Internet Archaeol. 17. Willis. Samian supply to the Roman military

6.3 Samian supply to the Roman military

Hurst's discussion of the supply to the Neronian Roman military site at Kingsholm typifies views on the nature of provisioning the Roman army in Britain with commodities, including samian, coming from a long distance (Hurst 1985, 124-6). He viewed long-distance supply arrangements to the army in Britain as likely to be 'basically an extension of those used for the army in the Rhineland' and that the systems for military supply 'did not relate to the marketable value of the goods or cost of its transport' (1985, 124). These interpretations seem reasonable for the Conquest period at least.

Middleton (1979) contended that the motor for supply to Britain was the provisioning of the Roman army. He stated 'long distance trade ... was almost certainly directed to the "needs" of the military market which was the goal of those supply lines. There is no evidence that civilian markets generated long distance trade' (1979, 81). He further states that, 'entrepreneurial activity of private traders (negotiatores) was almost certainly geared to the satisfaction of military wants and was parasitic on the regular shipment of official goods' (1979, 90). By inference, from this perspective samian arrived in Britain mainly through the requirement to stock the standing army in Britain, and the possibility of doing so through 'piggy-backing' on the established supply mechanisms to the military (Middleton 1979, 92). This thesis has never been fully assessed.

Certainly the army in Britain consumed samian on a large scale. Attention to Tables 23, 24, 27 and 28 demonstrates that military sites have consistently high proportions of samian compared with other site types (cf. Section 7). As discussed below, military sites have a much higher proportion of decorated samian vessels (as opposed to plain vessels) among their assemblages, which is testimony to the fact that the army consistently received selected supplies of samian. In addition, as also discussed below, military sites must have received and used samian from regular supplies. Scrutiny of the composition of contemporary samian groups from military sites and other site types suggests towns, for instance, have older samian stock in use (or more residual material) than military assemblages, which are more up-to-date (cf. Willis 1998a). This trend is confirmed emphatically by Table 5 comparing data for Drag. 29 bowls from forts with that of towns, and other sites, and has implications with regard to dating (cf. below). In the round it is likely that there existed separate marketing/distribution networks across the same geography: one supplying the army, another supplying civil sites.

That, proportionately, samian is most frequent at military sites supports the Middleton thesis. Whether this means it was the motor of supply to the province though is questionable. The evidence for the supply of samian directed to the civil province of the 2nd and 3rd centuries is strong; it cannot be seen as a spin-off of supply to the military who, by then, were located miles to the west and north. Middleton cites the Chichester area in support of his thesis. He states that samian is well represented in military horizons at Chichester (as it is at Fishbourne in the early military horizon, though he does not mention this) but rare in the surrounding countryside. He concludes that: 'The implication seems to be that terra sigillata/samian was known and admired, but difficult to obtain once the primary military market had moved on. The small amount of terra sigillata/samian which did reach the area ... is best understood, I suggest, in terms of the parasitic entrepreneurial activity always associated with military supply lines' (1979, 92). However, much is overlooked here, for instance, the large Flavian, and thus post-military, samian assemblage from Fishbourne ('Fishb2' which includes 280 vessels from the excavations of the 1960s) and the richly furnished graves of the 'East Hampshire tradition' associated with civil sites, which include samian, a number having a range of samian vessels present (cf. Appendix 9.1). The evidence shows that samian streamed into this area after the Roman army had left. There was no determining association with the military and its presence.

In addition, it is now clear that London was a major point of entry for samian ware through the 1st to 3rd centuries (e.g. Marsh 1981; Bird 1986a; Symonds 2000). London was almost certainly a redistribution centre on a large scale for non-military hinterland sites in southern Britain (cf. Sections 6.4 and 6.6; cf. Rhodes 1986, 202). As Dickinson states of finds from the rural site at Chells, Hertfordshire, 'The two [Montans] vessels on this site were presumably distributed from London' (Dickinson 1999a, 84). The quantity of samian from across Britain is too large and non-military demand too great for it to be mainly a military 'piggy back' trade. Samian deeply penetrated Britain geographically and in terms of social life in the era, both near to, and away, from Roman military foci.

Recently, Jeremy Evans has outlined evidence for the existence of five military supply zones in the north of Britain and Wales; that is, distinctive areas within the zone of sites associated with the Roman military (Evans in preparation). Making links with other studies he has pointed out that sites associated with the Roman military along King Street display similarity in terms of their pottery assemblages and other artefact types (cf. Wild 2002), while, for instance, sites at the western end of Hadrian's Wall have a different ceramic 'finger-print' (cf. Bidwell and Speak 1994; see Sections 7.2.2, 7.3.5 and 7.3.6). Forts in Antonine Scotland show, too, a composition that is different again (cf. Section 6.6). Evans states that this patterning in assemblages might not necessarily be military supply as such from a centralised pool, but result from specific sources supplying the military. The patterns endure through time (cf. Section 7.2.2) and appear independent of garrison unit and changes to garrison as different units move in and out of forts and areas. Hence the evidence, Evans contends, points to the organisation of supply on a regional basis, probably with tied workshops within the zone and specific contract arrangements for external supplies such as samian ware.

Attention to the distribution of East Gaulish samian in Britain (Section 6.7) shows that its presence in quantity is biased towards sites associated with the Roman military and a component of supply must have been organised by, or with, the military. Nonetheless supplies arrived in London, presumably for a civil market, and presumably these shipments were independent of the military.