Internet Archaeol. 17. Willis. Examining Samian and Supply

6. Examining Samian and Supply

6.1 Introduction; previous studies and theories concerning samian supply

Samian pottery is remarkably widespread in Britain, occurring at virtually every investigated site of the period. In aggregate, great quantities were supplied to Britain from the various South, Central and East Gaulish sources. The broad chronology of these industries is established (e.g. Tyers 1996; cf. above).

Questions about the supply of samian to Britain have remained an important domain for those interested in Roman Britain, as well as those who specialise in ancient economics, pottery and samian studies. Examination and discussion of the nature and mechanisms of samian supply to, and distribution within, Roman Britain have generated a considerable literature. Perusal of any publication that considers the economics of the province reveals that the authors sooner or later evoke samian importation as one of their main threads of available evidence. Rather unincisive historical narratives have sometimes followed, outlining the developing sequence of supply and not moving beyond this story. The subject is a big one, and not as straightforward as it may at first seem. This perhaps accounts for the infrequency of discussions looking at samian supply analytically, overall, in a 'macro' manner. Samian supply to Britain has tended to be seen in a normative manner by those writing general accounts of the era, a commercial and contractual venture, a matter of markets and money, and cast in a light that owes much to 20th century capitalism system approaches; supply to the army is, however, often seen differently, as in some way an 'administered distribution' (cf. Renfrew 1977). There is much about these types of approaches that might be valid.

Marsh's (1981) paper on samian supply was a substantive statement. It was not exactly ground-breaking, nor was its use of quantitative data with regard to samian stamps and decorated ware original. Its great assets were a systematic comparative approach to site data combined with attention to the known development of the samian industries. Marsh demonstrated the level of export of samian to Britain was not constant over time and thus placed the fluctuating histories of the production sites centre stage. His study delineated fluctuations in the presence of samian from different sources over time in a quantitative manner, and related this to the general pattern of output of the source industries, showing that fluctuations were not necessarily a function of specific site histories in Britain. His work confirmed that it is important that samian present within site phase groups be understood ('calibrated') in terms of the wider pattern of samian availability and interpreted accordingly; (see proviso, Section 6.5.1). The importance of considering the archaeology and the practices at the production sites has been recently underscored by Polak (1998) and Dannell (2002). Meanwhile, attention to more specific studies within the subject, researching, importantly, the invaluable basics of, for instance, a particular potter/workshop has continued. Other serious engagements with aspects of supply exist and have moved studies forward (e.g. Bird 1986a; 1993; Darling 1977; Fulford 1991; Going 1992a; Greene 1982; Hurst 1985). Data collected for this project present an opportunity to examine and characterise aspects of the nature of samian supply to Britain over time.

How samian and other imported pottery arrived in Britain was a matter that generated discussion around the late 1970s and early 1980s, but subsequently this has been an under-explored area. To what extent samian arrived in Britain as 'bulk commercial consignments' (Dickinson 2000a, 204; Symonds 2000) or in somewhat piecemeal lots assembled by middle-men (cf. Millett 1993b) is a matter for investigation. Interestingly, both the Flavian Pompeii hoard and the likely wreck deposit of Antonine date from Pudding Pan Rock, off the north Kent coast, which lacks (to date) decorated samian, seem to lie between these two definitions, being sizeable specific groups of samian drawn from a reasonably broad pool of contemporary output (Atkinson 1914; Walsh 2000). Evidence of the paucity of stamp die duplicates at sites, among the Pudding Pan Rock material and at St Magnus House (cf. Rhodes 1986, 201) suggests it is unlikely that bulk consignments were supplied more or less complete direct from the workshop to sites, or at least that this was not invariably the means of supply. These questions are considered further below (cf. Sections 6.4.6, 6.6.3 and 6.6.6). Middleton had suggested that supply to the Roman army was the motor of supply to Britain of commodities such as samian (Middleton 1979). This hypothesis has not previously been systematically reviewed and the opportunity is taken to discuss this aspect, with regard to samian, below (Section 6.3).

Turning to trade and acquisition of samian at the local level, the observation of Fulford and Huddleston's review of the state of pottery studies in Britain, specifically that our understanding of the marketing of pottery at this level is at an elementary stage, still holds true (1991, 40). It does seem, though, from destruction deposits at Wroxeter, Colchester and elsewhere, that samian was often sold from shops or stalls in towns and other centres (e.g. extra-mural sites) which also sold other specialist and (probably) moderately expensive items, including glass and mortaria, which themselves had also 'travelled' (cf. Rhodes 1989). Coarse wares, conversely, do not seem to have been marketed in this manner.

A striking aspect of samian distribution is its wide diffusion, even of vessels from the same workshop bearing the same maker's die. This can be seen, for instance, with the later Antonine work of Maximinus i of Lezoux. His die 2a occurs on a Drag. 31 from the area of the principia (Period 5) at South Shields, Tyne and Wear (Bidwell and Speak 1994), on two Drag. 33 cups from a grave at Sompting, West Sussex (Appendix 9.1) and among the group of samian from the St Magnus House waterfront site, London (Hartley and Dickinson 1994, 207; Bird 1986a, 192, no. 3.116).

The dynamics of supply and distribution of certain samian types such as East Gaulish samian in Britain might be examined via a mapping technique such as Trend Surface Analysis (cf. Allen and Fulford 1996).