38 A maquette of Bishop Odo’s Romanesque cathedral at Bayeux.
Despite the bishop’s many and prolonged absences while performing the roles of Earl of Kent and regent of England, he maintained his involvement in Bayeux cathedral business as bishop. He continued to accumulate property for the cathedral and protected his clergy when necessary. Several members of the Bayeux chapter held land in England. Nine of Odo’s protégés obtained bishoprics either in Normandy or in England and four became abbots. By the 1080s he would have had an impressive network of contacts spread across the Anglo-Norman Church and beyond. Charters also show that during the 1070s he made grants to abbeys within his diocese exempting them from some aspects of episcopal authority.
Perhaps the most unexpected aspect of Odo’s work as bishop was his patronage of the arts, beyond the rebuilding of his cathedral. Undoubtedly, this sponsorship was undertaken in order to increase his prestige and that of his See, and perhaps it is overly sentimental to suggest that it shines a different light on the character of a churchman who was able to compartmentalise his activities to the extent of waging savage warfare at the same time as promoting artistic enterprise. By far the most important example of his artistic enterprises was the Bayeux Tapestry. It would have been in keeping with what we know of the man if it had been commissioned specifically to display at the consecration of his cathedral in 1077. In a year when there were several important consecration ceremonies it would have been a typically audacious stratagem to make Bayeux stand out conspicuously from the other Norman churches by producing his great hanging at the consecration.
Odo’s Church activities extended to England; not so much in his role as bishop, but as Earl of Kent he protected and endowed St Augustine’s Abbey in Canterbury, St Albans Abbey and Rochester Cathedral, all of whom acknowledged Odo as a benefactor. He also played an important role in the translation of relics of St Adrian at St Augustine’s, with a care that is contrasted to Lanfranc’s attitude to the relics of Saxon saints at Christ Church, which some historians characterise as cavalier. There is no evidence that Odo transferred his interest in education to England either, but his protégés from Bayeux were enthusiastic educators. For example, Archbishop Thomas of Bayeux established a schoolmaster at York between 1070 and 1100 and more generally schools were founded at many cathedrals in the following century (Orme 2006, 46–50).
• Odo’s Castles •
Unlike the great Domesday holdings of other tenants-in-chief, Odo’s estates were never formally arranged into baronies. This is probably because he built up his empire very quickly and his land was forfeited before 1086. Nevertheless, it is possible to detect the beginning of a geographical organisation around major central manors which can be seen as proto-baronies. A Kentish barony based on Rochester, a south-Midlands barony based on Deddington and an East Anglian/Lincolnshire barony possibly based on Castle Rising/Snettisham. The chroniclers attribute castles at Dover, Pevensey, Tonbridge and Rochester to Odo, and although he is associated in one way or another with all of them, this list is faulty. Dover Castle was granted to Odo along with his earldom of Kent, but although it was in his hands in 1067 during Count Eustace’s abortive uprising, after that his name is not associated with the site. Indeed, records, both archaeological and documentary, for the castle in the century after Hastings are very elusive.
The great age of Norman keep castles came after Odo, and most of the castles he was involved with were neither keeps nor motte and baileys. They were large open enclosures, as they had been during the Carolingian era, capable of housing large armies. Dover Castle certainly fell into this category, although there may have been two early stone towers there which were demolished in the eighteenth century. Only Tonbridge seems to have originated as a motte and bailey, the construction of which was begun in 1067 by Richard fitz Gilbert. It occupied a strategic site to the north of a bend in the River Medway, overlooking an ancient river crossing. A flat-topped motte was surrounded by a circular moat, with one ditched bailey to the south-east and a second, slightly larger, bailey to the north. Tributary streams to the west of the castle were channelled and diverted to feed the moat. A triangular market place lay immediately to the north-east of the castle. It was this site that Odo held briefly in 1088, after which it was captured and burnt to the ground. Subsequently, fitz Gilbert was pardoned for his role in the Barons’ Revolt and the castle was rebuilt, partly in stone (Kent County Council 2003).
39 Tonbridge Castle, a motte and bailey castle controlling the crossing of the River Medway in Kent. Extract from the Ordnance Survey 2nd Edition, 1895
40 Plan of Pevensey Castle.
William of Malmesbury claimed that Odo held Pevensey Castle, but it had actually been under the control of his brother Robert since 1067. Nevertheless, Pevensey Castle figures conspicuously in Odo’s story – as the landing point of the invasion fleet, as the headquarters of his brother’s castlery, and where Odo was besieged during the Barons’ Revolt of 1088. A Roman fort was built here, with massive fortifications, in the late third century. There was more or less continuous occupation of the 4ha site up to the Norman Conquest, when there was a market town with fifty-two burgesses contained within the walls, a harbour and a considerable salting industry here. The defensive walls were probably strengthened and an earth and timber structure built at the eastern end of the enclosure, thus creating an inner and an outer bailey. This is likely to have been banked and ditched with a substantial timber palisade, which is indicated by the levying of regular services of ‘heckage’, due from local manors. There may have been an early free-standing masonry tower of possible eleventh-century date within the enclosure, but there does not seem to have been a motte here. The inner bailey area contained more than 1m’s depth of dark earth midden deposits in the Saxo-Norman period, before the building of the keep (c. 1200), some of which can be attributed to the Norman invasion forces of 1066. There was evidence of concentrations of latrine waste, which reflected both a greater intensity of human occupation and the use of the general area for penning and stabling animals (Fulford & Rippon 2011, 1–3, 125). Pevensey appears to have suffered badly in the immediate wake of the invasion, and when it passed into the hands of Robert of Mortain in 1067, the number of burgesses fell by half. However, by the time of the Domesday survey it had risen to 110 and a mint had been established, reflecting Robert of Mortain’s establishment of a planned borough here, outside the castle walls. It has been argued that this was the harbour settlement to the east of the fort, but it could also have been at Westham, whose layout has all the characteristics of an early new town. In 1088 Odo and his brother Robert were besieged at Pevensey. The Anglo-Saxon Chronicle records:
The king with his host … learnt that the bishop had gone to the castle at Pevensey; and the king with his host followed after, and besieged the castle on all sides with a very great host for fully six weeks … Thereafter food ran short inside the castle; then they asked for a truce, and surrendered it to the king.
(Lyne, 42–3)
Only Rochester in Kent seems definitely to have been in Odo’s hands. John of Worcester says that ‘Odo carried off booty of every kind to Rochester’ and other writers confirm that Odo swore to surrender his castle at Rochester to William Rufus in 1088. Rochester Castle was built soon after the Conquest, when the city was in Odo’s hands. It stood between the River Medway and the Saxon cathedral on a slight hill, guarding the bridge over the river. The first reference to a castle here was in Domesday Book, which recorded that the Bishop of Rochester had been given land in Aylesford, ‘in exchange for land on which the castle stands’. The Textus Roffensis (c. 1122) enigmatically noted that the land on which the castle stood was said to be ‘the best part of the city’. The first castle was in the form of an earth and timber ringwork, which lay in the south-west corner of the Roman town of Durobrivae, and was comparable with the earliest phases of other important castles such as Exeter, the Tower of London and Winchester. The ringwork enclosed an area of around 1.7ha;
there is no evidence of a motte here, although there does appear to have been a second, somewhat smaller, outwork or bailey to the south, known as Boley Hill. Excavations in the 1970s identified a massive rampart, incorporating a section of Roman wall and a ditch, which was almost 7m deep. Nothing is known of the internal plan of Odo’s castle, but there would have been a wooden hall, numerous ancillary structures and possibly a timber tower (Flight & Harrison 1978). It would have been in this earth and timber castle that Odo and the barons held out against William Rufus in 1088. It was the site of Odo’s last stand in England (Paul Drury Partnership 2009, 12–17). Odo’s castle was replaced by a castle built by Gundulf, Bishop of Rochester in the late 1080s; this in turn was superseded by the great stone keep built by William de Corbeil, the Archbishop of Canterbury after 1127, which stands today.
41 Rochester Castle in the eleventh century (after Paul Drury Associates) and Deddington Castle (from Ordnance Survey 2nd Edition, 1895). The eastern enclosure at Deddington is probably not contemporary with the main castle bailey, which appears to have been constructed by Odo.
It has been convincingly argued that another of Odo’s castles was at Deddington in Oxfordshire (Ivens 1984, 118). Deddington lay at the centre of Odo’s spread of Buckinghamshire and Oxfordshire estates (see colour plate 23) and within one of the most concentrated blocks of his holdings. Furthermore, Deddington, along with Hoo (Kent), Bramley (Surrey) and Snettisham (Norfolk), was the richest of all Odo’s manors, valued at £60 in 1086, and was maintained in demesne by the bishop. The castle lies on the eastern edge of Deddington village, about 100m to the east of the church and market place, which it clearly pre-dates. The layout of Deddington is that of a planned borough, which was laid out in the twelfth century on the line of the Oxford to Banbury road. The castle earthworks consist of impressive banks and ditches forming a sub-rectangular enclosure of about 3.5ha (8.5 acres) with an inner enclosure of about 0.4ha (1 acre) at the eastern end. The inner enclosure was subject to intensive excavation in 1947–51 and 1976–79, and although some initial results have appeared in print, a definitive report on this work is still awaited (Jope & Threlfall 1946–7; Ivens 1984).
The size of the earthworks of the large outer enclosure had led to speculation that they might be Iron Age or possibly late Saxon; however, as late Saxon pottery was found underneath the rampart at the eastern end an early Norman date for this impressive earthwork now seems most likely. Deddington could have been Odo’s caput headquarters where he would have assembled his south Midland forces. The reports of the inner bailey excavations indicate a complex sequence of medieval structures which have been seriously disturbed by continual rebuilding and stone robbing. The earliest structure, dating to the eleventh century, is an earthen ringwork, inside which a small L-shaped stone hall was built, with a garderobe pit at its west end, all of which appear to be contemporary with Odo. A recent geophysical survey of the inner bailey indicates that the ringwork is a regular rectangle or square and calls into question a secondary motte postulated by the excavators. Traces of earthworks to the east of the castle may represent a second bailey or an enclosed park of similar size to the main enclosure, giving a butterfly shape in plan to the whole complex. There is evidence of a line of fish ponds of medieval date running to the south of the castle earthworks.
Odo’s third possible caput castle lay at Castle Rising within the manor of Snettisham, close to the north coast of Norfolk. Snettisham had belonged to Archbishop Stigand before the Conquest and was one of the bishop’s most valuable estates, worth £85 in 1086. In the mid-twelfth century the village of (Castle) Rising was chosen as the site of William de Albini’s grand keep with its imposing ringwork defence and associated planned town. It is tempting but dangerous in the light of the incomplete archaeological evidence, to presuppose that Odo had an earlier enclosure castle on this dramatic site. Nevertheless, excavations in the 1970s did identify considerable pre-keep stratification dating from the eleventh century. This included what might have been a wooden bow-sided hall, which had been demolished well before 1140. The excavators reflected that ‘it would not be surprising if part or all of the location was already set aside within a manorial enciente’ – was it in fact Odo’s East Anglian caput? (Morley & Gurney 1997, 133).
42 Aerial view of Castle Rising, Norfolk, the large outer enclosure could have been built by Odo and represent his headquarters in the east of England. English Heritage
The two castles in England that are most closely associated with Odo, and which he was probably responsible for building, Rochester and Deddington, have somewhat similar designs. Both consist of large heavily defended earth and timber enclosures. Neither appears to have had a motte, but could have had stone or wooden towers. Both are capable of holding large numbers of soldiers and horses and of being used as a base for large-scale armed raids in the surrounding region or for garrisoning an army before a set battle. It is tempting to see them as the bases from which Odo won and controlled much of his English land. The absence of a third castle of similar design and dimensions in East Anglia or Lincolnshire may be because the bishop obtained these lands at a later date and did not have time to construct a major fortification here. It is possible that the earliest phases of Castle Rising performed the same role as a caput for the bishop as the enclosure castles at Deddington and Rochester.
Odo’s early castles, along with those of the king and other magnates, played a decisive role both in achieving and consolidating the Norman Conquest. They were capable of withstanding a full-scale siege as well as delaying or blocking an enemy’s progress. They acted as bases for controlling the surrounding territory, administrative centres, high-status dwellings, sites for dispensing justice and entertaining, and permanent demonstrations of power. Nevertheless the basic simplicity of Odo’s castles compares strikingly with his cathedral and other contemporary ecclesiastic buildings. It was in the decades after Odo’s death that castles became grand stone keeps and far more elaborate affairs. Castles and palaces grew closer together in design, as seen, for example, in the Bishop of Winchester’s palace at Wolvesey.
• Robert of Mortain •
The Conqueror’s other half-brother, Robert, also enjoyed considerable rewards in England after 1066. Henry of Huntingdon wrote that Roger [sic] de Mortain ravaged the country around Pevensey before the Battle of Hastings, and he was certainly granted the rape of Pevensey with its castle in 1067. He was also granted the castle at Berkhamsted, William’s base before the surrender of London, which may have been his headquarters in England. After helping subdue some of the early attempts to unseat William, notably, his role in destroying Danish forces in Lindsey in 1069, Robert appears to have spent most of his time in Normandy. In England, Robert had estates valued at over £2,100, making him second only to Roger de Montgomery among the post-Conquest lay magnates. Most of his land was in the south-west, where he was by far the largest landholder in Cornwall and he is sometimes referred to as the Count of Cornwall. Robert also held extensive estates in twenty English counties, with concentrations in Somerset, Devon and Dorset. Unlike Odo, Robert held his English estates at the time of the Domesday survey and as a result it is a little easier to trace his activities, which show that he often acquired land in questionable circumstances.
Robert, like Odo, was no respecter of monastic lands and according to Domesday Book twenty-four institutions lost land to the count, notably, St Albans, Glastonbury and Christ Church, Canterbury. Robert acquired the most valuable estate in Sussex at West Firle from Wilton Priory. In Somerset he acquired the manor of Bishopstone, which had belonged to an English owner called Tovi, who had found a miraculous cross on a hilltop called St Michael’s Hill on his estate. The cross was transferred to a new monastery that Tovi had founded in Essex at Waltham and became a sacred symbol for King Harold and the English at Hastings. Tovi granted Bishopstone to Athelney Abbey and after the Conquest Robert appears to have acquired the estate by sleight of hand. Robert built a massive castle on St Michael’s
Hill, which was perceived as a calculated insult by the English. The site, known as Montacute Castle, was besieged in 1068/69 as part of the western uprising, which was suppressed by Geoffrey of Coutances (Golding 1990, 119–44).
Robert had created new towns, markets and fairs in Normandy before the Conquest and he continued this activity on his English estates. Boroughs were established next to his castles at Pevensey, Berkhamsted, Montacute and Launceston. At Launceston Domesday records that, ‘The count of Mortain took away a market from this manor [St Stephen’s by Launceston] and put it in his castle.’ Another Domesday entry, for St Germans, Cornwall, records that, ‘In this manor there is a market on Sunday, but it is reduced to nothing by the Count of Mortain’s market, which is nearby in his castle [Trematon] on the same day.’
If Robert had no reservations about depriving English monasteries of land, he was generous towards Norman houses. In particular, the family abbey of Grestain was favoured and granted ten estates in England by Robert with a value of more than £50. He also gave land to the priory within his castle at Mortain and to the abbey of Fécamp.
The Man Behind the Bayeux Tapestry Page 13
