With this background in mind, let us turn to the representations themselves. Richard Green (1991, 42) believes that: ‘‘The clearest and most convincing example of depictions derived from the theatre is the series which seems to relate to Sophocles’ Andromeda.’ He considers five Attic red-figure vases dating from just after the midfifth century and later. On them appear different moments drawn from the exposure of Andromeda as an offering to appease the sea monster (ketos), who has been terrorizing her land. Each vase chooses one segment of the action. A hydria in London (No. 1; see List of Objects at the end of this chapter) shows Andromeda held by two youths, as the stakes are planted in the ground on the right. A pelike in Boston (No. 2) chooses the next moment with Andromeda's right hand already attached to one of the stakes (plate 7.1). A third vase (No. 3), a white-ground calyx-krater in Agrigento, has the end moment with Andromeda fully attached and exposed, as her rescuer, Perseus, contemplates her on the left. Green argues that the vases form a coherent group dated roughly to the 440s and, significantly, are from different workshops, hence likely to have been created independently of each other (Green 1991, 43).
Yet the motif of tying a person to stakes occurs at least as early as the sixth century BCE and therefore before Sophocles wrote his play. Similarly, Persian dress for Andromeda in the second half of the fifth century bce need not go back to a staged performance, because before the time of Sophocles it had already become part of the wardrobe of the wealthy (Miller 1997, 164). Showing successive moments in the
Plate 7.1 Andromeda. Attic red-figure pelike attributed to the Kensington Painter and Kensington Class. ca. 450-440 bce. Boston, Museum of Fine Arts 63.2663. Photograph: Courtesy, Museum of Fine Arts, Boston. Reproduced with permission. © 2004 Museum of Fine Arts, Boston. All rights reserved.
Tying of Andromeda to the stakes likewise does not require a theatrical source, for an even greater variety of moments is shown in Theseus’ fight with the Minotaur, which would not have appeared on stage (Brommer 1982, plates). All of which brings us to the major problem: the play by Sophocles has survived only in snippets quoted by other classical authors. As a result, it is disputed what parts of Andromeda’s story Sophocles used, especially since the literary tradition about Perseus and Andromeda goes back to the sixth century bce (Phillips 1968, 1-2). Finally, the two alternate traditions for the exposure of Andromeda - tied to stakes or to a cliff - co-exist. More importantly, the Darius Painter (No. 4), one of the more notable and more ‘‘learned’’ of Apulian vase-painters, uses both motifs. In conclusion, I think that neither a text nor a particular performance can be cited as the source for these vases in the current state of our knowledge.
Oliver Taplin (1993, 22-23) takes an intermediate position on the dependency of artists on texts. He examines one of the rare examples where we have both text and pictures: Medea by Euripides with its possible representations on South Italian vases. Consider an early Lucanian calyx-krater (No. 5), now in Cleveland and dating to around 400 bce, with Medea flying off in her dragon-drawn chariot after having killed her children. Medea and her chariot are enclosed by a rayed nimbus, a symbol of her grandfather Helios (plate 7.2). Below on the right lie her two slain sons draped over an altar, as their white-haired nurse stands in mourning, pulling at her hair. Just behind her stands a tutor. Jason appears on the left and holds a staff. Seated on the upper left and right are matching female demons, probably the Erinyes. No stage is portrayed. Taplin (1993, 22) makes the connection to Euripides on the basis of the depiction of Jason as helpless, ‘‘since this is where the whole emphasis lies in Euripides’ closing scene.’’ Taplin readily admits that the story predates Euripides’ Medea, produced in 431 bce, but notes that the extant visual representations date to after the play. At least seven other playwrights, all later than Euripides, wrote about Medea, yet Taplin ignores the possibility of a lesser playwright inspiring a painter. He proceeds with his analysis and gives six differences from Euripides’ Medea:
1 Jason is naked or half-naked.
2 The children are on the altar and not in the chariot.
3 Dragons drawing the chariot are not mentioned by Euripides.
4 The nurse has been added.
Plate 7.2 Medea. Lucanian red-figure calyx-krater attributed to near the Policoro Painter. ca. 400 BCE. Cleveland, Museum of Art 91.1. Photograph © The Cleveland Museum of Art, 2004, Leonard C. Hanna Jr., Fund, 1991.1.
5 The tutor has been added.
6 The Erinyes are alluded to by Jason (Medea 1389-90), but are not present in this scene.
The question is whether these differences are plausible additions or contradictions of the action. Taplin suggests that some of the discrepancies are due to artistic reasons. Actors always appeared dressed on stage, but figures in art were often in heroic nudity. Hence each medium follows its own conventions. Nonetheless, I am not sure that one should consider heroic nudity irrelevant in judging whether a particular picture reflects a particular play, because phlyax vases and three Sicilian vases with indisputable stages have dressed figures (Small 2003, 52-61). The other differences are more complicated. Although Euripides does not bring the Erinyes on stage, it is, of course, possible that they were part of the overall story, if not the Euripidean version. And that is the crux of the problem. When does the depiction depend on the ‘‘general’’ tradition, and when does it derive from a specific literary rendition?
Taplin maintains that he has a decisive answer that is also an advance in our ways of dealing with such material. He asks, ‘‘Does the image ‘call for acquaintance’ with a text or performance of a text?’’ And he responds, ‘‘The most important point is that the viewer’s pleasure would have been enhanced by an observant recollection of the play, and yet not offended or diminished by substantial departures from any particular scene’’ (Taplin 1993, 22-23). I find this reasoning very slippery. I can think of few cases in which more knowledge does not enhance one’s appreciation of either a work of art or a text. As one notes the divergences, one thinks that this or that is a nice touch or wonders why in the world the artist or writer put such and such into the scene. And this is not just a modern reaction on my part. Pausanias compares what Polygnotus has painted in his Iliupersis in the Lesche of the Cnidians at Delphi with what Homer has said. For example (Pausanias 10.25.4, trans. Jones 1935): ‘‘These names too are different from those given by Homer in the Iliad, where he tells of Helen going to the wall with her slave women.’’ Here Pausanias’ knowledge of Homer definitely enriches his view of the painting.
Furthermore, Taplin’s suggestion lacks any critical standard for accepting or rejecting a particular artistic rendering as dependent on a text. He basically says that differences between picture and play do not matter, which is true only if you are not trying to figure out how the two media work both together and separately. If you want only to enjoy either the play or the picture or both, Taplin’s approach is admirable and, actually, would strike a classical person as eminently reasonable. Taplin believes the crucial feature is the ‘‘helplessness’’ of Jason, since that is the crux of the end of Euripides’ play. In contrast, I am particularly disturbed by the fact that the artist of the Cleveland vase has focused visually not on Jason, who is off to the side, but on Medea first and the children next. Even more striking to me is the element of the children on the altar with Medea flying away without them. To me, Euripides focuses not on Jason’s helplessness, but on Medea’s not allowing Jason to touch or bury his sons. She is taking them with her. Euripides does not say whether the children are visible during the final scene. The text can be read either way. Nor does he mention an altar, nor whether Jason has a sword, much less whether he waves it at Medea.
In conclusion, I certainly agree that this vase shows Medea and Jason. I agree with the interpretation that Jason is ineffectual. I find that a number of the differences from Euripides’ version are plausible. Yet the vase portrays a different outcome. Medea clearly abandons her children, and Jason gets to bury them. This element violates the story line that Euripides has set up. It implies compassion for Jason and gives him what he pleaded for most of all in the Euripidean version. Finally, my argument about the central importance of plot finds support in Aristotle {Poetics 1450a38-39, trans. Halliwell 1986), for he says: ‘‘The plot-structure [muthos] is the first principle and, so to speak, the soul of tragedy, while characterization [ ta ethe] is the element of secondary importance.’’ As a result, the ‘‘call for acquaintance’’ argument must be discarded for the following reasons: it does not follow the logic of the plot, which makes it un-Greek in concept, and it specifically does not work for Taplin’s best-case example.
World History
