EXCAVATIONS AT
Based on notes written by Jeff Wallis, Roger Ainslie
and others. Compiled and edited in
2005 by Bob Eeles
The Abingdon Area Archaeological and Historical Society were
invited to complete a Trench (Trench 1) begun by the house owner in an area on
the eastern side of the rear wing. This trench was enlarged and excavated to
undisturbed natural gravel. The first c. 0.9 m. of mixed soils contained
recent backfill. Undisturbed archaeology was reached at a depth of between 0.9
m. and 1.2 m. A corner of what is thought to be a 16th century well
construction pit and remnants of a collapsed well lining occupied the southwest
corner of Trench 1. Much of the deposit above had slumped down into this
feature and because of its instability was not fully excavated. A segment of a
large Medieval circular pit occupied the eastern part
of Trench 1. It cut a Roman gully and a circular Middle Iron Age pit. The latter contained sherds of 9 bead rim vessels (or perhaps 9
sherds of a bead rim vessel) and fragments of daub. An early Roman
V-shaped gully with a squared slot in its base was cut by the Medieval pit and cut the Middle Iron Age pit. At 90o to the V-shaped gully, and cut in
part by the well construction pit, was a U-shaped gully, probably also of early
Roman date. In Trench 2 a spread of coal and compacted earth indicated the use
of the garden as a 19th century coal yard. Undisturbed levels were
exposed below this. A line of limestone rubble surrounded by brown stony earth
ran north-south along the trench. The nature of this stonework did not suggest
that it came from structural features. This part of the garden was undisturbed
archaeologically and so the trench at this depth was sealed with plastic
sheeting and backfilled. Another AAAHS excavation occurred between 1985 and
1986. Unfortunately from the information that is available it is not possible
to produce a more detailed write up than the one given here.
Thanks are
due to Abingdon Area Archaeological and Historical Society (AAAHS) members such
as Roger Ainslie, John Carter, Rachel Everett, Alison Gledhill, Kerry Johnson,
Godfrey Jones, Jackie Smith, Roger Thomas and Jeff Wallis (there were many more).
Excavations would not have been possible but for the late Miss Dorothea
Pickering who provided shelter, sustenance and facilities for excavation and
post-excavation work.
Much of
the following information was written in a variety of unpublished documents
shortly after the original 1980 and 1981 excavations. The 1980 excavation
report and (I believe) the finds descriptions were written by Jeff Wallis in a
style, and using pottery descriptions, similar to that used at the time by the journal
Oxoniensia. Roger Ainslie wrote the notes and pottery descriptions for
the 1981 excavation in a style more suited for inclusion in CBA9
Newsletter/South Midland Archaeology. For various reasons (see below) it is
now no longer possible to write this work up properly and submit it for
publication to a refereed academic journal. Instead the various notes have been
compiled, revised/updated to some extent, references have been added, and the
whole edited by Bob Eeles (hereafter, Ed.) who was not present at the time of
the excavations and who has had to rely on the old archive and slide
collections (all the latter being under-exposed and out-of-focus – I have
scanned copies of these). The Ed currently holds much of the existing archive
material so far as he is aware.
This note
provides additional information for the site, which has been excavated a number
of times before (e.g. by John Maddicott in 1966, 1967, 1968 and George Lambrick
in 1968). A hand-written note from 1968 (author unknown) lists residual Romano-British,
Medieval (“very small amounts”) and 17th-19th Century
pottery from a trench dug across the driveway. Unfortunately and following the
death of Miss D. Pickering all excavated finds, once housed at
30 East St
Helen Street (St Ethelwold’s House) is situated on first terrace gravels (OD.
51 m.), in the heart of Medieval Abingdon, in what is now recognised as also
having been the centre of a medium-sized Roman town, whose origins date back to
the Early Iron Age[2].
The site lies on the northern bank of the
Architectural
surveys in the 1960's identified (above-ground) house features dating between
the 14th and 19th centuries[6].
Exploratory trenches excavated to the rear, side and cellar of the house[7]
produced evidence of a Roman building (daub and a roof tile fragment) and early
Roman ceramics in the vicinity. A Roman roof tile fragment and early Roman
ceramics were also found c. 100 m. away at
A trial trench situated 45 m from the river edge, begun by Miss D. Pickering, measured 2.5 m x 1.5 m and exposed an orange clay loam 1.3 m below the modern ground surface. This cut through rubble and sand mixed with garden soils and then grey-brown soil containing a range of Medieval and Roman sherds together with roof tiles and Stonesfield slates. This trench was widened by the AAAHS due to the depth of archaeological deposits and for safety reasons. The top 0.9 m was very loose and crumbling. A trench 3.3 m x 2.4 m was laid out with the original trench in its centre. The northern edge abutted a wooden conservatory and the western edge a corridor of stone and timber construction, which was a modern addition to the east wall of the rear wing. The enlarged trench was within 1.2 m of standing structures so could not be excavated further north or west. Enlargement in an easterly direction was avoided because previous archaeological investigations had removed deposits to within 1 m. of the trench.
Layer
2, 30 cm in
depth, proved to be recent and consisted of disturbed garden soils,
grey-black earth, lenses of sand and fragments of roof tile. Finds included
Roman, 18th and 19th century ceramics. Two water pipe
trenches had been cut into Layer 2. A lead main supply running
north-south, parallel to the house whose construction trench penetrated to a
depth of 6.5 cm and at a slightly higher level an iron
pipe, crossing the trench east-west taking a water supply to an adjacent
outhouse, whose construction trench had been sunk to 1.3 m. Both had been
backfilled with rubble clinker and gravel mixed earth. Finds included fragments
of Medieval glazed tiles. These were noted in previous
excavations at number 30, and next-door at number 32, and may have been removed
from Abingdon Abbey after the dissolution (Ed.). The backfill material was said
to have come from a previous excavation in the house cellar floor. At the
southwest corner of Trench 1 the water pipe trench cut into a
rectangular area consisting of a layer of gravel with tightly compacted rubble.
This feature was first recognised at 2.5 m and was seen to cut Layer 2.
The function of the compacted rubble is thought to have been a soak away as
indicated by broken ceramic drainpipe. It occupied the top part of an earlier
well, the stone rubble probably being the remains of a lining or packing
material between the construction pit and lining.
The
remaining deposit, occupying the eastern quarter of the trench, consisted of
loose earth with gravel and occasional rubble. It appeared to occupy part of a
large circular feature or the edge of backfilled archaeological trench that was
sited within the confines and east of this excavation. Finds were Medieval and
Roman. Brick fragments and cardboard packaging at a depth of 2.3 m confirmed
that this constituted previous backfilling material.
After
removal of modern disturbance several features were seen. The edge of Pit 15
was observed at the eastern edge occupying the entire section, with its curving
form jutting 75 cm into the excavation. This feature cut Layer 6,
a grey-brown earth containing limestone, charcoal and Medieval
and Roman sherds. Also cutting Layer 6 was the corner of a square
well construction Trench (12) seen at a depth of 2.5 m in the
southwest part of the excavation. The well lining was in a very disturbed
condition, its true edge unrecognisable and its courses disarranged probably
after stone robbing during construction of the soakaway. The feature was
excavated for 40 cm in the hope of obtaining datable artefacts but because of
its instability excavation was taken no further. Dating is, therefore, not firm
but it is likely to be Post-Medieval. Tripod pitcher sherds and bones were
recovered from the loose mortar flecked fill which became very loose, with many
voids, indicating the presence of a well shaft whose dimensions were impossible
to tell. Layer 6 was cut by modern disturbance in its
northeastern sector that continued to cut down through this layer as if a
continuation of the water pipe trench. This layer was perhaps a Medieval
occupation soil some 15 cms deep that originally covered the whole of the
trench at a depth of 1.7 m below site datum (Site Datum = 59.11 m).
Pit
15 had an estimated diameter of (Unreadable measurement given - Ed.) and
cut 1.64 m into natural gravel with a total depth of 5.12 m below Site Datum.
This feature had been partially excavated, undisturbed pit fill being
encountered at 2.8 m from the surface. Descriptions of its layers directly
under the deepest modern penetration were as follows:-
15/1 A coarse brown loose textured earth with
10% gravel (98 cm max. depth) seen on the eastern side of the feature. Previous
investigations appeared to have removed part of it from the western end, thus
exposing Layer 15/2.
15/2 A green-stained earth with gravel and
charcoal flecks slumping down each side of the thin central area, which was
1.06 m deep. Limestone fragments were common and randomly
distributed within this layer that overlay 15/3 and 15/4.
15/3 This dipped east under 15/2. The
only difference being a grey colour, 36 cm deep.
15/4 This became deeper towards its western
end. It comprised a pale-green stained earth with gravel and charcoal and a
scatter of limestone.
15/5 Alternating bands of fine powdery ash,
grey, brown and green in colour. The layer
(max. depth 20 cms) dipped towards the west.
15/6 Was 5.2 cms deep and consisted of fine
brown earth directly beneath the ash. The only feature within it was a tip line
of sand that predated the ash deposit and penetrated the layer below.
15/7 Was 15 cms deep showing no evidence of
piling up or dumping. The layer make up was similar to the post-glacial orange
loam (8) but contained the ‘parent’ gravel. The bottom of the pit was
generally flat but uneven with a vertical edge. Its general impression
suggested a large rubbish pit. It contained large quantities of ash.
After
removal of Layer 6 an area of dark brown loam was seen to spread
over the undisturbed surface of the trench. The soils were divided into two
layers by a slight textural change, which blended without perceptible edges.
Layer
7 (85 cm deep) consisted of a grey-black charcoal flecked stony loam
that was seen in the southern part of the trench and abutted the south section.
Finds from Layers 6 and 7 indicate a Post-Roman occupation
layer.
Post Hole
(10/1) and its recut (10) were circular, c. 20 cm in
diameter, with the first cut being smaller. Feature 10 was 9 cm
deep with its semi-circular base 1.31 m below Site Datum. The original hole was
of similar depth. Both features cut through Layer 9 (46 cm deep)
and into the top of the red-orange loam (8). 10 contained the
remnant of post packing in the form of six ovoid limestone fragments c.
4 cm across, the rest being filled with a loose black earth with 5% gravel. The
position of the packing and its relationship to the edge of the feature
indicated an ovoid or circular section post.
The
removal of Layers 7 and 9 exposed first a dirty orange
then a clean orange-brown fine sandy loam (8) which has been identified
at other sites in and around Abingdon. The loam (8) was cut by the well,
and associated construction Pit (12), and by Pit 15.
The earlier trial trench located part of Layer 8 whose surface
was horizontal at 1.34 m below Site Datum. Two small animal burrows (labelled
as ‘mole’ on a section drawing, Ed) were found running west from the southern
section.
Feature
13 consisted of a gully running west-east from the western trench
section. It was filled with sandy-grey silt with charcoal flecks blending with Layer
7 above. U-shaped in profile it was 45 cm wide x c. 20 cm at its
south-western end. The base was 1.33 m below Site Datum. Feature 13
was positioned at 90o to a similar but larger Gully (14).
The relationship between these was unclear because a corner of the well
construction Pit (12) cut both features at the point of
intersection. However, in one section it appeared they were contemporary and
converged, both merging into a single gully running south-east and seen in the
southern section as Feature 14.
Feature
14 cut diagonally across Trench 1 and was traceable for
2.34 m where it was seen in the northern trench section. Part of its fill had
been removed during construction of Pit 15. It contained a similar sandy-grey silt to 13 but had a slightly
more clay-like consistency without charcoal flecks. It was on average 70 cms
wide x 58 cms deep. It contained a small squared slot (14 cm wide x 5 cm deep)
with a similar fill to the gully. It was traceable for 98 cms at the bottom of 14
but not observed along its entire length. The bottom of the northern end was
1.69 m below Site Datum. and at its intersection with 13,
1.74 m below Site Datum.. It also penetrated Layer 8 and cut into
gravel. A green stain within it may have represented traces of a copper or
bronze object. At the southern section the top part of this feature, at the
point where it merged and changed direction, was truncated by the base of
modern disturbance and its western lip was cut by the well construction Pit
(12). 13 and 14 were certainly Pre-Medieval, post 2nd
century BC and probably early Roman. The slot in 14 would indicate Roman
work.
Feature
(Pit) 16 was a segment of an Iron Age circular pit originally 1.2
m in diameter. It had been cut by Pit 15 on its northeast side
and Gully 14 on its west leaving c. one third intact. A
half section across it was seen in the side of Pit 15. Pit
16 was 0.75 m below the top of Layer 8 where it was
observed in plan. Its sides flared slightly leading down into a slightly hollow
base. The fill was a grey-brown silty loam containing limestone fragments, 5%
gravel and patches of orange loam derived from Layer 8. Twenty
cms from its top was a wedge shaped tip of black soil and charcoal with a
maximum depth of 6 cms. This feature contained the sherds of nine bead rim
bowls dating from the 2nd - 3rd Century BC. Other finds
were two burned quartzite pebbles, three pieces of burned limestone and two
burned bones. A tooth and three bone fragments were unburned. Four pieces of
daub indicate a possible structure in the vicinity. The pit (16) cut 37
cm into natural gravel.
Because of
the restricted nature of the excavation only a limited interpretation is
possible. The earliest feature was Pit 16 dating to c. 200
BC with contents implying domestic activity nearby. It has many local
parallels, e.g. Thrupp[11]
and nearby at
Gully 14
and Gully 15 are thought to be contemporary; 14 clearly
seen to cut the Iron Age Pit 16. Gully 14 with its
distinctive slot was probably Roman. Slotted gullies with dateable finds have
occurred before in Abingdon.
Identification
of Iron Age and Roman features provides additional evidence of occupation
within the limits of the modern town. Evidence of Roman architecture has been
recorded previously at
Layer 7
and Layer 9 the grey-black (‘Roman Black’) loams may be seen as the
latest Roman or a post-Roman 'occupation soil' that has been observed on many
urban sites in this country.
The only
other datable feature was the Medieval circular Pit
15. Fragments of iron slag and distinct bands of ash near its bottom
indicate it was used for disposal of metal working waste and latterly for
rubbish disposal.
The
rectangular nature of the edge of the presumed well and associated construction
pit, and the loose limestone rubble and voids between them suggest this
function originally and its later re-use as a convenient soak-away. It remains
undated but is likely to be c. 16th century in date.
Fragments of glass from this area and 18th century ceramics from
above would support this conclusion. Deposits above the well reflect more
recent history. The many sand lenses were perhaps remnants of a child's
recreation area when St. Ethelwold's House was a school between
1964-66.
Future
work should examine (and, in fact, did, but inconclusively as the river bank
appears to have migrated many metres southwards after many centuries of rubbish
dumping, Jeff Wallis, Pers. Comm. to Ed.) areas near the waterfront.
Waterlogged timbers of early wharves may be sought and the character of early
riverside stratigraphy could be examined without the pressures of rescue
archaeology.
The finds
descriptions for the Medieval pottery (Table 1) make
use of a system of identification that is not now in vogue, seemingly the
system used in, e.g. the
Complete
Medieval pots in fabrics identical to those found at
Descriptions:- Dec. = Decoration. Fab. = Fabric. Int. = Interior. Ext. = Exterior. Frag/s. =
Fragment/s. Temp. = Temper.
|
TOPSOIL |
DESCRIPTION |
QUANTITY |
|||
|
Roman |
Coarseware |
1 rim
sherd |
|||
|
|
Pink
fabric with orange slip, small roulette dec. |
1 body
sherd |
|||
|
|
Large
necked jar, fine hard grey fab. |
1 rim
sherd |
|||
|
|
Greyish
fab |
3 body
sherd |
|||
|
Medieval |
Fab. A,
cooking pot = OXAG = Abingdon A = |
1 rim, 1
base, 3 body sherd |
|||
|
|
Fab. B =
OXBF? |
1 body
sherd. |
|||
|
|
Fab. J = |
4 body
sherd |
|||
|
Post-Medieval |
Medieval/Tudor
roof tile, mottled green glaze int. & ext., grey fab. |
1 frag. |
|||
|
|
Shallow
dish with large internal flange, fine hard grey fab. |
1 rim
sherd |
|||
|
|
Brown
int. & ext. glaze |
1 body
sherd |
|||
|
|
Clear
int. glaze, red fab., with grey core |
1 body
sherd |
|||
|
Undated |
Oyster
shell |
2 frags. |
|||
|
|
Clay pipe |
1 frag. |
|||
|
|
|
1 |
|||
|
|
Burned
bone |
4 frags. |
|||
|
|
Bone |
77 frags.
Including 1 tooth |
|||
|
|
Limestone |
5 frags. |
|||
|
|
Stonesfield
slate |
1 frag. |
|||
|
LAYER 2 |
|
|
|||
|
Undated |
Oyster
shell |
6 frags. |
|||
|
|
Glass |
Unknown
number of frags. |
|||
|
|
Burned
bone |
1 frag. |
|||
|
|
Bone |
66 frags.
Including 1 jaw & 5 teeth |
|||
|
|
Charcoal |
|
|||
|
|
Limestone |
2 frags. |
|||
|
|
Roof tile |
2 frags. |
|||
|
|
Unidentified
pottery |
4 sherds |
|||
|
LAYER 4 |
|
|
|||
|
Iron Age |
Carinated
bowl, burnished |
1 sherd |
|||
|
Late-Saxon |
9th-11th |
OXB (?)
with shell 2 mm. Thick, pinkish ext., |
1 sherd |
||
|
|
Green
glazed tripod pitcher with vertically applied stripe |
2 sherds |
|||
|
|
Sandy
unglazed fab. |
1 sherd |
|||
|
Medieval |
Uncertain date |
Brown
glazed |
1 sherd |
||
|
Uncertain
date |
Brownish-red
smooth sandy fab. |
1 sherd |
|||
|
14th |
Cistercian
type |
(No
additional data) |
|||
|
15th |
Bowl,
pinkish, clear int. glaze |
1 sherd |
|||
|
15th |
Buff
fab., 6 mm. Thick |
1 sherd |
|||
|
Post-Medieval |
18th-19th |
Clay pipe |
1 piece |
||
Uncertain
|
Black
cooking pot (Iron Age or Medieval?) |
20 sherds |
|||
Undated
|
Oyster
shell |
1 frag. |
|||
|
|
Glass |
1 frag. |
|||
|
|
Bone |
25 frags.
Including 1 jaw & 3 teeth |
|||
|
|
Red roof
tile |
1 frag. |
|||
|
|
Patterned
paving tile |
1 frag. |
|||
|
|
Charcoal |
|
|||
|
|
Burned
cobble |
1 |
|||
|
|
Limestone |
1 frag. |
|||
|
| |||||