This volume provides a collection of fragments, preceded by a substantial (62pp.) essay in which the material is surveyed in broadly chronological order. A few works preserved in extenso are referenced, but sensibly not reproduced in full: Aristotle’s Poetics, Platonius on comedy, Satyrus’s life of Euripides and Plutarch’s comparison of Aristophanes and Menander (Dio’s Oration 52, comparing three versions of Philoctetes, is a surprising absence). Some fragmentary authors are treated selectively (including Aristophanes of Byzantium, Didymus, Eratosthenes and Lycophron), as are certain categories of text (see below). The collection of fragments follows a familiar format: authors are arranged alphabetically; each author is given a number; and the fragments of each author are themselves numbered. So if the not-very-brief abbreviation proposed by B. is adopted, citations will take the form (e.g.) AntTrDr 13F5. As it happens, AntTrDr 13F5 turns out to contain a discussion of hydraulic organs, which may prompt the reader to ask: what have hydraulic organs to do with drama? Or the gargantuan trumpeter reported by Amarantus of Alexandria (2F2)? The criteria for inclusion in this collection evidently merit closer investigation.
First, despite the title, this book is exclusively concerned with Greek treatises on drama. Latin authors are not covered; Varro, for example, gets a passing mention in the introduction (p.20 n.8: not indexed) but has no place in the collection of fragments. Secondly, despite the first sentence of the preface, the book is only concerned with ancient treatises on drama. ‘Byzantine’ texts are excluded (although that term is not defined). Platonius is admitted on the grounds that Kaibel thought him pre-Byzantine (p.72: a complete misunderstanding — in the passage cited Kaibel is referring to Platonius’ source; in CGF he says of Platonius only ‘Tzetzis fratribus… aliquot saeculis antiquior’), but otherwise the prolegomena on comedy, including the Tractatus Coislinianus, are excluded.
Thirdly, the collection is concerned with treatises. This means, first of all, that it is limited to written texts. That restriction seems innocuous, until one considers what it means for the various current opinions on drama mentioned by Aristotle: in each case, B.’s project requires a decision whether the opinion had been written down. Such a procedure is inevitably arbitrary: Ariphrades (12**F1 = Poet. 1458b31-9a4) is allowed in (admittedly with asterisks indicating serious doubt), but there is no mention of the possibility that the recognition attributed to ‘Polyidus the sophist’ ( Poet. 1455a6, b10) was proposed in a critical treatise; and opinions which Aristotle cites anonymously are ignored. But picking out those opinions which Aristotle attributes to a named individual who we conjecture may have written the opinion down in a treatise on drama gives a very incomplete picture of the evidence for the state of literary discussion in Aristotle’s time. It does not even help us address the question how far that discussion had been conducted orally and how far in writing — to make judgements about that, we need to see the larger picture.
With later authors we can be more confident that their sources were written, but we cannot be sure that the written source was a treatise rather than (for example) a commentary. The collection therefore excludes virtually everything preserved anonymously. When it cannot be traced back to an identifiable treatise, the mass of material relevant to drama to be found in scholia, lexica and other compilations must be omitted. The question must then be asked: what is the point of devoting a book to this somewhat arbitrary slice of ancient writing on drama? B. never explicitly addresses the rationale of his project. One possible answer might be that it gathers together the evidence for the genre ‘treatise on drama’. But that raises further questions: was there such a genre? And does this collection succeed in gathering the relevant evidence?
First, what counts as a treatise on drama ? Taken strictly, it would exclude material drawn from treatises in which drama is a (perhaps significant) subject, but not the sole or main subject. In fact, B. is not so restrictive. He admits, for example, works on famous prostitutes, although a reading of book 13 of Athenaeus shows that these works did not restrict themselves to prostitutes named in comedy, and that the range of sources on which they drew was very wide. If these are treatises on drama, then so are many kinds of writing that B. leaves out. When a work of diverse content is included, it is not always easy to discern the logic of B.’s selection from the fragments. Aristocles’ On Choruses is a case in point: fragments on satyric (13F3) and tragic (13F4) dance manifestly earn their place, but if hilarodes (13F1-2) and those hydraulic organs (13F5) are admitted, why are
Some works seem, at first sight, to pose no such problems. It is self-evident, for example, that all the fragments of Apollodorus On Sophron deserve inclusion. But in fact there is a problem: does this mean that all the fragments of Apollodorus on Sophron are to be included? B. includes fragments which our sources explicitly attribute to On Sophron, but omits FGrH 244F238, a gloss in which Sophron is cited but which is not explicitly attributed to that treatise. Likewise, fragments of mythological lore transmitted under the name of Asclepiades of Tragilus are printed if they are explicitly attributed to the Tragoidoumena, but omitted if there is no explicit attribution. (There is one exception: no text is printed for 21F14, only a reference. I hope that the reason for the omission is not that, since the source is not on the TLG CD. would have had to type the text out himself; but I cannot think of any other explanation.) The principle of excluding fragments that probably, but do not demonstrably, derive from a treatise deemed to be on drama would not (I think) be a wise one; but further investigation reveals that B. is not following any such principle. Consider Hieronymus: an anecdote about Sophocles (56F2 = F31 Wehrli) is included; another, about Sophocles and Euripides (F35 Wehrli), is omitted. The reason for the omission is presumably that our source attributes it to the
More fundamentally, it is clear that B. has failed to think through the demands of the particular kind of fragment-collection that he is trying to create. When one is trying to collect all the fragments of a given author, the more cautious and the less cautious approaches are both defensible; either way, all the evidence will be made available to the user — only its position in the collection will be affected. But neither model is satisfactory for a collection such as B.’s. Following Jacoby means that we are not given a full picture of the evidence for Apollodorus’ work on Sophron, nor of the evidence that is potentially relevant to the treatise entitled On Sophron. In the case of Hieronymus, following Wehrli means that we are not given a full picture of the evidence for the circulation of anecdotal material about dramatists in antiquity; nor of the evidence for Hieronymus as a source of such material; nor (crucially for B.’s project) of the evidence for Hieronymus’ treatise On Tragic Poets — since B. does not alert us to the possibility of doubt about the assignment of 56F2 to On Tragic Poets, and does not present the evidence that would enable us to discover that possibility for ourselves.
The following observations (culled from a reading that I cannot pretend was rigorous and exhaustive) will give an indication of the level of care with which the collection of fragments has been prepared:
2F2: the reference is wrong (read ‘414e-415a’). This fragment overlaps with 68F1, but there is no cross-reference: B.’s policy on cross-referencing seems inconsistent, and his practice is not always accurate (15F6 is cross-referenced as 15F5 under 10F2, and as 15F7 under 49F2: exactly the same extract from Athenaeus is printed in each case, but the record for such duplication is taken by 3*F6 = 4*F1 = 8F1 = 10F1 = 15F5 = 49F1).
8F3: the words
13F1: another dittography: fifteen words (
20**F2: the extract from the scholia to Wasps has been truncated in a way that leaves it unintelligible.
28**F3: the text as printed is unintelligible: to work out why
31*F1: the reference is wrong (read ‘1095a’); the last four words of the sentence preceding the extract have been mistakenly included.
31F7: B. unwisely follows Wehrli (F84) in treating an explanatory expansion by Thomas Magister as if their were part of Dicaearchus’ text.
43F17: the square brackets do not balance.
53F2: a lacuna is marked, although B. does not believe that the text is lacunose (see p.30).
71F1: the lemma is not marked off from the body of the scholion.
The introductory essay must also be treated with caution. But critical readers will in any case be put on their guard by the very first page, with its futile conjectures about the contents of Sophocles’ On the Chorus. Doubts about the existence of a work attested only in the Suda are dismissed (p.11): ‘Immerhin kann anhand der biographischen Quellen, die Sophokles als Erfinder technischer Varianten zum Drama vorstellen, als gesichert gelten, dass er eine Arbeit über die Bühnentechnik verfasste.’ The logic of this escapes me totally. The rather better attested second book of Aristotle’s Poetics is, by contrast, dismissed without argument (p.22). There is no reference to Janko at this point, an omission which makes p.23 n.25 a masterpiece of misleading perversity: ‘Zu Theophrast als mögliche peripatetische Quelle für den Tractatus Coislinianus vgl. R. Janko… 48ff.’ On p.16 unnecessarily heavy weather is made of an anecdote in which Antisthenes cites a chreia of Sophocles (not otherwise attested — but so what?); Antisthenes 9**F1 is certainly not a fragment of a treatise on drama. The classification of Heracleides of Pontus as a Peripatetic ceases to surprise when one reads B.’s discussion of the fragments (p.16f., p.30): H. Gottschalk Heraclides Ponticus (Oxford 1980) is cited, but its lessons have evidently not been absorbed.
Recently Thomas Schmitz (BMCR 99.4.20: my review of the same book, forthcoming in CR, reaches similar conclusions) has posed ‘the disturbing question whether the German academic system can continue to flood the shrinking market of scholarly monographs with unrevised dissertations of questionable value’. The present book, which originated as an Italian rather than a German dissertation, raises the same question. It is not the only volume in Beiträge zur Altertumskunde that I have recently had for review in which the scholarship proved to be flawed in easily discoverable ways (see JHS 118 (1998) 216-7). The last one, volume 64, was published in 1996; this one, volume 111, was published in 1998 — a statistic which may throw some light on the editorial failure.