The text, as a tissue of signifiers, makes certain demands on the textual analyst, and provides the material for analysis. That material is by no means an empty space, a vacancy into which we pour whatever we like; instead, the text itself participates in the process of signification. It reproduces or reiterates meanings, which always come from outside, and are not at the artist's disposal, any more than they are at ours. The work of art is in that sense always citational, constituted, as Barthes puts it, of quotations. Tarquin and Lucretia reiterates a familiar tale, quotes an existing narrative, and, in turn, the classical sources themselves re-inscribed a story already in circulation. But even when there is no narrative to cite, the text invokes intertexts: new female nudes signify in relation to existing female nudes, just as domestic interiors allude to other domestic interiors, and landscapes are intelligible in relation to the tradition of landscape painting. There is no moment of ‘origin’, but only breaks with what went before. In that sense, every iteration is always a reiteration. Research involves tracing these intertexts, and reading them attentively too, to establish the difference of the text in question.
At the same time, texts can only ever quote with a difference. ‘Iteration alters, something new takes place.’ In Limited Inc Derrida makes the paradoxical point that, while a repetition is the same as the original utterance - or it would not be a repetition - it is also the case that a repetition is never the same as the original, or it would be the original itself. In other words, every time anyone uses a familiar mark or image, they shift its meaning very slightly in the process, precisely by quoting its previous occurrences, as well as changing its setting. In that sense, every text breaks with what went before. Maybe artists shift the meanings or break with the past more radically at times, but changing meanings is not the same as making them up.
The possible meanings of Tarquin and Lucretia, then, are to be found - or perhaps more accurately, supposed, hypothesised - in the relation between the painting and the viewer who is its destination. And each party - the picture and the spectator — contributes to the process of making it mean. The viewer faces the picture from a place outside it, and examines from that location the internal relations on the surface of the canvas.
As an instance of one elementary imposition of limits exerted by the painting itself the optimum physical place of the spectator is fairly sharply specified by Tarquin and Lucretia. The picture is not visible, for example, from the back. What is more, it is not clearly visible from one side or the other, because the lines appear distorted from there. And from too close up, the trompe l'oeil effect disappears. The folds of the fabric, for instance, dissolve into lines on a flat canvas. An image in monocular, fixed-point perspective, as this is, addresses a viewer who stands in a specified place, directly in front of it, and with one eye closed. Only from there do we fully grasp the three dimensions simulated on a two-dimensional canvas. In this instance, Tarquin, Lucretia and the properly placed viewer form a triangle.
The text exercises certain constraints, and yet we are not entirely at its mercy. A good textual analyst would be aware of the text’s requirements on us, in this sense, but the same good analyst would also acknowledge that we might deliberately refuse the position the text offers, might choose to look at it from somewhere else. On close examination, Titian poses a puzzle: the very precise mimetic effects are apparently achieved by techniques that, viewed from close to, appear quite impressionistic, compared, for example, with Bellini. How, we might want to know, is it done? An analyst of Titian’s brushstrokes would reject the optimum position of the viewer, would get close and lose the three-dimensionality in a good cause. Alternatively, the composition lines might draw my eye to the violence, but I remain entitled to look at the bedlinen. Engaged in dialogue, the textual analyst retains a certain independence.
How does this account of the reader’s relative self-determination square with Barthes’s account of a reader who is no more than the destination of the text? Barthes helps to locate meaning where it belongs, in the signifier, not in the head of either the author or the addressee. But at the same time, his essay retains a trace of structuralism. Fully alert to the differences that make up the text, Barthes here ignores the differences between readings. If the textual analyst understands all the quotations in the text, grasps all its intertextual allegiances, every trace by which it is constituted, how is it possible for interpretations to differ? How can we, after all, say anything new?
