Skip to Content.

Kent Archaeological Review extract

The Discovery of the 'Classis Britannica" and "Saxon Shore" Forts at Dover --
(An interim Report on the 1970 Excavation).
by Brian Philp.


To visitors and residents alike the vast surge of port-traffic through the centre of the town must seem like an eternal nightmare. Happily, a solution in the form of the York Street Diversion is now becoming a reality and it was even possible that work would start on it during 1970. Demolition of schools, taverns, shops and houses has continued at a steady pace on the line of the proposed road. It it unfortunate that the proposed line must cut through and destroy very large areas of the Roman and Medieval Lover. Accordingly, at the invitation of the New Dover Group and on behalf of the Ministry of Public Building and Works, an extensive programme of rescue-research work was launched in 1970 under the CKA emergency-scheme. This involved a very large number of volunteers, a full-time team of supervisors and mechanical aids. The latter included a variety of mechanical excavators in use throughout most of the excavation and these were supplemented by bull-dozers and a fleet of lorries. Attention was confined largely to the central section of the new road where larger areas were available for excavation and where there would be total destruction of all archaeological deposits. Although important Neolithic, Saxon, Medieval and later deposits were revealed, this report deals only with the Roman military aspects of the site. A second phase of work, on behalf of the Department of the Environment, will be undertaken during 1971.


The work in 1970 proved to be, with one exception, a splendid example of co-operation, help and encouragement by many different organisations, groups and individuals. Substantial financial help came from the Ministry of Works, Council for Kentish Archaeology, Dover Corporation, Kent County Council, Marc Fitch Fund and by private donation. The helpful interest and the considerable and spontaneous support of Dover Corporation and, in particular, of its Engineer's Department, must be publicly acknowledged. Next, to the Reculver Excavation Group who not only cancelled its own major excavation, but who provided the main-stay of the volunteer force and most of the equipment and tools. Other tools and equipment were kindly provided by the CKA.

Thanks, too, must be given to the Faversham, Fawkham, Otford, Ruxley, Sittingbourne, Springhead, Canterbury University, Upchurch and West Kent Border Groups for sending contingents of volunteers under the umbrella of the CKA emergency-scheme. Finally, thanks are due, last but not least, to the New Dover Group for its support and help throughout what proved to be a long and difficult operation.

Of more than 140 people employed on the various sites special mention must be made of those who worked for extended periods. In particular, the deputy director, Mr David Ellwood and the main area-supervisors Gerald Cramp, Edward Connell and Ralph Mills; the site-supervisors Miss E Mynott, Gerald Clewly, Alan Gidlow, Roger Walsh and Cliff Ward. Of the volunteers Mrs J Clithero, Mrs N Roberts, Mrs C Robertson, Miss K Fielden, Miss P George topped the ladies' list; Messrs J Clithero, D Crellin, H Davies, I Deeks, D Garrod, R Garnett, J Gaunt, D Harrington, A Mills, C Read, A Ross, D Thomas and J Williams; all must be sincerely thanked for their very hard work under difficult and dangerous conditions (plate 1). Grateful thanks are also offered to Dover College, Dover Grammar School, Queen Elizabeth's School at Faversham and Bishop Thomas Grant School at Streatham. for sending contingents to help with the work.

The officers of the CKA kindly manned an exhibition of finds to which more than 2,000 members of the public were given free admittance and guided tours of the site. Special guests included the Mayor and Council of Dover, Norman Cook and David Kelly. The highlight of the programme was a protracted visit by Sir Mortimer Wheeler who was himself directly concerned with a study of the Roman military aspects of Dover more than 40 years ago.

DIAGRAM: Outline of Queen Street area showing Roman buildings and defensive walls.

Figure 1. Outline of Queen Street area showing Roman buildings and defensive walls (Miss Pamela George).

The "Saxon Shore"Fort (Figure 1).

The Notitia Dignitatum, a civil and military list, probably compiled in the fourth century AD, lists nine coastal forts under the command of the Count of the Saxon Shore. Eight of these forts, including that at Reculver, were identified with extant remains many years ago at sites ranging from Norfolk to the Isle of Wight. Only one, that recorded as Dubris, which must have been at Dover has never been conclusively identified. Following some observations by a local man named Amos, it was possible, in 1929, for Sir Mortimer Wheeler to offer a tentative outline of a fort on the north side of Queen Street and covering part of the Market Square. Several excavations after the Second World War, aimed at proving the suggested fort, failed totally and by the last day of June, 1970, the enigmatic fort was once more struck from the records. Two days later the CIB corps arrived and within half an hour of starting the south wall of the Saxon Shore fort was finally located.

PHOTO: Excavations by men, and machine starting in the Queen Street area.

Plate 1. Excavations by men, and machine starting in the Queen Street area.

More than 13 metres of the south wall (Plate 2) were exposed and several important sections recorded on what had been a car-park site. The wall was 2.42 metres thick and built of squared tufa and chalk blocks set in a hard white mortar. It was badly cut by graves and pits of medieval and later date. The wall ran roughly parallel to Queen. Street and it must be the same wall as seen in Market Lane in 1915 and at another time in Goal Lane, both to the cast. It is thus now possible to trace the south wall of the fort for a minimum distance of about 120 metres.

PHOTO: The south wall of the 'Saxon Shore' fort .

Plate 2. The south wall of the "Saxon Shore" fort (2 metre rods).

As excavation continued on the adjacent burial-ground site to the north, it became clear that about 20 metres of the west wall of the Saxon Shore fort had crossed that site from north to south. The wall itself was almost totally robbed, but a broad foundation-trench containing rubble foundations and identical mortar showed its exact alignment. Equally significantly, several feet to the west and parallel to the robbed wall was a huge V-shaped defensive ditch dug in late-Roman times and silted and almost invisible by late-Saxon times. There can be little doubt that this ditch, possibly one of a pair, related to the Saxon Shore fort in the normal way.

A short section of the west wall of the fort was eventually found on the car-park site and this fixed the position of the south-west corner of the fort. It thus gave the west wall a minimum length of about 45 metres.

In addition to the south and west walls of the fort and the wide defensive ditch two bastions were also found. At the south-west corner the south wall was found to project several metres beyond the external face of the west wall. The exact arrangement here could not be determined owing to the presence of surviving buildings, but it is clear that some form of external projection forming an integral part of the fort had existed. There can be little reasonable doubt that this was an external corner bastion as found at several other Shore fort sites. About 30 metres along the west wall was the massive chalk and flint foundation of another bastion, probably semi-circular in plan, sitting on the lip of the ditch and clearly added to the original wall.

No clear trace of related masonry buildings was found within the small area of the fort's interior which was examined. There was, however, the outline of a circular hut of wattle and daub and also deep deposits of domestic rubbish, including pottery and large numbers of coins, the latter generally in very poor condition. It seems reasonably clear that the fort was constructed in the third century AD, but exactly when must await a detailed study of the many stratified finds.

It is clear that the south and west walls of the fort were not built to meet at right-angles, but to intersect at an angle of about 100 degrees. From this it seems certain that the fort was not planned as a rectangular structure and this may account for some of the earlier difficulties in tracing the defensive walls. The shape of the fort was probably trapezoidal and this would anyway fit into the the typological sequence of shore-forts. The earlier forts in the series, at Reculver and Brancaster, were perfectly rectangular in plan, whilst the later forts included trapezoidal and irregular forms. The two trapezoidal plans, at Burgh Castle in Norfolk and at Bradwell in Essex, are immediately reminiscent of the implied plan at Diver. The bastions too, absent at Reculver and Brancaster, are again features of the later forts in the series. On purely typological grounds, therefore, a construction date in the second half of the third century seems most likely for the "Saxon Shore" fort at Dover.

The Classis Britannica Fort (Figure 1)

Over the past 200 years many traces of Roman Dover have come to light. There is evidence of Roman buildings, mostly with chalk-block walls, at different points and some evidence of road-metalling, but never has any of this fallen into a coherent plan. Particularly interesting have been more than 40 fragments of tile stamped with the initials CLBR (or variant), representing the CLASSIS BRITANNICA, or British fleet. Similar stamped tiles have also been found at eight other sites in Kent and Sussex (KAR Number 21, page 25). The presence of the tiles at Dover had suggested to some that Dover was one of the bases used by the fleet.

DRAWING: Tile-stamps of the Classis Britannica (CLBR).

Figure 2. Tile-stamps of the Classis Britannica (CLBR).
(Drawn by Mr. D. M. Grifilths.)

As could be expected more Roman masonry buildings (Plate 3) and roads were located during the excavation and large numbers of stamped tiles were recovered (Figure 2). It was at once clear that all the buildings and roads pre-dated the "Saxon Shore" foundations, which, on both the south and west sides, had been dug through everything almost regardless. For the first time, owing largely to the very extensive nature of the excavation, it was possible to see that these earlier roads and buildings formed a regular rectalinear pattern. At least six major structures can now be identified, all with chalk-block walls, clay floors and some with internal plaster. Flanking and dividing the buildings were at least four roads, some re-metalled at least six times and often containing a succession of drains, gutters and water-pipes. The whole complex was at once reminiscent of the layout of the military buildings at Reculver and on an equally large and impressive scale. Finally, towards the end of the excavation a defensive wall (Plate 4), delimiting the roads and buildings on the north side, was uncovered for a distance of more than 35m. This wall, built of mortared chalk and tufa blocks was just 1.17 metres thick and fronted by a slight V-shaped ditch. At the west end was a small guard-room (Plate 5) partly destroyed by the great ditch of the "Saxon Shore" fort. It now seems certain that this complex of buildings, roads, defensive wall, ditch and guard-room all formed part of a hitherto unknown Roman fort. The very numerous stamped tiles found in direct association make it virtually certain that the fort was built and occupied by the Classis Britannica.

PHOTO: A section of one of the Roman buildings relating to
the Classis Britannica fort.

Plate 3. A section of one of the Roman buildings relating to the Classis Britannica fort.

As regards dating it may be said, provisionally at least, that this fort was constructed during the second century AD. Several periods of rebuilding are represented, but precise dating must again await a detailed study of the finds.

Of special interest was a small bronze hand holding between its thumb and first finger a circular ball, or orb, upon which perched a rather subdued version of the imperial eagle (Figure 3). This "victory" symbol may have been mounted on a frame or wooden stall and used on ceremonial occasions. It was found by Ralph Mills in a second or third century destruction level and almost certainly related to the Classis Britannica installations.

DRAWING: The Roman bronze hand holding the Imperial eagle.

Figure 3. The Roman bronze hand holding the Imperial eagle (height 16 centimetres).
(Drawn by Miss Rowena Chatters.)

PHOTO: Detail of the external face of the Classis Britannica fort-wall.

Plate 4. Detail of the external face of the Classis Britannica fort-wall.


The results of the excavation seem to have justified the large amount of time, energy, effort and money put into the operation. In addition to the two Roman forts even earlier Roman foundations and features were found over a wide area, but their precise function and date have yet to be determined. The excavation revealed, for the first time, clear evidence of Neolithic and Saxon settlement and also medieval and later structures.

PHOTO: The foundations of the guard-room of the Classic Britannica fort.

Plate 5. The foundations of the guard-room of the Classic Britannica fort.

The discovery of the Classis Britannica fort, with its internal roads and buildings, proves the existence of a major naval base at Dover during at least the second century AD. This must have stood adjacent to an adequate harbour with extensive quays and, perhaps, repair yards. Indeed, evidence of such installations has been found over many years to the cast of the Market Square. Topographically the site was, to a large extent, ideally situated at the mouth of a river flowing into a wide bay protected by high cliffs on each side. Add to this the shortest crossinging point from the Continent and the site is at once unique. Then consider the massive medieval and 19th century fortifications and the long history of this worldfamous port and the presence of major Roman military installations becomes almost predictable. That Dover was the headquarters of the fleet in Britain there can now be a little doubt and the presence of the great masonry light-houses above each flank seems to add the final clue. How could the Romans miss "The Key to England (Britain)"!

The modest defensive wall and ditch of the first fort were probably never intended to withstand assault from what must have been a friendly shore. These were mere "token" defences much like those of the Roman fort at Cripplegate, London and indeed, the similarity may not end there. Certainly elsewhere the form of military installations was kept to a rigid plan and it seems that not even the marines were allowed to deviate.

That the Classis Britannica fort was largely or entirely superseded by the "Saxon Shore" fort is now beyond all reasonable doubt. That the construction of one coincided with the destruction of the other is, however, far from clear. Only more work can tell us this. What is clear is that only a segment of the original fort was enclosed within the new defences which perhaps ran rather closer to the harbour installations. The new defensive wall was massive and to it were added bastions and a wide defensive ditch. Not now the "token" defences of the second century, but earnest fortifications in many ways typical of third century forts elsewhere. Nor any clear trace of substantial internal buildings, but a simple circular hut and the usual great mass of domestic rubbish. Just when and how this fort expired is a problem for the work in 1971.

In the meantime serious thought must now be given to the eventual home for the 20,000 finds recovered from the excavation. The prehistoric pottery and flint implements; large quantities of Roman pottery, imported and local fabrics; coins, stamped-tiles and many small-finds; Saxon loom-weights and pottery and important groups of medieval material. All this now forms a unique collection and one which must surely always remain at Dover. The present small museum is already bursting at its seams and what better opportunity than now to create a fine, new archaeological museum worthy of this famous town. Stop the passing pilgrim in his tracks and show him clearly that Dover, then as now, is truly the "Gateway to England."

Close Window
Accessed this page via search engine or bookmark? Full K A R Index here.