Bryn Mawr Classical Review 95.05.02

M.J. Osborne and S.G. Byrne, A Lexicon of Greek Personal Names, Volume II, Attica. Oxford: Clarendon Press, 1994. Pp. xxi + 510. $75. ISBN 0-19-814990-5.

John S. Traill, Persons of Ancient Athens, Volume 1, A- to Alexandros. Toronto: Athenians, 1994. Victoria College, Toronto, ON, M5S lK7, Canada. Pp. xxv + 339. $125 + $10 postage & handling. ISBN 0-9692686-2-9.

Heiki Solin and Olli Salomies, Repertorium nominum gentilium et cognominum Latinorum. Editio nova. Hildesheim/Zurich/New York: Olms-Weidmann, 1994. Pp. x + 508. DM 148. ISBN 3-487-07986-0.

Reviewed by Kent J. Rigsby, Duke University (

1994 witnessed the appearance of two lists of persons in Attica. While Kirchner's Prosopographia Attica (1901-03, two volumes, ca. 16,000 entries) included only Athenian citizens and only the period down to Augustus, the Oxford and the Toronto projects are substantially more ambitious: all of antiquity, and far more generous definitions of "Athenian" (which itself had been a political football in classical Athens). A simple statistical comparison of the two is not possible; the Oxford entries are not numbered (the authors offer the total 62,360) while the Toronto entries are not numbered consecutively but leave gaps for future entries.

Lexicon II continues the project, headed by Peter Fraser, to tabulate persons named in Greek lands. The first volume, encompassing the Aegean islands and the Cyrenaica, appeared in 1987. This dense matter has been given a clarity of presentation for which all users will be grateful: well-chosen fonts and styles distinguish the categories of information. Despite the inevitable discomforts that the volume of data imposed (large pages, small type in three columns), the entries are easily readable and even elegant. The price is a bargain.

The oddities of the overall plan (described in the first volume) persist here in modified form. This project is called a lexicon but its content is a prosopography; its target is not the possible names in the language of a place, but the persons attested in that place (a "lexicon" would enter DHMH/TRIOS once, not 782 times). The authors have therefore quite properly labored to distinguish persons, preferring to err in the direction of fission rather than fusion. The present volume differs from the first in adding a reverse index (of names, naturally, not of persons).

The oddities of scope and goal are unproblematic for this coherent material from one city, material moreover which has been worked on across a century by excellent scholars. We know why we want a list of persons in Athens. It will be recalled too that neither IG II2 nor I3 has an index. This volume will be unique in the series because of the significant role played by literary sources and the complexity of attestation of so many individuals; elsewhere most Greeks are attested only once, and by a document. This Attic volume will also have far more users, with more diverse agendas, than any of the others in the Oxford Lexicon.

The entries are highly compact: deme or tribe when known, date, bibliographical citations, immediate kin; quotation of the Greek text (name only) when there is restoration or a spelling variant. Because function is so rarely indicated ("archon epon.," "epeng.," "het."), a searcher can find the right homonym only if forearmed with knowledge of one or more of these items. Thucydides the historian does not leap to the eye from among the 18 homonyms (no. 11); many users of this book will know his father's name as a guide, but fewer his deme, which is the first piece of information in these entries. Spelling variants are normalized tacitly (there is no cross-referencing) but not pronunciation variants; XRUSA/RIN is separate from XRUSA/RION, E(STIAI=OS from I(STIAI=OS. Persons with Roman tria nomina are listed under the cognomen, without cross reference for the other names (so the student of Vibius will readily find Julius Vibius, but not Co. Vibius Crispus, who keeps company with Crispi).

In keeping with policy the volume excludes metics whose origin is known (and who will therefore be included in the relevant non-Attic volumes), even while perforce including those whose origin is not known (and who thus cannot be placed elsewhere). The authors are sufficiently uncomfortable with this decision that they promise soon a separate volume that lists the foreigners in Athens, which will be welcome. Slaves are reserved for a common grave in the sixth volume. Fragmentary names are excluded except where the restoration seems to be regarded as certain (an extreme case is DIE/M[POROS]); in this particular this is a lexicon rather than a prosopography.

The most problematic elements of their criteria for admission are their headings "Athens?" and "Athens*": the first signals "some likelihood of being Athenians," the second "the residents of Attica who do not appear to be Athenian, whose place of origin is not known, but who are not slaves or merely visitors to the city" (p. x). The reasoning behind these two judgments cannot be spelled out for each case, given the compact format, and often remains obscure (e.g. Marathon the father of an ephebe of Roman date is "Athens*"; claimed as a heroic name at Philostr. 553). "Mere visitors" evidently include the sophists (whose origin if of course usually known and who will be listed in the relevant volume); but the many brief mentions of persons in Attica by literary sources make this a difficult distinction to draw. I note for instance the absence of Nicostratus alias Clytaemnestra at Diog.Laert. 4.18: overlooked, or taken to be a mere visitor?

Fictitious characters are included ("fict."), which has its uses for reconstructing an Athens of the mind; but only access to the computerized data would produce a list of them, short of reading continuously for this (or any other) detail. Here as elsewhere the addition of a question mark ("fict.?") is an inadequate guide to the editors' thinking: EU)RU/TH: Lang labeled this name "mythological," however that was intended; PRAC<I/>NH on the same vase draws no question mark. O)RSIMH=S: this labels a warrior at Troy not known from literature: surely fictional, despite the question mark, but in what sense Athenian? O)RSI/LOXOS: the scholiasts to Ar. Lys. 725 assume this was a real person but offer only speculation about his character; it is the great virtue of a book of this sort to reveal a second instance of the name, on a fourth-century grave stele.

New discoveries and original contributions are scattered throughout (e.g. EI)DA=S, EU)TUXI/S 12). The authors have included the names in a set of unpublished squeezes from Attica kept at the Institute for Advanced Study (see p. ix). The treasures of the volume are endless. While Athenaios and Athenais are popular, Atthis is quite rare and late, apparently not felt primarily as a name (Athenaia is of course entirely avoided). Again, Arsinoe is rare, despite the extraordinary popularity of Philadelphus' wife. A number of names in the defixiones remain attested only there (e.g. Apistia, Memphades). Kopreus, the herald in Eur. Heracleidae named in the dramatis personae but not in the text, has often been taken as an Alexandrian scholar's deduction from the Kopres at Iliad 15.639; so it is interesting to find a Kopreus once in Athens, on a pot of the fifth century B.C. (for the type see Robert, Noms indigènes 53). The lone Strophe undermines an old classroom joke.

John Traill's project had its origin in the work of Benjamin Meritt to maintain current data on Athenian prosopography. Meritt (whose account of this enterprise, written in 1984, is given as a Foreward) early on saw the potential of the computer to make his paper resources at the Institute for Advanced Study dynamic and expandable, and by 1972 Traill was at work entering data.

Twenty volumes are projected, the first fourteen of which are described in this first as "in press" (the second has now appeared). The total cost will exceed $2000; the publisher, "Athenians," seems to be an office in Victoria College, University of Toronto, apparently with no other publications and little means of distribution; the book is not listed in the 1994/5 Books in Print. It seems clear that, in practice, libraries and individuals will acquire the Oxford U.P. Lexicon and not the Toronto volumes.

Persons goes beyond the scope of Lexicon II in listing all foreigners: metics, slaves, and indeed "foreigners who have been honored by Athens," which proves to include some who never set foot in Attica (e.g. several Macedonian kings in the present volume). In fact, all persons named in Athenian documents are intended -- so the Athenais merely mentioned as mother of an honorand, Ariobarzanes of Cappadocia. Thus the difficult and to a degree arbitrary decisions about inclusion made by Lexicon II are not a problem here. Persons also includes fragmentary names, on which, at Athens, it will be possible to build, given the rich context that the volume constructs around them. Spelling has been normalized, and variants are in the alphabetized list only to supply a cross-reference (phrased illogically as "see also") to the normalized name; in the quoted Greek, variants, however familiar, are marked with !, which quickly becomes tiresome (A)GRIPPEI=NA!); this includes normal dialectic variants like the genitive FILOKRA/TEUS! on Lesbos (102715).

Even more aggressively than Lexicon II, Persons aims for fission over fusion: "The result has been a large-scale dismemberment of the traditional identifications in Attic prosopography" (xvi); but also new equations, and new dates. Persons is the more skeptical book, inserting far more question marks about readings, dates, kinship. On the other hand, Persons is less concerned to canvass the literary sources for possible Athenians (so omitting e.g. Lexicon II's Agathion 3, "Athens?," from Philostratus). "Fictitious" is not a category, yet some literary characters are admitted; some odd disparities result between the two projects, so a hetaira-name Habrotonon enters Persons only from Menander, Lexicon II only from Lucian. But in both projects the thinking behind the decisions and question-marks is left to the reader to reconstruct, a task far easier in Persons because of its extensive quotation of sources.

For purposes of illustration, I quote each volume on Habron of Boutadai:

Lexicon II, under A(/BRWN:
-- Boutadai (18): s. iv BC Plu., Mor. 843a; 843e-f; IG II2 463, 36; 1492, 123; Reinmuth, Ephebic Inscr. 12, 72 (PA/APF 15) (s. LUKOU=RGOS IV, KALLIOPW/ (Bate))
[All entries per deme are printed run-on.]

ephebe, lochagos, ca 330a. Status A. Treated in APF 9251 p352. Possibly the same as 101575.
1.1 H S8 p274, line 8 (cat eph).
= 1.2 ReEI p42 12, line 8
2.1 H S8 p274, line 70 (cat eph).
= 2.2 ReEI p43 12, line 73
[where D = APF; A = citizen; H = Hesperia; ReEI = Reinmuth (omitted from the abbreviations, p. xxiv).]
[where * = by deduction. This entry gives a description of him as litigant, followed by seven literary references each with substantial quotations from the Greek; as politician (one passage); as a treasurer twice (one passage each); and as dedicator of portraits (one passage).]

Every user will be grateful for the generous quotation of sources in Persons; but the cost is substantial, for the two entries on Habron occupy three-quarters of a page where Lexicon II takes four lines of one column. There is some needless swelling: the double allusion to APF, ", line" where a period would do; above all, separate lines giving (selectively) the publishing history of individual inscriptions, each time repeating the line number. In the end, the entries do not make clear the basis for distinguishing ("possibly") two men where others have seen one.

For all its expansiveness, Persons neglects to give some valuable and easy information; above all of persons honored by "decr forgn" it would have occupied comparatively little more space to identify the government; likewise "decr club." (Also, it often omits dots in quoting Greek.) Neither project remotely approaches the argued detail of Davies' APF, which readers will always consult for the wealthy of Athens.

I note the following details: 013040, the status entry after the husband of this woman from Ancyra appears intrusive; 102575 with inconsistent AGAQ[I / A)GAQI [ repeats a typographical error from IG (the second is correct); of the one A)GGE/LHS we have both a question mark and "name doubtful"; this apparently reflects no more than the question mark printed in IG II2, where again the intention was unclear; 102710, read "99/8" (both Lexicon II and Persons rightly accept the lower dating of the stephanephoric coinage, citing Habicht).

It is distressing that these two volumes should appear simultaneously and, despite abundant public preview over recent years, without reference to each other (the Lexicon editors thank Traill among others on p. viii but do not mention his project). A great deal of labor has been duplicated to produce incompatible databases. The authors must understand that (especially in the current fiscal climate of the academy) scholars and libraries will be forced to choose.

In 1988 Solin and Salomies gave us their important volume listing Latin nomina and all cognomina; six years later Olms has reissued this with an attached supplement of addenda and corrigenda. The original book is unaltered, except that its one page of addenda has been subsumed into the new attachment; this supplement is printed in a different and better typeface which distinguishes upper- and lower-case letters and can print Greek. It is continuously paginated with the original; the names listed in this addition occupy 22 pages (the whole is 34 pages when reverse-indexes, preface, and empty sheets are counted). Owners of the original volume can be expected to bridle at the $100 price of this improvement. It is a pity that the supplement was not issued separately (as was done with the far more substantial addenda to Broughton's MRR). It is a greater pity that a project which from the start was computer-based did not use the power of the computer to integrate this new material with the old and produce a coherent volume. Because gentilicia and cognomina are listed separately, the searcher of a name will now look in four lists, each followed by its own reverse-index.

The following remarks are confined to the supplement. As in the original, the authors admit as gentilicia all names that are found so used, but as cognomina only those that are linguistically Latin (quite rightly eliminating thereby the many Greek names that become cognomina with the gaining of Roman citizenship). The authors' knowledge of the Latin material is formidable and current, and sometimes based upon unpublished texts; their control of the Greek evidence is somewhat bookish and haphazard. The entries offer more information than those in the original lists, which is much to the good; but there is much randomness in what is included: usually place, sometimes date, rarely character ("soldat," "Gladiator"); bibliography is usually the source but sometimes an encyclopedia article.

Because of their generous definition of novelty, there is less new in these lists than meets the eye. One criterion for inclusion is names (marked by an asterisk) that have proved erroneous and are to be deleted from the original book. It would have been far better for the computer to eliminate these from the original and generate a new text, a task of minutes rather than months. The same criticism can be made of names for which some particular testimony must be deleted (marked with a minus sign); names about which the authors now conceive some doubt (marked ?); and those for which testimony as been added (+) or corrected (!). The user must now check the supplement, watching for these signs, regarding every name consulted in the original. The brevity of exposition, moreover, often means that the user must resort to the original publications of the relevant texts to understand the basis of the addendum.

There remain names not found in the original volume; here the authors distinguish new entries (marked by bold-face type) from feminine cognomina where only the masculine had been known (not marked at all, e.g. Allina when the original had Allinus); the original was inconsistent on the gender question, normally giving only masculines, citing the feminine alone when that is all that is on record, but sometimes listing a masculine as a deduction from the feminine (e.g. Emerentianus).

As to genuinely new entries, which will represent the progress of the past six years, rather more are corrigenda (from texts overlooked) than addenda (from texts newly published). Many are merely spelling variants, a necessary feature given the conventions of the project, but of limited interest because the ways of variation are familiar -- thus geminations (Appulleius, BIH/SSIOS) are listed as new entries. Rarely, the normalized spelling of an already attested name is referred to with =, thus "Autillus (= Auct-)," sometimes (oddly) with "Vgl.," thus "Aceronius Vgl. Acheron-." More often there is no such reference and the user is left to deduce equations, in the first instance by reverting to the authors' original lists (whose great value cannot be overpraised). Thus KAILH/RIOS(?) is simply Celerius (the question mark is not explained); similarly Biturianus is Veturianus, KESSIANO/S(?) is Caesianus; I)OUBINIANO/S (= Iuvenianus?) is rather the attested Iovinianus; Pacideianus and PAKIDIANO/S are given as separate entries; I)STIME/NIOS at Aizanoi, listed separately from Istimenius at Rome and without comment (the original volume had already Istimennius and Istiminius): the editors of the Aizanoi inscription rightly remarked that this is merely the prothetic iota, for Steminius. Many additions are of quite late date, into the seventh century, when one begins to wonder about the proper definition of Latin nomenclature.

I note the following. A)KLA/N[I]OS is given without comment from an inscription from Bulgaria; IGBulg cites suggestions of Etruscan Aclani and Latin Aclenius; but is it not simply Aeclanius? FLABOULH/IOS in an inscription at Claros: this is FLABOULH/IOS BA=SSOS, a Neocaesarean from Pontus; as the same text shows abbreviated Claudius and Ulpius, it seems possible that this is Flav. Uleius Bassus. SEPI/LIOS at Cnidus: the editor W. Blümel reasonably suggested emending to SEP<T>I/LIOS; the whole grave-stone reads SEPILI/OU NIKI/A MNI/AS XA/RIN: SECTILI/OU A)NIKH/TOU; as the two deceased seem related, it is tempting to emend even further, SE<CT>ILI/OU. O(RDEA=NOS at Ephesus: correct reference is SEG 39.1223. Albucio is listed as a cognomen in a Christian inscription and then called "falsch"; inspection of the original explains neither the classification (the name is isolated) nor the judgment. Tutilla: the entry has then fallen out; one wants to be sure that this is not the familiar Tutilia.

Greek inscriptions are often cited from out-dated editions (despite the claim on p. viii). Usually this will not affect readings; but it puts the onus on the user to check the best edition. I have noticed that A)GUL-, cited from CIG and Le Bas-Waddington, in fact disappeared in 1883 (the proper citation is I.Tralles 140). In short, the reader is often obliged to go to the sources to obtain a clear accounting of the situation, and on the Greek side to do the homework of bringing the citations up to date.

The simultaneous appearance of these admirable books invites uncertain thoughts about the computer and publishing. The control that the computer allows upon printing meant, for example, that Lexicon II and Persons could take into account the second fascicule of IG I3 and Clairmont's Tombstones (both 1993), and the Repertorium is similarly current. Lexicon II by its complex exclusions has sought to avoid redundancy in a project that will eventually take in all Greek lands (and so the redundancy of two "Athenians" projects is all the more baffling). But redundancy in that sense is a meaningful concept only when deciding in which printed volume to give an entry: the electronic searcher of the whole will not care whether Protagoras should be "printed" with Athens or with Thrace.

The convenience of in-house typesetting has in the Athenian case produced a David and Goliath situation. We must wonder how many readers will locate an obscure press and spend $125 in order to own 5% of the Athenians available from Oxford UP for $75. But which will be David and which Goliath?

The personal computer broke the publishing industry in half. The complex traditional tasks of setting and inking type, producing pages, and assessing the result can now be done by authors or editors rather than professional craftsmen, using small, cheap, versatile equipment in private space. The other half of publication, what is done with those pages -- multiple copying, book manufacture, binding, marketing, distribution -- depends in its various stages upon large, expensive, specialized machinery and upon established commercial contacts and legal structures, and these remain in the hands of trained specialists of diverse sorts with little connection either with authors or with one another. This second half of the industry is now in the process of being rendered superfluous by electronic publishing. When the Internet comes to replace the stop-gap technology of the CD-ROM, academics in humanities will need to face squarely what it means to publish scholarship, who will guard the gates, and how the results are to be judged by those responsible for hiring, promotion, and grants.

Some assessment of these three books on their use of the computer is therefore in order. The least can be said for the Repertorium nominum. Here the computer was used as a flexible typewriter; it is inexcusable that in 1994 the pertinent service that the computer could have rendered, resorting the new material into the old, was not done. Moreover, the fuller information offered by the supplement is added so randomly that access to the electronic data to search would rarely ferret out interesting comparisons. The selfless labor that Solin and Salomies have put at our disposal has not been given the rich format that their unrivaled knowledge of Latin nomenclature deserves.

Lexicon II speaks cautiously of its names as "being subject to the type of analysis which a database makes possible" (p. vi). The scope of the project, all names in Greek lands, and the meager information appended for each (chronology, geography, immediate kin, "fict.," very occasionally other items such as "het.," "Jew," or gender) mean that the database will support only limited analyses, e.g. probability studies on the distribution of names in time and space. The recognition of epichoric names will be facilitated; but in the end such studies are valid only when all Greek lands have been covered, and that seems a distant prospect (the "Middle East," according to Lexicon I, will form a "second project"). Peter Fraser, speaking on the intended Lexicon at an APA meeting some years ago, indicated that the data would not be released to the public in electronic form; plans no doubt change, but Lexicon II contains no hint how or whether the database might be made accessible.

Persons from the start was designed with the computer search in view. "It is the goal ... to disseminate this material in both electronic and conventional formats" (xi); of electronic dissemination nothing more specific is said here, doubtless because basic decisions wait upon the state of technology when the project is completed. In the database a name is endowed with information in as many as fifteen categories (function, tribe, kin, etc.), and sources characterized with as many as seven (decree, metrical, etc.); this rich detail will be priceless to future research. Traill has worked on computer generation of stemmata, but the gain from this database will not be confined to family prosopography. We will eventually be able to generate lists of ephebes of the mid-third century B.C., metrical epitaphs of women, origins of spouses of foreigners, pottery owners from coastal demes, etc. Few scholars will or should buy the paper copy of Persons, because of its high price and the rivalry of the completed Lexicon II -- but mainly because it has doomed its own commercial viability by its intended electronic format. But no one will doubt that the future of these studies belongs to John Traill's project.