RECENT DISCOVERIES AT RECULVER
By BRIAN J. PECLLP
INTRODUCTION
OE the Roman Fort of REGVLBIVM, situated when buUt perhaps a
mile inland from the Thames estuary, barely half remains today ; the
rest and the land to the north having been washed away by the sea
during many centuries. This constant erosion has also been responsible
for revealing, both in the past and at present, large numbers of antiquities
in the debris of the destroyed cliffs.
Work of a rescue nature carried out by the author between April,
1952, and December, 1954, resulted in the recovery of much such
material and the gleaning of useful information, the recording of which
forms the subject of this report. As the finds were not restricted to
the vicinity of the fort the work has, for simplicity, been divided into
two parts, each dealing with specific areas. A careful study of the plan
(p. 168) should provide readers with a useful guide.
The writer wishes to extend his sincere thanks to the staff of the
GuUdhall Museum, London, for many generous services : in particular
to the Keeper, Mr. N. C. Cook, for his invaluable help with the preparation
of the report, for the note on the pre-historic finds and for numerous
kindnesses and privUeges about the Museum ; to Mr. R. Merrifield, for
his able work of cleaning and dating the coin finds, and to Mr. and
Mrs. Noel Hume for much valuable help and encouragement. Thanks
are also due to my friends Mr. D. Mould of Keston and Mr. D. Hicks,
Ministry of Works Custodian at the Reculver Ancient Monument, for
theh help on various occasions.
SUMMARY OE RECORDED FACTS
From the Notitia Dignitatum we learn that the Fort formed part of
a system of coastal defences known as the Litus Saxonicum (Saxon
Shore). The system incorporated at least nine shore-forts in Britain
and was founded late in the thhd century A.D. to thwart Saxon phate
raids on the mainland. The fort at Reculver was constructed on a low
hiU guarding the northern mouth of a narrow channel which in Roman
times flowed between the present-day Isle of Thanet and the mainland.
Another fort, Richborough, was constructed at the channel's southern
entrance for a simUar purpose.
167
00
KfcCULVt^
*oo too 300 *oo MM
I Ot UXM.O.T - VOW o * WCH VM1&K tftkK
OP exmuty no*s
SPOT Htl&MT* S^OWU 1UUi- »2« («**•)
PROXIMITY Oh BMlErLY'5
tlSURMS* MJD VDUUDAT10U51
• #
. k f t a a j
ucovtdts n J
U.W.M.O.T
L.W.W.O.T
& * * & OP CU*JCfcU7
fc*t,oM
- j p ^ - ' i
*"««« if*
j s s »<
/SS/'""
,,.„tlll''
DRMU&Cfc
LOW LYIUO r J i S T U t LAUD
( S U t OF-' UfclOl&l kMCAOCKCE)
[Crown Copyright Reserved
Ground-plan of Reculver, Kent, showing location of discoveries 1952-54
RECENT DISCOVERIES AT RECULVER
Of finds in the locality coins representing almost the enthe Roman
occupation of Britain and a small number of Republican and Ancient
British coins have been recorded. Pottery evidence affords us a
somewhat closer dating. The Rev. J. Battely, writing late in the
seventeenth century (Antiquitates Rutupinae) records large quantities
of relics being uncovered to the north of the fort. He mentions in
particular pottery of second and thhd century date being recovered
from rectangular sumps or, as he knew them, " cisterns." Mention is
also made of " foundations of great bulk " being revealed complete with
hypocaust system, and probably representing the remains of a bathhouse
or villa. The impression given by his work, and further confirmed
by material evidence, is that a fixed form of settlement had existed to
the north of the fort, in the second and thhd centuries A.D.
Since Battely's time not a few attempts have been made to establish
the early history of Reculver,1 yet significant results were not forthcoming
until 1951, when Mr. F. H. Thompson, working on behalf of the
Ministry of Works, cut a section through the surviving south wall of
the fort. In his summary of results2 he reports finding slight yet
distinct traces of a native population under dhect Belgic influence and
with lingering Iron Age A traditions. An unknown length of time had
then elapsed before the construction of the fort, indicated in his section
by a mortar-mixing floor. The upper layers had contained a few
insignificant potsherds of shore-fort date.
THE AREA WEST OE THE FORT
The local geological formation consisting of sedimentary rocks has
a covering stratum of Thanet sand.
The secondary layer is sandstone ; the coastal portion of this
constitutes a wave-cut platform when the less resistant layer above is
removed by sea erosion. This platform is itself being steadUy denuded
(Fig. 1).
It was whUst examining this particular part of the foreshore (roughly
300 to 900 feet from the fort) that the writer noticed a number of
neatly cut rectangular pits (Fig. 1, Stage 2). It was apparent that these
represented the lower portions of wells, the upper sections of which had
been removed by erosion, and whose original depth would have been in
the region of 14 to 18 feet (Fig. 1, Stage 1). Each had originaUy been
sunk to penetrate the sandstone layer thus tapping the water-table.
The depths of these remaining portions vary according to theh position
1 For useful summaries see Mr. R. P. Jessup's paper in Antiquity for June,
1936 ; and V.C.H., Kent, Vol. 3, p. 19.
2 Arch. Cant., LXVI, 52.
169
RECENT DISCOVERIES AT RECULVER
in relation to M.H.W.M.1 Of the six examined2 all were lined at the
bottom with a layer of clay ; presumably deposited as an extra precaution
against any percolation which might occur. Five contained,
or had contained, a filling of a tacky vegetable matter and four an
accumulation of pottery and other refuse. This sUting indicates, and
is confirmed by the pottery evidence, that the wells had remained open
for some considerable period. The same five had remained in a water-
STA6E 1 STAGE 2 STAGE 3
«*|iti|dttliiw'«n\'iiiltw>i\t|
"Ju QRlGINAL LAND --SURFACE N ia
\ THANET SAND\O
\\X£ ^ \
\
\ \ \ " N
WATER M.H.W.M WAVE CUT PUATFORM
w / V7, ,
S A N D S T O N E z Fig. 1. Diagrammatical Section through the cliffs west of the Roman Fort at
Reculver, showing effects of sea erosion.
logged condition since theh final filling and it is this which accounts
for the remarkable preservation of certain organic materials such as
leather, wood and bone.
The analysis of the pottery finds show at least three weUs were in
use during the latter part of the second century, and a fourth at a
shghtly later date. Of these, one had been filled towards the end of the
second century, whUst the other two were stUl open in the fourth.
It was not possible to date the remaining weUs with any accuracy.
It is important to note that the majority of the pottery finds from
these weUs belong to the later Antonine period and early thhd century.
1 Mean High Water Mark.
2 Other wells are known to have existed in this area.
170
RECENT DISCOVERIES AT RECULVER
THE WELLS1
Wells A and B. Situated 580 and 630 feet respectively west of the
fort and in ahgnment with each other and C. Each was approximately
40 feet below M.H.W.M. and some 70 feet from the actual
chff face. They were rectangular in shape, measuring 3 by 4 feet ;
both contained traces of the clay lining and thick vegetable sUt,
indicating a prolonged period of use. No finds recorded, owing to
theh washed condition. Both totally erased by beach-surface erosion
by August, 1954 (Fig. 1, Stage 3).
Well C. Position 740 feet from the fort, and measuring 3 by 3 feet.
Depth 10 inches. The only find recorded was a single sherd of late
second-thhd century Castor ware found adhering to its side.
Well D. This well, about 810 feet west of the fort, and 15 feet
below M.H.W.M., formed a perfect 5-foot squa.e (horizontally).
When excavated in December, 1953, its over-all depth was 26 inches,
but when re-opened in the May of the foUowing year it was barely
18 inches deep, giving an indication of the rapid rate of beach-surface
erosion. Excavation, although hampered by tide and subsequent
shingle movements, secured a quantity of interesting material. The
large pottery deposit, dating from the mid-second century and including
some thhd and fourth century types, could be chiefly assigned to the
second half of the second century. Other finds include leatherwork
(parts of a sole and belt) and numerous animal bones.
Well E. Situated 840 feet west of the fort. Two feet above
M.H.W.M. and 24 feet from the cliff. It measured 3 feet 6 inches by
4 feet, and was 42 inches deep. The clay lining was again present, but
the silting so characteristic of the others was missing. This suggests
that the weh was only open for a short period before its final filling.
The only datable find was a complete, though badly worn, Samian
bowl, Form 31, of late second-century date.2
Well F. 880 feet from the fort and 16 feet above M.H.W.M. It
was rectangular in shape, measuring 4 feet 6 inches by 5 feet ; the clay
lining and vegetable deposit were again present. Its depth was just
5 feet. The pottery from it could be dated from late Antonine times,
whilst thhd- and fourth-century types were again present.
There can be httle doubt that these wells are analogous with the
" cisterns " described by Battely as being to the north of the fort, some
400 or 500 feet from the proximity of the current finds.
A seventh well (Well O) was discovered late in 1954, its outline
being easily traced in the exposed cliff face. Situated some 470 feet
from the fort, it differed markedly from the uniformity of the others.
1 Measurements taken midsummer, 1954. Distances given are approximate.
2 This well was emptied by my friend Mr. A. W. Jan of the London Numismatic
Club in whose safe keeping the Samian bowl now rests.
171
RECENT DISCOVERIES AT" RECULVER
It had been dug roughly chcular with a diameter of 3 feet, and measured
13 feet (from the present surface) in depth, just failing to penetrate the
hard sandstone layer. There were no indications of lining or peat-like
sUt. The upper 11 feet of the filhng, consisting of a sandy-brown
deposit, contained but two potsherds of Antonine date, and indicates
that the well had soon faUen into disuse and was then deliberately
filled in. The bottom two feet contained a mixed deposit of the, hon,
animal-bone (including two crushed oxen skuUs), an urhdentifiable coin
and a large accumulation of pottery. This deposit has been dated
A.D. 140-180.
Further indications that this area was extensively used is shown by
the number of rubbish-pits, of irregular size and shape, that have been
revealed by the subsidence of the cliffs. Unfortunately many of these
were destroyed by the action of the sea and theh contents scattered.
A sherd of early second century Samian was, however, recovered from
one (Pit 1), whUst another (Pit 2) produced a smaU amount of Antonine
pottery and corroded iron. From the remaining portion of a thhd
(Pit 3) came material of late first-early second century date. It was in
such a pit along this same stretch of shore that the clay figurine of a
hooded dwarf was discovered in 1949.1
At another point some 750 feet from the fort, a complete coarse
ware vessel (Fig. 5, No. 29) was exposed in the chff face. Located at a
depth of 2 feet 6 inches, it could not be placed in association with any
regular pit or guUy. It was standing in an upright position and
contained a deposit showing signs of calcination which suggests that
it had formed part of a cremation burial. It can be dated with confidence
to the mid-second century A.D., when paraUel types were being
produced at kilns near Gravesend. Other complete vessels recovered
from this part of the beach in recent years may represent burials of a
similar nature.
It is significant that no foundations have been revealed in this area,
although tile and other debris, including pottery, mostly Roman, is to
be found hberaUy strewn around in the top soU. A shaUow hearth,
devoid of finds, was however noticed.
With the constant erosion of the land it is not unnatural that the
majority of the finds are listed in the unstratified category ; much
material from this source has, however, been badly defaced by weathering
and the action of the sea, and cannot be identified. The floods of
January, 1953, were responsible for the bodUy removal of many feet
of the Reculver cliffs, and it was during the foUowing weeks in particular
that many objects were recovered from the foreshore.
The finds include some 64 coins ; many others known to have been
found by visitors and casual coUectors go unrecorded. As wUl be noted
1 Arch. Cant., LXVI, 86.
172
RECENT DISCOVERIES AT RECULVER
from the appended list (Fig. 2). the coins represent nearly three centuries
of the Roman occupation of Britain. It reveals, however, an almost
complete lack of first-century coins. Similarly the late fourth century
is poorly represented, particularly when considering the numbers found
at the Saxon shore fort at Richborough. The greatest incidence lies
perhaps in the second half of the second and early thhd centuries, but
insufficient numbers have been found to make this conclusive.
Coins recovered by
the Author
Ancient British
Vespasian or Titus ..
Trajan
Hadrian
Antoninus Pius
Faustina I I ..
Marcus Aurelius
Julia Domna
Septimus Severus
Unassigned 2nd cent.
Caraculla
Severus Alexander ..
Philip I
Gallienus
A.D.
—
69-81
98-117
117-38
138-61
141-75
161-80
193-211
193-211
—
211-17
222-35
244-47
253-68
—
1
1
2
1
2
3
3
1
4
1
2
I
2
at 1 4^
O
1
—
—
—
1
—
1
—
1
2
—
—
—
1
Coins recovered by
the Author
Claudius I I ..
Victorinus
Tetricus I
Carausius
Unassigned late 3rd
century
Constantine I
House of Constantine
House of Valentinian
Unassigned 4tb cent.
Indeterminate Roman
Saxon Sceatta
Nuremburg Jetton ..
Totals
A.D.
268-70
268-70
270-73
287-93
—
306-37
337-61
364-92
—
—
e. 600
o. 1580
2
ID
,J3 o
2 —
1 —
4 —
1 —
2 1
3 1
4 1
2 —
3 —
4 1
1 —
1 1
52 12
Fio. 2. Analysis of coin finds from the Foreshore at Reculver.
The unstratified pottery deposits provide somewhat simUar evidence
to that of the coins ; a complete lack of first-century material being
again apparent, whUst the Antonine-early thhd century is again well
represented. Comparatively little pottery of the late thhd and fourth
centuries has been recovered from this source, perhaps as it is not so
readily seen as the bright Samian ware1 of which the earlier deposit
was predominantly composed.
THE FOBT
The highest point of the hUl upon which the fort stands is now
crowned by the picturesque ruins of Reculver church, which is some
35 feet above the present beach. The land to the westward drops
suddenly (in the region of the weUs) before graduaUy rising again in
the dhection of Heme Bay. To the east he the marshes, once a wide
tidal channel.
Barely half (about four acres) of the original area enclosed by the
1 Three potters' marks are recorded on this ware.
173 15
RECENT DISCOVERIES AT RECULVER
wahs1 of the fort survives today, despite protection from breakwaters
and sohd concrete aprons. Of the 200 yards of the fort's interior thus
exposed, in section, only about 180 feet now remain unprotected ;
this being the portion nearest the east waU of the fort.
Here, too, the heavy storms of early 1953 caused large portions of
the cliff to coUapse on to the beach. SimUarly, investigation amid the
debris revealed a number of important finds. From the remnants of a
/ \
O
O
o
3
xr-s
y.
Fig. 3. Samian Ware (§).
shaUow pit or gully (Pit 5) came a deposit of Antonine material. This
included Samian pottery Forms 33 and 37, the latter in the characteristic
style of the Lezoux potter CINNAMVS (Eig. 3, No. 3) ; coarse
pottery of the shaUow pie-dish variety ; an undamaged bone-needle ;
some pieces of bronze ; animal bone and, significantly, tile and a
fragment of daub bearing the key impression for the adhesion of plaster.
The deposit indicates that the site was definitely in use during the
Antonine period, whilst the presence of building materials suggests a
fixed form of settlement hereabouts at that date.
1 It is unnecessary to include here a detailed account of the wall's construction
as previous observations are adequate. Cf. Dowker, Arch. Cant., XII, 1 ; Home,
Arch. Jour., LXXXVI, 260.
174
RECENT DISCOVERIES AT RECULVER
From the unstratified deposits came smaU amounts of pottery of
both second and fourth century date.
It is noticeable that very few foundations are visible in the exposed
section of the chff, presumably accounted for by the exploits of later
builders. The remains of a smaU hut with a the floor (bearing considerable
signs of burning) buried beneath its faUen clay waUs, was
revealed. SimUar remains have been located at other shore fort sites
such as Richborough, Pevensey and Brancaster.
Hard by the eastern waU of the fort and beneath the internal
earth-bank a smaU substructure was noticed. It was composed of
hewn blocks forming a narrow floor three feet wide and with low
retaining waUs about 20 inches high, and probably represented the
external corridor of a buUding (perhaps wooden) later destroyed.
Being situated below the earth-bank, and assuming the waU and
bank to be contemporary, it would appear that some form of structure
existed here before the foundations of the fort were laid.
Yet another small pit (Pit 4), some 14 inches in depth and 28 inches
in length, was located in a large chff-fall shortly after the 1953 floods.
It underlay the clearly visible Roman occupation layers by between
10 and 12 inches. From its compact fihing of daub came fragments of
four or five separate vessels of pre-historic manufacture.
This group has kindly been examined by Mr. N. C. Cook, B.A.,
F.S.A., and is dated by him to the Early Iron Age. The fine carinated
beaker in particular shows very close affinities1 to the so-eaUed Halstatt
pottery introduced by the first Iron Age settlers and which is known
from a number of sites in south-eastern England.
The other sherds are clearly of native manufacture, being particularly
coarse and Ul-made ; these in turn bear a relationship to the
beakers of earlier cultures.
The finding of daub in association with this pottery suggests that a
fixed form of settlement had existed here during the fifth century B.o.
We have here, then, ample evidence to support the inference of
" a native population with lingering Iron Age ' A ' traditions."
1 Compare with C. Curwen, Archaeology of Sussex, 2nd ed., 1964, PL XXV,
Nos. 2 and 4.
175
RECENT DISCOVERIES AT RECULVER
APPENDIX T
THE SAMIAN POTTERY
From the area west of the fort :
WELL D
1. Form 33. Orange glaze. External median groove. Height
of waU 52 mm. Cf. 0.
Previous
Previous
A New Coin of the Kentish Rebel Eadbearht Preen
Next
Next