deposits (ibid., 62), superimposed on the Eocene strata of north
Kent, particularly in the Great Stour valley and flood-plain, along the
Swale catchment area, and in the Upper Medway (in the Weald). Mainman
(forthcoming) has demonstrated that the Head Brickearths and London Clay
in the vicinity of Canterbury were both suitable materials for potting,
producing sandy wares with, in the former case, chalky inclusions. The
Brickearths of Kent have been exploited exhaustively for the manufacture
of bricks, a further indication of their merit.
Sandy wares were of little or no importance in the century
before the Claudian conquest. They comprise less than 2 per cent (by
vessel rim equivalents) of the pottery from a late first century
B.C./early first century A.D. deposit at Rochester, to which a further 13
per cent of grog-sand and flint-grog-sand wares may be added. The earliest
levels at Canterbury, possibly of the mid-late first century B.C., contain
virtually no purely sandy wares, but flint-sand and flint-grog-sand
fabrics are present. Grog-tempered ware undoubtedly prevailed in the first
half of the first century A.D. however, comprising 44 out of 54 sherds of
one early layer on the Marlowe Car Park site III (Pollard forthcoming, d),
and an even higher proportion of the groups from the lower fills of three
out of four of the known late Iron Age ditches in south-east Canterbury
(Frere 1954, Cellar L; Pollard forthcoming, d, layers MI 1260/1280 and MIV
732B). The evidence from Faversham (Philp 1968, 76-81) and West Wickham
North Pole Lane (Philp 1973, 70-2) supports the hypothesis that sandy
wares without a large grog or flint admixture were not produced in the
earlier first century A.D.
The east Kent evidence for pre-Conquest sand-tempered wares
has been reviewed here in order to provide a general background to discussion
of the problem of their emergence in the west. It may be inferred from
occurrences of 'Gallo-Belgic-derived' platters and cups in sandy
wheel-finished (if not wheel-thrown) wares throughout west Kent (see Fig.
17) that these wares were being produced by the early
|
|
A. D. 70s at the latest, as the popularity of Gallo-Belgic forms waned in
that decade. Pre-Flavian to Flavian examples occur in pre-Flavian/Flavian
contexts at Cobham Park (Tester 1961, no. 12), Rochester (unpublished),
Fawkham (Philp 1963, no. 15) and Farningham Calfstock Lane (Philp
1973, nos. 304-5), and in a Flavian deposit at Southwark (Sheldon 1974,
no. 15). The pottery from pre-Flavian Southwark includes several
other forms in reduced sandy wares, including bead-rim jars (Tyers and
Marsh 1978,IIA 5-6), carinated and sub-carinated shoulder necked jars
with bead or short-everted rim (ibid., IIC 2, lID, IIN-Q; q.v. nos.
93-4, 97 here), butt-beakers (Tyers and Marsh 1978 IIIA), and 'Surrey
bowls' (IVK: no. 99 here). The dating of these forms at Southwark is
vital to the interpretation of material from West Wickham Fox Hill (Philp
1973, nos. 159, 161-3, 175, 177-80), and Chariton (Elliston Erwood
1916, fig. 21 nos. 3 and 16, fig. 22 nos. 66-7). The internal dating
evidence for these two sites is poor; close typological links with the
Southwark material allow the suggestion to be made that sandy ware necked
jars, at least, were circulating in north-west Kent within a
quarter-century of the Conquest. The cordoned, necked jar or bowl is a
widespread first-century form in Kent (cf. e.g. Fawkham: Philp 1963, nos.
13-14; Rochester: Harrison 1972, nos. 1, 5 and 11), and lasted well into
the second century. Unfortunately, examples in the area between the Cray
and the Medway estuary cannot be dated firmly to the pre-Flavian period,
and it remains a possibility that, apart from 'fine' forms of
Gallo-Belgic derivation, sandy wares were not used in this area until
mid-Flavian times. This applies to bead-rim jars as much as to the necked
forms. The 'Surrey bowl' has not been recorded in Kent apart from a
single example in the east, at Canterbury (Frere 1970, fig. 10 no. 1, in a
later second-century context).
The evidence for pre-Flavian mortaria in west Kent is
extremely limited. Claudian wall-sided
|