The Prima Porta Garden Archaeological Project. Terra sigillata from the Villa of Livia, Rome. Consumption and discard in the early Principate

Sammanfattning: This study examines a corpus of Italian sigillata from the Villa of Livia outside Rome, and presents the excavations during which it was found. Most of the material stems from archaeologically secure contexts, a refuse tip and dump dating to the late Augustan and Neronian periods respectively. These prove to be valuable sources of information concerning the influx of terra sigillata from different places of manufacture to Rome during the early Principate. Unlike other sites around the Roman Empire, where the central Italian stamps and/or fabrics amount to less than five percent of the finds, about 60 percent of the stamps in the Augustan contexts were found to be central Italian. In the light of this, it is argued that the central Italian production of terra sigillata has been considerably underestimated. Special focus is directed on this production as a manifestation of the non-agrarian economy in the hinterland of Rome. It is suggested that these workshops operated by means of locatio conductio contracts and were mainly supplying the capital. An analysis of forms and fabrics reveals that products from the Arezzo workshops were more or less barred from the market of Rome during the same period. In context of this evidence, it is suggested that these local manufacturers were an important factor in a competitive environment of terra sigillata production in Italy. It is argued that Arezzo's establishment of regional and provincial branch workshops should be seen in relation to this, and not solely as a means of seeking export markets. Ocular inspection has identified six central Italian fabrics, a division that can only partially be confirmed chemically, but techniques of MGR-analysis make it possible to locate the production centres for several potters of unknown provenance. Several methodological issues are raised here and it is argued that attempts to match fabrics between sites risk a high margin of error, due to their dependence upon a blend of subjective observations, inference and an unwarranted faith in 'scientism'. I argue that the excavated material can also be interpreted in behavioural terms, linked to patterns of consumption and refuse-management. A distinction between cenae and/or convivia during the late Augustan period, and servant-related activities during the reign of Nero, is postulated. During the former period it seems that terra sigillata was much appreciated even at the very heart of the Roman Empire, whereas by the time of Nero it had taken on a devalued, functional and more utilitarian aspect. I argue that the vessel forms Consp. 23.1, 26.3, 28.1-2, 33.4 and 33.5 should be tied with a higher degree of certainty to the late Augustan period. The evidence also implies that the traditional dating of Consp. 23.2 is too late, and I propose that its introduction should be pushed back to the late Augustan period. It is further suggested that the idea of terra sigillata services with one plate and two cups can be explained by the higher discard rate of the latter in relation to the former, and that the mending of vessels was done irrespective of replacement costs.