Cite this as: Tyler, S. 2015 The Saxon pottery, in M. Atkinson and S.J. Preston Heybridge: A Late Iron Age and Roman Settlement, Excavations at Elms Farm 1993-5, Internet Archaeology 40. http://dx.doi.org/10.11141/ia.40.1.tyler2
A total of 6.83kg (366 sherds) of Early Saxon pottery was recovered from sixty-eight contexts in all. The most significant assemblage came from deposits representing the infill of a Saxon re-cut (14529) of Roman well 14984. In particular, three contexts (14528, 14558, 14613) between them produced c. 2kg of pottery (30% of the total recovered from all excavated features). Other contexts producing Early Saxon pottery comprised the fills of sunken-featured buildings, pits, post-holes and the upper levels of Roman ditches. Saxon pottery was also intrusive in small quantities in some of the prehistoric and mid-Roman features. Late Roman features producing both latest Roman and Early Saxon pottery were looked at in some detail in order to define the inter-relationship between the two groups. The Saxon features present themselves as a scattering across the site with two obviously cohesive groups forming a cluster around sunken-featured buildings (Buildings 67-9), which perhaps represent a focus of activity during the Early Saxon period.
The nature of the Saxon settlement is problematic; but, as on other Essex sites, it may be that many of the shallower remains have been removed leaving only the deep features (the wells and large ditches) and a random scatter of the shallower features to tell the story of the Early Saxon settlement.
This report presents a catalogue of the diagnostic elements of the pottery assemblage, followed by a discussion of the date range of the pottery and its inter-relationship with the latest Roman material. Full details of quantification by context, sherd count, weight and fabric can be found in the archive.
Seven fabric groups have been discerned (some subdivided) and these are detailed below.
Fabric 1: A small quantity of pottery (8g), found in only three contexts, exhibits no added tempering. It was made from the local brickearth, which incorporated common small inclusions of micaceous quartz-sand.
Fabric 2a: This is the predominant fabric (4.85kg) and was found in sixty-three contexts. This fabric is hard to medium-hard and has abundant small to medium quartz-sand with a varying amount of mica present. It is difficult to assess how much sand occurs naturally in the clay and how much was deliberately added as a tempering agent. Diagnostic forms in this fabric are represented by nos 1, 2, 4-13, 15-18, 21-31, 33-41, 45 and 47.
Fabric 2b: A variant on the sandy ware (Fabric 2a) which has a moderate amount of haematite present; this generally gives a slightly softer fabric. A small amount (73g) came from four contexts.
Fabric 2c: This sandy ware is differentiated by the application of a slip containing large particles of quartz-sand on the outer surface of the vessel, referred to as schlickung. This is seen as an indicator of an early 5th century date. It occurred in six contexts, total weight 176g.
Fabric 3a: Organic-tempered with abundant temper, resulting in many voids in the surfaces. A softer, more friable, fabric than Fabric 3b. A comparatively small amount (463g) occurred in fifteen contexts, three of which are diagnostic forms, nos 14, 20 and 46.
Fabric 3b: Medium-soft to medium-hard fabric with sparse organic temper. A small amount (73g) was recovered from four contexts.
Fabric 4a: A medium-hard to hard fabric with sparse to common finely chopped organic temper and sparse to common small to medium quartz-sand. It is difficult to ascertain whether the sand is a deliberately added tempering agent along with the organic matter. It may simply occur naturally in the clay source. A total of 76g was recovered from seven contexts.
Fabric 4b: A medium-hard to hard fabric with sparse to common haematite in addition to quartz-sand and organic temper of Fabric 4a. A single context, pit fill 15862, produced 42g, including two diagnostic sherds, nos 43 and 44.
Fabric 5a: A fairly substantial amount (490g) of pottery from six contexts has a vesicular appearance; in addition to a common to abundant amount of quartz-sand, it is tempered with small chalk particles that have leached out on the surfaces. Pit fill 14613 produced a diagnostic rim, no. 32.
Fabric 5b: As Fabric 5a, but with sparse to common haematite in addition to quartz-sand and chalk. It occurred in two contexts, with a total weight of 16g.
Fabric 5c: A variant of Fabric 5a has chalk temper and an outer surface with schlickung. It is present in fairly small amounts (148g) in only two contexts, pit fill 14528 producing a diagnostic bowl, no. 19.
Fabric 6: A few sherds (207g) from three contexts have common to sparse grog temper, usually alongside some quartz-sand.
Fabric 7a: A small amount (172g), which, in addition to quartz-sand, has some shell in the fabric. This may be an intentionally added temper or it may occur naturally in the clay source. Pit fill 15862 produced a decorated sherd in this fabric, no. 42.
Fabric 7b: As Fabric 7a but with surfaces treated with schlickung. It is found in only one context, pit fill 14528 (weight 21g).
The following catalogue comprises the diagnostic sherds, ordered by context. Each entry in the catalogue is taken as a separate 'vessel', although instances occur where variations in fabric composition among several entries are minimal and it is conceivable that they do in fact represent one vessel.
1. Rim, everted, rounded. Fabric 2a. Dark grey throughout. Outer burnished. Post-hole fill 34, Group 933, Structure 58, Area W, Period 6
2. Rim, everted, rounded. Fabric 2a. Dark grey throughout. Post-hole fill 34, Group 933, Structure 58, Area W, Period 6
3. Rim sherd from a bowl. Crudely made with uneven, flattened rim. Dark brown. Fabric 2a. Post-hole fill 73, Group 932, Building 69, Area W, Period 6
4. Rim, high-shouldered jar. Thick-walled. Inner abraded. Black. Fabric 2a. Post-hole fill 73, Group 932, Building 69, Area W, Period 6
5. Rim, splay-sided bowl. Dark brown. Fabric 2a. Post-hole fill 73, Group 932, Building 69, Area W, Period 6
6. Rim, everted and slightly flattened on top, from a large jar. Outer part-burnished, especially on top of rim. Fabric 2a. Fill 4944, Pit 4931, Group 269, Area K, Period 2
7. Rim, two joining sherds, everted, rounded, from a wide-mouthed jar. ?Cooking pot. Fabric 2a. Dark grey throughout. Fill 8807, Pit 8748, Group 44, Area P, Period 2A
8. Squat, ?globular jar with corrugated outer walls. This vessel would appear to have been thrown on a slow wheel. It is fairly thin walled with characteristic corrugations, although still rather crude in appearance. Black-brown. Fabric 2a. Cleaning layer 9720, Group 8012, Area D
9. Rim sherd, slightly everted, rounded, from a jar. Fabric 2a, particularly well fired. Dark black-brown throughout. Layer 10514, Ditch 25027, Group 838, Area F, Period 5-6
10. High-shouldered jar, three non-joining sherds, with hollow neck and everted rim, flattened on top. Outer has incised haphazardly intersecting lines on lower half of vessel. Surfaces dark reddish-grey. Core grey with reddish-brown margins. Fabric 6. Maximum rim diameter, 110mm. Fill 10781, Pit 10782, Group 832, Area E, Period 6
11. Jar rim, everted, rounded. Black. Surfaces burnished. Fabric 2a. Machining layer 11000, Area N
12. Jar rim, everted, upright. Black-brown, reddish core. Surfaces burnished. Fabric 2a. Fill 11506, Pit 11507, Group 700, Area N, Period 6
13. Rim, everted, angular, from a jar. Fabric 2a. Sooting on outer surface suggests use as cooking pot. Dark grey ware, outer surface smoothed. Fill 12045, Ditch 25271, Group 969, Area R, Period 5-6
14. Rim sherds, two joining, everted, rounded. Dark grey throughout. Fabric 3a. Fill 12052, Group 973, Building 67, Area R, Period 6
15. Two joining rim/neck sherds and one lower body sherd from a small jar. Rolled-over rim of uniform thickness, slight neck hollow. Fabric 2a, unusually hard, well-fired, dark grey ware. Lower body sherd has beginnings of base profile which is slightly sagging. Rim appears to be wheel-thrown. Possibly a Saxon imitation of a Roman vessel or a hybrid 'Romano-Saxon' form. Fill 12059, Ditch 25271, Group 969, Area R, Period 4-6
16. Rim sherd, slightly everted, rounded. Black-brown. Fabric 2a. Fill 12059, Ditch 25271, Group 969, Area R, Period 4-6
17. Rim sherd, flaring, rounded. Black-brown. Fabric 2a. Outer part-burnished. Fill 12059, Ditch 25271, Group 969, Area R, Period 4-6
18. Four body sherds, possibly from the same vessel. Fabric 2a. Outer dark reddish-brown. Inner and core black. Cleaning layer 14516, Area L
19. Tall bowl with slightly inturned, flattened rim. Three non-joining sherds from the same vessel. Outer is part-burnished around the rim and in a band c. 22mm further down, and part-roughened using the schlickung technique on lower body, underneath the band of burnish. Surfaces black-brown, core dark reddish-brown. Fabric 5c. Base missing. Fill 14528, Pit 14529, Group 722, Area L, Period 6
20. Body sherd (upper) from a large decorated jar. Outer burnished, inner smoothed. Dark brown to black throughout. Fabric 3a. Decoration comprises two incised concentric lines beneath which are the beginnings of pairs of opposed diagonally incised lines. Fill 14528, Pit 14529, Group 722, Area L, Period 6
21. Body sherd (upper) from a decorated jar. The neck is hollow and there would probably have been an everted, rounded rim. Outer burnished, inner smoothed. Surfaces black, core black-brown. Fabric 2a. Decoration comprises an incised concentric line, beneath which is an incised diagonal line. Fill 14528, Pit 14529, Group 722, Area L, Period 6
22. Jar with everted, rounded rim (cooking pot?). Rim and upper body sherd only. Surfaces smoothed, some burnish on inside of rim. Dark brown-black throughout. Fabric 2a. Carbonised ?food residue in neck hollow. Fill 14528, Pit 14529, Group 722, Area L, Period 6
23. Body sherd. Inner and core black, outer buff-brown. Fabric 2a. Decoration on outer surface comprising vertically combed striations covering all of the surface. Fill 14528, Pit 14529, Group 722, Area L, Period 6
24. Bowl rim, inturned, rounded. Outer burnished, inner smoothed. Black-brown throughout. Fabric 2a. Small areas of carbonised food residue on outer surface. Fill 14528, Pit 14529, Group 722, Area L, Period 6
25. Jar rim, everted, rounded. Surfaces burnished, partly abraded away. Black-brown throughout. Fabric 2a. Fill 14528, Pit 14529, Group 722, Area L, Period 6
26. Jar rim, slightly flaring, flattened. Black. Inner and top of rim burnished, outer has traces of burnish. Fabric 2a. Fill 14528, Pit 14529, Group 722, Area L, Period 6
27. Body sherd from a decorated jar. Black-brown throughout. Surfaces smoothed. Fabric 2a. Decorated with two incised concentric lines, infilled with circular stamps (Briscoe's type A1, single circles). Fill 14528, Pit 14529, Group 722, Area L, Period 6
28. Jar rim, everted, rounded, flattened on inner edge. Dark reddish-brown. Inner burnished. Fabric 2a. Fill 14528, Pit 14529, Group 722, Area L, Period 6
29. Rim sherd, slightly everted, angular. Fabric 2a. Surfaces dark reddish-brown. Core grey. Fill 14528, Pit 14529, Group 722, Area L, Period 6
30. Globular jar with slightly everted, rounded rim. Dark grey throughout. Patches of burnish on outer surface, particularly on and immediately under rim. Beneath the shoulder the outer surface is much rougher with slight finger-tip impressions. Fabric 2a. Maximum diameter at rim, 176mm. Fill 14558, Pit 14529, Group 722, Area L, Period 6
31. Rim sherd, everted, rounded, from a globular jar with hollow neck. Outer black-brown, part burnished, inner reddish-brown, smoothed. Core grey with reddish-brown margins. Fabric 2a. Fill 14613, Pit 14529, Group 722, Area L, Period 6
32. Bowl with slightly inturned, rounded rim, three joining sherds. Surfaces vesiculated. Outer burnished, abraded. Inner smoothed. Surfaces black. Core dark reddish-brown. Fabric 5a. Fill 14613, Pit 14529, Group 722, Area L, Period 6
33. Rim sherd, everted, rounded. Surfaces orange-buff. Core dark grey. Inner part-burnished. Fabric 2a. Fill 14613, Pit 14529, Group 722, Area L, Period 6
34. Rim/upper body sherd, everted, rounded, from a large jar. Probably a cooking pot. Fabric 2a. Surfaces reddish-brown, smoothed. Core grey. Fill 14613, Pit 14529, Group 722, Area L, Period 6
35. Body sherd. Black ware. Fabric 2a. Outer burnished. Decorated with incised concentric lines underneath which are vertical incised lines. Fill 14613, Pit 14529, Group 722, Area L, Period 6
36. Rim, flaring, rounded. From a medium-sized jar, probably biconical. Black Fabric 2a. Fill 14613, Pit 14529, Group 722, Area L, Period 6
37. Rim, everted, angular. From a large globular jar or bowl. Black Fabric 2a. Fill 14613, Pit 14529, Group 722, Area L, Period 6
38. Globular jar, three joining sherds. Outer surface top half is part-burnished black, bottom half is reddish-buff with combed vertical striations. Inner and core dark reddish-grey. Inner smoothed around rim. Fabric 2a. Maximum diameter at rim, 176mm. Fill 15353, Pit 15354, Group 701, Area M, Period 6
39. Small carinated bowl with everted, rounded rim. Complete. Single perforated lug above maximum girth. Fabric 2a. Dark reddish-brown ware, inner blackened, outer some blackening around base and lower body. Maximum rim diameter, 83mm. Fill 15694, Gully 15688, Group 472, Area M, Period 6
40. Carinated bowl with facets and impressed dots, eighteen joining sherds. The facets are at the carination and are depressed so that a small boss is formed in between them. Each boss has a centrally impressed dot. Above are seven incised concentric lines. Rim is everted and rounded. Base is flat. Black-brown surfaces have patches of abraded burnish. Core reddish-brown. Fabric 2a. Maximum diameter at rim, 118mm. Fill 15862, Pit 24456, Group 701, Area M, Period 6
41. Carinated bowl, everted, rounded rim. Thin-walled vessel. Areas of burnish on brown-black surfaces. Inner and core reddish-brown to black. Fabric 2a. Maximum diameter at rim, 90mm. Fill 15862, Pit 24456, Group 701, Area M, Period 6
42. Body sherd. Decorated on outer surface with two deep concentric finger grooves, the ridges formed are emphasised by concentric incised lines on either side, underneath are beginnings of incised stehende bogen. Reddish-brown surfaces. Core black with reddish-brown margins. Fabric 7a. Fill 15862, Pit 24456, Group 701, Area M, Period 6
43. Bowl rim, upright, rounded. Surfaces reddish-brown. Core dark grey. Fabric 4b. Fill 15862, Pit 24456, Group 701, Area M, Period 6
44. Neck sherds, four joining, from a jar. Outer burnished black over reddish-brown core and inner. Fabric 4b. Decorated with three finger-depressions delineated with incised lines. Fill 15862, Pit 24456, Group 701, Area M, Period 6
45. Squat globular jar with slightly everted, rounded rim and combing on outer. ?Cooking pot. Outer surface is patchy reddish-orange to dark brown with some blackening. Three shallow grooves under neck hollow, intersecting combed lines on lower half of vessel. Inner patchy reddish-orange to black with carbonised food residue. Fabric 2a. Maximum diameter at rim, 127mm. Fill 15969, Pit 15968, Group 141, Area M, Period 2B
46. Body sherd from a globular jar. Outer patchy reddish-brown to black. Inner and core black. Surfaces abraded. Outer decorated with shallowly incised lines comprising two concentric lines with at least one oval shape beneath. Fabric 3a. Fill 16073, Pit 16088, Group 559, Area H, Period 4
47. Rim sherd, rounded and flattened on top, from a hollow-necked jar. Surfaces patchy reddish-brown to black. Core dark grey. Fabric 2a. Fill 20020, Pit 20019, Group 711, Area L, Period 4
The assemblage contains several elements that are typical of a late 5th century group; the decoration is simple, typically incised concentric necklines with further incised schemes below (nos 20, 21, 35, 42). Where stamped decoration does occur it is of the simplest design, Briscoe's A1 or single circle motif (no. 27).
The most common forms include hollow-necked globular and biconical jars (nos 10, 22, 29, 31) and globular bowls with upright or slightly everted rims (nos 19, 24, 30). Two examples of straight-sided bowls (nos 32 and 43) are atypical of the group. A single faceted, carinated bowl (no. 40) can be dated to the late 5th century. All the previously mentioned pottery forms are commonly found on Early Saxon settlement sites of 5th to 6th century date. However, at Elms Farm three vessels stand out as unusual. Two vessels are wheel-thrown with deeply corrugated surfaces; the squat globular jar from cleaning layer 9720 (no. 8) and a body sherd from well fill 12246. The squat globular jar could be a Frankish import as it resembles examples of Frankish jars found in Essex at the 6th-century cemetery site at Prittlewell, Southend (Tyler 1988, 91-116), and the body sherd could be from a similar vessel. The third oddity is a small jar with rolled-over rim from ditch fill 12059 (no. 15). It does not resemble Frankish forms and is the best candidate for a Saxon imitation of a Roman jar, perhaps a hybrid 'Romano-Saxon' form.
Surface decoration is consistent with the late 5th century date range. Deliberate roughening of the surface by the application of a slip containing large quartzite particles - schlickung - is common (no. 19). Burnishing (both inner and outer surfaces) is also common. Finger-rustication does occur, but only in small quantities, whereas combing is comparatively common (nos 10, 23, 38, 45). The dating implications of the surface treatments listed above have been analysed by Hamerow (1993) in her discussion of the dating of the Early Saxon pottery from excavations at Mucking, Thurrock. The most obvious dating indicator from the distribution of the various surface treatments is the pattern of occurrence of the sherds with schlickung, which were found in the earliest (5th to early 6th century) areas of settlement.
This is a key group in understanding the Late Roman/Early Saxon transition. The group has characteristics typical of a late 5th century assemblage; the decoration is simple, typically incised concentric necklines with further incised schemes below (nos 20, 21, 35). Forms include hollow-necked globular and biconical jars (nos 20-2, 25, 31, 33-4) and globular bowls (no. 19). Surface decoration includes deliberate roughening of the surface by the application of a slip containing large quartzite particles (schlickung), burnishing (both inner and outer surfaces) and finger-rustication (the latter only in small quantities). All these characteristics are typical of a late 5th century assemblage.
The pottery from the wood-lined ditch is, with one notable exception, typical of a 5th century assemblage, in the main comprising undecorated hollow-necked jars, mostly in Fabric 2a (tempered with quartz-sand and including some mica). However, the exceptional vessel is a small jar (no. 15) that appears to be a Saxon copy of a Roman form; the rim appears to be rolled over and is surely wheel-thrown with a minimal neck hollow. Although the vessel is far from complete, three sherds survive, one of which (a lower body sherd) does not join with the other two. Enough of the vessel survives to reconstruct its profile; a small jar-shaped ?drinking vessel with a sagging base. This form is unusual in the Early Saxon pottery repertoire and is, I believe, a very strong candidate for a hybrid vessel.
The pottery from this feature (total weight 161g) is utilitarian in form, with rather crudely made globular jars and bowls in the sand-tempered Fabric 2a, which is the predominant Early Saxon fabric from the site. The large high-shouldered jar (no. 4) is most likely a cooking pot. There is nothing exceptional in this group: it is a typical assemblage for a 5th to 6th century sunken-featured building. The range of pottery forms can be paralleled on many Early Saxon settlement sites, and many can be found in the Mucking volume (Hamerow 1993).
Fill 15694 of a short, curving and enigmatic Saxon ditch contained a complete carinated bowl with a perforated lug (no. 39), but no other pottery in this fill, Saxon or otherwise. This vessel appeared to have been set in an upright position on the floor of the gully before infilling, which suggests deliberate or structured deposition.
The Saxon re-cut, 14529, of Roman well 14984 contained by far the largest assemblage of Early Saxon pottery. The stratigraphic relationship between the Early Saxon material in this feature and the latest Roman material is discussed in the section on Roman pottery and need not be repeated here. However, it should be stressed that the comparatively large amount of Early Saxon pottery shows this feature to have been in use during the late 5th and most probably through into the early 6th century, indicating a break between Roman and Saxon occupations. The practice of Early Saxon reuse of late Roman features, in particular deeply cut features such as wells, has been noted on other settlement sites with both Late Roman and Early Saxon occupation, but in most cases it is difficult to prove anything other than reuse following a break.
On several excavated sites in the county well-groups account for the bulk of the recovered pottery of Early Saxon date; these wells can be of Saxon construction or are Saxon re-cuts of Roman features. For example at Slough House Farm (Wallis 1998, 57; Tyler 1998, 157-8) nearly all of the 2.4kg of Saxon pottery recovered came from the fills of two wood-lined wells of Early Saxon date. Excavations adjacent to Elms Farm at Langford Road in 1994 recovered some 1.38kg of Early Saxon pottery, most (1.2kg) from the final infilling of a Roman well (Langton and Holbrook 1997, 29). This is close enough to Elms Farm to be considered as part of the same settlement; the excavators at Langford Road found no evidence of Saxon structures, indicating that the main settlement focus was probably towards the east, at Crescent Road (Drury and Wickenden 1982). In this connection, it is worth noting that two clusters of features containing Saxon pottery, centred on Building 69 in Area W and Buildings 67 and 68 in Area R, are in relatively close proximity to the sunken-featured buildings uncovered at Crescent Road and to further Saxon remains identified immediately adjacent to this in 2002 (Roy 2003).
The date of the Elms Farm Early Saxon pottery assemblage supports the argument for a site-wide break in occupation during the first half of the 5th century, contra Drury and Wickenden (1982, 31). The research archive contains a study on sherd brokenness and the association of pottery types within a range of deposits. The average sherd weights, percentage EVE and percentage weights for selected Late Roman and Saxon assemblages were tabulated for comparative purposes. The exercise was repeated for other site assemblages, for which investigations of a similar nature have been attempted. The study has produced interesting results that would appear to confirm the proposed break in occupation at Elms Farm, and has gone some way towards establishing pottery deposition patterns during the late 4th and 5th centuries.
Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.