LME/11 Romano-British Pottery from Long Melford (Katie Anderson)
A large assemblage of Roman pottery, totalling 546 sherds, weighing 3137g and representing 2.16 EVEs (estimated vessel equivalent) was recovered from 21 test pits across Long Melford. All of the pottery was examined and recorded in accordance with the guidelines laid out by the Study Group for Roman Pottery (Darling 1994). Sherds were sorted within context by fabric, with unsourced wares of the same type e.g. greywares grouped together. Details of form, decoration, use, wear and date were recorded along with any other information deemed important.
Assemblage Composition
The assemblage comprised primarily small sherds with a low mean weight of just 5.7g, although the nature of recovery (test pits as opposed to excavated features) is likely to have influenced this. A number of sherds were noted as being abraded/heavily abraded, suggesting that they had not been deposited straight after breakage, or else may have been redeposited from earlier features. The assemblage broadly dates from the mid/later 1st century AD to the 4th century AD, with an apparent peak in the 2nd-3rd century AD.
A range of vessel fabrics were identified (see Table 1), with coarsewares dominating the assemblage, representing 95% in total. This is typical for Roman rural sites. Sandy greywares were the most commonly occurring fabric type, representing 85% of the assemblage. Within this category there were a number of different greyware fabrics identified, many of which could be sourced, although it is likely that these were produced locally, as is the case for most Roman coarsewares. Therefore for the purposes of this report, the unsourced greywares have been amalgamated into two groups; coarse sandy greywares and fine sandy greywares. Sourced coarseware fabrics included a small number of Horningsea greywares and Wattisfield reduced sandy wares, both of which could be considered as local products, as both were fairly widely distributed in the Roman period.
Fabric / No. / Wt(g)Black-slipped ware / 8 / 39
Buff sandy ware / 21 / 64
Central Gaulish Samian / 11 / 51
Colour coated ware (unsourced) / 5 / 5
Coarse sandy greyware / 451 / 2646
East Gaulish Samian / 2 / 28
Fine sandy greyware / 11 / 42
Grog-tempered ware / 2 / 24
Horningsea greyware / 2 / 69
Nene Valley colour-coat / 4 / 9
Oxidised sandy ware / 13 / 37
Pakenham colour-coat / 3 / 5
Red-slipped ware / 3 / 5
Reduced sandy ware / 3 / 31
South Gaulish Samian / 1 / 3
Shell-tempered ware / 1 / 41
Wattisfield reduced ware / 1 / 9
Whiteware / 4 / 29
TOTAL / 546 / 3137
Table 1: All Roman pottery by fabric
Finewares represented just 5% of the assemblage by count, including both Romano-British wares and a small number of imported wares. 14 sherds of Samian (82g) were recovered, with sherds from each of the main industries represented. Central Gaulish Samian sherds were the most commonly occurring with 11 sherds in total, followed by two East Gaulish Samian sherds and one South Gaulish sherd. Most of these sherds were small and abraded, with the one exception being a medium sized East Gaulish decorated sherd from Test pit 34. This sherd had part of an animal (possible a lion) and ovolo decoration, and dates 2nd-3rd century AD.
A small number of Romano-British finewares were identified, including Nene Valley colour-coated wares (4 sherds, 9g) and Pakenham colour-coated wares (3 sherds, 3g). There were also five unsourced colour-coated wares identified.
The range of vessel forms identified was fairly limited, with 84% of the assemblage comprising non-diagnostic sherds, which is not unsurprising given the condition of the assemblage. Of the forms that could be identified, jars were the most frequently occurring totalling 61 sherds (973g). Within this category there were a variety of different types of jars, including everted rim jars as well as necked jars with beaded rims. Rim diameters ranged from 10cm to 32cm, highlighting different functions from storage to the preparation and serving of foodstuffs. Other vessel forms were not as well represented as jars. Seven beakers were identified, including an early Roman poppyhead beaker (Test pit 30, context 5) and an indented beaker with roughcast decoration, dating 2nd-3rd century AD (Test pit 5, context 7). Two greyware straight-sided dishes dating 2nd-3rd century AD were recovered from Test pits 25 and 29, contexts 3.
Form / No. / Wt(g)Beaker / 7 / 37
Bowl / 4 / 53
Cup / 3 / 32
Dish / 6 / 47
Flagon / 1 / 8
Jar / 61 / 973
Lid / 2 / 21
Mortaria / 1 / 6
Unknown / 461 / 1960
TOTAL / 546 / 3137
Table 2: All Roman pottery by form
Several vessels were noted as having limescale on the interior, which is an indication that these particular pots were used for holding/boiling water. A further vessel had evidence of sooting on the exterior suggesting that it had been used over a fire.
21 sherds (3.8% of the assemblage) were noted as being decorated, with combing and rouletting being the most commonly occurring types. The relatively small number of decorated wares is not unexpected from rural Roman sites.
Test Pit Analysis
21 Test pits contained Romano-British pottery; albeit in varying quantities (see Table 3). Two test pits stand out as containing large quantities of pottery. Test Pit 5 contained 114 sherds, weighing 427g, from ten different contexts. The material from this test pit was mixed in date with some examples of Early Roman pottery, including a single South Gaulish Samian sherd (context 3). The majority of sherds were however, 2nd-3rd century AD in date, including three Central Gaulish Samian sherds (from contexts, 5, 8 and 10). It is of note that there is no apparent stratigraphic relationship between the different contexts, with the earlier dating sherds occurring alongside later dating sherds.
TP No. / No. / Wt(g)1 / 2 / 2
2 / 17 / 49
3 / 2 / 5
4 / 16 / 40
5 / 114 / 427
7 / 1 / 55
9 / 32 / 145
10 / 3 / 3
17 / 4 / 23
18 / 1 / 1
25 / 51 / 201
29 / 55 / 313
30 / 21 / 352
32 / 38 / 243
33 / 8 / 24
34 / 113 / 907
36 / 26 / 123
37 / 4 / 61
38 / 12 / 68
39 / 25 / 88
40 / 1 / 7
TOTAL / 546 / 3137
Table 3: All Roman pottery by Test pit
113 sherds weighing 907g were recovered from eight contexts within Test pit 34. The pottery although mixed in date, is predominately 2nd-3rd century AD in date. This included two East Gaulish Samian sherds from context 8. As with Test pit 5 there is no apparent difference in date between material recovered from the bottom of the test pit versus pottery from the top. This suggests that the pottery had either been redeposited due to the recutting of features, or else the pottery had been left on the surface for a period of time before being deposited, hence the mixed date of the assemblages from these two test pits. This is a pattern that is seen in many of the test pits, for example, in all cases, early Roman pottery always occurs alongside later dating material. There were no test pits which are solely early Roman in date. Most of the pottery came from test pits in the centre of Long Melford, with no evidence of Roman pottery coming from outside of this cluster.
Discussion
The assemblage of Romano-British pottery recovered from the test pits at Long Melford is of great interest in terms of the quantity and also the nature of the pottery. It suggests a Roman settlement(s) with origins in the early Roman period (mid 1st-2nd century AD), which peaked between the 2nd-3rd century AD. The assemblage is somewhat typical of a Roman rural settlement. The fabrics and forms identified suggest that although most of the pottery was obtained from the local area, the site did have access to goods from outside of the immediate locale, including a small number of imported wares. Given the quantity of material recovered from test pitting (and in particular Test pits 5 and 34); there is a suggestion of a fairly large settlement in this area during the Roman period.
References
Darling, M Ed. 1994. Guidelines for the Archiving of Roman Pottery. Study Group of Roman Pottery.