From Colchester Archaeological Report 1
(iii) The accounts of the foundation and early history of St John’s Abbey
There are four key texts relating to the early years of St John’s Abbey, of which two contain material concerning the relationship between Eudo, Colchester, and the castle; one of these is in the British Library, Cotton MS Nero D viii, ff 22-5. The former has been printed in transcript and translation (Astley 1903) and the latter in translation only (Rickword 1923). (Hereafter these will be referred to as ‘Nero’ and ‘Gough’ respectively.) The Nero manuscript is well known, its credibility having been in the past the subject of considerable and sometimes bitter debate. Dr J H Round’s opinion of the manuscript was characteristically derisive (Round 1922) but Mr W Rye claimed that ‘there is no Monastic Chronicle in existence which is so well substantiated by extraneous evidence as this one’ (Rye 1922, 12; see also Rye 1921).
The truth seems to lie somewhere in between for not only have recent excavations in St John’s Abbey grounds corroborated certain points in the text (see Chapter 4) but also there are three other accounts with which Nero shares a limited but not complete correspondence and of which both Rye and Round were unaware. The relationships between the four texts are discussed below and illustrated in Figure 28.
The castle is not mentioned in the Nero account but the first and third sentences are of particular value to this present study. ‘King William the Younger gave the town of Colchester with its appurtenances into the care of Eudo, who was the royal major-domo… Moreover, Eudo had received this honour from William the Elder in recognition of his own and his father’s devotion to the royal family’. The Gough text is much shorter but Eudo’s association with the castle is explicit. The relevant part reads ‘Now, when Eudo had obtained the custody of Colchester castle, with all that belonged to it…’ (Rickword 1923, 123).
Excavations in the grounds of St John’s Abbey *
A series of small-scale excavations of varying quality took place between 1971 and 1977 within the precinct of St John’s Abbey and in St Giles’s Church. The principal discoveries were a small church predating the foundation of the abbey, some medieval burials, a large quantity of dumped soil, and some burnt buildings with walls of sandy clay. All these can be related to the written evidence.
Little is known of the monastic buildings since the remains of these, shown on Speed’s map of Colchester in 1610, were almost completely destroyed during the siege of 1648 (Morant 1748,2,36-7). A drawing of the southern elevation of the abbey’s church exists (Morant 1748, 2, facing p 36) but is suspect since it shows no traces of claustral buildings. Some foundations south of the abbey gate are indicated in Cutts 1888 (facing p 34) and according to Hull were observed as turf or crop marks (Hull 1958, 295).
Early historical background of the site
The four texts which relate to the early years and foundation of the abbey have been discussed in Chapter 2 with the result that the essential trustworthy details can be set out as follows. The monastery was measured out on 29 August 1095 and dedicated on 10 February 1115. Its founder, Eudo Dapifer, died on 16 February 1120 and was buried in the monastery, his body being brought over from France. In 1133, the monastery and a large part of Colchester were burnt and all the offices and workshops (officine), which were originally on the north side of the abbey church, were transferred to the south side. Although much of the Nero text is of dubious value (pp 28-30), there are two places where it appears to include threads of fact which do not appear in any of the other extant texts but which have been borne out by recent archaeological work, namely the existence on the site of a small church pre-dating the foundation of the abbey and the existence of large quantities of dumped soil. The relevant parts are cloaked in highly coloured and suspect detail, and can be summarized as follows. The first part records that the site chosen for the abbey was where a priest named Sigeric had his house and a church which was built of wooden planking and dedicated to St John the Evangelist. The church, situated on the northern slopes of a little hill, was reputed to be the scene of a miracle and a place where inexplicable voices were often heard and strange lights frequently seen… No further mention is made of the building. The second part of interest relates to the monastery itself. Abbot Hugh, tired of the noise and bustle of the town, decided, sometime between c 1104 and the dedication of the abbey (1115, according to the other sources), to transfer the offices and workshops (officine) and the monks’ dwellings (habitula) from the north to the south side of the abbey church. To do this, a little hill which overshadowed the abbey church was removed and the soil produced spread over the area north of the latter to make a cemetery.
The parish church of St Giles was outside the northern boundary of the abbey precinct. However, this was not always the case for originally it was situated in the monastic cemetery, as indicated in a number of early charters, eg in a document of William of St Mary, Bishop of London 1199-1222 (Cart St J A, 62). Of particular importance in this respect is a charter of Bishop Gilbert of London (1165-71) confirming that St John’s held all of St Giles’s Church which was ‘founded’ in their cemetery (Cart St J A, 87). This implies not only that the church post-dates the foundation of the abbey but that the church also belonged to it. Further corroboration of the last point is provided by a charter of Pope Innocent III dated to 1202 (Cart St J A, 67).
In 1971, several exploratory trenches were dug in the north-east corner of the abbey precinct (trenches Tl to T4, Fig 36). Trenches 1 and 2 were excavated by machine and revealed several metres of dumped soil and debris behind the abbey’s precinct wall. This redeposited material accounted for much of the 5m difference between the ground levels on either side of the wall. In trenches 3 and 4, burials were found which from the associated sherds were mostly late medieval at the earliest. In 1972, the north-east corner of the precinct was lowered by machine for a new roundabout and as a result the inside of the abbey wall was exposed. For much of its length the inner face was well preserved and contained a series of original putlog holes.
In 1973, part of the graveyard of St Giles’s Church was lowered in order to construct a private car park. The deepest area of the contractor’s excavations was in the south-eastern corner of the graveyard where the following observations were made in section. It was clear that the burials continued eastwards under the present graveyard wall and linked up with the series of burials found in trenches 3 and 4 in 1971. The burials were cut through a thick deposit of soil and sand which reached a depth of at least 1-75m and sealed a lime-pit and an original land surface containing spreads of charcoal with soil and occupation debris. Lying on the old land surface and at the base of the dumped material were large burnt lumps of grey sandy clay which derived from the walls of nearby demolished buildings (Fig 36). By a comparison of levels, it was apparent that the foundations of St Giles’s Church were substantially higher than the buried land surface and therefore must have been dug through the thick deposit of dumped material.
In 1975, archaeological excavations directed by Mr N A Smith for the Colchester Excavation Committee (now the Colchester Archaeological Trust) took place in the nave of St Giles’s Church. The area and depth of the work had to be strictly in accordance with the proposed building operations. The northern wall of the original nave was located and found to have been replaced with an arcade when the northern aisle was added, probably in the 14th century. As only the foundations survived of the north wall of the Norman nave, it could not be determined whether the wall had contained a doorway opposite the one in the southern wall. Several burials were found north of the Norman nave but were sealed by the construction level of the 14fh century aisle, implying that the cemetery had initially enclosed the church on all four sides. From this it is concluded that the abbey’s precinct wall had originally extended across the northern side of the church so that St Giles’s lay wholly inside the cemetery. The Norman foundations were found to have been dug into a layer of dumped soil, a discovery which is in accordance with the evidence noted in the church graveyard in 1973 and discussed above. The addition of the aisle in the 14th century was accompanied by the construction of a tower at the west end of the nave. At the same time, the precinct wall adjacent to the church was possibly rebuilt further south (Fig 36). The hypothesis that this part of the precinct wall was rebuilt at this time is based on similarities in the stonework of both the wall and the tower (Cater 1919, 218).
The excavations of 1975 were confined to the nave but, as part of some minor building alterations in 1972, an eastwest trench for a heating duct was dug along the length of the chancel and a straight north-south foundation observed in section by the architect for the works, Mr C M H Barritt. This discovery represents the original east end of the chancel and indicates that the latter must have been rectangular rather than apsidal. The part of the present chancel which lies east of this north-south foundation contains a lancet window in its southern side (RCHM 1922, 43) implying that the east end of the chancel was rebuilt and extended in the 13th century. In 1972, during salvage excavations to the east of St Giles’s Church, a substantial part of another church was uncovered. The work had to be carried out very quickly under considerable pressures and constraints caused by both the rebuilding schedule and redevelopments elsewhere in the town. This in part explains the odd shape of the excavation (Fig 37).
The walls and foundations of the excavated building were of broken Roman building materials, the foundations being simply rubble coursed with layers of sand (Figs 38 & 39). The sand was not in reality decayed, weak mortar but was natural redeposited from the foundation trenches. With the exception of one small section of wall (Fig 37), the building had been demolished down to foundation level and in some places the tops of the foundations had been robbed out. The remains were sealed by a layer of dumped soil and broken building materials which derived from the demolished church. Cutting this deposit were burials, the earliest of which were in stone- and tile-lined graves (Fig 37). In addition the church sealed part of a 3rd to 4th century inhumation cemetery.
The building was a three-celled structure. The middle compartment was almost square and this suggests that it had been a central tower. The junctions at X and Y (Fig 37) were sectioned to examine the relationship between each compartment. The results of this were inconclusive but it was felt that on balance the western cell butted on to the middle cell and that the latter was contemporary with the apsidal cell. If this interpretation is correct, then the building would have been of two phases, the earlier comprising probably a tower-nave with a stilted apsidal chancel. The possibility, however, that the apsidal cell was butted on to the middle compartment cannot be discounted. If this was the case, all the butt-joints may simply be a result of the order in which the various sections of the church were erected. The distinctive method of constructing the foundations of all three cells suggests that they all belong broadly to the same period and that, since no Roman foundations of this type have been found in Colchester, they are all post-Roman. The closest parallel to these in Colchester is the foundations of the nave of St Giles’s Church which consist of broken Roman building materials set uncoursed in a matrix of sand.
The small fragment of wall which survives (Fig 37) indicates that the opening between the western nave and the ?tower was at most 3 0m wide. In plan, the foundations of the western nave were offset to those of the ?tower. This need not rule out the walls of the latter being flush with those of the former. The offset in plan presumably is no more than a result of using narrower foundations in the western nave. Figure 36 shows the walls of the church reconstructed on this basis. Early in 1977, a series of small trial holes was dug along the line of the nave walls to establish their western extent (ie trenches T5, T6, and T7: Fig 36). Unfortunately, the remains of the demolished church proved to be too deep for the scale of the excavation and the project was abandoned. However, it became clear as a result of this work that the soil and debris found in 1972 sealing the foundations of the church increased substantially in thickness to the west and was the same deposit as that observed in the graveyard of St Giles’s sealing the old land surface and cut by the foundations of St Giles’s. Thus from stratigraphic considerations, it seems that the construction of St Giles’s postdates the demolition of this second church and that, as shown in the conjectural
section A-A in Figure 36, the layer of dumped material sealing the foundations of the demolished church is probably the same as that up against the precinct wall. As a result of the various trenches and earth-moving operations, the gound level prior to the deposition of the dumped soil can tentatively be reconstructed as shown in Figure 36.
Reconciliation of the archaeological discoveries with the written evidence
The demolished church found in 1972 is almost certainly that described in Nero although the rubble walls implied by the foundations are at variance with the ‘wooden planking’ unless planks were somehow incorporated in the structure higher up the walls. More likely, however, the wooden planks may have as little factual basis as some other parts of the Nero account (eg the miracles inside the Anglo-Saxon church), and perhaps stem from an attempt to demean the structure of the building to justify its demolition. The burnt walls of sandy clay are to be equated with the monastic buildings originally sited north of the abbey church but rebuilt on the south side of it in 1133 after the fire which destroyed the monastery and a large part of Colchester. The substantial quantities of dumped soil can be attributed to the levelling, as described in Nero, of the ‘little hill’ which stood next to the abbey church although clearly it should be dated to 1133 not to the abbacy of Hugh as the writer of the Nero account thought.
If the ‘little hill’ mentioned twice in Nero refers to the same feature, then it must have been situated to the south of Sigeric’s church and to the east or south-east of the abbey church. In view of the Roman cemetery discovered under Sigeric’s church in 1972, the mound may have been a Roman barrow. Alternatively, a suggestion made by Hull is that the earthwork may have been the mound of a Roman temple since in 1891 a bronze votive plaque dedicated to Mars was found in the area (Hull 1958, 248). However, it is clear from recent excavations that the volume of redeposited soil was so large that the feature from which the material was derived was most likely natural. The land surface today rises gradually to the south, and in 1102 this slope may have been considerably greater. The 12th century earth-moving may simply have been done to create a terrace into the natural slope south of the abbey church for the new monastic buildings. Certainly the description of Sigeric’s church being on the northern slope of a little hill can be considered apt in terms of the natural topography of the area (Fig 36) since from Colchester’s town centre the buildings must have been prominent on the southern landscape.
In conclusion, then, the work of 1971-7 can be broadly reconciled with the accounts of the abbey’s early years. In particular, the discoveries of an Anglo-Saxon church, large quantities of dumped soil of early date, and burnt early buildings between the abbey church and St Giles’s are compatible with the written evidence. In detail, the correlation between the sources, written and archaeological, is by no means complete but the essential elements are consistent.
1 The artist may have drawn the church as he believed it might have looked from the south had there been no such buildings.
2 Little useful pottery was recovered from the main deposits.
3 We are grateful to the Masonic Hall Co Ltd for permission to excavate and to their architect Mr C M H Barritt for his assistance and co-operation.
4 The date of the aisle is taken from RCHM 1922, 43. 5 The doorway could not be detected by an examination of the floor deposits since excavation was not permitted in the relevant part of the nave.
6 I am grateful to Mr Barritt for this information.
7 For a plan of this, see Crummy 1974, 29. These form part of the extensive series of burials which flanked the southern side of the Roman city (Crummy 1975, 9).
8 Tower-naves are rare, examples being Earls Barton and Barnack in Northamptonshire, Barton-onHumber in Lincolnshire, and Sompting in Sussex.
9 ‘The Mount’ is a small mound in the south-east corner of the precinct (Fig 44). It apparently contains much human bone and traditionally is fancifully regarded as being a product of plague or the siege of Colchester in 1648. Its true character is unknown but the mound is too far from ‘Sigeric’s church’ to be the little hill referred to in Nero.