Internet Archaeol. 17. Willis. The incidence of the samian flagon, Stanfield 67

8.6 The incidence of the samian flagon, Stanfield 67

Samian flagons of any form are rare in Britain (e.g. Hermet 15) and hence the occurrence of examples is noteworthy. The samian flagon type known as Stanfield 67 (Stanfield 1929, Form 67) is a small but distinctive vessel of globular character which, although rare, has a widespread distribution in Britain. Its distribution is examined here as it appears that a high proportion of finds of the type appear to be associated with burials or structured deposits, with indications that the form was treated differently from other samian types. No previous study of the distribution of Stanfield 67 has been undertaken and hence this examination constitutes an initial view.

The incidence of examples of Stanfield 67 is listed in Appendix 8.3. Twenty examples are recorded from 14-15 locations. The majority of the find-sites are in southern Britain, but examples also occur in the Midlands, at York and at Birdoswald. The type appears to be Antonine to mid 3rd century. Of the twenty recorded examples (Appendix 8.3) a high proportion come from 3rd century or later deposits and may be residual; the Brighton, Springfield Road, example is believed to come from an Antonine group, while that from the burial at Crabtree Lane, North Lancing, could have been interred at a similar date. Of the 20 examples recorded in the course of this study, ten are of Central Gaulish fabric and seven are East Gaulish; in two cases the fabric/source is not reported. King states of the example from Silkstead Sandpit, Otterbourne, Hampshire, that it is 'possibly South Gaulish samian of the late 1st or 2nd century AD' (King 1992, 50). It is more probable that this item is from Central Gaul, and since it was complete it may not have been possible for King to examine the fabric in detail. Bird states that, 'there is a closely similar flagon [to the Stanfield 67 type] in South Gaulish ware of mid-later 1st century date from Tharros in Sardinia (British Museum), indicating a long if sporadic tradition for the form' (Bird 1992, 86). Hence a South Gaulish origin is perhaps possible for the item from Otterbourne, though it is more likely this vessel was from Central Gaul.

Stanfield 67 flagons are recorded from settlement deposits at six sites. Despite their overall rarity in Britain, five of these six sites have produced more than one example of the type. This clustering may be a function of supply, or indicate that they were used in pairs, or that certain reporters were more adept at recognising examples, or there may be some other explanation. Eleven of the examples of the vessel type come from these occupation areas at a number of different sites. Two examples, in Central Gaulish fabric, were recovered from the fort at Birdoswald, while three examples in East Gaulish fabric came from the as yet unpublished 1969-81 work at Piercebridge, either from the fort or vicus (Table 57; Appendix 8.3). Examples are also reported from London and the 'Small Town' at Alcester (Appendix 8.3). The vessel from Great Witcombe villa came from beside the villa building and could be reconstructed to be almost complete, suggesting it may have, unusually, been deposited whole (Leach 1998; Dickinson 1998, 63, fig. 17, no. 172). At the rural enclosure settlement at Oakridge, Hampshire, a pair of Stanfield 67 flagons, of Antonine date, were deposited whole in a well shaft along with a Drag. 37 decorated bowl bearing a very unusual Bacchus scene (Appendix 8.3). These finds appear to be part of a votive deposit within the well (cf. Bird 1992; Willis 1998a, 119); the deposit dated to the late 3rd or early 4th century, indicating that these samian vessels had been curated over decades.

Three of the twenty Stanfield 67 flagons, at least, occur in graves, which is a high frequency. The three cases are: Brighton, Springfield Road, Crabtree Lane, North Lancing, and Heybridge, by Bouchernes Farm, where the exact context is not certain, though circumstances indicate a very high probability that this item is from a grave (cf. Table 57; Appendix 8.3). In each case the vessels were essentially complete; that from Brighton, Springfield Road, was minus its handle and had been truncated at the neck (Dudley 1981), though these aspects are consistent with deliberately 'wasted' or 'altered' funerary pottery (cf. 9.6). Coarse ware flagons were commonly employed as grave goods in southern Britain in the early and middle Roman era, and the employment of samian flagons is evidently an extension of the practice. Of the remaining six cases two are old finds of rims only, from London, and so are probably neither from burials nor unusual deposits; the vessel from Telscombe near Brighton is fragmentary and contextual details are unclear (Table 57).

On the other hand the other three vessels from Dymchurch, Otterbourne and York are complete. Since complete vessels are extremely rare among normal settlement deposits, the prospect that these three items are also from burials or structure deposits must be considered. The Dymchurch flagon, said to be from the sea wall, is an enigmatic find from a natural boundary location which may be significant. Construction of sea defences at Dymchurch in 1844-6 lead to the discovery of extensive occupation of Roman date lying between the sea and the marsh, including burials (Issacson 1846; Cunliffe 1988); hence there is a strong chance this example of a Stanfield 67 is from a burial. The find from Silkstead Sandpit, Otterbourne, is associated with other complete vessels, and it is possible that these are from burials at this site or nearby, or from a postulated temple at this location (Denford 1992, 51-2). The flagon from York, found more than a hundred years ago, may be from a grave; many graves were disturbed at York in the 19th century (and before) through development, with vessels collected in the museum. Finally, returning to the vessels from the Baromix factory site, Alcester, it is noteworthy that while no burials are known from the vicinity of the Stanfield 67 vessels recovered from this site in 1972, a ritual or symbolic association is possible. At this site, as Jeremy Evans (pers. comm.) has noted, the presence of a copper-alloy moulded finger from the site (Lloyd-Morgan 2001, 76), in a phase H context, is likely to be from a tap, and could indicate a spring, as indeed could the stone-lined drain from phase F, while the possibility of a baths building adjacent to the excavated site is also of potential interest in this connection (Booth and Evans 2001, 90).

Stanfield 67 flagons occur at many types of site. That the majority of the find sites are in southern Britain may be a function of archaeological practice and recovery, or may, indeed, be revealing a genuine aspect of the distribution of the form which raises questions about its potential contents and use. A curious fact is the 'coastal', near coastal, or estuarine occurrence of many of these vessels in southern Britain. Locations with examples of the type falling into this category include London (New Fresh Wharf), Heybridge in Essex, Otterbourne in Hampshire, Dymchurch, Kent, and Brighton (Springfield Road), North Lancing (Crabtree Lane), and Telscombe in Sussex (Appendix 8.3).

Whether these vessels of Stanfield Form 67 were employed as grave furnishings or in structured deposits because they were unusual distinctive samian forms, or, more simply, because they were of a basic generic form class (that is to say, flagons and jugs) that was more or less routinely used in practices of this manner is of interest, and for investigation. The balance of evidence demonstrates that these vessels were evidently treated (or at least deposited) in a way that differed from the general treatment of other samian forms. A further observed trend that underscores the probability that they were ascribed a special status is the fact that many of the vessels appear to be have been curated over long time periods before entering the ground. These aspects raise questions both about the contents of these vessels and their function/s. That they are unusual forms occurring in what were probably sensitively charged contexts places them among examples of a trend noted elsewhere: that is for unusual samian types, of potentially some value, to appear in graves and special deposits, particularly at rural sites and at the smaller civil centres.