Skip to main content

The fragments explicitly attributed to Aeschylus’ Σεμέλη ἢ Ὑδροφόροι are four in total (221-224 Radt), all from scholiastic or lexicographical sources. Their aid in reconstructing the plot is minimal. More helpful is a scholion to Apollonius of Rhodes (1.636a Wendel) according to which Aeschylus showed on stage a pregnant and possessed Semele and a group of females (surely the chorus) also became possessed when they touched her; this notice probably refers to the Semele (Schütz). However, Radt’s conservative approach to the attribution of fragments and papyri may have left the Semele somewhat short-changed.

It is now commonly agreed that P.Oxy. 2164 (frr. 168-168b Radt), despite containing two lines attributed to the Ξάντριαι by an ancient commentator (schol. Ar. Ran. 1344b Chantry), is actually a manuscript of the Semele, as first argued by Latte (see e.g. Lloyd-Jones, Mette, Lucas de Dios, Sommerstein). It preserves the end of a choral ode mentioning Hera and Semele, with a prayer in the latter’s favour and apparently a reference to her union with Zeus, followed by a hexameter monody sung by Hera in the disguise of an agyrtis (as several sources recall). These hexameters are a hymn to the Nymphs, listing their beneficial powers over humankind, especially in the context of marriage – an effective strategy if Hera wanted to gain access to the unsuspecting Semele in order to provoke her downfall (cf. Ov. Met. 3.269-74, Apollod. 3.4.3, Nonn. D. 8.171-6, and others).

Two smaller papyrus fragments are also potential candidates for attribution to the Semele: P.Oxy. 2248 (fr. 451e Radt) becauseof its close physical similarity with P.Oxy. 2164 (Mette after Lobel), and P.Oxy. 2249 (fr. 451f Radt) given its apparent reference to Hera and her arrival (Snell). Their connection with the Semele is uncertain, and their contribution to our understanding of the tragedy is slight, though if fr. 451f belongs to this play we know at least that one character knew of Hera’s presence and warned another accordingly.

A further manuscript can tentatively be related to the Semele. P.Oxy. 2881 (tr. adesp. fr. 659 Kannicht-Snell) consists of two iambic fragments reminiscent of Aeschylus’ style. Fr. a mentions a frightening dream in the first person; fr. 2, a bull (Page). The few scholars who have engaged with this fragment so far have attempted to connect it with Europa, Io (Luppe), Pasiphae, or the Minotaur (Kannicht-Snell). However, bulls have well-known Dionysiac associations: Semele may well have dreamed of a bull representing her divine offspring-to-be. Nonnus depicts Semele as experiencing precisely such a dream (D. 7.141-54), though in his poem this happens before Zeus conceives his lust for her. Fr. a may thus represent Semele’s announcement of her dream, and fr. b (whose position respective to fr. a is uncertain) her narrative of its content.

This supposition allows closer investigation of the play’s title. Comparing titles such as Χοηφόροι, one would expect the women of the chorus (presumably house-slaves) to have entered the stage in the act of carrying water. It has been suggested that they were doing so in order to wash the newborn Dionysus (Ahrens, Maass) or for Semele’s purification after childbirth (Latte), but this does not sit well with a myth in which Semele’s delivery was precipitated by her own actions (and probably Hera’s) and thus cannot have been known to be imminent already at the beginning of the play. However, purification with water is widely attested after an ominous dream (A. Pers. 201-2, Ar. Ran. 1339-40, and others, cf. also Nonn. D. 7.175-7). One can therefore suggest that the chorus had been summoned to bring water for Semele’s purification after her dream and entered the stage in this capacity.

Despite its largely speculative nature, a close investigation of the fragments that can be related to the Semele or Water-bearers is well worth attempting in order to shed further light (however tentatively) on what must have been one of Aeschylus’ most tantalising dramas.