|
the lack of investigation conducted into the economics of ceramic
production. This aspect of the study owes much to that author's more
recent work on the ethnography of pottery production (Peacock 1982). On the
whole, coarse pottery does not reflect differential site status (Pollard
1983a, 444-61), except that towns and military bases are distinctive in
the variety of wares of minor quantitative importance that tend to be
included in assemblages. The cases of the selective marketing of BB2 in the
early second century to the more prosperous sites, and of pre-Flavian
Canterbury sandy wheel-thrown ware to the city and military base(s), are
rare exceptions to this rule. Others may be obscured by the inadequacies of
chronological refinement. Higher value wares may, in their proportions
relative to coarse wares, provide information on site prosperity;
occasionally, a limited distribution, such as those of pre-Flavian
colour-coated wares and third-century Moselkeramik, may reflect
selective marketing to (or purchase by) wealthier individuals or
communities.
It is the present author's belief that the regional study of
the whole network of pottery production, importation and distribution is a
valid approach in archaeology. How then might future studies be conducted?
It is plain that the quantification of assemblages is of vital importance to
the furtherance of such studies. The initiative must come from
post-excavation teams, for much material that lies in museums is patently
worthless so far as quantification is concerned. Hodder's '30 sherd
threshold' was, it would seem, adopted in the absence of a more
appropriate measure to test his theories against available evidence. Hodder's
(1974b) own reservations on the assemblages with which he was dealing, and
the low value accorded to sherd-count statistics by Orton (1975; 1980, 156-67)
imply that this technique is a poor substitute for the full, or
statistically sampled, quantification of recovered assemblages prior to
disposal. The application of well-funded sampling strategies to
post-excavation sorting should
|
|
also pay dividends in time saved and in producing comparable data. A third
improvement will be aided by sampling; it is important that a greater number
of assemblages be examined from each site if inter-site comparison is to be
made, in order to level out possible biases due to
assemblage-differentiation according to place-function within each site.
A regional study must be followed up by the asking of specific
questions: D.F. Williams' work on petrological characterisation of BB1 and
BB2 (1977) and Monaghan's study of 'Upchurch' wares (1982) are
excellent examples of the kind of issue that can be pursued. Mainman's
(forthcoming) petrological work on late Roman and Saxon coarse wares in
Canterbury provides a third case-study, and serves to highlight the need for
fabric and technological analysis as well as form-decoration studies, if the
thorny problem of the transition from 'Romano-British' to 'Saxon'
pottery is ever to be solved (Chapter 4.VI). The fabric groups defined on
visual examination by the present author also need to be examined in more
detail; in particular, it would be of interest to learn whether there are
any consistent differences in inclusion-composition between Canterbury,
Swale, Cliffe and Mucking grey wares that might confirm the visual isolation
of a 'Canterbury' group from the remainder. Other aspects of the present
study that should be worth following up are the cross-Channel trade in
coarse pottery (cf. Fulford 1977a and Chapter 4 above); the movement of Nene
Valley wares southwards along the Anglian coast to east Kent and its
relationship to the development of the Saxon Shore system (cf. Young 1977b;
1980), and the general trade along the Channel coast of Britain (cf. e.g.
Williams 1971; Cleere 1974; Fulford 1975a, 1975b; Williams 1977). The latter
topic would shed some light also on the mutual isolation of Sussex and Kent
that is apparent (Pollard 1983a, 251-306). The two regions have much in
common - physical geography, the early development of villas, and the
|