Caesar's Camp, Keston

CAESAR'S OAMP, KESTON By the late MRs. N. PmRoY Fox, B.A., F.S.A. !NTRODUOTION EXTENSIVE excavations were carried out by the writer, with the as&istance of Dr. M. W. Thompson, M.A., Ph.D., F.S.A., of the Inspectorate of Ancient Monuments, during the years 1956-9. A preliminary survey of the Camp was made to decide the best site& for excavation, permission having been given by the landowners-Seismograph Service (England) Ltd.-and by the then Minfatry of Works, who also provided four experienced workmen for a period of two weeks in 1958, under the direction of Mr. J. S. Wa.cher, B.Sc., F.S.A. No previous excavations had been carried out at the Camp. My thanks are due to Dr. M. W. Thompson, for generous assistance throughout the excavation; to the numerous helpers whose hard work made the excavation possible, in particular Messrs. M. Snow, L. Smith, Clark and Hedges; to Seismograph Service (England) Ltd., for permission to excavate, help and co-operation, including the loan of an office; to Professor G. W. Dimbleby, B.Sc., M.A., D.Phil., for visiting the site and carrying out the pollen analyses; to Professor S. S. Frere, M.A., F.S.A., for assistance in the interpretation of the evidence; to Dr. I. W. Cornwall, B.A., Ph.D., F.Z.S., for the analyses of soil samples; Mr. M. Snow, for surveying the Camp and drawing the sections; and to Mr. I. J. Bissett, for drawing some of the pottery. DESORIPTION OF TBE SITE The Camp lies on the Blackheath Pebble-Beds and is approximately 430 ft. A.O.D. at the northern end, rising to approximately 500 ft. at the southern extremity and occupying an area of 43 acres. On the southern slope beyond the Oamp lie outcrops of the Woolwich Beds, Thanet Sands and the Upper Chalk. The Camp h􀃲 long been known, and an inaccurate plan was published by Hasted in 1775. The Camp was surveyed for the Society of Antiquaries by Thomas Milne and engraved by James Ba.sire in 1790, when the defences were almost complete. The engraving was published in Vetusta Mooumenta, vol. iv, pl. x (Fig. 2). Shortly after this survey, William Pitt the Younger, who lived at Holwood, levelled the northern and eastern sides of the Camp, which left only the defences on the western side intact, the southern defences having been levelled at least a century earlier. 186 16 CAESAR'S CAMP, KESTON No finds have been recorded from the interior of the Camp. The defences as shown on the 1790 plan consisted of three banks and two ditches on the western side, two banks and one ditch on the northern and eastern sides; but on the strategic southern sector, which controls the only level approach to the Camp from the chalk country to the south, only the line of the defences is recorded . .An e;x:amination of the ground shows that in the southern sector also the defences consisted of two banks and two ditches. The Camp is strategically sited in relation to the natural tra.ckways, and at least six gateways or breaches are shown on the 1790 plan. Plate I, A and B, show the defences before excavation. THE FmsT SEASON'S EXCAVATION IN 1956 · This was planned to answer three questions: (1) How were the defences constructed1 (2) What was the date of the defences1 (3) Were the numerous gateways shown on the 1790 plan original entrances or later breaches1 THE SECOND SEASON IN 1957 This was planned to confirm the evidence obtained in 1956 and to explore the principal entrance-the west gateway. The site chosen for the 1956 excavation was the south-western angle of the Camp. A section was cut right through the three banks and two ditches which form the defences on the western side of the Camp. The section was 140 ft. long, with a minimum width of 8 􀄅-; but across the inner ditch, which is 29 ft. wide, the section was 12 ft. wide and was cut through 12 ft. of filling to reach a vertical depth of 14 ft. 9 in., in the V of the ditch (Fig. 3). A second section was out through the back of the inner rampart. The defences consist of an impressive flat-topped inner rampart, basal width 56 ft., with an inner ditch 29 ft. wide and 14 ft. 9 in. deep; then an outer rampart, basal width 23 ft., with an outer ditch 22 ft. wide and 8 ft. deep and a counterscarp bank (Fig. 3). The maximum height of the existing inner rampart above the Iron Age turf line is 8 ft., and the outer rampart 4 ft. 6 in. In addition to these defences there are two natural banks and valleys which form parallel outer defence lines on the western side of the Camp. Around the whole site there are protective belts of woodland growing on clay, marshy valleys and sloping hill-sides, which together must have ma.de the Camp an immensely strong fortification. The inner rampart is dominant and is sited to take adva.ntage of whatever slopes the dissected promontory pla.teau provides except on the south side, where the defences cut across the flat plateau, leaving the pro• montory headland outside the defences. 186 CAESAR'S CAMP IN 1790 CAESAAS CAMP HOLWOOD PARK KESTON KE NT SECTION THROUGH THE WESTERN DEFENCES SITE I. 1956 MILES FI Cl I CO UNTERSCARP BANK OUTER RAM PART INNER RAMPART to r·c. 2. WEST EAST 24 18 PERIOD IB PERIOD U 18 12 12 FIGURE. 3 6 6 Ol---"'-----''-----''-----''---''-----''---'---'---J.--, __ , __ .. __ . __ -'----'----'----L----"------'------'---...i..---1----1----l...---l...---l----l----L----L----L---..l---􀀶O 6 12 18 30 36 48 60 66 72 78 90 96 102 108 114 120 126 132 138 150 156 162 168 174 180 186 192 SCALE IN FEET Fros, I, 2 and 3. lfacc p. 186 PLATE I Pholo: Dr. E. V. Piercy Fox A. Western Defences: the inner Ditch bctwoon Ramparts, outer Ditc-h on the left. B. Defences on the Xorth-East. f'hoi,o: nr. F.. I". f'ierc11 Fo:c [ face p. 18H PLATE II Photo: Dr. E. V. Pim:y F'o:r A. West Gateway, showing Flint Revetting, South Side. Phol.o: nr. E. J'. Pitrr-v .Fox B. West Gateway, showing Flint Revetting, North Side. CAESAR'S CAMP, KESTON The principal entrance-the west gateway-is most impressive, and the inturn is greatly exaggerated by siting the gateway in a small natural valley which runs 450 ft. into the interior of the Camp. The excavation has shown that the inner rampart is a complex structure; all the evidence points to three periods of construction designated Period I, Period IB and Period II. Iron Age B sherds were recovered from both sections in 1956. When the sections were completed, the site of the south-western gateway shown in the 1790 Plan was examined The inner ditch was found and followed for 75 ft. This proves that the inner ditch continued beyond the south-western gateway, and crossed part at least of the strategic southern sector. The south-western gateway as shown on the 1790 Plan is most probably not an original entrance. THE 1957 EXCAVATIONS-SITE III The main section, which was cut through the inner rampart and inner ditch close to the principal entrance, the west gateway, confirmed the evidence obtained in 1956 The inner rampart, which is here most impressive, still stands 9 ft. 6 in. above the Iron Age turf line. It has two structural periods: Period I -The building of the first rampart. Period IB-An addition to the first rampart, to make it higher and flat on top. Period II -A major addition to raise the back of the rampart and form a flat crest at least 8 ft. wide. A hearth was exposed on the Period I surface on the inner side of the rampart. Sherds were found and, as in 1956, all recovered were Iron Age B. The pottery is important. The sherds include coarse A-derived wares; curvilinear B sherds from two vessels; other B sherds of notably good quality; but no Belgic wares. Four B sherds were sealed under the Period I turf line, which is itself sealed by the ramparts of Period IB and Period II (Fig. 3). B sherds occurred at all levels in the Period IB and Period II ramparts; and it follows from this evidence that the inner rampart in all three stages belongs to Iron Age B. It is clear that each rampart had a life of its own, for the stages are marked by welldeveloped turf lines-one of which has yielded pollen showing major changes in vegetation. Tri 1958 EXCAVATIONS-SITE IV The third season was devoted to the excavation of the west gateway -the only entrance of the Camp which has survived undamaged. The 187 - 00 00 NORlll WEST GATEWAY. N + I I ______ ,_L NMTlir.i1\QH1 ---- - - -- It=ST ___ I ______ _ -- 􀀃 D- ._..--_ 􀀢+---l-----==.1-l------1----·•_AA __ ,N_f£rr_ _ __..., 􀀉 Flo. 4. CAESAR'S CAMP, KESTON gateway is situated in a :flat-bottomed valley running 450 ft. into the interior of the Camp. The causeway between the ends of the inner ditch is 40 ft. wide, and the flat bottom of the gravel valley here forms a natural trackway some 16 ft. wide. Later wear has hollowed out a, narrower roadway. At the end of 1957, an exploratory section was dug between the lips of the inner ditches, cutting an ancient chalk road and a chalk drain. A grid was laid out to facilitate excavation, and later extended eastwards. The two weeks' excavation of the gateway by Mr. Wacher showed three phases: In the first phase, only the front of the inner bank was revetted with flints, which were laced with a rather haphazard structure of timbers. The ends of the bank forming the gateway seemed in this stage to have been revetted with timber only, and a series of postholes ')a,n be related to this phase. In the second phase the flint revetting was carried round into the gateway; and again a further series of post-holes relating to this phase was found, allied with post-holes set further back in the substance of the bank This would seem to suggest a system of timber ties. In the third phase, the flint revetting was carried fμrther into the gateway, and now enclosed the intum banks-again with timber lacing. Unfortunately, this was only visible on the north side, the south side inturn bank having been cut away by an estate road which ran through the gate. Only the tail of this bank had survived. The gateway was never much more than 15 ft. wide, but appeared to be fairly long. Much of the timber lacing in the banl-: had been burnt; can this be related to a violent destruction1 The excavation was continued with an exploration of the inturns. The most striking feature is the north inturn, some 85 ft. long, with the burnt timbers and severely comminuted :flints of a massive revetting destroyed by fire ·in situ. THE 1959 ExoA.VA.TIONS The grid was extended to cover the whole of the northern inturn, and the area behind the much shorter southern inturn, where an occupation deposit was cliscovered. Sections were cut into the inner rampart on both sides of the gateway, to study the structure of the gateway and the rampart as a whole. The work was concluded in 1960 with the mechanical filling-in and restoration of all the sites. In 1956, a group of Belgic sherds was discovered about a quarter of a mile down the south-east slope of Holwood Hill. In the third year trial trenches on a valley crest inside the Camp cut through a very heavy scatter of Mesolithic flints (Site V). The other side of the valley which lies on Keaton Common, was searched and, in a corresponding 189 CAESAR'$ CAMP, KESTON position, a floor exposed in section was found. From this floor a Thames pick and other Mesolithic material has been found. In 1962 a drainage trench was dug completely across the interior of the Camp from north to south. None of these trenches revealed any Iron Age material. Only Iron Age B sherds have been found associated with the con· struction of the defences, and in the interior of the Camp. TlTI!.D' LINES-SEQUENOE AT CAESAR'$ CAMP, KESTON Site I: Period I Inner Rampart Site II: Site III: Site IV: Period II Period I Period II Period I Period II Period I Period II Phase IA Phase IB Phase IA Phase IB THE POTTERY (Fig. 5.) Inner Rampart Inner Rampart West Gateway 1. Very fine solid paste, with micaceous specks. The surface is black, well-polished, with incised lines and punched dots in a curvilinear pattern. The incising tool has slipped. The profile is very rounded. (Site II, Period II.) 2. Dark brown, quite hard fabric, very roughly tooled on exterior. Interior of lip reddish and well finished. Paste fine, with micaceous flakes and fine grits. (Site IA, Period II.) 3. Ten sherds (two rims, one base), from the rampart jacket. 4. Finely-polished leather brown ware. Fabric, sandy with slightly reddish outer surface. An occasional grit, not quite uniform, softer in one place at least. Grit is vecy fine gravel. (Site IA, Period II.) 5. Large vessel, good reddish colour in places. Vecy smooth fabric, with grit dissolved out. (Site IA, Period II.) 6. The fabric is black with a rather attractive reddish finish. The black core has a similar texture to shell-grit ware, but is all black and very hard baked. The pale terracotta exterior has numerous dissolvedout shell-grits; but it is a hard, fine and quite pleasing surface and 190 CAESA.R'S CAMP, KESTON Fro. o. (i} emphasizes the difference between the site and Crayford. (Site II, Period II.) 7. Rim. (Site IIIA, Hearth, Period I.) 8. Eight brown sherds, very friable grey paste, tooled and smoothed SU1face, with scored pattern. Typical red surface under slip on both sides of fa.bric. (Site IV.) 9. Coarse ware, black hard shell-grit. Very poor, like much of Crayford. (Site IA, Period II.) 10. Polished brown base. (Site IA, Period I.) ll. Polished over a rather coarse fabric. The vigorous curvilinear design is incised or gouged shallowly. Paste is grey, with micaceous specks, and occasional grits. Inside mouth and lip reddish under polished surface. Tool marks visible. Hard and well fired fabric. 0£. A1·ch. Gant., lxxxi (1966), 153, nos. 25-6. (Site IA, Period I.) 12. Base of Iron .Age B ware, of. Site m, Period I. (Site IV.) 13. About forty sherds of coarse ware on brown jacket-rim, score zone and base. (Site IV.) 191 OAESAR'S CAMP, KESTON 14. Fragment of rim-polished thin Iron Age B ware. (Site IV.) 15. Numerous fragments of chevron pot; rim and shoulder. (Site IV.) (For parallels, cf. Arch. Cant., lxxxi (1966), 149-52. Ed.) The sma-11 finds included fragments of whetstones and of a loom weight (Sites I and II); a small piece of silver from a harness or belt, found in brown turf line just below chalk road (Site IV); a fragment of a bronze terret (Site IV}; a fragment from a loom weight, found in the north inturn (Site IV). REPORT oN Son, S.AMPLES FROM RAMP A.RT SECTIONS By DR. I. W. CORNWALL, B.A., Ph.D., F.Z.S. Eleven samples were taken, representing visibly distinct layers in the stratification of the rampart sections (Site II and Site I, the Middle Bank). The material was very pebbly coarse sand, derived from the Blackheath Pebble-Beds on which the site rests. Minerals present are silica, ferric oxide and hydroxide (limonite), and little else. All the samples were completely non-calcareous. pH values and the concentrations of humus and of iron oxides were determined for each sample, with the following results: Alkali-soluble Ir.on, a,s Fe208 Sample No. pH Humus (mg./100 g. (%) Dry Soil) Site II: 8 5·4 1·8 235 7 5·5 0·3 90 6 4·5 2·2 875 5 4·7 0·25 300 4 4·1 2·8 435 3 5·1 0·17 455 2 4·7 0·5 610 M idrlle Bank Site I: 4 5·7 0·4 90 3 6·2 0·02 90 2 5·2 0·34 515 1 4·4 0·05 475 The figures in the table occupy the positions of the corresponding deposits in the sections, those of the Middle Bank representing the surface-soil of the time of its construction (the land-surface before any Iron Age structures). The lower three of Site II, an apparent soil, if 192 CAESAR'S CAMP, .KESTON proved to be so, must have formed on the surface of an early low bank. The remainder represent the modem soil formed on the material piled up over this in the completion of the rampart as seen today. The soil-type throughout is a well-developed podsol, with very acid humus, so that not only water-soluble and alkaline-earth bases are washed out, but even the sesquioxides of iron, manganese and aluminium are mobilized and washed down the profile. The pod.sol is characteristic of initia.lly base-poor siliceous parent rocks under a cool, moist climate, frequently under heath plants or birch-pine forest. The deciduous trees now occupying the site were probably largely planted in the eighteenth century, and are not the natural flora of the prevailing soil-conditions; indeed, many are of species not indigenous to Brit,ain, such as Robinia pseuaacacia among others. Dr. Dimbleby's pollen-investigation will probably throw light on the floral sequence shown by the section. The pod.sol character of the soils is well shown by the experimental results. At the summit we have (no. 8) the modem A1, dark humus-horizon, very acid with accumulated, poorly-humifi.ed plant-remains (I· 8 per cent.) and somewhat impoverished in iron. The iron-content of the parent material of this soil is shown by no. 5, a value of 300. mg./100 g. This is distinctly lower than the 475 mg./100 g. of no. 1 from the Middle Banlt, which is the presumably unweathered 'natural'. This suggests that the make-up material of the reconstructed rampart consists, at least in part, of already weathered and leached soil-material, scraped up from the surface nearby. Sample no. 7 is clearly a bleached .A..z,-horizon, poor both in humus and in iron. Only the humus actually in transit through it is found, and practically all the iron has been dissolved out and carried further down the profile. A marked concentration of both is found again in no. 6, the B-horizon. The decrease in humus and iron in no. 5 shows that we have reached the parent-material of this profile, the O-horizon, material somewhat weathered before the rampart was thrown up. Nos. 4, 3, 2 represent what seems to be a second podsol-profile in the body of the bank (Site II). The pH and humus figures bear out this conclusion, indicating that they a.re A1, A2 and B-horizons, respectively. The iron concentration in no. 2 supports this also, but the iron figures for nos. 4 and 3 are rather high in comparison with the corresponding values for the modem soil. This perhaps indicates that the second profile is less mature than the modern. This would fit admirably with the theory that it was a soil of a few decades, or even a century or so, in age-representing the weathering of an interval between the first construction and that of a major reconstruction and strengthening of the defences. If so, its parent material must 193 CAESAR'S CAMP, KESTON have been more ferruginous than that of the reconstruction, perhaps more like no. I, the subsoil (0-horizon) of the buried soil below the Middle Bank. On the other hand, the relatively high iron contenL in nos. 3-4 could be, at least in part, secondary-due to some percolation of iron-bearing solutions from above. It seems to be established by a 3 points to I majority, however, that this really is a soil-profile (pH, humus and the concentration of iron in a B-horizon at no. 2). In that case, it does not show fully the typical podsol iron-leaching, and so is immature. The Middle Bank samples 4, 3, 2, I, represent a third podsol; this time the mature profile of a soil, for the formation of which some 8,000 years of post-glacial time are available, in existence at the site before any Iron-Age disturbance of the natural soil-regime. The horizons repr(>,aented are the A1 (humus), A2 (bleached), B1 (humus illuvial) and 0, the relatively unweathered Blackheath Pebble-Beds. The ·figures are quite typical. In view of the marked bleaching shown by the iron-tigure.