The central claim of this pair of volumes is that Classical tragedy was more diverse than is commonly accepted; or in its stronger formulation, that the tragic genre had no essential unity, but was a fundamentally heterogeneous affair. W. aims to highlight this diversity by providing a survey of fragmentary authors and plays accessible to an undergraduate or general reader. Although much remains speculative, and certain points of detail fail to convince, overall the work manages a useful balance between the introduction of a considerable body of largely discontinuous information and a lively picture of fragmentary tragedy of a kind that should encourage wider interest in the field.

W.’s books stand within a growing body of scholarship dedicated to fragmentary drama, spurred on in the case of tragedy by R. Kannicht, S. Radt and B. Snell's monumental Tragicorum Graecorum Fragmenta (TrGF). Specifically, these studies represent a contribution to making the field of lost tragedy more accessible to students and non-specialists, adding to work such as the Loeb Classical Library's editions of the fragmentary plays of the ‘triad’ of Aeschylus, Sophocles and Euripides. Within this context, W.’s first volume on fragmentary playwrights, which includes the first English translation of a meaningful portion of TrGF vol. 1, in particular helps widen access to the field.

W. organises his material in a variety of ways, so that Agathon receives his own dedicated chapter, while more than 40 tragedians with no extant fragments fill up another. The largely chronological order is broken in one chapter that groups together playwrights with identifiable family connections. In spite of this varied presentation, the author remains the basic unit of analysis throughout. This has the effect that the lengthier unascribed fragments are left aside, whether from unknown authors (TrGF vol. 2) or by the triad but from unknown plays. The two valuable chapters that conclude the second volume break away from the author-centred focus and provide thematic discussions, in one case looking at variation within different treatments of the same myth and in the other at distinct moments of stagecraft among the fragments.

The first volume opens with a discussion of pre-Aeschylean dramatists, linked by their particularly poor textual transmission. W. introduces Rossi's concept of ‘“submerged” literature, by which [he] mean[s] texts that were maltreated from the very origin of their transmission’ (L.E. Rossi, ‘L'autore e il controllo del testo nel mondo antico’, SemRom 3 [2000], 165–81, at 170, my translation). The concept is helpfully applied to early tragedy and to Thespis in particular, but in fact it has a much broader importance for W.’s overall subject than is allowed. Even the tragedies known to have been lost represent only a fraction of the total number of plays produced at the Athenian festivals, and our knowledge of total output for most tragedians is negligible because canonisation happened early on, and much was lost already by the Hellenistic period. W. regularly broaches this subject, noting for example that significant variants indicate a period of textual fluidity between a play's first performance and its integration into a Hellenistic catalogue (vol. 2, pp. 188–9). Particularly intriguing is W.’s suggestion that a play for which fragments are preserved may nevertheless have been ‘lost or hard to access’ for Hellenistic scholars (vol. 2, p. 184); unfortunately, the example he gives of Euripides’ Licymnius is not convincing, since Callimachus seems to have had access to some text of this play (Call. fr. 455 = Schol. (b) Ar. Aves 1242 = TrGF 5.43, test. iib.2 ταύτης τῆς διδασκαλίας οὐδὲν μέμνηται ‘[the phrase] is not used in this production’). In spite of occasional comments, the reader would have benefited from a more sustained account of the complex but fundamental topic of pre-Alexandrian transmission and of the ways in which our access to the lost plays comes pre-filtered by the work of ancient scholars. Surveys of fragmentary tragedy often imply that the triad dominated tragic competitions already in the fifth century, if not by virtue of wins, at least through greater productivity; a specific counter-argument taking account of early transmission would have gone a long way towards reviving the position of so many now obscure names.

The juxtaposition of a large body of often disparate material helps emphasise how great a variety of styles and approaches coexisted under the umbrella of Greek tragedy. W. suggests that Pratinas’ hyporchêma fragment may ‘belong either to a tragedy or to the shady genre of ‘proto-tragedy’ (vol. 1, p. 16). Phrynichus and Aeschylus are seen as operating in a milieu of theatrical experimentation by staging landmark contemporary moments entering mythical time (vol. 1, pp. 23–7). W. reads Agathon as chief exponent of this theatrical avant-garde, inserting ‘deliberately aporetic’ maxims into his plays and engaged in ‘formal experiments with dramatic structure’ (vol. 1, pp. 77, 80). Agathon concocts the embolima or portable choral interlude and produces some play, the Anthos or Antheus, with a ‘completely invented’ plotline (vol. 1, pp. 78–80, 81), as opposed to a mythological one. These are intriguing suggestions, and W.’s view of Agathon as a conscious innovator, subject to a lawless Muse (TrGF 1.39 F2 ἀθέμιστοι μοῦσαι), is largely convincing. However, it is a distortion of Aristotle to claim that Agathon ‘once attempted to fit the material of an entire epic poem into a single tragedy’ (vol. 1, p. 80). In fact, Aristotle implies that a number of poets including Agathon (Poet. 19.1456a18 καὶ Ἀγάθων) failed on the grounds of having plots with multiple narrative strands, and this speaks against the notion of a unique experimental attempt at a new type of play structure. At this rich banquet of tragedies, even a ‘realistic or naturalistic style’ appears distinctive, as in the case of the playwright Sthenelus, criticised for his overuse of props and his characters’ unpoetic language (vol. 1, pp. 178–80). It is remarkable that Sthenelus, of whose work not a word remains, was well enough known in his time that he probably featured in a comedy (Plato's Props?) about the theft of his theatrical equipment (costumes, props), a narrative plotline ‘uncannily reminiscent of the plot of Cratinus’ Pytine’ (vol. 1, p. 179 n. 7, building on a hypothesis of J.R. Green).

Among fourth-century authors, another forceful impression is left by the description of the tyrant Dionysius’ tragedies as autobiographical (following recent work by A. Duncan). However, in his attempt to reframe Dionysius’ plays in a more positive light, W. also unexpectedly argues that they were no more political than any other tragedy. This is difficult to imagine with lines like ‘Tyranny is the mother of injustice’ (TrGF 1.76, F4), especially once it is allowed that an authorial persona of the tyrant himself may have appeared as a character within the play. Similarly, W. does well to emphasise how the tragedies by Diogenes of Sinope (the Cynic philosopher) are the work of a ‘provocateur’ (vol. 1, p. 159); yet there is little reason to think that this approach was meant to lead to ‘reasoned discussion’ or ‘more complex or nuanced assessment’ (vol. 1, pp. 155–6) of topics like cannibalism or incest. Given W.’s particular emphasis of what is bold, interesting or striking among the lost works, surprisingly little is made of Chaeremon's Centaur, which Aristotle singled out as a bizarre experimental work (Poet. 24.1460a1 ἀτοπώτερον) on account of its metrical innovations.

A similar variety is emphasised among the lost plays of the canonical triad. For example, W. identifies ‘a distinct sub-category of tragedy based on weird or monstrous subject-matter’ (vol. 2, p. 61), exemplified by the naked and crazed witch-doctor Medea of Sophocles’ Root-Cutters. The fragments indicate a play set in an uncanny, Halloween atmosphere, replete with voodoo doll prop and a chorus of witches (vol. 2, p. 115). W. is at his best when proposing evocative readings of plays in spite of their fragmentary state. Although any real evidence is lacking, he may well be correct that the domestic chorus of ‘wool-carders’ in Aeschylus’ homonymous play ‘may also grotesquely evoke the pulling-apart and dismemberment (sparagmos) of Pentheus’ (vol. 2, p. 48) or that Euripides’ chorus of male gymnasts in the Alope indicates ‘the milieu of the athletics ground’ and the central role of Cercyon as lethal wrestler (vol. 2, p. 155).

Readers familiar with W.’s earlier work will recognise his views on quotation culture and anti-prize mentality as well as his more controversial belief in a well-developed fifth-century readership for literature. We learn, for example, that the injunction against the re-use of Phrynichus’ Sack of Miletus may imply ‘that as early as 492 there was also a small but significant reading audience for texts of tragedy’ (vol. 1, p. 26). A more likely explanation of the injunction, in view of Herodotus’ vague χρῆσθαι (6.21), is that it refers to a ban on reuse of the same subject matter in future tragedies. W.’s faith in early textual dissemination also informs his more attractive proposal that Helen's self-harm with a pencil in Sophocles’ Demand for Helen's Return (TrGF 4, F177) is a metapoetic gesture of self-effacement (vol. 2, p. 88).

In terms of methodology, W. presents two somewhat contradictory inclinations. Although a self-confessed sceptic, he evokes some fairly extreme staging scenarios. Thus we find W. hesitating to see reflections of specific scenes from tragedy in vase paintings on methodological grounds, whereas he readily allows that petrification (vol. 2, pp. 265–7), animal metamorphosis (vol. 2, p. 37) or the murder of the entire chorus (vol. 1, p. 33; vol. 2, pp. 99, 259) might have occurred on ancient stages, rather than (if at all) in messenger speeches. While W. often maintains the distinction between what is merely conceivable and what is likely, there is also a regular slide between the notion that a given scenario is plausible and that it is correct. The benefit of W.’s bold approach to the material, where it does convince, is that it can offer exhilarating interpretations of the lost plays. Sophocles’ Niobe, for example, is no doubt correctly identified as the busy production Aristotle negatively contrasts with Aeschylus’ austere handling of the same myth, and W. convincingly suggests that Niobe's children were murdered on stage, probably with a chorus of the sons’ male lovers looking on (vol. 2, pp. 259–62).

W. is at pains to rehabilitate our appreciation of lost poets who have been attacked in ancient and modern scholarship. The categories of ‘“good” and “bad” poetry, or “major” and “minor” status’, he writes, are ‘banal distinctions’ (vol. 1, p. 137). Few scholars today would object to withholding value judgements of this kind, but W. also tends to hold back from making larger interpretative claims of any kind. In spite of a thorough survey of fifth-century tragedy, ‘No overall patterns or tendencies seem to suggest themselves’ (vol. 1, p. 29). Similarly, with respect to fourth-century tragedy, ‘there is no sign of any general trends or patterns at all’ (vol. 1, p. 120). Somehow it remains possible for W. to claim that ‘the overall character of fourth-century tragedy … was not significantly different from that of fifth-century tragedy’ (vol. 1, p. 120). Discussion of Aeschylus’ fragmentary plays shows that ‘it is impossible to discern such a thing as a “typical” Aeschylean myth or plot structure, or a “typical” Aeschylean hero or heroine’ (vol. 2, p. 14), and a similar sentiment is expressed about Sophoclean drama (vol. 2, p. 67); it is notable, however, that W. does not make the same point about Euripides, where he is on surest ground.

An appendix to volume 1 provides a translation of much of TrGF vol. 1, and this will be one of the most consulted aspects of the book. It is regrettable that there is no facing Greek, since the target audience for this book may not have easy access to the source text. In general W.’s translations are fit for purpose, although I note the following corrections: Phrynichus (TrGF 1.3): F1 should be in square brackets, following W.’s scheme elsewhere (similarly Achaeus, TrGF 1.20, F45); F5 ‘contingent of men’ would be clearer than ‘body of men’; F14 for ‘killing’ and ‘chopping’ read ‘killed’ and ‘chopped’; F16a read ‘a messenger more of good tidings than of ill’ (‘delivering’ is not in the Greek). Aristarchus (TrGF 1.14): F3 read ‘O death, punishment for those who lack judgement’ (σωφρόνισμα is in apposition to θάνατε); F4 the meaning of τιμωρέω is obscured, read, for example, ‘[I say this to you] not of my own accord, but in retribution’. Ion (TrGF 1.19): F43a αἰκισάμενος ἑαυτόν is not ‘having disguised himself’ but ‘having disfigured himself [as a disguise]’. Achaeus (TrGF 1.20): F2 for ‘begging you to put a stop’ read ‘in the hope that you put a stop’; F3, line 1, while ‘religious ambassadors’ is conceivable, ‘spectators’ is more natural; line 3, for ‘Boeotia’ read ‘They're Boeotian’; F23a in place of the antiquated ‘humour’ read ‘temperament’; F52 Hesychius indicates that the meaning of the adjective includes the notion of ‘sexual assault’ rather than merely ‘pursuit’. Philocles (TrGF 1.24) F1 and Xenocles (TrGF 1.33) F1 both leave out daggers or other indication of textual problems, against W.’s more general practice. Critias (TrGF 1.43): F11, lines 2–3 are unclear, read ‘no public speaker could ever manage to corrupt an honest character, whereas he might often twist the meaning of the law’. Diogenes of Athens (TrGF 1.45): F1, line 11 for ‘with the choruses’ read ‘with the dances’, as normally with unqualified instances of the plural χοροί in drama. Diogenes of Sinope (TrGF 1.88): F7, line 4 read ‘even if one occasionally fares less well’ (πράξων is concessive); line 11 for ‘one needs must bear up’ presumably ‘one must needs bear up’, although a less antiquated translation would be preferable, for example ‘one must endure’; line 13 for ‘I do not choose’ read ‘I would not choose’.

Typographical errors are few and do not lead to confusion.Footnote ¹ However, throughout the first volume, books are cited in full where a page reference would have been expected.

This Rashomon picture of tragedy, in which competing playwrights rework the same myths in contrasting and irreconcilable ways, is bold and enlivening. Those unfamiliar with fragmentary tragedy will learn much in consulting these books, and scholars of the fragments will find suggestions that, if not always compelling, are always stimulating and worth exploring.