Introduction
Marlowe sites
Sub-Roman
Three Saxon phases
Medieval Canterbury


Introduction

Perhaps the most remarkable of the many discoveries made by Sheppard Frere in his long seasons of excavations at Canterbury was the discovery of Anglo-Saxon sunken huts in the ruins of the Roman city. These have become a major part of archaeological mythology, for it was assumed that these were the huts of the foederati and the Anglo-Saxon settlers arriving with Hengist and Horsa, and they have become the prize evidence for that magic word 'continuity' between Roman and Saxon cities. Since 1975, however, Tim Tatton-Brown and the Canterbury Archaeological Trust have been carrying out further excavations in Canterbury and have found many more Saxon huts, and they have been able to show that these are separated from the Roman levels by a period of desertion and that in this part of Canterbury at any rate there was no continuity between Roman and Saxon; and since this was the very centre of the Roman city, it is a reasonable assumption that Canterbury did not survive as a town between the Roman and Saxon periods.

The excavations at Canterbury really date back to 1 June 1942 when the Germans carried out a major air raid on the city, missing the cathedral but destroying a large area of the eastern quarter. Between 1946 and 1960 excavations were carried out in this bombed area by a young schoolmaster from Lancing College called Sheppard Frere, who spent much of his spring and summer holidays in Canterbury. He found not only the Saxon sunken huts but also the Roman theatre and the adjacent public baths. However, much of this area situated around the Marlowe Theatre has still not been redeveloped, and thus when Tim Tatton-Brown began work in Canterbury and the Canterbury Archaeological Trust was established in 1975, one of their major priorities was to carry out large-scale open area excavations in the car parks surrounding the Marlowe Theatre to supplement the earlier trenches of Professor Frere. There is also an extensive series of sites on the other side of the road often known as the 'Cakebread Robey' sites after the first site excavated. However, we must begin at the beginning and look at the Belgic and Roman town for here too major new discoveries have been made.

^top

We can begin with some good news for the continuity addicts, for if they have lost continuity from Roman to Saxon, they have probably gained continuity from Belgic to Roman. It has long been suspected that there must have been Belgic occupation at Canterbury, but now more extensive evidence of occupation has been discovered. Thus on site Marlowe II a round house was discovered with a porched entrance. More significant, perhaps, were the defences that ran through sites Marlowe I and IV, where three concentric ditches were discovered. These appeared to form a staggered entrance to a defended enclosure and perhaps resembled the Belgic dykes that have been found at Colchester, Silchester and other Belgic capitals. The finds, too, were interesting for in addition to imported wheel-made pottery, there was also a clay mould which may have been used to produce coins, though this is controversial. A large number of Belgic coins have also been discovered, notably potin (that is, tin) coins; indeed out of a total of 66 Iron Age coins found in Canterbury in the last five years, 38 are potins, and quite a large number are rare continental coins.

And thus we arrive at Roman Canterbury itself, where we can pause to remark that, then as today, Canterbury evidently had something of a continental atmosphere about it. And its heart was the massive Roman theatre still only known from Sheppard Frere's excavations. Only two other roman theatres are known in Britain, one at Verulamium which is small and late, and one outside Colchester at Gosbeck's Farm at the centre of pre-Roman Camulodunum. (Stop press: Yet another theatre has just been discovered, this time inside Colchester itself.) The theatre did not stand alone but was part of the ceremonial centre of the town. Adjacent to it were the public baths discovered by Frere, while opposite there was a large open area which probably contained a classical temple which is the major Roman discovery of the current excavations. It forms an analogy to the theatre/temple complex which is well known in Roman Gaul but only paralleled in Britain at Colchester.

^top

The temple lay on the other side of the modern road where five separate sites were excavated under the direction of Paul Bennett. Unfortunately the temple itself was not discovered and its presence could only be deduced. The excavations covered three sides of the Roman peninsula and in each case the foundation for an impressive Roman portico was revealed, facing on to an open courtyard. Nothing was found inside the courtyard except on the western side at 69a Stour Street, where a small shrine was discovered. However, many of the excavations produced impressive architectural fragments; some of them were parts of carved corinthian capitals and fluted columns which came from some impressive classical building, and a classical temple would be the obvious answer. There were also many fragments of imported marbles. Unfortunately a large modern office block was erected in the middle of this insula in the dark ages of Canterbury's archaeology, between the departure of Sheppard Frere in 1960 and the arrival of Tim Tatton-Brown in 1975, and we must therefore assume that the temple was either destroyed by this building or possibly – hopefully – it lies under its adjacent car park.

^top

Marlowe sites

But the main sites excavated were in the car parks surrounding the modern Marlowe Theatre, where Kevin Blockley and Marion Day excavated four large areas known as Marlowe I-IV. The Belgic occupation already mentioned began here at the beginning of the 1st century AD and continued for some 20 years after the Roman Conquest. The only sign of the Roman Conquest indeed was a pit which produced a fine group of about 40 tinned bronze horse harness fittings. This clearly belonged to a Roman cavalry man and adds further evidence for a military phase at Canterbury. This is followed by a series of timber buildings, possibly with three wings round a central courtyard, which was probably a series of tenements.

In the Flavian period (AD 69-96), at a date that would fit in well with Tacitus's description of the laying out of towns by Agricola around AD 80, the town centre was formally laid out and a new road was laid down over the top of some of the earlier timber buildings. In this new layout, Marlowe II was occupied by part of the public baths with a portico surrounding it, as indeed Frere had discovered in his earlier trenches. (In retrospect, one of the interesting side lights from the recent excavations has been to prove the value of trench excavations, especially when carried out by someone as perceptive as Sheppard Frere.) Only the corner of the baths could be excavated, occupied largely by a swimming pool and adjacent courtyard. The lead pipe which drained the bath to a large sewer was found still in position.

^top

In the next insula to the south (Marlowe III) the corner of a fine town house was discovered. At first this was merely a series of rooms but in the late 3rd century this was rebuilt to form a private bath house which had a number of unusual details in its heating system. Instead of the floors being supported on stacks of pilae tiles, they were supported on box-flue tiles which would normally have been used for channelling hot air up the side of the walls; having used the box tiles on the floor, the builders then had to devise a new method of heating the walls, so they used some ceramic flue spacers, held out from the walls on iron 'T' staples. This private bath wing was separated only by a narrow lane from the open exercise ground of the public baths, which were also rebuilt around this time when the swimming bath was replaced by a more conventional hot room.

But now we come to the interesting question: what happened at the end of the Roman period? The breakdown begins towards the end of the 4th century, say around 350, when Roman efficiency finally lapsed: the sewer system clogged up. All over Marlowe II a layer of grey silt was laid down, 15cm thick, infiltrating the baths, the portico, the private bath wing to the south, and extending over the main street. However, life went on after the flooding and the portico of the baths became a row of shops whose frontage encroached out over the buried street – a further indication of the decline of Romanitas. These shop fronts were an archaeologist's dream, for lines of grit and dust had fallen between the floorboards and were still perfectly preserved in the silt. However, activity continued down into the 5th century, for the coins continued down to the very end of Roman coinage, and there were even some later levels.

^top

Sub-Roman

However, the best indication of conditions in the early 5th century comes from a burial found on the other side of the road, on the 'temple' site at Adelaide Place. The very fact that a burial was found within the Roman walls is significant, for burials were, of course, forbidden within Roman towns and it must date from a period when Roman order had collapsed. The burial was of a whole family: father, mother, two daughters and the dog, all buried in a deep pit lined with straw or grass, and possibly covered by a wooden structure. The parents were buried in a seated position, the mother held one daughter in her arms while the other lay at her feet, and the dog lay across the father's lap. There was a mixture of jewellery, consisting not only of late Roman bronze, silver, bone and ivory bangles together with some keys, but also some more barbarian objects, such as glass and amber beads that are perhaps continental imports of the early 5th century. Examination of the skeletons by Dr Peter Garrard showed that one of the children may have died as a result of a blow to the head, and we can only surmise that the rest of the family died during one of the many epidemics that swept through the countryside during the late Roman period. The overall impression is of an early 5th-century ritual burial, containing both Roman and Germanic objects and showing a mixture of both traditions at a time when the Roman town was in decline.

Decline was followed by the fall, and everywhere on the Marlowe sites the sub-Roman occupation is followed by desertion. Everywhere there is a layer of black earth suggesting that the sites had become at best gardens, more likely perhaps just a jungle. This humus accumulated for perhaps up to a century and it is into this layer that the sunken huts are dug. The stratigraphical evidence is conclusive, but there are no finds either that can be dated earlier than 550.

^top

Thirty-seven of these sunken huts have now been discovered in Canterbury. Three of these can be eliminated as being not so much sunken huts as cellars of the late Saxon period, but of the remaining 34, 9 were discovered by Sheppard Frere and the remaining 25 by the Trust. The position of these is significant, for all save two were in the courtyards of Roman buildings, notably Marlowe IV, where a large number congregated in the open space in front of a substantial Roman house. The Roman buildings were not robbed out until the 12th century, though there were two exceptions where a sunken hut overlay Roman buildings which must have been demolished in the 4th century; the Saxons must have been most surprised when they dug down and found a nice Roman floor.

An elaborate typology has been established of these sunken huts. The majority are of the well-known 2-post or 6-post type, but there were a number of irregular types including one where there were 10 post holes, situated not in the pit, but surrounding it in an irregular form.

^top

Three Saxon phases

The Saxon age at Canterbury can be divided into three phases. The first is the phase of the sunken huts, though whether this is the settlement that St Augustine found on is arrival in Canterbury in 597 or whether it sprang up after his arrival, is uncertain. The middle Saxon phase is a mystery, for only one structure of this period has been found, a 6-post hut, though there are numerous finds including quite a few 8th-century sceattas. The Marlowe area of Canterbury must therefore have been occupied and we can merely assume that the buildings were all above-ground structures which have left no trace. The late Saxon phase is marked by three further sunken huts, though these are more cellars, being larger, more rectangular and considerably deeper than the earlier structures. There was also a large 'earthfast' building of the 10th century, but the main structures are more elusive, for by this time a new street layout had sprung up which lasted down to the present day, and thus most of the late Saxon remains have been destroyed by medieval and subsequent buildings.

Moving down into the medieval period, we should note that the Roman masonry buildings apparently survived in a ruinous state until the 12th/13th centuries, when they were extensively robbed for their stone: pottery of this date continually appears in the robbing trenches. The major medieval structure to be excavated on the Marlowe site was the Church of St Mary Bredin, one of the 22 parish churches in the city, which was dedicated in the 12th century, totally re-built in 1867 and destroyed in the air raid of 1942. Part of the west end of this church, built and re-built from the mid 13th century onwards was excavated on the M IV site, together with its surrounding graveyard. A fine late 17th century grave slab of Joshua Webster ('some time Merchant in London, tradinge to ye Levant seas: but at the Time of his death an inhabitant in this Parish') was found re-used within the fabric of the Victorian church.

^top

Medieval Canterbury

The main medieval excavations in Canterbury took place elsewhere. There were considerable excavations, for instance, around the Cathedral in the Mint Yard where the King's School have been building a new house. Here the remains of the Almonry Chapel built in 1324 were discovered overlying an interesting sequence of both early and late Saxon boundary ditches. More extensive work has been taking place in Stour Street on the site of the Poor Priests' Hospital which was dissolved in 1575 and then used successively as a Bluecoat School, a workhouse, a regimental museum, and most recently a health clinic. It is now about to be turned into Canterbury museum, and already the first exhibit has arrived – the steam locomotive Invicta built by Robert Stephenson in 1829 for the Canterbury and Whitstable Railway. The archaeologists have not only surveyed the building but also carried out excavations and showed that the main hall pre-dated the advent of the Poor Priests in 1220, and may well have been the hall built by the moneyer Lambin Frese when he moved there in 1174. The whole building with its splendid open hall is clearly going to make a fine museum.

But the major medieval discovery has been in the south-eastern corner of the city where there is a large mound known as the Dane John. This is situated up against the city wall and is today set in the middle of a public park as a result of extensive landscaping in 1790. In popular belief Dane John is the burial mound of a Dane called John, though historians have long suggested that the name is probably a corruption of Donjon and that it is probably a Norman motte. The trouble with this interpretation is that Canterbury already had a fine early Norman castle in the south-western corner of the town by 1086; the keep was discovered by the Royal Archaeological Institute in its summer meeting in 1929 serving as a coke store for the local gas company. Recently the outworks have been excavated by the Trust and the fabric surveyed, showing that it must have been built around 1100.

^top

However, the solution to Dane John came unexpectedly in the excavations between two fine rows of Georgian houses fronting on to the park opposite the mound. Here a very large ditch was discovered, more than 12 feet deep, that is clearly the ditch of the outer bailey of the motte and bailey castle formed by Dane John. The ditch does not appear to have been opened for very long and it would seem that Dane John was established by the Normans almost immediately after the Conquest and abandoned only 20 years later. But once again an apparently unpromising excavation has succeeded in solving one of the mysteries of the history of Canterbury.

Back to Archaeology in Canterbury

^top