If the current structural approaches to the interpretation of Greek vasepaintings have a common denominator, it is the model of structural linguistics. In this view, the imagery as a whole is taken to be a system of communication analogous to language and, like language, a social construct that makes no direct reference to an objectively existing reality. An important question arises from these premises: do the different systems operating in the same culture relate to one another, and how? In particular, granted that their structures are parallel, is there also a direct connection between language and visual imagery?
The Saussurean model does not preclude that there may be. If verbal signs refer not things but to concepts, and visual signs also refer to concepts, can their respective terms of reference be utterly discrete, when both kinds of representations circulate in the same ambience? To state the question in this simple form does not intend to suggest that the answer is either simple or obvious, but to expose the fact that the issue has been avoided, by and large, in investigations into the meaning of images. There seems to be some reluctance to venture into that forbidden zone, where the problem of how images are used in actual social exchange, whether as “natural symbols” or as socially constructed ones, must eventually be encountered.
On the basis of the premise that verbal and visual signs ultimately refer the same patrimony of concepts, the one that lies at the heart of the culture, it is proposed here that a semantic connection of the two is normal and that it takes the form of the bonding of word and image that is the essential feature of metaphors. Metaphor here is understood not as the substitution of one term for another, having primarily the function of amplification and ornament, but in the sense in which it has come to be defined in the course of the past fifty years by philosophers and linguists; a fundamental mode of cognition
and communication, which makes abstract notions comprehensible and theoretical discourse possible.
The figure on which I rely to explain my hypothesis is very common in Attic vases of Late Archaic and Classical date: the enveloping mantle, which encases the person like an impenetrable shell, leaving only head and feet exposed. It is most frequent in scenes painted on the “backs” of vases, normally column kraters, showing standing figures, generally three, of which at least one is often wrapped in the mantle. The traditional explanation, “conventional conversation scene”, acknowledges that the picture is commonplace, as its frequency and repetitivcness indicate. This is not to say, as the label implies, that the image is meaningless. On the contrary: by definition, what is commonplace is packed with meaning, so consistent and obvious to the viewer toward whom the image is directed as to be subliminal. To the modern viewer, on the other hand, the sense of these representations is all the more impenetrable for being given in stereotype, that is to say, reduced to a minimum of articulation, lacking emphasis.
Other scenes that include the enveloping mantle are richer and recognizably anchored to certain themes. In representations of courtship, it is the object of desire – boy, girl or woman – who is represented wrapped, whether subjected to longing glances, or offered gifts, or, rarely, embraced”.