The Abbey Church

The investigation of the abbey church has been at the heart of the project and has been directed since the 1980s by Sue Hirst and Sue Wright. The south transept (together with the rooms immediately to the south and the north-eastern portion of the cloister walk), the presbytery, crossing and choir, the eastern part of the nave and the north and south aisles, and part of the exterior eastern cemetery have all been excavated. Monographs detailing work on the south transept and on the east end and crossing/eastern choir have been published (respectively, Rahtz and Hirst 1976; Hirst, Walsh and Wright 1983); the eastern cemetery report is awaiting publication (Hirst and Wright forthcoming).

The structure survives up to 2m high, and the walls of the excavated eastern part of the church have been consolidated for public display (see Figure 4). The archaeological preservation is remarkable. Over a metre of stratified deposits exist in most areas of the church, which represents a sequence of seven separate floor levels and intermediate make-up and builders’ layers, stretching from the twelfth-century preparatory building operations to the final floor and Dissolution destruction debris. The archaeological stratification can be related to the architectural remains to give a detailed history of the building and its use in a way ‘which is probably without parallel’ (Hirst and Wright 1989, 297).

Combining the archaeological, architectural and documentary evidence requires a team effort, and in particular the architectural input from David Walsh and Iain McCaig has been fundamental. The surviving foundations and first few courses of walls, together with the excavated ex situ architectural remains, have been used to help reconstruct the wall elevations (see, for example, Figure 5); and, indeed, sometimes it is possible from ex situ mouldings to identify periods of building that are not represented in the surviving church fabric. This methodology was extended with the excavation of the north aisle and northern part of the nave, where the overburden, essentially the debris left after the demolition of this part of the church at the Dissolution, was dissected on a grid to give a more precise location for all architectural fragments. This has allowed David Walsh to suggest how the elevations of the northern nave arcade changed from east to west. These excavations have also produced some unusual sculpture and decorated tiles (see, respectively, Figure 6 and Hirst, Walsh and Wright 1996; below and Stopford and Wright 1998).

Another example is the excavation of the north-eastern part of the cloister walk which demonstrated that the cloister was redesigned and rebuilt in the early fifteenth century. It was possible to work from the stratification and the plan, and to use the mouldings found in the demolition debris to produce a reconstruction of the elevation (Figure 7 and Walsh 1979); this in turn formed the basis of a display in the visitor centre.

The evidence, then, can give us a sophisticated chronology which not only produces a structural sequence, but also provides an unusually accurate date for the use of decorative elements in the church, such as a particular type of floor tile.

We can only give a brief indication of the range and quality of the evidence coming from the church, and this may be best done by choosing three topics: firstly, information about the building and how it changed; secondly, the treatment of space within the church; and, lastly, the evidence from the burials of members of the monastic community and of the laity.

The changes identified in the church appear to be partly a result of ground conditions and partly a response to changes in the monastic community and its economy. In one sense the church should be regarded as the most sensitive indicator we have of the fortunes of the monastery. Considerable information exists about the construction of the church. There is a suggestion that the first church was of timber, and we can see that the first stone structure was laid on pebble rafts (some of which perhaps had a symbolic as well as a functional purpose) and the walls set out using a consistent module, the ‘Burgundian foot’ (Walsh 1980; Hirst, Walsh and Wright 1983, 222—9). The builders’ levels, drains and systems of scaffolding illustrate the technologies used to build the structure.

The form of the stone church remained basically unchanged from its completion probably in the 1150s. The Romanesque church was of the so-called Bernardine plan, with a plain, square east end and three chapels in each of the transepts (see Figure 5; Walsh 1983; Walsh 1994). The nave was probably of nine bays (and c.40m long), with simple piers, and the austere appearance would have been emphasised by an earth floor covered with reeds. Such simplicity was modified within a few generations. In the first half of the thirteenth century, following the burning down of the timber night stair, a new stone stairway was built, and new internal arrangements in the south transept, south aisle and crossing emphasised a different access to the choir stalls (see below). The church appears to have become more elaborate - ceramic tiled floors were laid, windows were probably glazed and walls were plastered and painted with false jointing.

There were, in total, five main building periods subsequent to the completion of the Romanesque church. For over a century the monastic community had to deal with structural problems caused by instability. The church was built on a hillside and there may have been a spring in the vicinity (Hirst, Walsh and Wright 1983, 49—50). The western part of the church had been terraced into the side of the hill, but not the eastern, downslope, part which may have started to tilt or slip. Great clasping buttresses were added at the northern and southern corners of the east end, and in the third quarter of the thirteenth century the north-west crossing pier was replaced (and perhaps the western and northern crossing arches), and the crossing tower may have been raised. The worst damage occurred in the early fourteenth century when the same crossing pier collapsed, and a large coursed section of masonry fell and was embedded in the floor. The rebuilding which followed included a new north-west pier, connecting crossing arches and probably a new crossing tower. The last major building programme occurred in the early fifteenth century, when the nave - including apparently the clerestory - was redesigned and rebuilt in a coherent scheme, as was the cloister arcade (see above, and Figure 7). While there is evidence of new floors being laid in the later fifteenth century, there does not seem to have been any major building work between the early fifteenth century and the Dissolution. The excavations have also demonstrated the process and sequence of the demolition of the church, particularly for the north aisle and nave (see Hirst, Walsh and Wright 1996; Stopford and Wright 1998).

The building materials themselves have enormous potential for analysis and this is demonstrated by the decorated floor tiles. The tiles occur in situ in floors and grave settings and ex situ in destruction and builders’ levels (see, for example, Figure 8 and Stopford and Wright 1998). They thus represent an exceptionally well-seriated and independently-dated assemblage, and have been extensively studied by Jennifer Stopford (Stopford 1990). By analysing their methods of manufacture and decoration she has been able to assign the tiles from the abbey church and the gateway chapel to production groups; she has also related these groups to tiles from other ecclesiastical sites. The dating evidence for the church floors has been incorporated to demonstrate that the production groups form a coherent, chronological sequence. The neutron activation analyses of the groups has also assisted in the distinction between locally manufactured tiles and more distant imports (Leese, Hughes and Stopford 1989; Stopford, Hughes and Leese 1991; and see below). The tile evidence has also been used to discuss methods of tile manufacture, and how tile industries were organized, and how that organization changed (Stopford 1992; Stopford 1993a; Stopford 1993b).

The well-preserved and deep stratification in the church has ensured the survival of ephemeral features which allow us to see how space was used within the building, bringing us nearer, for example, to charting change in liturgical practice. Such features as the partition or screen of c.1300, shown by three square-sectioned voids at the junction between a dirt and tile floor, and located between the piers separating the crossing from the south transept, give an indication of the quality of the evidence (Figure 9). We can thus see how for almost two centuries the transept and chapels were used as a special area of burial (see below) and how, by the early sixteenth century, the south transept chapels had gone out of liturgical use and had been blocked off, as was the south door (see Hirst and Wright forthcoming). The abandonment of the south transept chapels might be related to a change of use elsewhere, that is, the partitioning of the nave and aisles in order to create small chapels (see Figure 4, above).

The choir stalls also make a crucial contribution to this theme of space. The series of timber slots for short joists which supported the floor and stalls of the twelfth century occupied the first two bays of the arcade; this arrangement would have allowed the choir monks to move through the south transept and crossing and enter the stalls at their east end, the superior introitus, as was usual. In the thirteenth century the first bay of the arcade was blocked and a much more extensive set of choir stalls constructed. These were based on large east-west timbers, with room for two rows of choir stalls on each side and returns at the west end for the abbot’s and prior’s stalls. More remarkably, these stalls extended eastwards so as to occupy most of the crossing, which would have made it difficult to use the superior introitus. Instead, the monks may have entered the stalls at the west end (the inferior introitus) via the south aisle; indeed, a very well-worn path was recorded across the south aisle floor, which survived to the end of the fourteenth century and suggests this remained the principal route throughout the fourteenth century. The stalls were thus at their maximum extent in the thirteenth century and this must also indicate the time when the monastic community was at its largest.

At the beginning of the fourteenth century the choir stalls were again changed. The foundations were now constructed of stone, and the stalls had contracted, leaving the crossing free and occupying only the first two bays of the arcade (with the second bay now also blocked providing a support for the backs of the stalls) (see Figure 10). This must reflect a decrease in the number of choir monks, but might also be part of an increased architectural emphasis on the crossing space, with a tower externally and enlarged crossing piers internally. The stalls were rebuilt again (twice) in the fifteenth century, on roughly the same foundations (see Hirst and Wright 1989, 301—5, for the above two paragraphs).

Lastly, there is the issue of population. Over 100 in situ burials (and a number of other individuals represented by disturbed fragments) have been excavated at Bordesley up to 1994. Those buried inside the church - in the south transept (20 in situ), presbytery/crossing/choir (8 in situ), and nave and aisles (22 in situ) - are normally interpreted as being the lay patrons of the monastery, while those in the eastern exterior cemetery (56 in situ) were in an area thought to be used for the monks themselves. There is, therefore, the opportunity to compare the skeletal data and burial rites of the two different populations. While there are significant differences between the two groups of burials, the explanation may be more complicated than a patron/monk dichotomy. There were, for example, two children and up to three females buried in the exterior cemetery. (The figures in this and the next two paragraphs supersede those in Rahtz and Hirst 1976, and Hirst and Wright 1989; see Hirst and Wright forthcoming).

With the exception of a child buried during the primary construction phase, burial in the church began in the late thirteenth century, and this seems to match the national picture (Coldstream 1986, 157-9). The density of graves in the south transept (22 excavated graves with 20 in situ burials) is interesting. Certainly, during the second half of the thirteenth century considerable building work took place in the crossing area (with the result that the east end would also have been congested), the place of burial usually preferred by patrons. However, the south transept continued to attract burials during the fourteenth century and, in particular, the fifteenth century. The burials also appear to cluster around the north and, especially, the central transept chapels; three of the earliest burials were also found within the central chapel. The clustering might suggest family groups, and indeed there is some indication from the paleopathological data that this is indeed the case, with one grouping at the entrance to the north chapel and the other outside the central chapel.

Thus the south transept appears to have been designated an area for burial from the late thirteenth century and the central chapel continued to provide a focus for burial long after it was used as a mortuary chapel. In contrast, the burials within the presbytery/crossing/choir, while less numerous, appear to have been of a higher status, to judge from the greater proportion of graves that contained coffins (seven out of ten excavated certain graves, compared to seven out of 22 excavated graves in the south transept) and somewhat more evidence for new graves being marked at floor level (Hirst and Wright 1989, 307). There is a variety in the form of burials within the church, where both stone and wooden coffins were used.

The complex and deep stratification allows us to see from what level the graves were cut. Relatively few graves were dug during the life of a floor in the south transept or presbytery/crossing/choir. Burials were more often inserted from builders’ levels during, or more usually at the completion of, the building work, before a new floor was laid. Thus three (or possibly four) graves in the presbytery/crossing/choir area were dug from mid-fourteenth-century builders’ levels, after the reconstruction of the crossing, but prior to the laying of a new tiled floor in this area. Even more striking are the 13 or so burials (nine or ten in the south transept, and four in the crossing/choir) which were inserted during the period of activity, debris and make-up associated with the rebuilding of the nave and cloister around 1400.

The coincidence of burial with builders’ and make-up levels is a marked and unusual feature of the burial pattern within the abbey church at Bordesley. It could imply temporary interment elsewhere and translation to a permanent grave later. Such translation could minimize the disturbance caused by grave digging, and this may have been the intention when a large number of burials were accommodated around 1400.

Another aspect of this burial pattern is the lack of any surviving evidence in the case of many of the burials to show that these graves were marked at floor level, initially and especially subsequently. Even if the archaeological evidence underestimates the practice, it is unlikely that those graves which were originally marked in a floor continued to be so when floor levels were raised and new floors laid. However, the relatively few instances of superimposition or inter-cutting of graves and the generally orderly arrangement of the burials would suggest that the position of burial was remembered when the graves theselves were no longer visible, or even recorded. Some burials of course may have been commemorated on nearby walls.

These two unexpected characteristics of the burial pattern within the church - the cutting of graves from builders’ rather than floor levels, and the apparent absence of grave markers on particularly subsequent floors - sheds important new light on the nature of the relationship between patrons and the monastic community. It may cause us to rethink the common assumption that the monastery was the passive partner in such a relationship: the physical impermanence of the commemoration of burials emphasizes the relatively short-lived nature of the connection between abbey and patron (Astill and Wright 1993a, 132—6).

The eastern exterior cemetery shows more variation in the type of burial than is to be found in the church. Composite coffins (ie, constructed from several blocks) were the usual form of stone coffin; stone grave slabs and headstones also occurred, as did wooden coffins. The earliest graves frequently employed a distinctive form of burial where one or more, large, wooden, east—west covers, supported by short wooden cross pieces, were placed over the bodies; many of these timbers were evidently reused from an earlier structure. One cover (see Figure 11) has a dendrochronological date of 1150 ± 9; it has been suggested that this may have been the burial of one of the first monks, covered with a timber of special significance, perhaps even one saved from the original wooden church (Hirst and Wright 1989, 307—8; see now Hirst and Wright forthcoming).

The Gateway Chapel

The abbey church has for many seasons been the main focus of the project, but the site of the gateway chapel has also been examined (as part of a MSC scheme) and as a result the structural sequence and character of the abbey church can be compared to that from the gateway chapel (BAG, Figure 3, above). The chapel was the only part of the monastery to survive the Dissolution, serving as the parish church of Redditch. A new church was subsequently built in the centre of the settlement of Redditch and in 1805 the chapel was demolished. The first chapel was a single-cell structure which was built in the first half of the thirteenth century (perhaps contemporary with the alterations in the abbey church outlined above). In the last quarter of the thirteenth century, however, the chapel was transformed. The west end was slightly extended, but most effort was devoted to creating a new scheme of elaborate fenestration which was of the highest quality and surpassed anything that had so far been constructed in the abbey church (Walsh forthcoming). The chapel also yielded two groups of decorated floor tiles which were probably made c.1300. One group is highly decorated and represents the most exotic kind of tile that was then available and which, significantly in view of the chapel’s renovation, is poorly represented on the abbey church. The second group is similar to tiles from Hailes Abbey in terms of stamps and designs employed, and similar tiles were used in the abbey church (Stopford forthcoming). Both the architectural and the tile evidence show the high-quality work which was carried out on the chapel, and clearly demonstrate that the chapel was attracting patronage which was comparable to, and indeed at this time exceeded, that of the abbey church.

back to top