Rescue Excavations in the Outer Court of St Augustine's Abbey 1983 - 84
RESCUE EXCAVATIONS IN THE OUTER COURT
OF ST. AUGUSTINE'S ABBEY, 1983-84 *
PAUL BENNETT
WITH CONTRIBUTIONS BY
J. Bayley, G. Egan, P. Garrard, N. Macpherson-Grant
M. Metcalf, L. Sellwood and J. Shepherd.
INTRODUCTION AND SUMMARY
Salvage work undertaken in 1983-84 during the construction of a new
Students' Union building for Christ Church College, Canterbury,
revealed a complex sequence of archaeological deposits dating from
the Middle Anglo-Saxon period to the present. The Anglo-Saxon
features of eighth- and ninth-century date may relate to a small
secular settlement north of the precincts of St. Augustine's Abbey,
founded in c. A.D. 598. Agricultural levels post-dating this possible
settlement culminated in a period of industrial activity, which in turn
terminated sometime before the area was enclosed by the Abbot and
Convent for an Outer Court during large-scale developments in the
Abbey in the late thirteenth and early fourteenth centuries. The
extensive foundations of a Cellarer's Range, which originally separated
the new Inner and Outer Courts of the Abbey, were uncovered
during the machine clearance of this site, together with a complex
sequence of Outer Court metallings and service drains. The Outer
Court and its associated building ranges to the north and south were
probably demolished shortly after the Dissolution of the Abbey in
1538.
The grounds of the college were long known to have been part of
the extensive precincts of the abbey, but until very recently they were
• Published with the aid of a grant from the Historic Buildings and Monuments
Commission (England).
79
PAUL BENNETT
Fig. 1. St. Augustine's Abbey: Excavation Location Plan.
not included within the area of the scheduled Ancient Monument.
They are surrounded by extant abbey boundary walls and contain
ruined elements of at least one surviving service range (Plate I, D).
The first stage of the salvage operation began in November 1983,
when members of the Trust staff visited the site following notification
that machine work had already commenced. A bulldozer had by that
time exposed and cut into the foundations of a large medieval
building which later proved to be part of the Cellarer's Range
flanking the south side of the Outer Court of the abbey. Over a wet
and cold two-week period and with the agreement of the contractors,
the foundations of this extensive range were cleared and recorded,
together with an impressive sequence of Outer Court metallings and
earlier medieval and Anglo-Saxon features.
Early in 1984, areas on either side of the Union building, which by
that time was receiving its roof, were terraced. These terraces north
and south were for future extensions to the finished structure. A
number of main service-trenches and two large soakaways were also
cut at this time. Following the machine clearance of these areas time
was given to investigate the ground north of the Union building. Here
a small 'island' of intact undercroft floors was examined, together
with parts of the south wall and through-passage of the Cellarer's
Range. A number of Anglo-Saxon and earlier medieval rubbish pits
80
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
and a large medieval ditch were also located in this area. A brief
investigation north of the Union building enabled the recording of a
long section of the later medieval main drain which extended
diagonally across the Outer Court on a north-north-west to southsouth-
east line from the abbey Kitchen to discharge probably in the
city ditch. At least two subsidiary drains feeding foul water from a
service range north of the Outer Court into the main drain were also
recorded together with a section of terra-cotta water-pipe, which fed
fresh water from the abbey's private water supply to the northern
service range. A complex sequence of Outer Court metallings,
medieval features and a thick deposit of 'ploughsoil', containing
residual Roman, Anglo-Saxon and Early Medieval pottery, were also
briefly examined.
The construction of the new Union building completely removed
or severely damaged archaeological deposits which had remained
undisturbed since soon after the Dissolution of the abbey in 1538.
Those levels not completely removed during the preparatory levelling
of the site were severely truncated by an extensive network of deep
foundation pits and wall and service trenches for the new building.
The construction of this new structure represents the most recent of a
series of developments at the college dating back to its foundation in
1962 that have, at the very least, destroyed levels associated with the
Outer Court of the abbey established in the early fourteenth century.
The sequence of deposits and features exposed and truncated during
the various stages of the development indicate that parts of the
college grounds may have been occupied from the Middle AngloSaxon
period onwards. Indeed, there is some evidence to suggest that
a cemetery and extra-mural agricultural area may have existed here
in the Roman period.
Regrettably, the speed of the Union building development did not
allow for a thorough archaeological examination of the area and in
the final analysis an opportunity to fully examine a complex and
important archaeological site was lost.
PERIOD I: ROMAN OCCUPATION
Although no features of Roman date were identified, some form of
activity during the Roman period was tentatively represented by a
small collection of pot-sherds recovered from residual contexts and
from machine-disturbed soils. Other Roman period finds included
two fragments of glass, a coin of Cunobelin and a regular radiate; all
gleaned from residual contexts. A small quantity of fragmented
81
PAUL BENNETT
Roman tiles, including tegulae, imbrices and bricks, was noted in
machine-disturbed soils, in residual contexts and in exposed sections,
though this debris could have been from Early Medieval buildings
where Roman building materials were often re-used.
These residual and unstratified finds perhaps indicate sparse occupation
spanning the period from the first to the third or fourth
century A.D. and may relate to the presence of a known Roman
cemetery in the area later covered by the abbey. The Roman street to
Richborough leads eastwards from Burgate and runs through the
later medieval Lay Cemetery of the abbey some 150 m. south of the
development site. A second and as yet unlocated Roman street
probably led from Roman Queningate in an easterly or south-easterly
direction. It may have run just to the south of the present site, and
perhaps joined the Richborough road in the vicinity of St. Martin's
Church. To date, most of the known burials occur between the
postulated positions of these two streets, the furthest burial being
located some 210 m. east of the city wall. Most of the known burials
(both inhumation and cremation) were uncovered in the Lady
Wootton's Green area during sewer trenchin in 18681 and in minor
excavations by Professor S.S. Frere in 1951. A marble tombstone
fragment was also found in the Green outside the Abbey Gate in
1922.3 A number of Roman inhumation burials was located during
trial trenching in the Lay Cemetery of the abbey in 19724 and in 1981
two cremation burials of late first- to early second-century date were
recovered from service trenches some 75 m. south of the Students'
Union building in the Inner Court area. A 'clay kiln and circular
bronze-smelting furnace' were found in 1929 associated with late first
century A.D. pottery in the south-west corner of the abbey cloister.5
Although early deposits overlying natural head-brickearth were
exposed during the cutting of foundation trenches and pier-bases for
the new building, no identifiable Roman features were exposed. The
early soil horizon, a uniform light brown silty clay 15-25 cm. thick,
contained slight pebble and a small number of Roman tiles, isolated
scatters of oyster shell and a few animal bones, (Fig. 4: Section AB,
layer 65; Section CK, layer 105; Section HJ, layer 83, and Section
PQ, layer 87). Much of the material, which appeared at different
1 J. Brent, Canterbury in Olden Times, 2nd Edition, (Canterbury, 1879), 46 and Pl.
10, No. 1.
2 The Archaeology of Canterbury, vol. viii (forthcoming).
3 VCH (Kent) iii, (1932), 74 and JRS., xxvii (1927), 214.
4 Med. Arch., xxvii (1973), 144.
5 R.J. PoUard, 'Two Cremations of the Roman period from St. Augustine's
College, Canterbury', Arch. Cant., xcvii (1982), 318-24, and VCH (Kent), iii (1932),
74.
82
-------,I
I
I
. .
.··----....... ·..,,,---. . ·------..._
OIJITR CCURT .
:'i=-=: ... ,..
... /·--: •
.. :
(NORTH HOLMES>
ST. AUGUSTINE'S
ABBEY
WT W..ter•IDNlfr W,11,koo,m
H l-lol1 6c!:l We!b n -de.t.8,1 unc.erU.\J'I
K Kuhon
Suvie4, h,-,,'J Wall,do,,d,.oo!,nll..,i:,ha>e
8',u"" c:3 Approxunau, }<><8UJn
p..,..,,
S1ts d o:eavat.ion
'--' Exwnt roof
Pre•ccnqual church t.i.
-- Oro,n
CEU.AREJ{S GARDEN
50 lOO 300
50
----------------
- 1070-1091
CJ £1091-1124
:ll24•lJE,3
:116(!·1220
:!220-1283
=-l'l()9 - :IJ09-l375
:1375-1457
1ml :1457-1539
CJ :1...,.539,-_1650
CJ Un.ec.na1n
/· \ I,
'/ 100
-----
---------------· Marttn's Hfll - - . _
--------. ,-.___ ----
LONGPORT MANOR
c:;-r---·---..
./ f'QND """'°" «>I>'"'
Fig. 2 St. Augustinc·s Abbey: 1984-85 Survey, showing the Outer Court and Cellarer's Range.
RESCUE EXCAVATIONS AT ST. AUGUSTINE'$ ABBEY
levels within the deposit, often pitching at angles, suggested that this
primary 'made-ground' was disturbed soil perhaps consistent with an
agricultural horizon. Much of the ceramic material was worn and
abraded, a phenomenon again perhaps suggestive of residue
contained within 'turned over' soil.
An alternative explanation for the scatter of Roman period finds
may therefore be that the ground north of the postulated street
leading from Queningate was only utilised for agricultural purposes,
whilst the area south of it was, perhaps, a cemetery. The evidence for
this interpretation, however, remains equivocal. No early deposits
were excavated manually. The primary levels were severely truncated
by later features and all conclusions and assumptions are entirely
based on evidence hurriedly recorded at the base of machine-cut
trenches.
PERIOD II: THE MID TO LATE ANGLO-SAXON LEVELS
A number of features dating to the Anglo-Saxon period was recorded
during the salvage work. Residual Anglo-Saxon pottery of seventhto
eighth-century date was also recovered from machine-disturbed
soils overlying natural brickearth and from residual contexts.
The Anglo-Saxon and medieval ploughsoil
Overlying the primary soil horizon sealing natural brickearth was a
30-40 cm. thick deposit of turned-over brown loam, (Fig. 4: Section
AB, layers 64, 63; Section CK, layers 71, 70; Section EF-GH, Layers
146, 148, 79; Section HJ, layer 79; Section PQ, layer 86). This
loose-textured, heavily worm-casted layer contained slight pebble
and occupation debris consisting mainly of oyster shells and fragmented
animal bones. Occupation debris was noted throughout the
entire deposit, though no identifiable horizons were observed. The
interface between the primary deposit of 'made-ground' (probably of
Roman date) and the brown loam, though reasonably distinct in
terms of soil colmr and texture, was uneven and irregular with a
mixing of the two deposits evident at the junction of the two layers.
The brown loam appeared to be turned-over soil, perhaps ploughsoil,
which may have developed over a considerable period. This layer,
which extended across the entire redevelopment area, appeared to
seal a number of Anglo-Saxon pits, whilst other features containing
Anglo-Saxon material may have been cut from a low level of the
deposit. A number of medieval features dating from the late eleventh
century up to c. A.D. 1300 were almost certainly cut through the
83
PAUL BENNETT
deposit. Small quantities of Anglo-Saxon pottery dating from the
seventh to the early eleventh century were recovered from the loam
level during machine clearance together with Early Medieval sherds
dating up to c. A.D. 1180. The implications are that this secondary
deposit of 'made-ground' may have been associated with agricultural
activity spanning the period from the late ninth century to the late
twelfth century. This possible agricultural use of the area north of the
abbey may have superseded and destroyed traces of earlier Saxon
domestic occupation, of which only a small number of identifiable
deeply-cut features remained.
Anglo-Saxon Features
A considerable number of features was observed during the mechanical
reduction of the stratified deposits. The heaviest concentration
of features observed in the truncated horizon occurred in the area
immediately south of the Cellarer's Range and immediately north of
the range extending to the line of the main drain. The continual
presence of heavy machinery together with the high-priority
clearance of the Cellarer's Range made it impossible to excavate the
features north of the building.
A sampling strategy of stray finds churned up during the machine
reduction of the stratified deposits was adopted to ascertain the date
range of these intercutting features. The area chosen extended 8 rn.
west of Section AB, from the line of the eastern cross-wall of the
Cellarer's Range to Section AD. Three members of staff collected
pot-sherds from this area over a ten minute period. Of the 150 stray
pot-sherds collected, 5 per cent were of Roman date, 29 per cent
spanned the period from the seventh to the tenth century, and 66 per
cent were medieval, dating up to c. A.D. 1300. These percentages
compare favourably with the total ceramic assemblage recovered
during the salvage work.
The excavated Pits (Fig. 2)
In only one small area south of the through-passage of the Cellarer's
Range was it possible to excavate a number of intercutting features.
Of the seven rubbish pits located in this area, four were of eighth to
ninth century date (Fig. 4: Section PQ and a-b) and three from the
eleventh and twelfth centuries.
Pit 133
This square-shaped pit cut 65 cm. below natural brickearth and
capped by the loam 'ploughsoil' was cut by a large Early Medieval
84
RESCUE EXCAVATIONS ATST. AUGUSTINE'S ABBEY
ditch (130). The pit contained a lower fill of discoloured pale brown
clay mixed with layers of burnt carbon and pale grey clay (133B and
C). This primary fill was capped with lumps of yellow brickearth and
carbon (133A). The uppermost pit-fill was a subsidence deposit
consisting of stony brown loam mixed with pale brown clay (133).
The pit-fills yielded thirteen pot-sherds of eighth-century date.
Pit 157
Only a fraction of this feature was within the excavated area. The
feature was cut by an Early Medieval pit (158) and was sealed by the
loam 'ploughsoil'. The pit, cut 60 cm. below natural brickearth,
contained a uniform fill of pale brown clay flecked with carbon. A
small number of oyster shells and animal bones were gleaned from
the pit-fill. Although no datable sherds were recovered from this
feature, its stratigraphic position and backfill (similar to pit 133)
indicated that it was probably contemporary with pits 133, 122 and
129.
Pit 129
This feature cut by pit 122 was truncated during the cutting of a
soakaway for the new building. The sub-circular shaped pit, cut at
least 1.30 cm. below natural brickearth, was not fully excavated. This
feature may have been a well. The pit contained an upper fill (129) of
stony grey-brown loam which yielded fragments of Roman brick and
tile together with oyster shells, animal bones and eight sherds of
eighth-century pottery. The lower fill (159) of pale brown silty loam,
heavily flecked with carbon, contained only a few oyster shells.
Pit 122
This feature, which cut pit 129, was arguably the latest of the
sequence of Anglo-Saxon pits. This rectangular shaped feature, cut
95 cm. into natural brickearth, had a primary fill (122A) of finegrained
grey-brown clay, heavily fie.eked with carbon. This deposit
which contained traces of rotted organic matter, together with oyster
shells and animal bones, was capped by a layer of gravel and a dump
of pale brown clay, with oyster shells. These primary deposits yielded
forty-five sherds of eighth- to ninth-century pottery and a coin of
Aethelberht of Wessex and Kent (A.D. 858-866). The coin (Plate III
C) was recovered from the uppermost fill of the pit and may have
been deposited some time after the bulk backfill, perhaps during a
phase of subsidence consolidation.
85
PAUL BENNETI
Other aceramic Features of possible Anglo-Saxon Date
A number of features, buried by the 'ploughsoil' horizon and
identified only in machine-cut sections, may have been of AngloSaxon
date. These were: Feature 62, a shallow pit in Section AB;
Feature 149, a shallow pit in Section HJ and Feature 155, a ditch,
running parallel to and under Section EG.
Discussion
Although a handful of sixth- to seventh-century pot-sherds was
recovered during the salvage work, the greater percentage of AngloSaxon
finds dated from the eighth to the ninth century, and it is in this
period that the highest concentrations of features probably occurred,
these possibly connected with domestic occupation immediately
north of the Anglo-Saxon Abbey Precinct.
The line of the Roman street leading out from Roman Queningate
may have survived into the Early Medieval period, perhaps eventually
becoming the lane known to have existed in 1283 which ran
'between the door of the court of the Abbey' and its 'land at
Nordholm.' This lane probably lay to the south of the present back
lane leading to Christ Church College (the latter only became a lane
after the Dissolution) and may have been the centre of Anglo-Saxon
extra-mural settlement with houses built against and close to the
street frontage. This area to the north of the abbey, still known as the
North Holmes today, may conceivably have been in origin the site of
a small secular settlement, which developed soon after the founding
of the abbey by Augustine in c. A.D. 598 and continued in existence
into the late ninth century. The scanty evidence recorded during the
salvage work perhaps suggests that the occupied area then became
open ground again, perhaps used for agricultural purposes. The
settlement may therefore have been abandoned in favour of a
location within the city walls during the troubled period of Viking
incursions.
Although recent excavations have provided more evidence for.
occupation within the city walls in the Middle Saxon period, the most
significant ceramic assemblages (particularly imports) and finds have
come from outside the Roman town walls on the north-east, in
particular from the present site and most recently from a site near St.
Martin's Church. Although much more work needs to be done, the
implications of the pottery and finds distribution tend to suggest that
one area of Middle Saxon occupation lay outside the old Roman town
walls together with a separate intra-mural settlement perhaps centred
on the Cathedral. These separate intra- and extra-mural settlements
may relate to the 'Innan burhware' and the 'Utan burhware' (the
86
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
people who live inside and outside the burh) recorded in ninth
century documents and discussed recently by Dr Nicholas Brooks. 6
The initial siting of the present settlement may have been dictated
by the presence of the monastic establishment, indicating perhaps
that the occupants were retainers of the religious house. A similar
situation may have prevailed near St. Martin's Church, where a
number of Middle Anglo-Saxon rubbish pits and a metalled street
leading to the church may have been part of a settlement associated
with the small religious community based at the church.7 A larger
area of settlement may, however, have existed in the area north-east
of Canterbury and south of the ancient River Stour estuary ( covering
much of the later parishes of St. Martin's and St. Mary's Fordwich).
This area was probably called the 'wic', a word which comes from the
Latin vicus, and means in the Middle Anglo-Saxon period, a trading
settlement. It is very likely that most of the area from Fordwich to St.
Augustine's and St. Martin's was in the seventh to the ninth century
covered by this large open settlement. Fordwich is first mentioned in
a charter of A. D. 675, and later evidence tells us that the area to the
south was called Wic. 9 There were similar settlements outside
London, York and probably Lincoln, as well as at Dover and
Sandwich in Kent. With the Viking invasions of the mid ninth century
these large wics were probably destroyed, but from the tenth century
a series of dispersed settlements grew up again at Fordwich itself
(which became a separate borough probably in 1055), at Wic and St.
Martin's.
Although the presence of imported Ipswich-type wares recovered
from the Anglo-Saxon features during the salvage work may indicate
a direct relationship between the extra-mural settlement and the
abbey, an agricultural base for this community may be indicated by
the subsequent 'ploughsoil' development. The total assemblage of
mid to late Anglo-Saxon finds gleaned from the Union building site is
of singular importance for the study of this as yet little known period
of Canterbury's history. A thorough examination of these important
levels would have undoubtedly provided a better insight into the
nature of the occupation deposits summarily recorded here.
6 Nicholas Brooks, The Early History of the Church of Canterbury, (Leicester,
1984), 32,-3. A very recent (December 1985) trial excavation in the most northern area
of the Abbey Precincts has also produced Middle Anglo-Saxon pottery.
1 Ibid., 251.
8 P.H. Sawyer, Anglo-Saxon Charters (London, 1968), No. 7, which is almost
certainly a genuine charter.
9 First recorded in Domesday Monachorum.
87
PAUL BENNETI
PERIOD III: c. 1050-1300
Capping the Roman horizon and Anglo-Saxon features was a
uniform deposit of 'turned-over' dark loam, interpreted as possible
agricultural 'ploughsoil'. This deposit (Fig. 4: Section AB, layers 64,
63; Section CD, layers 71, 70; Section EF-GH, layers 146, 148, 79;
Section HJ, layer 79; Section PQ, layer 86), yielded residual Roman
and Anglo-Saxon material, together with pot-sherds dating up to the
second half of the thirteenth century.
This uniform deposit, located across the entire development area
was cut by a large number of features of which only five were
sampled. As previously mentioned, a very large number of features
was identified during the preparatory machine levelling of the site,
particularly in the area immediately north of the Cellarer's Range.
The sampling strategy undertaken west of Section AB, discussed
above, revealed that 66 per cent of the total assemblage recovered
from the machine disturbed soils in that area were of Early Medieval
date; the bulk of the finds centring on the period from c. 1100-1200.
Many of these truncated features, recognised only as soil stains, were
undoubtedly domestic rubbish pits. Other features contained
residues consistent with industrial activity. The only excavated
feature containing industrial debris proved to be a casting pit (47).
This feature, possibly one of the latest in this area prior to the
construction of the Outer Court and the service ranges in c. 1300,
may have been only one of a number of such features disturbed
during the course of this redevelopment.
The excavated Features
The ditch
The earliest dated feature was a very large linear ditch aligned
roughly east-west. This feature located in the area south of the
through-passage of the Cellarer's Range, was only partly excavated;
the lower levels of the ditch being sampled by means of digging a
narrow but deep trench east of the line PN to the position of pier base
16. The southern edge of the ditch was revealed for a length of
approximately 3 m. The sump of the ditch was not found, but the
feature must have been in excess of 3 m. deep with shallow-sloping
sides. The angle of the ditch backfill suggested that the centre of the
ditch lay within the excavated trench and that the ditch may have had
a flat bottom. Allowing for a narrow but flat sump, the ditch may
have been at least 9-10 m. wide.
Feature 66, at the south end of Section AB may have been part of
88
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
the backfill of the ditch. The casting pit (47), which was cut through
feature 66, may well have been inserted into the upper levels of the
ditch. It is perhaps a salient point to note that natural brickearth was
not encountered during the excavation of the casting pit. Intact early
stratigraphy (layer 65) was, however, noted approximately,0.09 m.
south of the wall of the Cellarer's Range (11). If the termination of
intact stratigraphy at this point represents the northern edge of the
ditch and layer 66 a remnant of its backfill, then a ditch width of
approximately 10 m. is again suggested.
The ditch (in the area south of the through-passage) was backfilled
with laminated deposits of clay and loam. The earliest fills, three
distinct layers of light brown stony clay (131, 130 and 127), yielded
pottery dating from c. A.D. 1050-1150. The overlying fills of brown
loam and pale brown clay (125 and 124) also contained sherds dating
up to the mid twelfth century.
The only deposit likely to be associated with a weathering of the
ditch edges during occupation was layer 131. This deposit of
discoloured grey-brown silty clay yielded seven pot-sherds dating
from c. A.D. 1050-1080. The remaining layers, probably associated
with a single period backfill, contained small quantities of eleventhcentury
material but the chronological emphasis in these layers lay in
the early to mid twelfth century. Despite the limited nature of the
excavation, the pottery recovered from the ditch backfill suggests that
this feature was probably cut in the third quarter of the eleventh
century and backfilled in the mid twelfth century.
The considerable size of the ditch, approximately 10 m. wide and in
excess of 3 m. deep, was over-large for a boundary and may have
been cut for defensive purposes. The dating evidence places its
cutting in the early Norman period, shortly after the Conquest. Quite
why a ditch of this size was cut in this area and at this time is difficult
to ascertain; only further excavation can evaluate this and the
accuracy of the interpretation which was based on extremely limited
evidence. One other possibility, however, is that the ditch was cut for
the great drain coming out of the Reredorter. The line of the ditch is
very close to the projection westwards of the probable site in the
Reredorter for a drain (Fig. 2).
The pits
Despite the number of disturbances observed cutting earlier deposits
in the development area, very few were sampled. Two rubbish pits
were partially excavated in the area of the through-passage of the
Cellarer's Range. Pit 158 was cut from the surface of the 'ploughsoil'
deposit (86) in this area and yielded no datable finds. Pit 140, also cut
89
PAUL BENNETf
from the surface of the 'ploughsoil', yielded a corpus of pot-sherd:
dating up to c. 1100-1125, and may conceivably have been cut anc
backfilled before the ditch (130) was infilled.
The unexcavated Features
A number of possible medieval features was recorded in machine-cu
sections. Two shallow features, possibly pits, were located in Section:
EH and HJ. Feature 154/145 yielded twelfth-century pot-sherd:
(gleaned from the section) and pit 81 contained no datable finds, bu
was cut through the 'ploughsoil' deposit.
A possible ditch (180), aligned roughly north-south, was revealec
in Section HJ. This feature, which cut the 'ploughsoil' deposits
contained an upper fill of stony brown silty loam, capping dirt:
pale-brown clay; it yielded no finds and may conceivably have been i
field boundary.
The Casting Pit (Fig. 3, inset plan and Plate III, B).
During the machine-clearance of the south-east corner of the site a1
extensive area of burnt orange clay mixed with carbon and vitrifiec
clay lumps was exposed. This feature, one of a number of heavil:
burnt areas containing industrial residue, was systematically exca
vated to the line of Section AB. Only the west end of this extremel:
large feature was excavated. The construction pit was probabl:
rectangular, being 7.5 m. wide and of unknown length. The feature
aligned roughly east-west was cut to an average depth of c. 90 cm
At the base of the pit, part of a rubble masonry foundation o
greensand and flints bonded in burnt orange-yellow sandy morta:
(46) extended on a north-east to south-west axis into Section AB. P
thickening of the wall (from c. 48 cm. to 62 cm.) at the point o
intersection with the trench edge indicated the presence of a wall a
right angles to wall 46, extending under the section. On the line o
Section AB the wall foundation had been demolished to a singl<
course of rubble masonry. The northern end of the wall, however
survived to three courses, standing to a height of c. 45 cm. A shallo"
terrace, 35 cm. deep, had been cut into the base of the constructio1
pit, east of the wall. The terrace extended approximately 50 cm
north of the wall 46, terminating in an eroded but roughly vertica
face. A third wall at right angles to wall 46 may have been located ir
this position, but all trace of it removed during a subsequent episodt
of robbing. The base of the terrace within the masonry structure wa:
heavily fired and a well-defined 'crust' of vitrified clay ( 45) abutte<
the fire-fractured and burnt internal face of wall 46. The flooi
terminated close to and at right angles with the end of wall 46. Thi:
90
RESCUE EXCAVATIONS AT ST. AUGUSTINE'$ ABBEY
Period II
Periods IV & V
30
100
/ .-J
_.,::f \
'Pip,.., \' \
'I \ '-- \
Fig. 3. St. Augustine's Abbey: Phased Plans of excavated and observed Features in
the Outer Court Area.
91
PAUL BENNETI
scar is perhaps supportive evidence for a wall at right angles to the
surviving masonry. The base of the construction pit north, south and
west of the internal terrace was sealed by a 1-3 cm. compacted tread
of dirty grey clay, mixed with mortar specks (47). This deposit may
have been construction residue. The layer was in turn sealed by a
dumped deposit of silty brown loam mixed with dirty pale yellow clay
( 44). This dump was probably residue from the backfilling of the
construction pit, after the masonry walls had been built. Three
stake-holes ( 48, 49 and 50), located at the base of the construction pit
and cut to an average depth of 10-16 cm., were of unknown purpose,
but may have been inserted from a higher level perhaps as ancillary
supports for a flimsy timber roof covering the masonry structure. No
trace of a possible stoke-hole for the structure was found in the
excavated area. This must have been located somewhere east of
Section AB.
Following an unknown period of use, the masonry structure was
demolished and robbed to its foundations. During the robbing the
burnt floor (45) was also disturbed. The robber pit was subsequently
backfilled with successive layers of burnt clay debris (including
numerous casting-mould fragments) and dark loam. Three successive
dumps of burning ( 43, 41 and 39) were discerned in the lower backfill,
separated by dumps of dark brown loam (42, 40 and 38). These
dumped layers were in turn cut by a later disturbance containing
further deposits of burnt clay and mould fragments (37). The
remaining hollow was then infilled with a deposit of dark brown loam
(36) which was in tum eventually capped by the undercroft floors of
the Cellarer's Range.
Only residual pottery dating up to c. 1150 was recovered from the
robber pit's backfill (layers 37 and 38). The casting pit was probably
built over a large ditch infilled by the mid twelfth century (130).
Subsidence of the backfill within the casting pit led to evident
consolidation of the undercroft floor. The subsidence of the pit
during the life of the later building may indicate that only a short
period separated the final infilling of the pit and the construction of
the range. The evidence for this is equivocal and a construction date
sometime after the mid twelfth century and before c. 1300 can be
confidently suggested.
Besides the small number of pot-sherds, the robber pit backfill
yielded quantities of metal-rich carbon (containing traces of copper
and tin), fired clay (some partly vitrified) and vitreous slag, which
could have formed in a hearth or furnace where metal was being
melted. The fired clay - residue from casting moulds - displayed
varying degrees of burning and almost all the fragments recovered
bore traces of copper alloy on their surfaces. When first discovered,
92
E
113
- - -
. .
B
Fig. 4. St. Augustine's Abbey: Principal Sections of stratified Sections in the Outer Court Area and inset Plan of the Casting Pit.
LJ L1;t loom
[IiI] Modrf\ topsoitfgorden loom D Lt9hl cloy
[I!JJ Pos\•dlsso\u\lon \oom - Oor\l: cloy
Q Robber lrtnch bockhll
-Mo.rtor floor
Courtyard mt-tollm;
lffiill]] ?A9nculturol soil
liiilliill) Oork toom
LJS•II
j::· <\-'::] t-tortor
IIIII[IJ Rammed chalk
h;f aur,u cloy
R. ·t Carbon
0 L...,el•12"S0m00
.R.
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
the feature was interpreted as a bell-casting pit, but an analysis of the
mould fragments by Justine Bayley (see below) indicates that the
fragments do not conform to the shape expected for the manufacture
of bells and that the industrial residues may conceivably have been
associated with the casting of monumental plaques for use perhaps in
the abbey church.
The total assemblage of finds and observed and excavated features
of Period III date indicate a complex occupation sequence from c.
1050 to 1300. Of particular interest is the evidence for industrial
activity. Verbal reports of features similar to the casting pit were
recounted to the author by members of the college staff, who were
present during the early 1960s and later phases of developments at
the college. Overall, the sequence of archaeological deposits predating
the establishment of an Outer Court for the abbey in the early
fourteenth century, indicates almost continuous activity from the
Middle Anglo-Saxon period, which culminated in a phase of industrial
activity perhaps associated with the manufacture of metalwork
for the abbey. This last period also coincides with the Norman
rebuilding of the abbey on a large scale (c. 1070-1150). The northern
boundary of the abbey at this time was probably immediately to the
south of the excavated site.
PERIOD IV: c. 1300-1550
The licence for the abbot and convent to enclose a block of land
measuring 150 ft. by 80 ft. immediately north of the abbey Precinct
(given in July 1300)10 was probably rapidly followed by the enlargement
of the new Outer Court for the abbey. An earlier stage of the
expansion (a licence granted in November 1283)11 saw the abbot
being allowed to enclose the lane between the door of the court of the
abbey and his land at 'Nordholm'. These licences represent just two
stages of a major phase of expansion of the abbey buildings which
effectively saw an enlarging of the Inner Court in the late thirteenth
century and a relocation of the principal service buildings in a new
northern Outer Court. Apart from probably relocating the great
gate, the abbey moved the Cellarer's Range from the west side of the
great Cloister to the north side of the enlarged Inner Court (a
building range which effectively separated the Inner and Outer
Courts). A new range of service buildings, almost certainly for the
1° Cal. Pat. Rolls Ed. I, (1292-1301), 527.
11 Cal. Pat. Rolls Ed. I, (1281-1291), 51.
93
PAUL BENNETT
Brewhouse and Bakehouse, was constructed on the north side of the
new Outer Court. (The ruined western end of this range still stands.
Plate I, D.)
During the initial clearance of the Students' Union site, the
foundations of a large masonry building, aligned approximately
east-west and later identified as probably being part of the Cellarer's
Range, were exposed and partly truncated by the bulldozer.
Levels, undoubtedly associated with floors within the building,
together with yard metallings to the north of it, were also exposed or
truncated. The recording of this structure was considered a high
priority and by negotiation with the contractors, a small team of Trust
staff and volunteers attempted to expose and clean the remains of this
building. The contractors were not prepared to delay machineclearance,
but fortunately since much of the area containing the
building had been reduced to formation level for the new structure, it
was possible to clear and record the foundations while machining
continued in the northern half of the site. During the latter stages of
development a small section of undercroft floors was examined and
further portions of the building were recorded.
The Cellarer's Range (Plate II, A, Plate II, B)
The foundations uncovered during the salvage work were for a
building of external width measuring 11.60 m. (internally 8.20 m.)
and of unknown length. The north wall (11) robbed to subfoundation
level, was 1.95-2.00 m. wide. Only one small portion of
intact superstructure of four chalk-blockwork courses bonded in a
pale-brown mortar and standing to a maximum height of 42 cm. was
located at the east end of this wall. This fragment of wall superstructure
was 1.50 m. wide and was rendered externally with plaster. A
single chalk-blockwork foundation course, on average 20 cm. thick,
overlay a sequence of rammed chalk and mortar sub-foundation
deposits extending down at least 1.50 m. (the full depth of the wall
foundation was only observed along Section PN, cut through the
south wall). The south wall foundation (13) was 1.25 ·m. wide. Only
sub-foundation deposits of rammed layers of chalk and mortar
survived. A well-defined robber trench 0.95-1.05 m. wide (12)
overlay the remaining portion of the sub-foundation. The construction
trench backfill, of dark brown clayey loam (114) yielded a small
corpus of residual late twelfth- to early thirteenth-century pot-sherds.
Two transverse wall foundations (1 4 and 15) 1.50 m. to 1.75 m.
wide, set 3.66 m. apart, indicated the position of a through-passage or
gateway passage joining the Inner and Outer Courts. These crosswalls
were of contemporary build with the north and south walls and
94
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
ofidentical construction. A single external buttress (18) was located
opposite the western cross-wall. The buttress, of contemporary build
with the main north wall, was 1.00 m. wide and extended 1.00 m.
north of the wall. The buttress sub-foundations were of rammed
chalk and mortar underlying a foundation of three chalk-blockwork
courses. The fragmentary remains of an in situ flat-chamfered
greensand block surmounted the buttress foundation.
An internal pier base (16) capped by a large 30 cm. thick, roughly
hewn greensand block, measuring 1.30 m. east-west, by 1.60 m.
north-south, was located in the centre of the building range, 40 cm.
east of the eastern cross-wall of the through passage. A corresponding
thickening of the main north wall (from 1.95 m. to 3.00 m.) east
of the passage walls indicated the position of the possible staircase,
entered from the north, leading up to the first floor, a similar
arrangement to the Domus Hospitum at Christchurch Priory 12 and
also probably to the still surviving ( though heavily restored) Guest
Hall on the west side of the abbey's Inner Court.
A third internal cross-wall (17), 10 m. east of the pier foundation,
suggested the existence of two major internal openings at undercroft
level. The foundations, 0.95 m. wide, were interrupted by openings
1.45 m. wide. The north and south elements of the cross-wall were of
contemporary build with the main north and south walls.
Intact construction levels were recorded in a small island of
undisturbed stratigraphy contained by the line of Sections PN and
LK. In this area, a construction horizon of trodden poured mortar,
mixed with crushed chalk (88) sealed the earlier sequence of deposits
associated with the backfilled ditch (feature 131). This deposit 2.4
cm. thick capped the remains of the construction trenches for the
principal south wall and the east passage wall and originally abutted
the wall foundations. The horizon was sealed by a 25 cm. thick dump
of redeposited brickearth flecked with mortar and crushed chalk
(Section LM, layer 106; Section KL, layers 134 and 141). Layers 134
and 141 yielded a large quantity of residual Roman, Anglo-Saxon and
mid to late twelfth-century pottery together with sherds dating up to
the late twelfth century. These construction deposits were capped by
the primary undercroft floor (118) of rammed and poured mortar,
2-10 cm. thick. The machine-cut section (Section AB) through the
building revealed a different sequence of deposits. Here an internal
construction trench for the north wall (1 lA) backfilled with
redeposited brickearth, was sealed by a thin lens of carbon and a
10-15 cm. thick deposit of rammed chalk and mortar (33). This layer,
12 The Archaeology of Canterbury, vol. iii, (forthcoming).
95
PAUL BENNETT
undoubtedly the primary undercroft floor, capped the final backfill of
the earlier casting pit (36). A small, shallow pit (34) underlying the
rammed chalk floor, located against the inside face of the north wall,
may have been cut during the construction process. The pit was
backfilled with light brown clayey loam, flecked with mortar and
chalk.
The possible Workshop, North of the Cellarer's Range
Of similar date to the Cellarer's Range was a small timber-framed
structure, possibly a workshop, located against the main north wall of
the range. This structure, 5.40 m. wide with its western end on line
with the eastern cross-wall of the range, was probably built after the
principal walls of the Cellarer's Building had been constructed.
Indeed, it is quite possible that the principal transverse members of the
framed structure may have been fixed into the fabric of the north wall.
A 70 cm. wide trench, filled with rammed dirty chalk (59) marked
the line of a foundation designed perhaps to receive a timber-plate
for the north wall of the workshop. The trench cut to a maximum
depth of 50 cm. below the contemporary ground surface, terminated
at its west end with a post-pit (30), 60 cm. in diameter filled with
rammed chalk. A soil stain for a post-pit approximately 22 cm. in
diameter was identified in the pit backfill.
Three posts divided the building longitudinally into two equal
halves. The easternmost post (58) located in Section AB, was set in a
post-pit 55 cm. wide cut 43 cm. deep. A soil stain for a post 30 cm. in
diameter was noted in the backfill. The post-pit was backfilled with
dirty brown loam heavily flecked with chalk and mortar. A second
post-pit (22) was located 2.50 m. east of Section AB. This subcircular
feature, contained a backfill of rammed chalk and traces of a
post 25 cm. in diameter. Two intercutting post-pits (21 and 24)
defined the west end of the structure. These pits, probably cut
simultaneously, contained a backfill of dirty rammed chalk, and soil
stains for posts 25 and 23 cm. in diameter, respectively.
The precise nature of the building frame was not established, but a
simple pent-roof possibly covered the structure. The row of internal
posts may have been for subsidiary roof-supporting members or may
relate to subdivisions or fixtures within the buildihg. The deposits
within the structure comprised layers and laminae of crushed chalk,
poured mortar, and loam containing stone and chalk chippings (60
and 61). These layers may have been rough working-floors associated
with a mason's workshop. The north wall of the timber building and
post 58 were abutted by these deposits, and were probably
aggregated gradually during the working processes. Post 58 was
96
RESCUE EXCAVATIONS AT ST. AUGUSTINE'$ ABBEY
withdrawn during the working life of the building and the remaining
void was filled with a deposit of dirty loam mixed with mortar and
chalk. The early deposits were capped by a levelling dump of light
brown loam flecked with chalk (57) and a new laminated 'workingfloor'
of poured mortar and mortary stone and chalk rubble, (55 and
56) developed over this new horizon. A further levelling deposit (54)
of bright yellow clay capped the intermediate floor, which was in turn
capped by a third and final floor of laminated white lime mortar (53).
The withdrawal and subsequent sealing of post-hole 58 suggests
that this feature at least may have been for an internal fitting or
division within the building rather than a load-bearing or structural
element of the frame. The remaining posts (all machine-truncated)
appeared to have been withdrawn when the building went into
disuse. The plate supporting the north wall was also removed when
the building was demolished.
Contemporary with the life of this structure was a huge, deep
feature (52) cut just outside the workshop's north wall. This disturbance,
which extended at least 2.00 m. into the excavated area and
continued to the line of Section AD, may conceivably have been a
clay quarry. The feature was backfilled with redeposited dirty brown
clay (52) late in the life of the timber building. This bulk infill was
capped by deposits of brown loam and pale-brown silty loam (51),
which also filled the void marking the line of the workshop north
wall, and covered the final working floors within the building. This
extensive levelling deposit was in turn sealed by dumped deposits and
metallings associated with the surfacing of the Outer Court (23).
The Outer Court Levels
Capping the early fourteenth-century ground surface north of
the Cellarer's Range and the possible workshop and clay quarry,
was a sequence of levelling deposits laid down prior to the
establishment of a metalled courtyard for the Outer Court. All these
deposits were removed during the machine-clearance of the site
and only a summary of the complex overlapping layers and
laminae appears in the published sections (Section AB, layer 51;
Section EH, layers 72-76, 153 and 156; Section HJ, layers 72-76,
77 and 78).
In Section AB the possible clay quarry arid workshop were capped
by thick deposits of light brown silty clay (51). These deposits were
cut through by the main drain (9), and a late medieval drain (8). The
dumped layers were capped by a 10-22 cm. thick layer of rammed
gravel mixed with brown loam (23). The gravel, containing quantities
of fragmented peg-tiles and oyster shells, was laid shortly after the
97
PAUL BENNETT
construction of the northern service range and Cellarer's Building as
a durable courtyard surface for the Outer Court. The courtyard was
undoubtedly remetalled or patched during its life; these episodes of
repair and maintenance showing as differently coloured layers of
gravel and isolated dumps or mortary 'hard core', oyster shells and
crushed peg-tiles.
A more complex sequence of dumped deposits culminating
in the deposition of courtyard metallings was observed in Sections
EH and HJ. Here depressions in the ground surface south of the
northern service range were infilled with layers of construction
debris, redeposited brickearth and loam (153 and 156). These
layers, 20-55 cm. thick, were also associated with horizontal layers
and laminae of 'upcast', perhaps residue from the cutting of
construction trenches for nearby buildings (76), together with
deposits of poured mortar (layers 72-75). The deposits of pale
yellow sandy mortar and off-white lime mortar may indicate the
presence of an extensive mortar-mixing area in the north-east
corner of the site. A disturbance cutting these horizontal deposits
was noted in Section HJ. This shallow, flat-bottomed feature,
40 cm. deep, cut for an unknown purpose, was backfilled with
layers of rammed chalk and light brown and dark brown loam
(layers 77 and 78). The sequence of deposits observed in these
sections was cut by the main drain (9) and a contemporary but
smaller interconnecting drain (144). The levelling deposits (and
drains) were capped by 10-30 cm. of dirty rammed gravel (151)
containing dumps of oyster shell, mortar and crushed peg-tiles.
These horizontal deposits represented the maintained and repaired
metallings for the Outer Court. Similar laminated deposits of
gravel, 'hard core' and mortar for the Outer Court metallings
(layers 66-69 and 24) were located sealing the early fourteenthcentury
ground surface (70) in Section CD. These deposits which
capped the construction trench for the main drain (99) were cut by
two Late or Post-Medieval features (25 and 150).
The main Drain (Plate I, A)
A large masonry-built drain (9) was uncovered and truncated during
the machine-clearance of the north-east corner of the Students'
Union site and in the subsequent levelling of the area to the north of
it. The drain, aligned approximately south-south-east to north-northwest,
was built prior to the laying of the Outer Court metallings
during the construction of the principal court ranges. The drain
probably fed foul water from the abbey Kitchen and associated
buildings diagonally across the Outer Court, west of the northern
98
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
service range and onwards perhaps discharging eventually into the
city ditch.
The drain, exposed for a length of 22 m. was trench-built in a
cutting 1.55 m. wide. A drain fall from south-south-east to northnorth-
west was recorded. The base of the construction trench was
lined with crushed chalk set in a dirty yellow mortar. Mortared chalk
rubble.walls on average 30 cm. wide were raised off this foundation to
a maximum height of 95 cm., terminating with a string-course of
mortared peg-tiles. The bedding deposit was sealed by a 1-2 cm.
thick layer of mortary pale-yellow sand which abutted the rubble
walls on either side of the construction trench. A number of regularly
spaced scars extending across the full width of the trench, set
approximately 60-70 cm. apart, indicated the position of stone slabs
lining the base of the drain. Well-cut 'facing' blocks of chalk
recovered from the robber backfill of the drain (101 and 147) were
probably residue from the internal lining of the drain, built over the
paving slabs and against the rough-face of the internal walls.
The string-course of peg-tiles bonded to the top of the rubble lining
walls marked the spring of a possible barrel-vault covering the drain.
The existence of a barrel-vaulted top was also indicated by the
presence of a number of voussoir chalk blocks recovered from the
robber trench backfill.
The secondary Drain (Plate I, B)
A secondary but probably contemporary drain (144A) aligned
roughly east-west was uncovered in the area north of the Union
building. The drain, designed to feed foul water from tl)e northern
service range into the Great Drain, was trench-built in a cutting 1.50
m. wide and 1.30-1.40 m. deep. The base of the cutting was lined
with a bedding of poured mortar (144B). A lining of faced and
mortar-bonded chalk blocks, set 20 cm. apart, was built off the
bedding deposits to a maximum surviving height of three courses.
The remaining trench area behind the facing blocks was backfilled
with mortared chalk rubble capped by redeposited brickearth (144).
The facing blocks were probably carried up to the full height of the
construction trench for the 20 cm. wide drain. The drain was
probably capped by stone or chalk slabs and sealed over by courtyard
metallings (151).
The later Drain (Plate I, A)
A brick-built drain (8) was located at the north end of Section AB
during the machine-clearance of the area. This feature, aligned
99
C
PAUL BENNETT
B
D
PLATE I
••\ Iii tIii- . .• . J .. •- ..;"$ I"
:: .. ..,,.; ·1,J. . 'I. f.t f: . A
tJ.
•4·' <;. . I: ·. .t.,....
-i •. .• •·-. . -. , .
. .. lf\ - '1,;..f-f . ,,.- .. ,:.:..'·_,.•.. . .\.'· I f..' :- V •..;!
). . ., '. • ,..t . • -, Lr• .. AMA \. -, ..... .,-..
St. Augustine·s Ahbcy. A. Main Drain 9 and Late Medieval Drnin 8. looking South.
(Scale: 2 m.) B. Service Drain 144. looking East. (Scale: 1h m.) C. Water-pipe 150.
looking North-cast. (Scale: ½ m.) D. Surviving Fragment North Service Range.
looking North-west.
100
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
east-west, was also probably designed to feed foul water into the
main drain. The drain, built late in the life of the Outer Court was set
in a construction trench 1.65 m. wide. Rubble brick walls 25 cm.
wide, bonded in a pale yellow sandy mortar, were built on either side
of the cutting over a bedding of poured yellow mortar. Chalk slabs
60-65 cm. wide were laid between the rubble walls and a lining of
whole, well-pointed bricks and occasional chalk blocks was built over
the slabbed floor against the rubble lining walls.
The Water-pipe (Plate I, C)
Cutting the court metallings in the north-east corner of the development
area was a Late Medieval water-pipe (150), laid perhaps to
supply fresh water to the northern service range. The pipe run, set in
a trench 50 cm. wide and 75 cm. deep, extended north-west to
south-east from Section CD to the line of the main drain. From this
point onwards the pipe was probably inserted into the construction
trench backfill of the main drain, perhaps diverging from the drain
just outside the west end of the service range. The ceramic pipes,
probably manufactured locally, were 39 cm. long with extended
tapering flanges 3 cm. long. The pipes, approximately 10 cm. in
diameter externally, were joined by flanges which were packed with
clay. The pipes were laid at the. base of the construction trench and
covered with sticky redeposited brickearth. An upper backfill of
brown loam (26) containing rubble chalk and stone blocks, together
with fragmented peg-tiles was in turn capped by a layer of dirty gravel
mixed with mortar and chalk (23).
The Undercroft Floors of the Cellarer's Range
Deposits associated with repairs and maintenance of the undercroft
floor of the Cellarer's Range were excavated in a small area east of
the through-passage (Sections LM and KL). A further sequence of
floor deposits were recorded in Section AB.
The primary undercroft floor east of the through-passage was
uniformly capped by a thin lens of grey-brown trodden occupation
loam (119). This thin layer was sealed over in isolated areas (see
Section KL) by later floor laminae comprising poured off-white
mortar (116), puddled orange clay (115) and occupation loam (111).
A dump of mortary brown loam containing rubble chalk lumps (117)
capped the primary mortar floor (118) and the later laminae. This 20
cm. thick deposit was in turn sealed over by a mortar floor (113). The
sequence of layers in this small area was completed by a rammed
chalk floor 3-10 cm. thick (10).
The corpus of pot-sherds, recovered from the matrix of the chalk
101
A
PAUL BENNETT
PLATE 11
St. Augu,tinc's Abbey. A. Cellarer·, Range. looking South-west. (Scale: 2 m.) B.
Cellarer\ Range South Wall. through-passage and Pier Base. looking South-cast.
(Scale: 2 m.)
102
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
floor (110) and from the laminated deposits underlying it, was of
early to mid fourteenth-century date. This material suggests that the
sequence of floors in this area represents a period of floor maintenance
spanning no more than fifty years or so. The rammed chalk
floor was worn and pitted with compacted occupation loam (109)
infilling shallow depressions in its surface. The rammed chalk horizon
may therefore have survived as a floor receiving little maintenance
from the mid fourteenth century up to the demolition of the
Cellarer's Range in the mid sixteenth century.
The floor deposits recorded in Section AB indicated that considerable
subsidence in the area overlying the casting pit had occurred
during the life of the building. Traces of repair in clay to the primary
rammed chalk floor (33) were apparent in the section. The primary
deposits were capped by a 10 cm. thick dump of compact mortary
loam (32) and a thick layer of crushed peg-tiles (31), undoubtedly laid
to counteract subsidence. A poured off-white mortar floor of two
phases (30 and 29) separated by a lens of occupation loam (29A)
capped the dumped deposits. The sequence of floors in this area was
truncated by a disturbance associated with the deposition of a topsoil
mound over the site in the 1960s. The floor levels east of the
through-passage were sealed by mid sixteenth-century demolition
deposits (112).
PERIOD V: THE POST-DISSOLUTION AND MODERN LEVELS c. 1550--1984
Following the Dissolution of the abbey in 1538, Henry VIII retained
the abbot's lodging and other buildings around the east, south and
west sides of the Inner Court and converted them into a small Royal
Palace. The Cellarer's Range north of the Inner Court was no longer
needed and so it was demolished. A new wall, interrupted by a gate,
was built probably in c. 1539-40 to cut off and reduce in size the
northern part of the Inner Court. 13
The occupation levels within the Cellarer's Range were sealed by a
30-40 cm. deposit of loose abraded mortar containing chalk lumps
and stone chips and fragmented peg-tiles (112). The walls of the
range were demolished to their foundations and the robber trenches
backfilled with mortary loam (layers 10, 12, 100 and 121).
The drains north of the Cellarer's Range were systematically
robbed from the latest surface of the Outer Court, and residual debris
thrown back into the robber trenches (the main drain, layers 6, 101,
13 The last ruins of the Outer Court buildings are shown on King's bird's-eye view.
103
PAUL BENNETT
PLATE Ill
:4
_
.. ....... ..,."
,
1i- .,,
- ....... _ A
B
St. Augustine·s Abbey: A. Cellarer's Range North Wall and Undercroft Floors (in
Sccrion AD). looking Eat. (Scale: 2 m.) 8. Casting Pil. looking North. (Scale: 2 m.)
C. Coin (obver-,e) Acthclbcrht of Wessex A.O. 8511-1166. (Scale 2: I).
104
C
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
101A, 147 and 152; the northern drain layer 143; the brick drain layer 7).
The demolition deposits and robber trenches were uniformly
sealed across the entire site by 20-40 cm. of pebbly brown loam
containing peg-tile debris together with deposits of mortar, crushed
chalk and rubble (3). Only three features were noted cutting postDissolution
'topsoil', Pit 25 in Section CD; Pit 19 and an extensive
soil disturbance (85) noted in Section PQ, possibly connected with
post-Dissolution terracing.
The 'topsoil' horizon was in turn sealed by a 10-20 cm. layer of
imported fine grained dark loam (2) possibly laid in more recent
times to level the site and bed an extensive grass lawn. This horizon
was eventually covered by a huge mound of soil (1), residue from
construction trenches for nearby developments within the college in
the 1960s.
THE POTTERY
N. Macpherson-Grant
For its size, this excavation produced some extremely useful assemblages,
particularly for the difficult and poorly represented Mid to
Late Saxon period. The pit group (122) discussed below is an
important key addition to the local ceramic sequence, and the
quantity of Ipswich-type sherds is unusually high. This pit group is
only the fourth of its type to be recorded from Trust excavations. Its
relative importance was immediately appreciated, but not fully
understood until our recent 1985 excavation at St. Martin's Hill
produced two further small pit groups and similar quantities of local
and imported wares. The present group is crucial for establishing the
chronological/formal links between Early to Mid and Late Saxon
Canterbury assemblages, but only comes into perspective when
viewed alongside the St. Martin's Hill and comparable Mid to Late
Saxon material from the city. Accordingly, a thorough examination
of the Pit 122 material, its formal inter-relationships and the implications
of the imported wares are deferred. Here only key aspects are
highlighted following an overview of the excavation's pottery. In
turn, there is a brief note on the sequence from the Early Medieval
ditch.
Summary of the Pottery
If the quantities of pottery (per period) are representative of the
105
PAUL BENNETT
adjacent unexcavated area, the following occupation sequence is
suggested:
Period I
Sparse Roman activity from the first to third/fourth century.
Period l/A
Saxon occupation or activity beginning in the seventh century with a
modest quantity of organic-tempered pottery.
Period IIB: 'Domestic Occupation'
If current dating of later Saxon wares is correct, a slight hiatus may
have followed the seventh-century activity, followed by definite
occupation from the mid or later eighth and well in.to the ninth
century, represented by Pit 122 (Fig. 6) containing purely Mid to Late
Saxon pottery, a coin of Aethelberht of Wessex and Kent (858-866)
and imported Ipswich wares. Later levels also produced residual local
wares and both fine and 'pimply' Ipswich wares.
Period l/C: 'Agricultural Activity'
Fabric quantities suggest a decrease in occupation during the Late
Saxon period, lasting until approximately the mid or later eleventh
century.
Period III
The bulk of the pottery from this excavation is of Early Medieval date
and from the late eleventh/early twelfth century occupation seems
intensive, possibly peaking in the mid to late twelfth century. From
the twelfth century occupation it appears continuous throughout the
thirteenth century, though with a possible decrease in the first half.
Period IV: 'The Cellarer's Range'
Definite late thirteenth-century occupation is represented by material
gleaned from construction and occupation deposits associated with
the Cellarer's Range broadly dated by documentary evidence to
1300. Occupation continued through into the mid to late fifteenth
century, probably terminating in the early sixteenth century (Period
V).
Period II: The Mid to Late Saxon Pits (Fig. 5)
Full fabric descriptions are omitted here, but ware headings are
106
RESCUE EXCAVATIONS AT ST. AUGUSTINE'S ABBEY
followed by their standard (C.A.T.) period letter-code and fabric
number-code (e.g. Early to Mid Saxon organic-tempered ware -
EMS 4). The coding (for discussion brevity) is based on a combination
of detailed thin-section analyses of Canterbury Saxon pottery
by Dr Ailsa Mainman and further refinements by the author. 14
Pit 129
Pit 133
1. Mid to Late Saxon coarse sandy ware (MLS 3). Jar. Pit
also contained one organic-tempered (EMS 4) and four
Mid to Late Saxon sandy ware (MLS 2) sherds, together
with one of 'pimply' Ipswich-type ware with characteristic
wide-spaced horizontal wheel-ribbing.
2. Mid to Late Saxon coarse sandy ware (MLS 3). Small
pot. Besides residual Roman material, 133 also had a
small quantity of EMS fabrics (including two EMS 4)
one transitional Mid to Late Saxon organic-tempered
sandy ware (MLS 1) and one sherd of MLS 2. Also:
3. Jar base in very hard, very coarse-gritted wheel-thrown
ware (problem ware - see discussion - e).
Pit 122 (Canterbury Key Pottery Group MLS 3)
Primary fill 122A
4. Mid to Late Saxon shell-filled ware (MLS 4). Cookingpot.
Heavy edge/face wear, soft, laminated.
Secondary fill 122
5-8. Mid to Late Saxon sandy ware (MLS 2). Nos. 5-6 are
large boss-decorated jars. On no. 5, the bosses are large
and repousse, on 6, incipient and applied. All pots
hand-made mostly knifed internally with varying
degrees of vertical external knifing, followed by patchy
14 The full rationale for both appears in The Archaeology of Canterbury, vol. v, Part
2, (forthcoming), but see A. Mainman, 'Studies of Anglo-Saxon Pottery from
Canterbury', in (Eds.) I. Freestone, C. Johns and T. Potter, British Museum
Occasional Paper no. 32, (London, 1982), 93-210.
107
PAUL BENNETT
FEATURES CUT BY PIT 122 :
PIT 129 : ,·: ·:--'
PIT 122:
PRIMARY
FILL 122A:
SECONDARY
FILL 122 :
6
I •
·, \ r
r .
.
--:,, - -·
i&+.M
10
),IJ/-Js..-•l.\1, z ;: 11 12
11l0tl•fll \ttl
'IIIIOflM JttfNT. lfflUOU\1 o, Wf.U«!M 1
MtlfHO l,Uffl •,HI ,
M.U.. lOH AUii)( •.CIM•tlO
CANlEROOOY
t