaking to the interests of students of the other artes — medicine, arithmetic and geometry, astronomy, music and dialectic — are referred to, and in some cases we have texts illustrating such material. [Plutarch]'s De arte musica, for example, compiles liberally from the works of Glaucus of Rhegium, Aristoxenus, and Heraclides Ponticus on the technical aspects of ancient Greek musical theory (West 1992: 5-6). Collections of compiled passages published with the needs of rhetoricians and rhetoric students in mind interest us here, because of the relationship between public speaking and the concept of the fully educated statesman. As we have seen, Cicero's model of the orator doctus influenced the ancient learner's concept of relevant scholarship and compilation. The student of rhetoric was taught to draw upon canonic literature both for models of style and for matter. These are the ends for which Dio (Or. 18) and Quintilian (Inst. 10) provide lists of authors useful for the public speaker. Dionysius' view of rhetorical mimesis as essentially a process of TOV rfjc; rroXupaOeiac, epavov ovXkeyeiv, links the practice with the act of scholarly compilation. The "flowers" of rhetorical excellence may be culled and displayed in a collection, to be drawn upon for private or classroom use. Diogenes Laertius refers to the practice of marking with a letter X any 181 word or passage seen as xpqcapov rrpoc, TCCC, eKXoydc, Kai KaXXiypathiac, (3.66), suggesting the development of a compiled literature of chrestomathies.3 The Imperial age recognized, then, the importance of much matter with which to fund rhetoric. Quintilian expects that the student lectione multa et idonea copiosam sibi verborum supellectdem compararit (Inst. 8. Praef. 28). Such copia is not for stylistic mimesis but for the acquisition of exempla, illustrative material with which to fill out and attractively to pad a declamation's bare structure and outline. Here we shall find polymathy at its best, for such exempla must be not only morally effective and attractive, but also new to an audience whose taste may be jaded. We need only mention Valerius Maximus' Factorum et dictorum memorabilia libri novem, specifically designed for the busy and committed student of exempla* by being arranged first according to the nationality of the main character (Roman or externus) and then by theme (e.g. stratagems, justice, hard work, old age, anger). Several centuries later Johannes Stobaeus will prepare a similar cop/a-collection arranged thematically. In Stobaeus' case, the passages are all compiled from earlier authors, most of them canonic, thus making the work an anthology. The pieces are arranged thematically, designed not so much as sources for exempla as easily accessed quotations. Both the Suda and Photius were familiar with Stobaeus' work. According to the former it provided evapera rravu Kai ycpovra rraoriQ TraiSeuoeioc, excellent [selections], chock-full of every kind of learning. Photius recognized the short CUtS it offered to otherwise rather tedious slogging through whole authors to find striking quotations: 8id ouvexouc, auTwv peAeTnc; O U K ev rroXXtp X P 0 V V TCOXXWV Kai KaXwv Kai TTOIKIXWV vonp&Ttov, 3 Proclus, however, is the earliest author to leave a work with this title (GGL: 881). ABreviter ... ab illustribus electa auctoribus digerere constitui, ut documenta sumere volentibus longae inquisitionis labor absit (I. Praef). 182 d Kai KE(baAatlb5r| UVfjunv KapTTlbaovTai [Readers] will reap a mental harvest, summary fashion, of many and various fine thoughts through brief but consistent work with them (Bibl. 167.115b). Photius adds that the collection is especially useful for speakers and creative writers. These are all general collections of compiled material answering to the needs of students of specific artes. But there remains a great deal of material which does not fit into any of the artes, material which is worth knowing and worth noting down but which must be seen as simply polymathic stuff, iiXn. The adult self-learner has acquired this material in the process of his pursuit of paideia, and it reflects his personal (biXoKaXia, cmou8f|, and rrovocj. It is this otherwise uncategorized harvest of amateur paideia-compilations which forms the basis of the miscellany. Although Sextus Empiricus criticized such polymathic information because it could not be reduced to a dialectic-style outline method, being simply due0o5oc. uAn (Math. 1.266.1), the Imperial miscellanists saw this quality as a positive aspect of their collections. The dialectic Texvn was a thing of the classroom, a device associated with the coercion, manipulation, and tedium of childhood education. But by focusing upon the unarranged, random, and various quality of their published notes, the miscellanists could offer the adult reader a different kind of learning experience: polymathic data presented in such a way that its consumption could be easily incorporated into the reader's lifestyle. Because the miscellanists themselves insist upon the disordered, even random quality of their collections, it is easy to look upon the Imperial miscellany as a kind of dustbin or grab-bag. Gellius, for example, who prides himself upon the disparilitas of his work, yet criticizes the other miscellanists for "sweeping together" (converrebanf) their 183 data like so much rubbish (NA Prol. 11). Yet when Imperial miscellanists describe their own activity, we find that they tend to structure their collections with several paradigms in mind. These are the paradigms applied to all uses of material compiled from paideia, models which suggest that the extracting of data from earlier works to create something new was viewed neither as a sterile and trite recycling of the past nor as an irresponsible pillaging of another's literary property. The first paradigm of compilation occurs in the fragmentary De imitatione of Dionysius of Halicarnassus. In seeking to define his concept of stylistic mimesis, whereby an author literally constructs a style by culling the best elements from the great authors of the past, Dionysius relates the story of how Zeuxis created his famous portrait of Helen. Zeuxis persuaded the people of Croton to let him see their daughters naked; given access to a variety of models, he selected out of this larger group a number of young women each with a distinguishingly beautiful trait. One had, for example, especially lovely elbows, another excellent hands, another a beautiful throat and shoulders: 6 5' rjv ac;iov Trap' eKdorri ypathfjc;, ec, uiav f|0poiaOri atouaToc, eiKova, KCXK TTOXACOV uepiov auAAoyfjc, ev TI cruve0r|Kev f| Texvr) TeXeiov eiSoc,. Toiyapouv Trdpeon K a i aoi KaSdrrep ev GedTpio rraXaitov atouaTiov iSeag ec;ioTopeiv Kai Tfjc; eKeivcov ipuxfjc, dTrav9iCea0ai TO Kpeirrov, Kai TOV Tfjc; 7roXuua0eiac, epavov auXXeyovn OUK eg~iTr|Xov X P ° V 0 P yevriaouevnv ekova TUTTOUV dXX' dOdvaTov xexvrig KdXXoc,. He incorporated into a single image each girl's separate quality which deserved to be painted. This collection made from many parts his craft composed into one perfect form. So you too can seek out the images of ancient forms as if in a theater, and can cull what is superior from the spirit of each. In gathering together this contribution-banquet of scholarship, you will fashion an image which will not fade with time but which will represent art's immorfal beauty.(irdL%. 31.1 Us.) 184 As Zeuxis made a work of ideal and lasting beauty by abstracting the finest qualities from a large number of female bodies, so Dionysius encourages the author who wants to develop a fine style to research thoroughly (ecjioropeTv) and to select with care (drtav9i£eo0ai, "pluck blossom"), not physical features, but the best assembly of scholarship (TOV xfjc; rroXupa9eiac, epavov) which authors of the past have to offer. Lucian presents a model of compilation which, like Dionysius', is drawn from the plastic arts. In Imagines, the speaker, Lycinus, struggles to describe to a friend a beautiful woman whom he has seen but whom he cannot identify. He finally succeeds in rendering her likeness by describing separate features from famous artists' works and applying these features to her. Starting with the head of Praxiteles' Aphrodite of Cnidos, r d pev dp4>i TT]V xopnv Kai peTcorcov 6(bpuwv re TO euypappov edaei exeiv cooTcep 6 npa^iTeXnc, erroinaev ... r d pfjXa 8e Kai ooa Tfjc, oipeooc, dvTiorcd reap' 'AXKapevouc, Kai Tfjc; ev Kqrroic, Xqiperai, Kai rrpoaen x e i P & v a K p a Kai Kaprriov TO eupu9pov Kai 6aKTuXwv TO eudyooyov eic, Xerrrov drroXfjyov rcapd Tfjc; ev Krjrcoic. Kai raura. TT)V 5e TOU rcavToc, rcpoawrcou rrepiypa^riv Kai rcapeicov TO drraXov Kai piva aupperpov q Aqpvia rcapecjei Kat <I>ei6iac; * en Kai oropaToc, dppoynv auroc; Kai TOV auxeva, rcapd Tfjc; 'ApaCovoc; Xaptov. [Reason] will keep the hair, forehead, and the beautiful shape of the brow just the way Praxiteles made them. The cheeks and face will be taken from Alcamenes' "Aphrodite in the Garden," as will the hands: , the proportion of the wrists and the slender, tapering fingers. The outline of the face, softness of the cheeks, the nose, Pheidias and the Lemnian Aphrodite will provide; Pheidias will use his Amazon as a model for the throat and mouth. (Im. 6) So successful is Lucian's collage that his addressee recognizes the referent immediately, and goes on to describe her spiritual and intellectual qualities using the same cut-and-paste method as has been applied to her physical beauty. Far from being an irresponsible or inefficient means of recreating an aesthetic experience, Lucian's description constructed from "compiled" features displays and ratifies both the speaker's great depth and range of 185 learning and the aesthetic sensitivity he has acquired through culture. He is successful at his description because he has the intellectual resources from which to extract its component parts. Finally, we may consider a series of models for compilation offered by a miscellanist. In the prologue to the Saturnalia, Macrobius describes the benefit to be gained from reading a compilation such as his. Drawing upon material extracted directly from Seneca's eighty-fourth Epistula, Macrobius presents a series of six images which apply both to the process of constructing a miscellany and to the reader's experiencing of it. First, Macrobius' own activity in gathering and working up extracts from earlier authors is compared to that of bees extracting honey from selected blossoms to make the honeycomb (Sat. Praef. 5). Next, the mind which has absorbed diverse material from a compilation is described as a kind of pickling-vat or keeping-room, in which many condiments are steeped into a unity while each preserves something of its original flavor, ut etiam si quid apparuerit unde sumptum sit, aliud tamen esse quam unde sumptum noscetur appareat so that even if you perceived the source of anything, it yet seems different from its recognized source (6). Again, the mind processes such various data in the same way as the body digests various kinds of food, and benefits from the variety in a holistic manner (7). The mind adds up information derived from a miscellany and arrives at a total, omnia quibus est adiutus abscondat, ipsum tamen ostendat quod effecit concealing all from which it has derived benefit while still revealing its effect (7), the way that we add up separate numerals to reach a final sum. Further, the process of assembling and reading various data is like that of mingling various scents to make a perfume (8). A compilation is also like a chorus 186 made up of many separate voices: una [vox] tamen ex omnibus redditur we hear one voice resulting from many separate voices (9). Macrobius summarizes both the process of compilation and its beneficial results thus: tale hoc praesens opus volo: multae in illo artes, multa praecepta sint, multarum aetatium exempta, sed in unum conspirata. Such is my goal for this work: that it contain information and scholarship about many subjects compiled from authors from many periods, all of it crafted into a unity (10). Macrobius insists upon the compilator's nurturing and responsible care and the compilation's benefits for the reader, all the while modeling the process by recasting Seneca's words to fit his new context. As an epavog xfjc; rroXupaGetag, the data forming the miscellany are seen by the miscellanist as intrinsically valuable both to himself and to his reader. The miscellanist's contribution lies both in his selection of data for inclusion and in his arrangement. He has no desire to reduce the value of his data-base by trivializing it through simplification, and to this end Gellius must warn his reader not to be intimidated by the technical nature of some of his chapters. One should note that it is the quality of the selection of the miscellanist's xpioig, his exXoyrj and not the arrangement of data, which Pliny criticizes in his assessment of rival miscellanists. Inscriptionis apud Graecos mira felicitas: Kqpiov inscripsere, quod volebant intellegi favum, alii Kepac, ApaXOdac,, quod copiae cornu, ut vel lactis gallinacei sperare possis in volumine haustum; iam Ta Mouacu navSexTat 'EyxeipiSia Aeiptov Hivac; Exe$icov: inscriptiones, propter quas vadimonium deseri possit; at cum intraveris, di deaeque, quam nihil in medio invenies! The Greeks display a breathtaking facility for creating titles. Some have called their work "Honeycomb, " others "Cornucopia, " implying that you might find in it any number of extraordinary things. They also have titles like "Violets, " "Muses," "Universal Compendium, " "Handbook," "Meadow, " "Register, " "Impromptu." Such titles seem to need no other guarantee. But when you 187 begin to read them, how worthless they'll be found! (HN Praef. 24: cf. Gellius NA Prol. 6-10) Pliny is quite specific about his own labors, as we have seen: he had extracted twenty thousand pieces of worthwhile data from two thousand books, and had reduced this extraordinary congeries to thirty-six volumina (HN Praef. 17). The effort he made in the rewriting is summed up thus: res ardua vetustis novitatem dare, novis auctoritatem obsoletis nitorem, obscuris lucem, fastiditis gratiam, dubiis fidem, omnibus vero naturam et naturae sua omnia. It's a difficult business to give a new lustre to old things, to give authority to what is worn out, glamor to the commonplace, elucidation to the unclear, grace to the repellent, trustworthiness to the dubious, its own nature to everything and all of its own to nature. (HN Praef. 15) In the prologue to the De natura animalium, Aelian likewise draws attention not only to his personal zeal in research but in composition as well: cbcj uev ouv Kai exepoic, ujrep xouxcov earrouSaorai, KaAtdcj oi5a • evto 5e epauTtp xauxa baa otov re f|v dGpoiaacj Kai 7repi(3a\cbv auxoicj Tf|v auvfj9r| Ae i^v, KeiufjAiov OUK dorrouSaorov eKjrovfjaai Tremor-euKa ... ei 5e em jroAAoic; TOIC, Trpioxoic, Kai aotboic; yeyovauev, uf) eaxio t/|uitoua ec, eTraivov f| xou xpovou Afjcic,, ei xi Kai auxoi 07rou5fjcj dijiov rrapexoipeGa Kai rfj eupeaei xfj TrepiTToxepa Kai xfj cbtovfj. / am well aware that others have researched these subjects. But I am quite confident that I have labored to create a keepsake well worthy of serious attention, painstakingly gathering together all that I could and clothing the results of my research in a familiar style .... If I have been born later than many of the best and the wise, let my allotment in time not result in a stinting of my praise, particularly if I too have somehow provided a study worthy of attention in terms both of my more extensive conception and of my style. The repetitive eorrouSaorai, OUK darrouSaoTov, oTrouSfjc, a£;iov make Aelian's point more than clear: both data and composition have required diligence and effort, and both contribute to the value of the De natura animalium as a KeiufjAiov, an object of value to be treasured. 188 Part of this effort in composition consists of retaining or recreating the spontaneity of the original act of compilation. Gellius describes this aspect of the miscellany style as the disparilitas quae fuit in illis annotationibus pristinis quas breviter et indigeste et incondite exauditionibus lectionibus variis feceramus the random quality which those original notes displayed. We had made the notes quickly, in a disorganized fashion, from a variety of things heard or read. (NA Prol. 3). Photius describes a similar policy on the part of Pamphila, whose thirty-three volume 'YrropvripaTa provided material for Diogenes Laertius, Gellius, and Favorinus: T a u r a be rcdvra, ooa Xoyou Kai pvn,pn,g auTfl a£ta eSoKet, eic; uTropvrjpaTa cnjppiyfj Kai ou TTpdcj Tag i5iag urcoOeoeig SiaKCKpipevov eKaorov SieXeTv, aXX' ourtog eiKfj Kai cog e raorov errfjXGev dvaypdipat, cog ouxi x a ^ E 7 r o v exouoa, thnoi, TO KOT' etSog aura SieXeTv, emTeprreoTepov be Kai xapieaTepov TO dvapeptypevov Kai TT)V rroiKtXiav TOU povet5oug vopiCouaa. Everything that in her reading she found remarkable and worth remembering she presented as miscellaneous notes, not with each matter arranged under proper headings, but she wrote each one up just as she happened upon it. She herself states that she did this not because she had any difficulty in arranging her material into categories, but because she believed that miscellaneity and variety were more pleasant and charming than uniformity. (Bibl. 175.119b) Pamphila's program of composition appears to be the same as Gellius'; she wrote up each chapter tog eKaorov errfjXGev. She was recreating the experience of a TrerratSeupevri using her leisure rightly, and attempting by this duplication to share with her reader her spontaneous amor discendi.5 Photius attributes a similar rroiKiXia to Sopater's 'EKXoyai Statbopai, a twelve-volume work which he reviews at Bibl. 161.105a: 5 Photius also relates how Pamphila claimed that she acquired data from her husband, by listening to her husband's dinner guests, and by compiling texts on her own. She has, that is, attempted to relate her polymathic data to the refined leisure of the educated elite. 189 TroXXrjv 5e TTJV xpciav TOUTO 8fj TO (biXonovriua ToTg dvayivcbaKouor jrapexerai.... ev oig Te eig TroXuuaGiav CK TOU eToiuou auvTeXel, Kai xrpog dpeTfjv Kai KaXorayaGiav TtXeiOTd eoTtv auTcov djravQiaaaGai, jrpog Te TO pnTopevJeiv Kai aocbioTeueiv (cog Kai auTog ToTg eraipoig ypdcbcov jrpooiuidg'erai) OUK eXaxiaTnv cbepei pOTrfjv, kavov eig Xpeiav KaGeornKev. rj 5e ctpdcng auTto TTOIKIXTI Kai ou uia TTJV iSeav. This careful study provides readers with much useful material .... When it comes to conveying a ready and comprehensive body of useful data, there is a wealth of material here for compilation with a goal to self-improvement and, as [Sopater] himself writes to his friends in his prologue, it makes a major contribution to rhetoric and scholarship. Consequently its value is established. His discourse shows "poikilia " and the style is varied. All three aspects of the Imperial miscellany come through in Photius' description: relevant and morally sound rroXuuaGia, the miscellanist's close connection with an adult reader (Sopater addressed his prologue to his "friends"), and variety of structure and style. Clement, aware that he is working within the guidelines of an established tradition in his Stromaleis, makes a similar claim to TroiKiXia and spontaneity: ev uev ouv TCO Xeiucovt TO dvGn jroiKiXcog dvQouvra KOCV TCO 7rapa8eiaco f) TCOV dKpo5pucov cbuTeia ou K a r a et8og eraaTov KexcopioTOi TCOV dXXoyevcov (f| Kai Aeiucovdg Tiveg Kai 'EXiKcovag Kai Kr |pia Kai IleTrXoug auvaycoydg (biXouaQetg jroiKiXcog eg*avGiaduevoi auveypaipavTo) • ToTg 6' cog exuxev erri uvfjunv eXGoucn Kai pr|Te Tfj ragei prJTe Tfj cbpdaei 8iaKeKa0ap|aevoig, 5iea7rapuevoig 5e eTriTr|5eg dvapi^, rj TCOV ExpcopaTecov fjpiv UTroTuTTcoorg Xeipcbvog Steriv TreTroiKiXTai. Flowers bloom variously in the meadow, and in the orchard fruit trees are not segregated from each other according to species. So certain people have also composed "Meadows" and "Helicons" and "Honeycombs" and "Peploses", men who love knowledge and compiled their collections from various sources. The structure of my Stromateis has been tricked out like a meadow with material which I happened to notice and which I did not arrange in terms of order or style, but which I scattered about here and there. (Strom. 6.1.2) Unlike Gellius and Pamphila, Clement is not clear whether he has retained the original order o f compilation; cog rruxev suggests as much, yet emTnSeg dvauig" implies that 190 Clement in fact expended some effort on arrangement in order to recreate the spontaneity of the original act of compilation (cf. Mahat 1966: 339-343). Macrobius too mentions the labor involved in seeming to be spontaneous and natural in one's scholarship. Taking his cue from Gellius, he adds an organic element to his description: nec indigeste tamquam in acervum congessimus digna memoratu: sed variarum rerum disparilitas, auctoribus diversa, confusa temporibus, ita in quoddam digesta corpus est, ut quae indistincte atque promiscue ad subsidium memoriae adnotaveramus, in ordinem instar membrorum cohaerentia convenirent. We have not piled up in a heap these memorable data; but the disarray of a miscellaneous collection drawn from a variety of authors and reflecting different times has been arranged into a kind of organic whole, in such a way that our random notes have come together into a coherent order, like the limbs of a body. (Sal. Praef. 3) In the epilogue to the De natura animalium, Aelian makes his position clear in regard to the ordering of data within the miscellany. For this author, rtouaXia requires the expenditure of considerable labor, and there is a baroque quality to the description of the "anti-structure" in the De natura animalium. oi5cx 5e on xai eKEiva OUK erraiveaovrai TIVECJ, ei un xa6' eKaorov TOJV t^ opoov drcEKpiva uou TOV Xoyov, pq5e iSia TO eKdorou eirrov dGpoa, dvepicja be Kai Td rroiKiXa rcoiKiXojg, Kai urcep rroXXoJv 8iet;fjX9ov, Kai rrfj pev drceXircov TOV rrepi Tojv8e Xoyov TOJV COJOJV, rrfj 5e urreorpeipa urrep Tfjc; aurojv (huaeojcj erepa eipojv....Tuj TTOIKIXOJ Tfjc, dvayvojaeojc, TO ecboXKOv 9qpoJV Kai TT)V £K TOJV opoiojv p8eXuypiav drco5i5pdoKOJV, oiovei Xeipojvd n v a fj oretbavov ojpaiov eK Tfjc; rroXuxpoiac,, OJC; dv9ea4)6pojv TOJV ^OJOJV TOJV TTOXXOJV, ojf|9qv 5etv Tfjv5e u<j)dvai re Kai 5iarcXec;ai Tf|v ouyypacbfiv. / know that there will be some who find fault with my work because I did not arrange my material according to subject, grouping all data in one section according to animal species, but rather mixed material up in a miscellaneous manner, now giving some a cursory review, now skipping over other matters and now retracing my steps and adding supplementary material. I believed 1 must interweave and implicate my study to achieve the effect of a meadow or of a garland made beautiful by varied hues, my many animals each contributing its blossoms. I was aiming at providing a pleasant reading experience through variety [TW JTOIIOXW] and at avoiding a repellent monotony. 191 rioiKiAia, then, is at the base of the Imperial miscellany tradition. Because the miscellanists themselves see it as a positive quality, they claim it for their work. But probably no other element has been more influential in lowering the ancient miscellany in the estimation of the modern critic as this claim to attracting and pleasing the reader through variety of content. However, to type these works as primarily "Unterhaltungsbucher" (Schmid 1893 vol. 5:6), as trivial, childish, or silly is, I think, to miss the point. The polymathic data-base of the miscellany provides for the needs and the tastes of one who, as a oAtoc; TreTraiSeuuevoc,, recognizes the need for the study of his literary culture but is also a responsible and functioning member of his society who must limit his scholarship to rare moments of leisure. For such an individual, paideia is experienced in small and varied units, without a consistent focus and goal but precious nevertheless. The connotations of jroiKiAia6 visually define such a person's encounters with paideia. nouaAicx implies the give and take of the symposium, the chance meeting of friends and old teachers at the bookshop, a quickly copied-out passage from a windfall of a text stumbled upon in a library. The miscellany does not answer the needs of the professional philosopher or rhetorician who can devote his career to the rigorous and attentive analysis of texts and examination of theses. When daily life must be largely devoted to 7rpaKHK&, however, the value of a TroiKtAr) ioropia in providing preselected data of proven worth is clear. 6rioiKiAia means variegated in the physical sense of a peacock's plumage or a dappled fawnskin, or in an object of craftsmanship such as an embroidery of many hues. Figuratively the word can be applied to the resourceful and omnivalent thoughts of an Odysseus or Prometheus. IloiidAoc can be used of a scholarly work which has been compiled from many sources or which is applicable to several areas of study (cf. Tolkiehn 1925: 2433). 192 Chapter 5 The 'AcjiocyrrouSaora in Aelian's Varia Historia The three previous chapters have focused upon the generic qualities which give the Imperial miscellany a literary identity. As I have attempted to demonstrate, the miscellanist concerns himself with a specific subset of paideia: the material which supplements the standard literary education and therefore may be termed polymathic. The miscellanist assumes a readership with whom he shares certain educative goals; because the miscellanist adopts the stance of a mature amateur scholar gathering data for a younger reader, he may reveal a patronizing tone in his paedeutic collection. Finally, the compilation which the miscellanist offers his reader is presented in a manner which reflects the Imperial attitude toward the cultured person's correct use of leisure. My purpose in the present chapter is to analyze Aelian's Varia Historia as an Imperial miscellany displaying these distinctive qualities. Because in dealing with a miscellany we are considering not so much an independent work of literary art as a collection of data accessed from a variety of kinds of written documents, an explication of the collection must therefore analyze the compiled data as reflections of the purposes, values, and goals of the collector in the act of selection. In the process of considering the categories of polymathic data which Aelian has gathered into this work I shall therefore attempt to isolate the moral tone which Aelian attempts to convey to his reader. As an educated adult, a oAtoc, 7TE7rai8eupevocj of the sort described in Chapters 2 and 3 above, Aelian has undertaken a compilation of the past for the uses of the present. Like other Imperial TTETratSeuuEvoi, however, he is constrained by his own circumstances 193 and by those of his reader to select from the secondary material available only what provides the greatest value in the briefest study time. An analysis of the contents of the collection should then reveal what Aelian has considered worthy of compilation and therefore worth including in the Varia Historia. For the purpose of this analysis we may divide this material very roughly into (1) anecdote, including here the chreia or apophthegm; ( 2 ) biographical sketches; ( 3 ) ecphraseis, ( 4 ) descriptions of the laws and customs of various peoples, including discussions of Greek states and of Rome along with barbarian nations; ( 5 ) natural history, which for Aelian consists both of anecdotes about and analyses of the behavior of lower life forms; (6) paradoxography; and (7) lexicography, consisting here of the explanation of etymologies and proverbs. Unifying this material is Aelian's particular standard of moral excellence, a paideia-based construct traceable at least in part to Isocrates, and which on occasion Aelian refers to as TO 'EAATIVIKOV : the Greek way. Moral Anecdote in the Varia Historia Most of the chapters of the Varia Historia consist of historical anecdote, a term which for want of a better we may apply to all short narrative patterns which Aelian presents as true — that is, as reflecting events believed to have actually occurred at some time in the past.1 Aelian's anecdotes are patterned in the sense that they have a definite beginning, middle, and end. Although he does not use the term OCVEKSOTOV, "unpublished" or "not made known," in reference to these little stories (the etymology at the basis of our current usage [Grothe 1971: 4 - 1 0 ] ) , Aelian does aim at recording in the Varia Historia 1 Aelian may, however, include a story which he or others may consider only probable; if so, he informs the reader e.g. VH 3.27. 194 material which is fresh and novel, as he suggests at, for example, VH 2.4 and 3.6. If he suspects that the reader has already encountered the anecdote, he will apologize, e.g. VH 3.35. Having a plot-line, the anecdote is a uuOog but it is a true uuOocj, at least in Aelian's usage. It relates an event the chief character in which is generally a well-known figure. We may include with anecdote the chreia, a narrative pattern which highlights not an event but a statement, one usually conceived as summing up in a sententious manner a speaker's character or a situation's import. As a narrative, the anecdote carries with it an immediate appeal. But it also suggests the presence of TO UUOIOSECJ, the quality which, as we have seen (pp. 88-89 above), turned the paradoxographers away from stories which one could judge to be fabulous. The tidier and more a propos an anecdote and the more artificial it appears, the closer it approaches to a mythic sequence. We often find the same anecdote or chreia applied to different characters by different authors. At VH 3.20, for example, a story is told of Lysander which at Athenaeus 14.71 is told of Agesilaus. Ancient authors were aware of this situation, and sometimes mention when they have found an anecdotal sequence referred to other characters in different sources (e.g. Diog. Laert. 6.25, Plutarch Ti. Gracch. 4; cf. Wehrli 1973: 195). The reliability of the anecdote is not supported by the ancient grammarian's use of the term historia to refer to any narrative-style explanation of a literary allusion or reference. As we have seen in Chapter 3, a major portion of the grammarian's lecture was devoted to historice, the elucidation of a piece of literature's background stories, a potentially endless source of trivia if the grammarian did not limit himself to his pupils' 195 immediate needs. In this sense, any story in, for example, [Apollodorus]' Bibliotheca would rank as historia; we may attempt to distinguish these background stories as more or less probable, i.e. not contradicting the known laws of nature, but we are still left with a less than straightforward approach to the truth (cf. Dietz 1995: 66-69). To complicate the. situation is Gellius' use of historia to mean an anecdote in the sense in which we apply the term to Aelian's assumedly true short narratives (though not to the chreia). There are some sixteen chapters in Gellius entitled historia, all concerning events in the distant Greek and Roman past. Although Gellius' historiae and Aelian's anecdotes are independent narratives related for their intrinsic interest, we do not usually encounter ancient anecdote in such an isolated context. Generally these brief stories are found as data supporting a developed argument, as illustration, or as evidence to support an assertion or thesis. They form the brief, digressive, often folkloric narratives — the gallant courtships, grisly acts of vengeance, noble gestures of friendship — that find their way into the works of the early historians, and are interpreted as part of the oral evidence necessary for comprehensive and responsible history writing. But anecdote had formed an integral part of traditional poetry as well, in a form which linked it with the rhetorical exemplum. Aristotle had divided inductive rhetorical persuasion into that effected by true stories (irapa(3oXai) and by fables (Aoyoi), the latter being easy to find, Aristotle asserted, but the former more effective because more realistic, i.e. acceptable, as persuasion. (Rhet. 1393a27-1394a8) Anecdotes to be used as rhetorical exempla were a compilable commodity in antiquity, as we have seen in the case of the collection made specifically for 196 declamation by Valerius Maximus (for other such collections cf. Litchfield 1914: 62-63). Much of the copia which Cicero and Quintilian urge the rhetorician to acquire through wide reading consists of such anecdotes. For Aristotle the historicity of an anecdote made it more valuable because more persuasive and compelling. The speaker who introduces historical anecdote into his argumentation can thus expect that his discourse will be more convincing. Isocrates adds that by selecting certain kinds of anecdotes the speaker can indirectly contribute to his status as a responsible individual. In the Antidosis he asserts that the study of the art of rhetoric makes people better citizens; in the process of composing a speech TCOV rrpcxcjcov TCOV auvTeivouacbv jrpdc, Tfjv UTroOecnv EKAecjeTai TCXC; jrpeTTcooeaT&Tac; K a i u&Aiora auutbepouaac; • 6 be TCXC; xoiauTac; aoveOic'ouEvoc; Oecopeiv Kai 8oKtuaCeiv ou uovov rrepi TOV eveorcoTa Xoyov dXXd K a i jrepi TCXC; aXXag npa^exc, TTJV auTfjv E%E\ Tai3TT)v 8uvauiv, coa0 ' a u a TO Xeyeiv a u K a i TO thpovelv TrapayevfiaeTai ToTg cbiXoaocbcoc; K a i cbiAoTiucoc; Trpoc; TOUC; Xoyouc; SiaKeipevoic;. He will make a selection of exempla (npaZpic) supporting his argument which are especially fitting and appropriate. Such a speaker, growing used to examining and assessing such accounts, will acquire this same capability not only in terms of his immediate argument but concerning other cases as well, so that to those who approach speaking with the desire to become wise and to win glory will accrue both speaking and thinking effectively. (277) The exempla a writer selects and the use to which he puts them not only show how serious he is in the study of rhetoric but also increase his rhetorical and intellectual capacities. That is, the quality of the exempla types the speaker. Aelian's practice shows that he is aware how one can thus be judged by the range of his exempla. In the Varia Historia Aelian compiles, for example, several anecdotes about hetaerae. Phryne's golden statue (9.32), an epithet of Lais (1.435), the clever way in which a young man escaped Lais' clutches (10.2), and a witty rejoinder by Gnathaena 197 (12.13). Similar stories can be found in Athenaeus' Book 13, the portion of the Deipnosophistae dealing with prostitutes. Athenaeus's stories, which claim to be drawn from a collection put together by Aristophanes of Byzantium, occasionally lapse into considerable ribaldry (e.g. 585a-c). Aelian's never do. A similar situation occurs in the case of Hippomachus the athletic coach. From references to him in Plutarch and Athenaeus, Hippomachus appears to have had a number of humorous anecdotes attached to his name. The one given by Aelian at VH 2.6 is relatively serious. Evidently Aelian purposely avoided those anecdotes about prostitutes or those chreiai which were particularly silly or irrelevant, and which would compromise the tone of his collection. In fact he seems aware that one can be judged by the type of behavior one mentions in a public forum. Commenting negatively upon the statue raised in honor of the voluptuous Phryne, Aelian qualifies his remark, insisting upon specifying exactly the object of his anecdote: <J>pi3vqv Tf|V eroripav ev AeX(j)oic, dveoTnaav oi "EXXnvec; em KIOVOC, ev pdXa uTrnXou. OUK epco 8e dTrXioc, roue, "EXXnvac,, cbc, dv pq 5oKoinv 5i' amac, dyeiv rcdvTag, oi)c, cbiXco rrdvTiov pdXiora, dXXd TOUCJ TCOV 'EXXqvtov aKpareorepoug. The Greeks set up a statue of Phryne the hetaera upon a high column at Delphi. No, I shall not just say "the Greeks," to avoid incriminating them all (1 am fondest of all of the Greeks), but just those Greeks who were overly dissipated. (VH9.32) Can we determine more precisely the tone at which Aelian was aiming? An examination of the more general qualities of the anecdotes reveals certain tendencies in Aelian's selection of detail and manner of presentation, tendencies which in turn suggest both his view of the intrinsic worth of the anecdotes and the message he sought to convey in collecting this material. For although Aelian's anecdotes do not function like the 198 rhetorical exempla in forming supportive material for a general thesis which shapes an entire discourse, they still demand an inductive or metonymic interpretation. That is, by allowing each anecdote to take up a position in Aelian's miscellany, we can begin to see a thematic pattern developing. If we segregate the anecdotes from the material dealing with customs, mirabilia, natural history, and the rest, we notice immediately that Aelian is not primarily concerned with the specific details of his narratives in terms of precise dates, places, and sometimes even characters. An Olympiad date stated in reference to an event (e.g. VH 2.8) is a singular occurrence. Nearly all anecdotes are datable through their context alone, presumably because a precise date does not concern Aelian; he is satisfied with a general approximation to the anecdote's correct time period. Moreover, few places more precise than the name of the city or state in which the anecdote occurs are given. Aelian must consider few to be very important. Even individuals may be referred to in a cavalier manner. At VH 9.9 Aelian mistakes Demetrius Poliorcetes for Demetrius of Phalerum; the recent editions of the Varia Historia do not hesitate to correct the text here. Aelian seems to have had particular trouble, or been particularly nonchalant, in keeping the Ptolemies straight. At 14.43 he begins an account of Ptolemy's addiction to dice with the words 6 pev riToXepaiog (baoiv (orcooog be auTOJV, edv 8et) K.T.X. They say that Ptolemy (whichever one he was does not concern us) etc.2 ' He repeats his parenthesis when dealing with another Ptolemy at NA 8.4: EpEO0E. Go ask somebody else which Ptolemy he was. OTTOcrroc 5E f | v O U T O C E K s i v o u c 199 a) Minor Characters in Major Events Despite Aelian's reticence, most of his anecdotes can be located in time and place because they tend to cluster around major events which had, by the second century AD, long been enshrined as paideia-monuments. Characters can be immediately given a frame of reference by their connections with four major "theaters": the expansion of Persia and its interaction with the Greek world during the later sixth century BC, the Persian Wars, Athens and Sparta in the second half of the fifth and first half of the fourth centuries BC, and the campaigns of Alexander of Macedon. That is, many of Aelian's anecdotes expand upon details already treated in the works of Herodotus, Thucydides, Xenophon, and the Alexander-historians. As we have seen, Aelian probably did not compile these authors directly. Although he cites Herodotus at VH 2.41, a glance at the parallel passage in Athenaeus shows that Athenaeus too quoted Herodotus in his text; Aelian and Athenaeus may have shared a source here. Again, chapters such as VH 6.1, an account of Athens' cleruchizing of the territory of the Chalcideans, differs enough from Herodotus' version (5.73-77) to suggest that Aelian was again accessing a source which had compiled Herodotus or referred indirectly to the passage in Herodotus' text. Aelian frequently relates situations and incidents to which the canonic historians may have only referred in passing: Xerxes' passion for a plane tree (VH 2.14), for example, Cyrus' relations with Aspasia of Phocis (VH 12.1), the low social status of Hyperbolus (VH 12.43), a near-death experience of Theramenes (VH 9.21), and a number of others. The impression here is that Aelian is supplementing the accepted historical accounts by highlighting deuteragonists 200 and tritagonists associated with events which have acquired an almost theatrical status.3 Only rarely in the Varia Historia does Aelian include a chapter completely devoted to characters and events thoroughly treated in a canonic historian. Chapter 3.25, for example, on the self-sacrifice of Leonidas and the Three Hundred, is exceptional in this regard, in fact more epideictic than anecdotal. In general, the anecdotes in the Varia Historia cluster around the above historical periods, only rarely focusing upon a prehistoric or legendary figure such as Neleus (VH 8.5), Lycurgus (VH 13.23), or Lepreus (VH 1.24). When Roman material is included (VH 7.16, 9.12, 12.6, 12.11, 12.14, 12.25, 12.33, 14.45) it does not deal with periods later than that of Augustus (VH 12.25). b) Paideia Icons in Anecdotal Situations In addition to those anecdotes focused upon secondary characters and incidents within important events and eras are anecdotes which take as their characters well-known figures in Greek culture. These figures may be statesmen and kings (both good and bad), poets, musicians, painters and philosophers. All are paideia-icons in the sense that they are encountered frequently in canonic literature. However, these figures stand in two different relationships to that literature: either as themselves producers of works of art, or as the subject matter of that art; in some cases they appear as both. All incidents involving the kings of barbarian nations, for example, came to Aelian as the contents of other people's histories. But in the case, for example, of figures like Demosthenes, Aeschines, and Critias, authors canonized as school authors for the student of rhetoric, Aelian will have both read of them in histories and studied their 3 Philostratus, we may recall, had defined the Imperial sophists' subject matter as being vnoQiaeic, set)' ocg rj iOTOpia dyei plots drawn from material provided by history [VS 481]; cf. p. 7 above. 201 original writings, as did his immediate sources; Aelian's own anecdotal material reveals that he encountered biographical scholarship and traditions of criticism in relation to such figures as Plato, Aristotle, and Epicurus. In considering the material which Aelian chooses to relate about these figures we must keep in mind two points. (1) Aelian is in most of these cases presenting a figure built up not only from a number of specific sources in traditional literature, but blended with an almost folkloric caricature derived from generations of mimetic literary treatment; (2) each figure derives his importance in Aelian's eyes, his position as actoc; pvr|pn.c,, from his relationship to paideia. That is, while anecdotes involve specific events, the figure in the anecdote is ultimately embedded within a multidimensional permanent record of paideia. Thus Aelian's paideia-icons perform and sometimes interact with each other on one plane, but do so surrounded by clouds of connotative tradition. We valorize the character as worthy of attention because o f this aura of tradition, but we focus upon the simple incident and allow the incident either to arise out of the character's tradition or to legitimize that tradition. An example might clarify Aelian's anecdotes of this type. In the second half of VH 4.9, Aelian narrates the following: (on ) nXdriov TOV 'ApioroTeXn exdXei IIioXov. ii 8e ePouXero auric TO ovopa EKEIVO; 5nXovoTi ibpoXoynTai TOV radXov, orav KopeoOfj TOU pnTpopou ydXaxTog, XaKTi^eiv TTIV pnrepa. pviTTero ouv Kai 6 nXdTcov dxapioriav Ttvd TOU ApioroTeXoucj. Kai yap exeivoc; TCX peytora eic; tJnXoootjnac; jrapd nXdriovocj XaPiov orreppaTa xa i ecb68ta, eira urcoTrXrioOeig Ttov dpioxtov Kai dtbnvidoag, dvTtpKo56pnoev aurto Siarpipfiv Kai dvnrcapecjriYaYev ev Tip jrepuraTip eTaipouc; extov Kai 6piXn.Tag, Kai eyXixero avrircaXocj eivai nXdrtovi. [That] Plato called Aristotle "Colt." What did he mean by giving him that name? Obviously the point is that the colt kicks its mother as soon as it has got its fill of her milk. So Plato was making an allusion to Aristotle's ingratitude. In fact, Aristotle took from Plato the most potent elements of and means toward his own philosophy. Then, filled with these excellent things and 202 rebelling against Plato, he built up his own school in opposition to Plato. He would walk in the Peripatos with his own pupils and try to compete with Plato. His goal was to be Plato's adversary. Aristotle's ungracious actions toward Plato are the point of the anecdote, but the import of the action is insignificant apart from the characters' biographical traditions. Proteges betray mentors frequently enough in daily life to make the incident trite. The event takes on more significance when we are familiar not only with the personal writings of each man involved (including an acquaintance with Aristotle's pragmatic definition of "friend," for example [EN 9.9], as opposed to Plato's sketches of Socrates' patience and loving kindness toward the young men who associated with him), but with their biographies and other authors' accounts of them, in which case the love/hate patterns such as those of Dionysius and Plato and Alexander and Aristotle also exert an influence upon a reading of the anecdote. This is, of course, precisely the way that the rhetorical exemplum operates, and the reason why the acquisition of a broad general culture was considered necessary for the ancient orator, c) Statesmen and Politicians The famous statesmen with whom Aelian deals in anecdote format are for the most part presented as either successful leaders or as tyrants; their historical records being in general so well known, it would have been difficult for Aelian to treat them in any other way. But by presenting anecdotal details of their lives Aelian not only fills out that record to make the figures more humane in their biographies; he also emphasizes their relationship to paideia (in this case, their positions in history) and thus the manner through which they are dcjioi 07rou5fjc;. 203 Thus non-Greek figures may be presented in incidents which emphasize their historical roles as aliens, as not having benefited from Hellenic paideia. Xerxes nearly dies of thirst because he.cannot drink water like any other person (VH 12.40). Anacharsis despite his exposure to Greek culture still drinks too much liquor (VH 2.41). The King of Persia prefers whittling to reading (VH 14.12); the brahmin Calanus chooses a painful and ritualistic death on a blazing pyre to the discomforts of an infirm old age (VH 5.6). When dealing with Greek statesmen, Aelian prefers incidents which contribute to the interpretation of an individual as a failure or success, either evaluation depending upon the larger moral point Aelian is trying to make. In general, a statesman's success or failure is already apparent from the historical tradition. Aelian presents that success as a paedeutic model to follow, the failure one to avoid; but to do so Aelian may have to redefine the terms of the tradition. Aelian tends to relate, for example, incidents in which statesmen make decisions or persist in some course of action which may be seen as injurious to themselves. Phocion is one of Aelian's favorites in this regard. The tradition presents Phocion as a man of limited financial means involved in politics during a time of turmoil, finally put to death by the state on a charge of treason. Is he to be regarded as a success or a failure? Aelian almost perversely insists upon his success, precisely because he willingly suffered so many apparent failures: though abjectly poor (VH 2.43, 14.10) he rose to power through the force of his personality and not through the influence of others (VH 12.43), yet was so outstandingly virtuous (VH 4.16, 3.17) that he rejected all the temptations of wealth (VH 1.25, 11.9) only to die at the hands of an ungrateful public which he had served 204 thanklessly but faithfully (VH 3.47). Timotheus and Epaminondas too enjoy a number of such anecdotes (e.g. VH3.47, 13.43, 11.9, 5.5). As negative models Aelian consistently chooses tyrants, focusing upon anecdotes which reveal their basic lack of excellence. In this regard Dionysius I and II are given much attention, in terms of the evil deeds they perpetrate (e.g. VH 1.20, 13.34, 13.45, 6.12, 9.8, 12.47), of their personal viciousness (VH 2.41, 6.12), and of their inevitable demise (VH 4.8, 9.8, 12.60). But in regard to both good and bad statesmen Aelian is especially interested in anecdotes which display political figures interacting with other paideia-icons. In this category we may include stories about statesmen's encounters (1) with philosophers: Dionysius' hospitality toward Plato (VH 4.18), Philip's financing of Aristotle's research (VH 4.19), Alexander's gratefulness toward Anaxarchus (VH 9.30); (2) with poets: Dionysius condemns Philoxenus to the quarries for his criticism of the tyrant's poetry (VH 12.44), Ptolemy builds a temple to Homer (VH 13.22); (3) and with painters: Alexander and Apelles (VH 2.3), Megabyzus and Zeuxis (VH 2.2). Aelian also relates a number of anecdotes dealing with a statesman's personal acquisition of paideia (e.g. Hieron [VH 4.15], Hipparchus [8.2], Dionysius [13.18], Alexander [3.32]). d) Philosophers and Poets Philosophers function as the main characters in a number of Aelian's anecdotes, but only rarely are their specific doctrines presented as closely related to the anecdotes in which they appear. Epicurus and his followers are the exception (VH 4.13, 9.12), Aelian showing himself in the Varia Historia as consistently negative toward this school as he is 205 in the fragments (e.g. fragments 10 and 61 Hercher). Only one chapter in the Varia Historia could be considered doxographic (Peripatetic doctrine on the physical location of the soul [ W 3 . l l ] ) . In other anecdotes featuring philosophers, Aelian is primarily concerned with displaying the effects of philosophy upon the individual's daily life. He is especially concerned, that is, with incidents which show that an action motivated by the love of wisdom will be a correct one. Here we may consider not only practical actions such as Meton's avoidance of conscription based upon his knowledge of the stars (VH 13.12) and the positive contributions made by a number of philosophers to the immediate needs of their homelands (VH 3.17), but the reactions of philosophers to the wrongheaded statements or acts of others: Socrates' responses to Alcibiades (VH 2.1) and Apollodorus (VHXA6), for example, or Plato's to Anniceris (VH2.21). Diogenes the Cynic is a special favorite with Aelian in this regard. Doubtless the availability of c//ra'tf-collections made the compilation of this philosopher's witty sayings much easier for Aelian (Stemplinger 1912: 222; Wehrli 1973: 195). His selection of only those chreiae of Diogenes which were sober and bitter (VH 4.11, 4.27, 9.19, 9.28, 9.34, 10.11, 12.56, 12.58, 13.26, 13.28, 14.33) rather than shocking and ribald (as seen e.g. in Diogenes Laertius), demonstrates Aelian's interpretation of what was worth remembering. Aelian may present a philosopher reacting positively to sickness or death (e.g. Socrates, VH 1.16, 2.6,2.36; Epicharmus, 17/2.34, 8.14; Aristotle, VH 9.23), or he may recount a future philosopher's conversion, the point in his life at which he chooses a philosophical career: Plato's rejection of poetry for philosophy (VH 2.30) and of military 206 service for philosophy (3.27), Aristotle's enrollment in the Academy after a brief career as a pharmacist (VH 5.9), Diogenes and Antisthenes (VH 10.16; on such anecdote as emphasizing the moment of conversion cf. Nock 1933: 180-185). In this regard we could add Timotheus's wistful resistance of the temptation to become a philosopher (VH 2.10). Aelian must have seen some literary potential in stories of philosophers interacting with each other within their communities, for two of his longer, more developed chapters are expanded anecdotes of this type. At VH 3.19 Aelian gives quite extensive treatment to the moment at which Aristotle is supposed to have thrown off his allegiance to Plato and driven the older man from the Peripatos, while VH 2.13 narrates the circumstances leading up to the performance of Aristophanes' Clouds as well as the occasion on which Socrates rose in the theater in order to acknowledge his identity with the figure in the play. Aelian's presentation of Aristophanes in VH 2.13 as a tool of the demagogues seeking to destroy Socrates is an unusually hostile one, as was Aelian's treatment of Aristotle in VH 3.17 and 4.9 (although elsewhere in Aelian Aristotle is presented in a more positive light). Aelian allows the demands of any given context to influence his portrayal of character. In general his anecdotal treatment of poets, like that of philosophers, focuses not so much upon their work as upon those aspects and events of their lives which in some way throw light upon the interpretatio