This then is the context for the ongoing research programme on the granges directed by Grenville Astill (see Astill 1994 for a summary of what follows). Bordesley’s estates were compact and were located within 35km of the precinct; they were grouped around 20 granges, a pattern that was established by the early thirteenth century (Figure 20). A few granges were leased in the later fourteenth century, but the majority were kept in hand until the mid-fifteenth century. Two granges, Hewell and Bidford-on-Avon, were retained as home farms until the Dissolution. The granges were located in three distinct zones which had a distinctive landuse and settlement pattern. The majority, some 14, of the granges were on the clay lands of the woodland pasture areas of north Worcestershire and Warwickshire - Arden - where a dispersed settlement was the norm. Three granges were sited on the river gravels of the Avon valley in south Warwickshire - the Feldon area - which was a champion, cereal producing zone dominated by villages. The last three granges were to be found on the lower slopes of the Cotswolds, an area usually associated with the rearing of sheep, but which also produced cereals. Through its granges Bordesley Abbey, then, was theoretically able to draw upon the resources of three different regions or pays. To test this hypothesis it is necessary to establish how the granges related to the precinct, and the extent to which the granges were keyed into the local settlement pattern and economy, and how this relationship changed through time.

The research plan is to identify the functional differences between the granges using a three-part methodology. Firstly, the granges and their associated lands have to be identified using a combination of place- and fieldname study and maps, usually of eighteenth- to twentieth-century date. The extent of tithe-free areas delineated in the tithe awards of the early nineteenth century often betrays the existence of a grange and its associated, twelfth-century, demesne. However, this information exists for only eight of Bordesley’s 20 granges; general topographic information from charters and estate maps has to be used to find the general location of the other 12 granges. With no extant cartulary, the means by which the lands were acquired has to be reconstructed from sporadic charter collections which are mainly of thirteenth-century date. These demonstrate that virtually all of the grange lands had been assembled from previously cultivated holdings; very few monastic holdings were the result of assarting. The fifteenth- and sixteenth-century leases which survive in the records of the Court of Augmentations in the Public Record Office provide some of the most detailed descriptions of the granges. The abbey’s lease of its property to a lessee will often detail the extent and condition of the lands and the services attached to them and sometimes an inventory of the buildings: leases survive for 14 of the granges.

The second stage of the programme is the location of the granges on the ground. Inevitably some of the granges have been destroyed by later development: the Birmingham conurbation, for example, has swallowed three of Bordesley’s grange sites (Houndsfield, King’s Norton and Kingsuch). The rest survive in a variety of forms which reflect the late monastic and post-Dissolution history of the granges. Some sites were deserted and survive as earthworks as a result of the partition of the grange lands between lessees which rendered the grange buildings inconvenient for the new holdings, as for example happened at Holway (see below) or Songar. In other cases the grange lands were divided in such a way that the buildings were retained as the farm for one part, while the other lands were worked from a new set of buildings, as seems to have happened at King’s Norton. These examples emphasise that granges, like other types of medieval settlement, cannot be assumed to be a stable element in the landscape. In other circumstances the grange passed entire to a family as an estate, and a country house was built, as happened at Hewell and Combe. While the medieval buildings failed to survive, some of the landscaped gardens around the country house could incorporate the earthwork remains of monastic-period enclosures and water systems (see below).

The third stage is the archaeological fieldwork which is designed to extract as much information as possible about the character and function of the granges from surface survey. Earthwork sites are surveyed and subjected to geophysical and geochemical survey. The treatment of space, the design and use of the water courses and the size of buildings will allow an assessment of a grange’s function, and of course the extended analysis of the precinct earthworks will aid interpretation. Fieldwalking is the main technique used on sites which are under cultivation, and will produce spatial information which can be compared to earthwork surveys in terms of distribution of buildings, activity areas and water courses. Fieldwalking also produces a sample of the material assemblage from the time when the grange was in use. A direct comparison can thus be made of the ceramics (both pottery and building materials), stone and metalwork recovered from the granges with those from the abbey precinct to ascertain how the granges were supplied with commodities. At its crudest, this exercise should make it possible to discover whether goods were obtained from the sources used by the neighbouring settlements or whether the local supply mechanisms were overridden by a monastic system which linked groups of granges with the main precinct. The following brief case studies illustrate in an interim way the kind of information which can be obtained using these techniques.

Sheltwood was one of the granges located close to the precinct, and appears to have been part of the original endowment. In 1291 it consisted of three carucates of land (c.144ha) and a dovecote. From 1369 until the Dissolution the grange was probably leased as a single farm. A lease survives for 1529, when the grange was farmed by William Cook, the third generation of that family to lease Sheltwood. The lease shows there was a mixed landuse, and details the elaborate coppicing arrangments of the woodland and stipulates the access for cattle. While arable land is mentioned, the emphasis appears to be on a woodland economy, heavily based on animal husbandry and woodland management, and this is also clear from the carrying services, which included 13 wainloads of wood to be transported to the monastery. The lease also records the buildings which had to be maintained: a hall with a gable, a bakehouse and a barn; a fishpond is also mentioned (Astill 1994, 545—6).

The site of the grange is presumed to be near Sheltwood Farm, which was built in the 1860s and probably replaced the grange buildings, for a tithe barn was demolished to make way for some cottages in 1864. Sheltwood is one of those granges for which no details of tithe-free areas exist in the 1839 tithe award. The farm overlooks a stream valley and, instead of flowing in the bottom of the valley, the stream occupies a course which has been cut higher up on the opposite side of the valley. Three regularly spaced earthwork banks straddle the valley bottom and act as dams for a chain of three fishponds. The diversion of the stream in order to make more intensive use of the valley is reminiscent of the diversion of the Arrow at the precinct. A remarkably similar example of water engineering has been recorded on the upper stretches of the river Arrow at another Bordesley property, New Grange. Here the river has been diverted into a contour canal on the valley side, and a substantial earthwork dam was thrown across the valley bottom to pond back the water; earthwork platforms at the east end of this dam suggest that the pond was used to power a watermill. After it had driven the mill the water was used to feed a string of small fishponds (Figure 21; Aston and Munton 1976, 24—6).

Holway is located c.12km south-west of the Bordesley precinct on the borders of the royal forest of Feckenham. It was a royal manor in the eleventh century and was granted to the monastery in c.1140. In 1291 the grange had three carucates of land and it is clear that the original endowment had been extended. The lands included large tracts of woodland and arable. Holway grange was first leased in 1323, probably as a single unit, but by 1400 the grange had been divided in two and each part had been separately leased (VCH Worcs 1913, 375—6). A tithe dispute of c.1530 makes it clear that the lands were partitioned, and the new boundary passed through, or very close to, the old grange site, which made it necessary for both lessees to construct a new set of buildings, and much of the land was turned over to pasture. The documentation, then, suggests that the site of the monastic grange was abandoned by c.1400 (Dyer 1991, 56). An estate map of 1731—2 records a tenancy as Hollow Grange, and virtually in the centre of that farm’s land a field called Hollow Court is marked. It was on a south-facing slope at some distance from water courses. The tithe award of 1840 records the extent of the Holway tithe-free lands, and this particular field was there called Little Hollow Court (Astill 1994, 547—9).

Christopher Dyer visited the site in 1982 and recorded earthworks which he interpreted as the site of the grange (Dyer 1991, 33—5). The plan shows a complex series of ditches which surrounded yards and at least nine building platforms. The southern boundary of the grange was marked by a lynchet beyond which was ridge and furrow. The northern part of this field had later been enclosed by ditches which were on a slightly different alignment from the other earthworks.

In 1983 the farm changed hands and soon after the hedges were removed and the land drained and deeply ploughed, and thereafter the site was ploughed regularly. In 1991 permission was given to fieldwalk the site. The majority of the site was walked on a 5m grid to gain adequate spatial control of the material on the surface, and the relict earthworks, areas of stone and differences in soil colour were also plotted. The south enclosure and the area of former ridge and furrow were sampled (Figure 22). A magnetic susceptibility survey and a phosphate survey were also carried out. The survey results confirmed the impression of the earthworks, namely that the grange did not have a regular layout. One stone building was located, which was associated with a dense spread of iron smithing slag, including many pieces from the bottoms of hearths. The building must have been an ironworking site or smithy. Other buildings were detected from distinct spreads of ceramic roof tile, some of which correlated with the earthwork platforms, while others did not, but all probably indicate the site of timber-framed structures. The finds assemblage from the site will offer an important comparison with that from the precinct.

Figure 23 Bordesley Abbey granges: pillow mounds at Combe.
© Bordesley Abbey Project

Fieldwork in the landscaped parkland on the side of the valley opposite Campden House revealed four distinct sets of earthworks which appeared to be small groups of buildings; nearby molehills yielded medieval pottery of twelfth- to fourteenth-century date. Just less than 1km further up the valley, near its head, three further groups of buildings were surveyed. These were located on artificial platforms, and were long buildings of a type which has been recognised as sheepcotes over most of the wold areas of the country (Dyer 1995). A bank running along the top of the valley was interpreted as the boundary of the grange lands. Five pillow mounds, one of which cut through the boundary bank, were also recorded, and these must date to a time when the lands were leased out, perhaps as late as the seventeenth century (Figure 23). Sites of quarries were also recognised. Combe still has a substantial medieval landscape surviving, some of which probably dates to the monastic period, and indicates a commitment to sheep breeding. The relationship between the small building groups and any potential medieval predecessor to Campden House has yet to be established.

Click here to see the image in full
Figure 24 Bordesley Abbey granges: plan of Kington.
(Drawn by Margaret Mathews and Steven Allen) © Bordesley Abbey Project

The present Kington grange is located in the extreme north of the grange lands and is a seventeenth-century timber-framed building. To the west and south of the house are the filled-in remains of a rectangular moat, one of whose sides was approximately 70m long; the present house may have been built in the north-eastern corner of the moated area. While it is possible that the house may contain medieval elements so far unrecognised, it is more likely that the grange buildings were centrally placed within the moated enclosure. A geophysical survey located a possible circular dovecote under the lawn to the south-west of the existing house. An outer enclosure, consisting of a slight bank and ditch, was attached to the western arm of the moat, with an area of ridge and furrow beyond. Kington grange appears to have had the residential buildings within a moated enclosure, with the agricultural buildings located in an outer court (Figure 24; Astill 1994, 549).