Ovid would be the first to agree with Foucault ( 1978a) that all human sexuality is constructed, arguing, however, that construction is not only inevitable but desirable. If the Roman gaze was primarily nude, so then is its object. If anything, it is the female object, rather than the female subject, that may be elided, in these pictures where men hold the mirrors (with the remarkable exception, I think, of the paintings from the Villa della Farnesina, which seem to me--and seemed to the "ragged imagination" of my Roman summer--to represent a female object for a female subject which might speak directly to women without male mediation). From the start, it had been the Roman male self (usually in a political or military role) that was the subject of Roman art. For the Roman male, it was perhaps more significant and noteworthy that in his erotic tabellae he gazed at, objectified, and took possession of himself, as he had been doing in other types of representation all along.
There may indeed be Roman pornography--those representations that speak a language of violence rather than sex, that construct a hierarchy of objectification, that equate female masochism with female sexuality, and that in themselves offer a sexist construction of reality--but I believe we will not find such pornography in these erotic paintings from first-century Rome.
Finally, as for the parva tabella Ovid claimed could be found in Augustus's own residence--for the reading vestris, "your houses," must certainly be correct--it would be nice to be able to report that Carettoni's excavations on the Palatine ( Carettoni 1983) had turned up one such painting in the austere cubiculum to which the emperor reputedly retired for forty years, winter and summer, scorning the extravagance of fancy villas while transforming Rome from brick to marble, or perhaps in the ruins of the technyphion, his famed upper-story retreat ( Suetonius Augustus 72.2-3). So far, no such picture has surfaced. But then, this would not be the first time that Ovid held a less than flattering mirror to the emperor's preferred image, and the digging on the Palatine is not done with yet.
Ancient works cited by abbreviation throughout the text and notes are HN = Pliny Historia Naturalis; Tr. = Ovid Tristia; AA = Ovid Ars Amatoria; QNat. = Seneca Quaestiones Naturales.
For the manuscript tradition of the Tristia, see S. G. Owen's introduction to his edition and commentary of Tristia 2 ( 1924) and more fully in the introduction to his 1892 Oxford