Hostname: page-component-76fb5796d-vvkck Total loading time: 0 Render date: 2024-04-28T22:34:48.593Z Has data issue: false hasContentIssue false
  • English
  • Français

DID EURIPIDES’ ANDROMACHE PREMIERE OUTSIDE ATHENS?

Published online by Cambridge University Press:  01 March 2024

Chiara Meccariello
Chiara Meccariello*
Affiliation:
University of Exeter
*
c.meccariello@exeter.ac.uk
Rights & Permissions [Opens in a new window]

Abstract

This article re-examines the scholium on Euripides, Andromache 445, which several scholars have used to support the claim that Andromache premiered outside Athens, and concludes that both the scholium itself and a remark in the play's hypothesis rather suggest that the play was produced in Athens as part of a dramatic competition.

Keywords

EuripidesAndromacheCallimachusDidascaliaeancient scholarshipGreek inscriptionsGreek tragedydramatic hypotheses
Type
Research Article
Information
The Classical Quarterly , First View , pp. 1 - 7
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
Copyright © The Author(s), 2024. Published by Cambridge University Press on behalf of The Classical Association

According to an ancient scholium, the Andromache of Euripides was not performed in Athens and therefore could not be dated precisely.Footnote 1 Taking this statement at face value, several scholars have concluded that the play premiered outside Athens. Accordingly, Andromache features prominently in the growing corpus of scholarship on theatre beyond Athens,Footnote 2 its performance located in such diverse places as Thessaly, Argos or the court of the Molossians in Epirus.Footnote 3 Yet scholarship contains little discussion of the scholium as a whole, and few attempts have been made to understand the relationship among the individual statements that are made in it. This paper reassesses the scholium, both in its own right and in the light of the fragmentary remains of Athenian didascalic records and ancient scholarship on them. My conclusion is that the scholium does not point to a non-Athenian premiere of the play, but rather offers clues to the existence of an Athenian performance during Euripides’ lifetime. The scholium, I argue, bears traces of an ancient debate and of Callimachus’ research on Athenian didascalic records. In its general outline, this view is not entirely new.Footnote 4 It is here revived with several new arguments, taking into account recent scholarly work on relevant materials, especially the inscriptional records of Athenian drama.

I report here the text of the scholium in its entirety with a minimal apparatus:

ὦ πᾶσιν ἀνθρώποισιν ταῦτα ἐπὶ τῷ Ἀνδρομάχης προσχήματί φησιν Εὐριπίδης λοιδορούμενος τοῖς Σπαρτιάταις διὰ τὸν ἐνεστῶτα πόλεμον. καὶ γὰρ δὴ καὶ παρεσπονδήκεσαν πρὸς Ἀθηναίους, καθάπερ οἱ περὶ τὸν Φιλόχορον ἀναγράφουσιν. εἰλικρινῶς δὲ τοὺς τοῦ δράματος χρόνους οὐκ ἔστι λαβεῖν⋅ οὐ δεδίδακται γὰρ Ἀθήνησιν. ὁ δὲ Καλλίμαχος ἐπιγραφῆναί φησι τῇ τραγῳδίᾳ Δημοκράτην. ἑξῆς δὲ αὐτοὺς εἴς τε τὰ ἄλλα καὶ φιλοχρηματίαν κακῶς λέγει. καὶ Ἀριστοτέλης δὲ τοῦτο ἱστορεῖ ἐν τῇ τῶν Λακώνων πολιτείᾳ καὶ τὸ ὑπὸ θεοῦ αὐτοματισθὲν προστίθησιν ἔπος ‘ἁ φιλοχρηματία Σπάρταν ὀλεῖ, ἄλλο δέ γ’ οὐδέν’. φαίνεται δὲ γεγραμμένον τὸ δρᾶμα ἐν ἀρχαῖς τοῦ Πελοποννησιακοῦ πολέμου.

δεδίδακται Cobet: δέδεικται codd. | εἰλικρινῶς … Δημοκράτην om. A

[Most hateful] to all [among mortals, inhabitants of Sparta]: It is Euripides who says this through the character of Andromache, blaming the Spartans because of the current war. For indeed they had violated the truce with the Athenians, as Philochorus records (FGrHist 328 F 124). Yet the date of the play cannot be clearly established, for it was not staged in Athens; but Callimachus (fr. 451 Pfeiffer) says that the name of Democrates was attached to the tragedy. In what follows he (sc. Euripides) criticises them for other reasons and especially their greed. Aristotle too relates this in his Constitution of the Spartans and adds the verse uttered by the god: ‘Greed will ruin Sparta, and nothing else’ (Arist. fr. 544 Rose = 550.1 Gigon, cf. Zen. Prov. 2.24, Diod. Sic. 7.12.5). The play was evidently written at the beginning of the Peloponnesian war.

The polyphony of the scholium was unravelled by Bergk.Footnote 5 Commenting on the anti-Spartan tirade that opens Andromache's speech to Menelaus (445–63), the scholium opens with a blunt statement: that the character is just a mouthpiece for the anti-Spartan sentiment of Euripides himself, who would be railing at Sparta ‘because of the present war’. The scholium then refers to the Spartans’ ‘violation of the truce’, presumably the Thirty Years’ Peace, the breach of which started the Peloponnesian War.Footnote 6 The remark on the uncertainty of the date responds to this political interpretation, which requires Andromache to be a product of the war's early years. After a further remark on Democrates, ascribed to Callimachus (more on which soon), the scholium goes back to Euripides’ criticism of Sparta. A final sentence returns to the date, concluding that the play was written at the beginning of the Peloponnesian War. The phrasing here, especially the vague φαίνεται and the ‘bookish’ γεγραμμένον, indicates that this is likely to be a secondary inference, and at any rate not a statement based on a record of the play's performance.

The central portion of the scholium, εἰλικρινῶς … Δημοκράτην, interrupts the linear progression from the previous discussion of Euripides’ anti-Spartan sentiments to the following reference to his denunciation of the Spartans’ greed. Therefore, the two remarks contained in this portion are likely to be additional takes on the same issue, and possibly responses to what precedes. This is clear in relation to the first sentence (εἰλικρινῶς … Ἀθήνησιν), as we have seen; and there is no reason that the second (ὁ δὲ Καλλίμαχος … Δημοκράτην) should introduce an unrelated topic.

The phrasing of the first sentence combines performance location and dating in a way that unmistakably indicates that the play was not to be found in the Didascaliae: as these recorded the results of the dramatic competitions at the City Dionysia and Lenaea year by year, any Athenian performance at those venues would have been datable.Footnote 7 Taken in isolation, the statement could indicate either (a) that the scholiast was somehow aware that the play was not performed in Athens, and concluded on those grounds that it would not be found in the Didascaliae, or (b) that the scholiast (or his source) had looked in vain for the play's date in the Didascaliae, and its absence had led him to conclude that it was not performed in Athens. At first sight, (a) seems to be literally what the scholium states. However, the fact that the scholiast does not say where the play was staged, but uses a negative sentence, strongly suggests that his conclusion is based on an argument from silence: it was the play's absence from the Didascaliae that indicated it was not performed in Athens.

The following sentence, ὁ δὲ Καλλίμαχος ἐπιγραφῆναί φησι τῇ τραγῳδίᾳ Δημοκράτην, confirms this interpretation. The meaning of ἐπιγραφῆναι is crucial here. While the verb itself may well indicate authorship of a work (LSJ V 2), a remark on authorship would be inexplicable here. This statement only makes sense within the logic of the scholium if it is related to the dating issue.

In the context of dating, and attempted dating through didascalic records, ἐπιγραφῆναι is likely to refer to the dynamics of theatrical production. Both ἐπιγράφω and its cognateFootnote 8 ἐγγράφω can indeed be found in several passages concerning Aristophanes’ collaboration with Philonides and Callistratus in the production of some of his comedies.Footnote 9 For example, the anonymous treatise On Comedy states that Aristophanes staged comedies for the first time during the archonship of Diotimus (428/427 b.c.e.) ‘through Callistratus’ (ἐδίδαξε δὲ πρῶτος ἐπὶ ἄρχοντος Διοτίμου διὰ Καλλιστράτου), and that Callistratus and Philonides produced his political plays and those against Euripides and Socrates respectively. Only after acquiring a reputation for himself was Aristophanes ἐπιγραφόμενος to his own comedies.Footnote 10 The inevitable implication is that Aristophanes was not ἐπιγραφόμενος to the plays he produced through Callistratus and Philonides, who, in turn, were ἐπιγραφόμενοι to several Aristophanic plays. In other words, ἐπιγράφω indicates the name officially associated with a theatrical production, that is, the name of the producer, which usually coincided with that of the author, but did not need to.Footnote 11

The Callimachean remark must refer to a similar case. Correcting the previous statement that Andromache was not performed in Athens, the sentence indicates that a Euripidean performance of Andromache could not be found in the Didascaliae because the name of Democrates was attached to it instead of that of Euripides. Neither a dispute about the play's authorshipFootnote 12 nor a reference to a later, post-Euripidean revival by DemocratesFootnote 13 would fit the logic of the scholium, which therefore provides not evidence that Andromache was not staged in Athens, but clues as to an Athenian performance, which someone had failed to locate in didascalic records because the recorded didaskalos was not Euripides.

This interpretation of the scholium implies that when poet and didaskalos were not the same person, only the name of the didaskalos would be attached to the performance in official records; or, at least, that this was the case with Andromache. Unfortunately, no recognizable example of how the Didascaliae treated such cases survives. A seminal article by Capps concluded that, in such cases, the official records would include the name of the poet;Footnote 14 but the validity of this conclusion is limited to the Victors’ Lists (IG II2 2325A–H), where a single example is available. This inscription probably records Aphareus and his two victories at the City Dionysia in the mid-fourth century b.c.e. (Ἀφα]ρ̣εύς followed by the number ‘two’ at 2325A.46), but we know from [Plut.] X orat. 839D (TrGF I 73 Test. 2.18–19) that he performed διὰ Διονυσίου. Moreover, the so-called ‘Roman fragments’ (IGUR 216, 215, 218), which record all finishings of Athenian comic poets in chronological order, seem to have reported, in one case, both the name of the poet and that of the didaskalos (IGUR 218.9 [- - - διὰ Ἀνα]ξίππου in the section on Anaxandrides, mid-fourth century b.c.e.).

Neither inscription, however, is indicative of how a performance would be recorded in the official records, or in the Didascaliae and Fasti. Millis and Olson have underscored the peculiarity of the Victors’ Lists in this respect, as their ‘implied audience […] would seem to be more interested in a summary accounting of the accomplishments of great figures from Athens’ literary and dramatic past than in pedestrian or technical questions’ on the performances, such as the identity of the chorêgos and didaskalos of each production.Footnote 15 As for the ‘Roman fragments’, this inscription is thought to derive from Alexandrian work, as the reference to plays’ survival at IGUR 216.9 makes likely. Körte even suggested that the inscription might be a copy of Callimachus’ Πίναξ καὶ ἀναγραφὴ τῶν κατὰ χρόνους καὶ ἀπ’ ἀρχῆς γενομένων διδασκάλων, mentioned in the Suda.Footnote 16 Therefore, neither the Victors’ Lists nor the ‘Roman fragments’ can be used to ascertain how the official records treated cases of substitute didaskaloi. Conversely, the Fasti are believed to reflect more closely the official records, as they were compiled earlier (between 347/6 and 343/2)Footnote 17 and record the chorêgos and the producer: a typical entry would include ‘X ἐχορήγει’ and ‘Y ἐδίδασκε’ for both the winning comedy and the winning tragedy of a given year. While, again, we have no clear example of whom the Fasti would record when the didaskalos and the poet did not coincide in the fifth century, the extant records do contain one instructive example. For the year 387, the Fasti report a victory of Araros, Aristophanes’ son, at the City Dionysia (IG II2 2318.1004), but the Suda (α 3737 s.v. Ἀραρώς = Araros PCG Test. 1.2–3) states that he won his first victory in the hundred-and-first Olympiad (376/372 b.c.e.). Araros is known from hyp. Ar. Plut. IV to have been in charge of the production of his father's late comedies Cocalus and Aeolosicon, and it is reasonable to conclude that in 387 Araros won with a play by his father. This is a good indication that, at least in some cases, the official records, which the Fasti are likely to reflect more directly than the Victors’ Lists or the ‘Roman fragments’,Footnote 18 did not include the name of the actual poet; but other more casual channels such as contemporary gossip, book circulation and later research could shed light on authorship even in such cases.Footnote 19 The original performance of Andromache seems precisely to be a case in which a Hellenistic scholar reconstructed the dynamics of the play's production by supplementing didascalic information with data drawn from other sources. That this scholar was Callimachus is perfectly in line with his compiling a chronological catalogue of didaskaloi and with his criticism of the Didascaliae in chronological matters, as known from a scholium on Aristophanes.Footnote 20 Callimachus, the Andromache scholium suggests, traced the slate of Euripides’ Andromache in the Didascaliae, which someone else failed to find because they did not realise that Democrates had acted as didaskalos for that performance, the name of Euripides not being reported in the official records on that occasion.

A further indication that Andromache was staged as part of a traceable competitive performance comes from the hypothesis prefixed to the play in the medieval manuscripts, where Andromache is said to be a play τῶν δευτέρων.Footnote 21 This is most reasonably understood as referring to the placing of the play in a theatrical competition.Footnote 22 The only parallel, τὸ δρᾶμα τῶν πρώτων, is found in the hypothesis prefixed to Hippolytus, part of a tetralogy that won indeed the first prize.Footnote 23 A similar use of πρῶτος in Athenaeus is unambiguously about placing in a dramatic contest.Footnote 24 At 9.373f the deipnosophist Myrtilus states that Anaxandrides’ play Tereus ‘was not judged among the first’ (μὴ κεκριμένον ἐν τοῖς πρώτοις). He then quotes an anecdote from Chamaeleon (fr. 46 Martano) about Anaxandrides’ habit of disposing of his plays if they had failed to win the first prize, and expresses surprise at the survival of Tereus μὴ τυχὼν νίκης (‘since it did not win’). The passage in its entirety clearly indicates that the sequence μὴ … ἐν τοῖς πρώτοις conveys the results of a dramatic contest, not a general qualitative judgement.

Since πρῶτος, δεύτερος, and τρίτος are regularly used in the hypotheses’ didascalic sections to indicate finishings in dramatic competitions, any other meaning of the ordinal number in the Hippolytus and Andromache hypotheses would be opaque. Moreover, each of the other options is otherwise problematic. A chronological connotation is unlikely, since Hippolytus, staged in 428, would not be chronologically far from Andromache, which modern scholars tend to date to ca. 425 on internal grounds; and with Euripides’ debut in the year 455, it seems hard to justify the use of πρῶτος for Hippolytus, unless we assume a bipartite periodization with the turning point from πρῶτα δράματα to δεύτερα δράματα conveniently falling between Hippolytus and Andromache.Footnote 25 Alternatively, the numeral could indicate the position of the play within its tetralogy, but there seems to be no particular interest in such a question in ancient scholarship.Footnote 26 A qualitative judgement has often been posited on the basis of the observation that similarly structured remarks in other hypotheses convey a general evaluation of the play. However, this is problematic for Andromache, as δεύτερος would convey a negative evaluation,Footnote 27 but the remark is immediately followed by a longer positive evaluation of several sections of the play.Footnote 28 To solve the inconsistency, Stevens suggests the meaning ‘of second-prize quality’, which he takes to be ‘a complimentary expression’,Footnote 29 but this is even harder to infer from the phrasing. As Stevens remarks, there are indeed numerous parallels, in hypotheses, for τὸ δὲ δρᾶμα τῶν + adjectival phrase conveying a positive evaluation, but all of them are unambiguous,Footnote 30 whereas δεύτερος meaning ‘deserving second prize’ is ambiguous both in isolation and in the context of hypotheses that regularly use ordinal numbers to indicate actual placing in dramatic contests.

Hence this detail of the hypothesis and the scholium on Andr. 445 concur in indicating that Andromache was indeed staged in Athens as part of a dramatic contest and won the second prize, with a producer named Democrates. On the reasonable assumption that the play premiered at the City Dionysia or, less probably, the Lenaea,Footnote 31 we can conclude that one or three more plays by Euripides were produced by Democrates—and therefore the reason for this choice need not lie in any feature of Andromache.

References

1 Schol. MNOA Andr. 445 = Eur. TrGF Test. 64, quoted below.

2 See e.g. Stewart, E., Greek Tragedy on the Move: The Birth of a Panhellenic Art Form c. 500–300 BC (Oxford, 2017), 139–44Google Scholar; Vahtikari, V., Tragedy Performances Outside Athens in the Late Fifth and the Fourth Centuries BC (Helsinki, 2014), 222–4Google Scholar (with aporetic conclusion); Easterling, P.E., ‘Euripides outside Athens: a speculative note’, ICS 19 (1994), 73–80Google Scholar, at 79.

3 Epirus: Schmid, W. and Christ, W. von, Geschichte der griechischen Litteratur (Munich, 1908), 1.343Google Scholar; Robertson, D.S., ‘Euripides and Tharyps’, CR 37 (1923), 58–60Google Scholar; Nilsson, M.P., Cults, Myths, Oracles, and Politics in Ancient Greece (Lund, 1951), 83Google Scholar; Butrica, J.L., ‘Democrates and Euripides’ Andromache’, Hermes 129 (2001), 188–97Google Scholar; Centanni, M., ‘Andromaca di Euripide’, AION 33 (2011), 39–57Google Scholar, at 41–3 (Dodona); Stewart (n. 2). Thessaly: Goossens, R., Euripide et Athènes (Brussels, 1962), 376–7Google Scholar; Bernand, A., La Carte du Tragique: La géographie dans la tragédie grecque (Paris, 1985), 238–41Google Scholar; Taplin, O., ‘Spreading the word through performance’, in S. Goldhill and R. Osborne (edd.), Performance, Culture and Athenian Democracy (Cambridge, 1999), 33–57Google Scholar, at 45. Epirus or Thessaly: Cairns, F., ‘Pyrrhic dancing and politics in Euripides’ Andromache’, QUCC 100.1 (2012), 31–47Google Scholar, at 39. Argos: Bergk, T., Griechische Literaturgeschichte (Berlin, 1884), 3.169Google Scholar, 486 n. 70 (but see below, n. 4); A. Nauck, Euripidis Tragoediae, vol. 1 (Leipzig, 18893), xvii; D.L. Page, ‘The elegiacs in Euripides’ Andromache’, in C. Bailey et al. (edd.), Greek Poetry and Life: Essays Presented to Gilbert Murray (Oxford, 1936), 206–30, at 223–8. Sicyon: A. Lesky, Die tragische Dichtung der Hellenen (Göttingen, 19723), 338 n. 90. A minor venue in Attica: E. Scharffenberger, ‘Andromache’, in A. Markantonatos (ed.), Brill's Companion to Euripides (Leiden, 2020), 1.139–57, at 144. For a summary of the vexata quaestio see W. Allan, The Andromache and Euripidean Tragedy (Oxford, 2000), 149–60.

4 See especially T. Bergk ap. Hinrichs, G., ‘Philologische Paralipomena Theodor Bergks’, Hermes 18 (1883), 487–95Google Scholar; this is a later stance than the one quoted in n. 3, despite its earlier year of publication. Other scholars have tacitly or succinctly supported the thesis of an Athenian performance: see e.g. Snell on TrGF I 124 Test. 2 ‘Callimachus in didascaliis invenisse videtur “Andromacham” Democratis actam Euripidis temporibus’. Allan (n. 3), 150–1 also states that the scholium, ‘if correctly reported and based on didascalic records, suggests that the play was produced in Athens’, but he is ultimately more inclined toward other possibilities.

5 Bergk (n. 4), 489–92. Bergk's identification of the scholars responsible for the different layers of the note, however, is highly speculative. Even more so is his thesis that the Fasti inscription IG II2 2318 (TrGF I DID A 1.118) preserves the Andromache slate under the name of [Me]necrates (422 b.c.e.), of which the scholium's Democrates would be a corruption.

6 The breach, in this case, would probably consist in the Spartan invasion of Attica (Thuc. 1.125.2). For alternatives, see Stevens, P.T., Euripides. Andromache (Oxford, 1971), 16Google Scholar.

7 The extant inscriptional records of the Athenian Didascaliae, IG II2 2319–24, are re-edited with commentary in Millis, B.W. and Olson, S.D., Inscriptional Records for the Dramatic Festivals in Athens (Leiden, 2012)Google Scholar together with the Fasti (IG II2 2318) and the Victors’ Lists (IG II2 2325). See also Millis, B.W., ‘Inscribed public records of the dramatic contests at Athens’, in E. Csapo et al. (edd.), Greek Theatre in the Fourth Century BC (Berlin and Boston, 2014), 434–40Google Scholar for description and discussion. Most references to the Didascaliae in literary and paraliterary sources, however, are likely to be references to Aristotle's Διδασκαλίαι (frr. fr. 618–30 Rose, 415–62 Gigon), which also reflected archival records of dramatic competitions. See Talin, M., ‘From Aristotle to the hypotheseis: Didascalic information between Athens and Alexandria’, Aevum 95 (2021), 39–68Google Scholar for a recent assessment and caveats.

8 Textual corruption cannot be ruled out; ΕΓΓΡ is an easy slip for ΕΠΙΓΡ.

9 A full discussion of this technical usage in Butrica (n. 3), 193–4.

10 Proleg. de Com. lines 38–41 Koster (PCG III.2 Test. 4): ἐδίδαξε (sc. Aristophanes) δὲ πρῶτος ἐπὶ ἄρχοντος Διοτίμου διὰ Καλλιστράτου. τὰς μὲν γὰρ πολιτικὰς τούτῳ φασὶν αὐτὸν διδόναι, τὰ δὲ κατ’ Εὐριπίδου καὶ Σωκράτους Φιλωνίδῃ. διὰ δὲ τούτων νομισθεὶς ἀγαθὸς ποιητὴς ταῖς λοιπαῖς ἐπιγραφόμενος ἐνίκα. Cf. schol. Ar. Eq. 507f ὅτι ὁ ποιητὴς τοῦτο πρῶτον τὸ δρᾶμα δι’ ἑαυτοῦ καθῆκε, τὰ δ’ἄλλα δι’ ἑτέρων προσώπων. ὑποκριταὶ γὰρ αὐτοῦ Καλλίστρατος καὶ Φιλωνίδης ἦν, δι’ ὧν ἐδίδασκε τὰ δράματα αὐτοῦ; hyp. Ar. Ran. lines 27–8 Chantry (text of M and E) Φιλωνίδης ἐπεγράφη καὶ ἐνίκα; schol. Ar. Plut. 179 Φιλωνίδην… τὸν ἐν τοῖς Ἀριστοφανείοις ἐγγεγραμμένον δράμασιν. The bibliography on the matter is immense. A reasonable assessment of the issue and further bibliography in Hubbard, T.K., The Mask of Comedy (Ithaca and London, 2011), 227–30Google Scholar; see also below, n. 19.

11 Similar cases are known for Eupolis’ Autolycus (PCG Test. 15 = Ath. Deipn. 5.216d), Plato Comicus (PCG Test. 7 = P.Oxy. 2737, line 44), Anaxandrides (PCG Test. 5.9 = IGUR 218.9, see below) and Aphareus (TrGF I 73 Test. 2.18–19 = [Plut.] X orat. 839C; see below). Another possible case concerns Iophon and Sophocles (TrGF I 22 Test. 5c = schol. Ar. Ran. 78).

12 For this interpretation see Jachmann, G., De Aristotelis Didascaliis (Gottingen, 1909), 28Google Scholar. See also the translation of the scholium in Csapo, E. and Slater, W.J., The Context of Ancient Drama (Ann Arbor, 1995), 15Google Scholar: ‘Callimachus says he found the name of Demokrates inscribed as author of the tragedy’. U. von Wilamowitz initially accepted the didascalic meaning (Analecta Euripidea [Berlin, 1875], 148: ‘docta per Democratem’), but his later, slightly more diffuse treatments (discussed in Butrica [n. 3], 189–91) simultaneously posit that Callimachus had access to a copy of the play under the name of Democrates.

13 Butrica (n. 3), 196–7 argues that Callimachus refers to a revival of the play in which Democrates was the main actor.

14 Capps, E., ‘Epigraphical problems in the history of Attic comedy’, AJPh 28 (1907), 179–99Google Scholar.

15 Millis and Olson (n. 7), 157.

16 A. Körte, ‘Inschriftliches zur Geschichte der attischen Komödie’, RhM 60 (1905), 425–47, at 443–7; Suda κ 227 s.v. Καλλίμαχος.

17 Millis (n. 7), 434.

18 For a tentative stemma showing the relationships among the different inscriptions, Aristotle's Didascaliae and Alexandrian work, see Starkey, J.S., ‘Aristophanes, Apollodorus, and the Dionysian actors’ contest’, ZPE 192 (2014), 45–58Google Scholar, at 53.

19 What little we know of Aristophanes’ collaboration with Philonides and Callistratus also suggests that during the early stages of his career only the producer's name was officially recorded, not the poet's, since Aristophanes’ entrusting theatrical productions to other individuals was, at least initially (i.e. before Knights), a way for him to remain behind the scenes. See especially Eq. 512–44 and Vesp. 1016–22 as interpreted in McDowell, D.M., ‘Aristophanes and Kallistratos’, CQ 32 (1982), 21–6Google Scholar, and Hubbard (n. 10). However, the fact that Aristophanes resorted to substitute didaskaloi later as well, including with Wasps, indicates that this choice could be due to different reasons, and that poets could choose how to manage such cases. See also R. Blum, Kallimachos. The Alexandrian Library and the Origins of Bibliography, transl. H.H. Wellisch (Madison, 1991), 38–9.

20 Schol. vet. Ar. Nub. 553 = Arist. fr. 621a Rose = 441 Gigon, Call. fr. 454 Pfeiffer: δῆλον ὅτι πρῶτος ὁ Μαρικᾶς ἐδιδάχθη τῶν δευτέρων Νεφελῶν. Ἐρατοσθένης δέ φησι Καλλίμαχον ἐγκαλεῖν ταῖς Διδασκαλίαις, ὅτι φέρουσιν ὕστερον τρίτῳ ἔτει τὸν Μαρικᾶν τῶν Νεφελῶν, σαφῶς ἐνταῦθα εἰρημένου, ὅτι πρῶτος καθεῖται ‘It is clear that Maricas was produced before the second Clouds. Eratosthenes says that Callimachus finds fault with the Didascaliae because they record Maricas two years after Clouds, whereas here Maricas is clearly said to have been performed first’. See Pfeiffer, R., History of Classical Scholarship (Oxford, 1968), 132Google Scholar.

21 Hyp. Eur. Andr. line 22 Diggle.

22 See e.g. G. Hermann, Euripidis Tragoediae, vol. II 2: Andromacha (Leipzig, 1838), viii; Welcker, F.G., Die griechischen Tragödien mit Rücksicht auf den epischen Cyclus (Bonn, 1839), 2.533Google Scholar.

23 Hyp. Eur. Hipp. line 30 Diggle. For this remark's probably later date than, and different intended readership from the one of the didascalic section see D.J. Mastronarde, Euripides: Phoenissae (Cambridge, 1994), 168 n. 2.

24 Ath. Deipn. 9.373f–4b (= Anaxandrides PCG Test. 2). See Millis, B.W., Anaxandrides: Introduction, Translation, Commentary (Heidelberg, 2015), 26–7Google Scholar.

25 At any rate, δεύτερος may not be the most natural adjective to use for chronological grouping: Plut. Comp. Ar. et Men. 853F uses τὰ πρῶτα, τὰ μέσα and τὰ τελευταῖα for Menander's plays. A chronological interpretation of the Andromache hypothesis remark is suggested by A. Garzya, Euripide, Andromaca (Naples, 19632), xii. Proposed dates, based on metre, style and other internal elements, range between c. 430 and 417: see the doxographical surveys in A. Giuliani, La città e l'oracolo (Milan, 2001), 156 n. 36; Allan (n. 3), 149 n. 3; also above, n. 6.

26 Verrall, A.W., Essays on Four Plays of Euripides (Cambridge, 1905), 22Google Scholar suggested that the phrase might indicate that Andromache was a sequel: ‘The note may refer to a list, distinguishing those of the plays contained in some collection which were known or conjectured to be sequels; and it may possibly signify that the Andromache is one of them.’ Later scholarship has often interpreted this suggestion as referring to the play's position in its tetralogy: see, for instance, Erbse, H., ‘Euripides’ ‘Andromache’’, Hermes 94 (1966), 276–97Google Scholar, at 276.

27 Cf. Erbse (n. 26), 276: ‘zweitklassig’.

28 It is not impossible that the hypothesis tacitly combined different views, but the evaluations of Phoenissae prefixed to the play show a clear demarcation between positive remarks and the single negative comment: see hyp. c Diggle τὸ δρᾶμά ἐστι μὲν ταῖς σκηνικαῖς ὄψεσι καλόν, ἔστι δὲ καὶ παραπληρωματικόν (accepting Kirchhoff's necessary ἔστι δὲ for the transmitted ἐπεὶ δέ or ἐπεί: the μέν in the previous clause ensures the presence of a responding δέ) and the preceding hyp. b Diggle, which contains only positive remarks. Hyp. Eur. Or. line 43 Diggle τὸ δὲ δρᾶμα τῶν ἐπὶ σκηνῆς εὐδοκιμούντων, χείριστον δὲ τοῖς ἤθεσι contains no appreciative remark beyond recording that Orestes was popular on stage, and the second δέ makes the commentator's stance unambiguous.

29 Stevens (n. 6), 27–8.

30 Hyp. 2 Soph. Ant. line 3 Xenis τὸ μὲν δρᾶμα τῶν καλλίστων Σοφοκλέους; hyp. 1 Soph. OC line 15 X. τὸ δὲ δρᾶμα τῶν θαυμαστῶν; hyp. Eur. Phoen. (Argumentum Thomanum) line 102 Mastronarde ἔστι δὲ τὸ παρὸν δρᾶμα τῶν ἄγαν ἐξαιρέτων; hyp. Ar. Ach. line 28 Wilson τὸ δὲ δρᾶμα τῶν εὖ σφόδρα πεποιημένων; hyp. Eq. line 25 Jones τὸ δὲ δρᾶμα τῶν ἄγαν καλῶς πεποιημένων; hyp. I Nub. line 24 Holwerda τὸ δὲ δρᾶμα τῶν πάνυ δυνατῶς πεποιημένων; hyp. III Pax line 27 H. τὸ δὲ δρᾶμα τῶν ἄγαν ἐπιτετευγμένων; hyp. I Av. lines 7–8 H. τὸ δρᾶμα τοῦτο τῶν ἄγαν δυνατῶς πεποιημένων; hyp. I Ran. line 39 Dübner τὸ δὲ δρᾶμα τῶν εὖ πάνυ καὶ φιλολόγως πεποιημένων; hyp. iii Ran. line 32 Dübner τὸ δὲ δρᾶμα τῶν εὖ καὶ φιλοπόνως πεποιημένων. That the structure does not need to convey qualitative judgements is shown by schol. Thesm. ante 1 Dübner τοῦτο τὸ δρᾶμα τῶν κατ’ Εὐριπίδου πεποιημένων; cf. hyp. Eur. Supp. lines 2–3 Diggle τὸ δὲ δρᾶμα ἐγκώμιον Ἀθηνῶν.

31 We have no evidence that Euripides ever competed at the less prestigious Lenaea, although this cannot be ruled out. However, if it is not a coincidence that Aristophanes’ Lenaea-plays do not use the mêchanê, whilst all of his comedies that do use it are certainly or probably Dionysia-plays (see I.C. Storey, Aristophanes: Peace [London and New York, 2019], 108), then we should be reluctant to postulate a Lenaea premiere for Andromache, which features a deus ex machina. Euripidean premieres at the Rural Dionysia are also unlikely, as the evidence for dramatic performances in Attic demes is more likely to refer to reperformances, and a performance at the Rural Dionysia not preceded (or at least followed) by one at the Great Dionysia is unlikely for well-established playwrights: Finglass, P.J., ‘Ancient reperformances of Sophocles’, Trends in Classics 7 (2015), 207–23Google Scholar, at 213.