The Geryoneis of Stesichorus survives only in fragments, mostly the fragments of a single papyrus (P. Oxy. 2617) except for a couple of quotations in Athenaeus, Pausanias, Strabo, and scholia. Lazzeri gives us all of these fragments, reviews previous work on the text, and reconstructs the plot of the poem as much as possible. The book is a painstaking, detailed study, using artistic, literary, linguistic, metrical, and papyrological evidence to interpret the fragments. Fans of Stesichorus or of Heracles will certainly want to work through it.
One of the papyrus fragments (S27, P. Oxy. 2617 frag. 7) includes a marginal line number, 1300, telling us that the poem is at least 1300 lines long, comparable to two or three books of the Iliad. Yet of these 1300 (or more) lines, only about 180 remain in any sort of readable form.1 Thus L's task is difficult: he aims to reconstruct the plot, style, and meter based on less than eight percent of the poem. Necessarily, then, his conclusions must be tentative, but he does an excellent job of wringing every possible bit of information from the fragments.
The story of Geryon and Heracles is known from various sources and allusions, in Hesiod, Pindar, Aeschylus, and other authors. Heracles must capture Geryon's cattle. Geryon is a three-bodied monster, and he lives far away, typically on an island far to the west, beyond the Pillars of Heracles. In some versions, Heracles erects the pillars on his journey to or from Geryon's island. To reach the island, Heracles sails in a golden cup given him by the Sun.2 All of these motifs appear in the fragments we have.
L deals first with PMGF S17, quoted by Athenaeus (11.469e), referring to the cup of the sun. The word is δέπας, which can be any sort of bowl or cup. L surveys the iconography, showing five vases dated from about 510 to 480 that picture Heracles in the δέπας. In some of them, it looks like a serving dish; in others, it seems to be a cauldron (p. 43-44).
The next fragments introduce Eurytion, who is Geryon's cowherd. S7 comes from Strabo (3.2.11), who tells us that it refers to Eurytion's birth in Tartessus. S8 is a papyrus fragment referring to the Hesperides, presumably telling how Eurytion's mother brought him there as a child. Since Eurytion is not essential to the main myth, L concludes "sembra ci troviamo davanti ad un excursus ampio" (p. 94), the sort of digression that is at home in epic poetry and consistent with the scale of the Geryoneis.
Next come a handful of fragments from speeches. Fragments S9, S10, and S11 seem to be an interchange between Geryon and the herdsman Menoetes. Of these only S11 is particularly extensive. S9 is some 42 letters on 10 lines, within which the word κεφαλάν can plausibly be read, perhaps referring to the Nemean lion's skin (p. 97). A few lines later ποκα is probably the adverb (= ποτε), and as L points out, quoting Barrett, this word is most likely to occur in direct speech (p. 98). Then ν ἥτορ in the next line is likely to be a vocative like φίλον ἥτορ. Papyrologists may take this sort of argument for granted, but to a non-papyrologist like me, it's amazing how much can be deduced from such a broken little fragment. Although the name of Menoetes does not occur in these fragments, in other versions of the story (in Apollodorus, for example), he tells Geryon that Heracles has come to steal the cattle, and these fragments seem to be that part of the story.
Fragments S12 and S13 are apparently spoken by Geryon's mother Callirhoe. She is encouraging him not to do anything stupid. L compares the speech to those of Hecuba to Hector in the Iliad (p. 146, 158), for its rhetorical strategy and its language. Since the speaker says πείθευ τέκνον ("obey, child") in S12 line 7, we certainly have a parent speaking to his or her child, and as she goes on to call herself ἀλαστοτόκος καὶ ἄλαστα παθούσα ("miserable in child-bearing and having suffered misery") in S13 lines 2-3, it must be Geryon's mother rather than his father. The feminine participle is certain; the compound is W. S. Barrett's supplement, accepted by L.
In fragment S14 we have a divine summit meeting. Zeus, Athena, and Poseidon are present, and Athena says something to Poseidon about Geryon and death. Poseidon is the father of Geryon's father Chrysaor, and Athena seems to be warning him that she will back Heracles in the coming battle.
The battle between Heracles and Geryon must have been one of the big moments in the poem, and several fragments come from this: S15, S16, and S21. According to L this is the most readable part of the papyrus (p. 193). Some 45 lines are preserved here, with about a dozen consecutive lines that are nearly intact. We see Heracles attack Geryon one head at a time, using a bow as well as his club. One of his arrows is coated with the Hydra's poison, making it particularly lethal. As Geryon's first head dies it is compared to a flower falling from a poppy plant (p. 254-268, S15 col. 2), in a simile related to, and probably modelled on, that of Iliad 8.306-308. L carefully compares the elements of the two similes, showing how Stesichorus has adapted the epic material. He also shows how the death of Geryon's first head is depicted on vases. Geryon generally looks like three men standing next to each other, fused into one, and when one head has been killed, it falls forward on its shoulders, leaving the body held upright by the other two bodies. Thus the head looks like a blossom drooping on its still-upright stalk (p. 257 and plates 14-16).
Another group of fragments involves Pholus (S19, S20, S85). Of these, S19 is from Athenaeus and S85 from Pausanias; both authors identify the material as from the Geryoneis. Exactly how Pholus fits into the story of Geryon is not clear. He is a centaur and thus belongs to the story of Heracles and the centaurs, which is a digression from the poem's main story. Perhaps Heracles met up with Pholus on his way to Geryon's island; as L and other scholars have observed (p. 288), it seems most natural that the Geryoneis should end with the death of Geryon, or with its immediate aftermath, so secondary stories like the Pholus story ought to come earlier in the poem.
Finally, L devotes a chapter to a handful of fragments that cannot be placed in the narrative. One is a note from a scholion to Apollonius Rhodius saying that Stesichorus mentions an island in the Atlantic called Sarpedonia. After a detailed discussion of other references to such an island, L tentatively concludes (p. 349) that Stesichorus probably mentioned it in a genealogical exposition, as opposed to having Heracles actually visit the island. Fragments S25, S71, S28, S29, and S55 are scraps of papyrus consisting of merely a few letters each, but it is in each case possible to work out one or two words, and L does a good job of teasing out the implications of those words. For example, S55 is (almost certainly) the name Hephaestus. Perhaps he made weapons for Heracles, or perhaps he constructed the cup of the sun.
By far the most important of the tiny scraps is S27, not for the words but for the stichometric indication in its margin: the number 1300. Actually the number appears twice, once crossed out and then again four lines later (p. 359-362), as if the scribe had miscounted the verses. Since we know the poem is in triads of 26 lines, and the papyrus has 30 lines to the column, this line number must be in column 44 of the papyrus. L does not say whether this means the roll is particularly large. If 44 columns is a great number, then the poem is not much more than 1300 lines long, but if this is a moderate size for a papyrus roll, then the poem may be somewhat longer. In any case, the poem is certainly at least 1300 lines and at least 50 triads long. For comparison, Pindar's fourth Pythian, the longest of his epinicians, is 299 lines and 13 triads, or roughly a quarter of this size.
Naturally, meter is an important constraint on reconstructing a verse text. We have enough fragments to reconstruct the metrical scheme of the poem, which is triadic, with nine lines in the strophes and antistrophes and eight in the epodes; the strophe may be four or five periods and the epode is probably three (p. 15-16). The lines (or cola) are 8 to 10 positions long, somewhat more than half the size of an epic line. The meter consists of alternating long and double-short, with most of the double-short elements contractible. There are no single short syllables anywhere. One would normally label such a meter dactylic, except that every line of the strophe and five of the eight lines of the epode begins with the double-short instead of with the long element. Moreover, both strophe and epode end short-short-long, never long-long. Thus through most of the schema, except for the middle of the epode, we have a rising rhythm, more like anapests than like dactyls. Hence it is customary to call this kind of verse "dactylo-anapestic." It can be distinguished from the dactylo-epitrite that Stesichorus also uses, for example in the Thebaid.3 Because there are no single-short elements, any word that can be used in epic can also fit into this meter, and as L demonstrates repeatedly, Stesichorus's vocabulary is close to epic.
The thorough bibliography runs to 30 pages. The book concludes with an index locorum and an index of scholars cited. There are also 22 black-and-white illustrations of vase paintings relevant to the story. It is unfortunate that the illustrations do not include photographs of the papyrus itself.
We cannot read the Geryoneis in its present state, but L's analysis gives us a good idea of what it must have been like. The book makes excellent use of linguistic, metrical, papyrological, and artistic evidence for Stesichorus's poetics and for the story of Geryon.
1. This is my own count of the lines in L's text that contain more than just a couple of isolated letters. The fragments are as numbered in Poetarum Melicorum Graecorum Fragmenta, ed. M. Davies, Oxford: 1991.
2. The story of a hero stealing cattle from a three-headed, six-eyed monster is possibly Indo-European, or at least common to Greek and Indo-Iranian; see Calvert Watkins, How to Kill a Dragon (Oxford: 1995), p. 464-468, and M. L. West, Indo-European Poetry and Myth p. 260-261, 300 (Oxford: 2007). As L points out (p. 31, note 57), there are Near Eastern parallels as well.
3. See M. L. West, Greek Metre (Oxford: 1982), p. 49-51, though he simply calls the Geryoneis-type "wholly dactylic" and the Thebaid-type "dactylic with iambo-trochaic elements," though later he notes this is "to all intents and purposes dactylo-epitrite" (p. 70).