Page 55 - John Barber's Oakham Castle and its archaeology

Fig. 4. Section (C–D) of rampart, 1954. 1 – Topsoil, and post-medieval disturbance of wall.
2 – Make-up of rampart (yellow clay and marl). 3 – Old ground surface (brown soil).

post-medieval wares. The purpose of this ditch was obscure, for it did not join the moat, and only some 5ft of it was
removed during building operations.

Work in 1954
Three tasks were therefore set for the second season’s controlled excavation. Firstly, a detailed investigation of the
rampart was required; and secondly, the ground surface beneath demanded searching for traces of earlier occupation.
Finally, the opportunity presented itself for a study of the adjoining gateway, to see if the remains of any gatehouse
could be found. The major part of the excavation involved the cutting of a trench C-D across the rampart, in the north-
east corner of the site where no major building operations had been allowed to take place (Pl. III). This extended
north of the castle wall, and was 22ft long, 8ft 6in at maximum depth, and 3ft wide. The make-up was found to
consist of layers of marl and clay (2), lying alternately; and rising from north to south, which suggests that the
rampart was originally rather higher at this point – Fig. 3 (plan) and Fig. 4 (section). The pottery found here consisted
of Stamford and St Neots wares, and four sherds of early medieval ware. Beneath the rampart was a clean layer of
loam (3), the old ground surface, which contained some sherds and bones, but no further trace of occupation.
The stratigraphy of the rampart observed elsewhere on the site in 1953 was therefore confirmed. Moreover, the
pottery found has considerable bearing on the argument of Mr Ralegh Radford quoted above.
All the Stamford ware from the rampart and beneath, which must be of one date, lacks the rich green glaze which
Mr Hurst now considers a ‘developed’ feature, found only from c. 1125. The question is, how much earlier can one
date the ‘undeveloped’ ware? As is well known, stratified deposits at Thetford carry it back to the late ninth to early
tenth century, or possibly slightly earlier, and there it lasts until the late eleventh century at least. At other East
Anglian sites it has been found in late Saxon contexts, but its main centre is the East Midlands, where, as Mr Dunning
has shown, it was probably made, but, on the other hand, has never yet been found in a pre-Conquest context.
As shown below (p 54), the Stamford ware from the rampart compares closely with that from Alstoe Mount, dated
10
by Mr Dunning to the early twelfth century. There were, also, four sherds of medieval ware, to support this, though
these need have little significance. A date nearer to 1150 than 1100 for the construction of the rampart might therefore
be suggested. On the other hand, Stamford ware could, a priori, be earlier at both Oakham and Alstoe Mount, while at
11
other sites associations of Saxo-Norman and medieval ware as early as the eleventh century are known. Finally, as
argued below, some features of the St Neots ware suggest that here too an eleventh rather than twelfth-century date is
to be preferred. This compares favourably with the date of the bailey (1075-1100) suggested independently by Mr
Ralegh Radford, and as such provides a dated group of Saxo-Norman wares for the East Midland area.
Considerable clearing took place at each end of section C-D (Fig. 3). To the south, a rectangle 14ft by 12ft was
excavated, on a house site which had disturbed the stratigraphy of the rampart. A number of pits and depressions were
found here containing post-medieval wares, one of which (1c) was useful for its association of pottery and clay pipes
(pp 54, 57, 59, 60). To the north, within the Castle grounds, stripping took place in an area 21ft 6in by 17ft. On the
rampart was a remnant of curtain wall built up with brick. As it approached the gateway from the west, and 9ft from
it, the wall turned sharply before joining the gateway itself.

53