Stamford in Domesday Book

Dr David Roffe 2018

Stamford and District Local History Society, 15 February 2018

Introduction

Thank you so much for inviting me to Stamford. It is a great delight to be back here again and to be invited to reflect on the history of the town. I first began thinking about its origins and early history in 1975. I had volunteered to work on the castle excavation directed by Christine Mahany. Many of you will remember the site. It was at the time one of the largest medieval excavations ever undertaken and certainly unprecedented in terms of a castle. The finds were quite spectacular, not that I had much to do with them: it soon became apparent, in fact by lunchtime on the first day, that I was not a natural digger. Bang went another career option. I could read Latin, though, and in the afternoon I was sent off to the office in Kings Mill House to try to make sense of the medieval documents relating to the castle and the town generally. The rest is history.

Chris and I presented a synthesis of the historical and archaeological evidence for the origins and growth of Stamford at the Battle Conference in 1982. The paper was published in Anglo-Norman Studies in the following year. Much of the conclusions was based on the first comprehensive analysis of the Domesday account of the town. Historians had long mined the source for information, but there had never been a comprehensive analysis in terms of the later history of the lands that it describes. Tonight I shall revisit that study for the first time in 35 years. The intervening years has seen much new research in the field of Domesday studies, but I am glad to say that, by and large, our analysis has stood up well.

The Domesday account of Stamford

Domesday Book comes in two parts. Volume 2 is known as Little Domesday Book and describes the three eastern counties of Essex, Norfolk and Suffolk. Volume 1, Great Domesday, covers the rest of the country south of the Tees and just a tiny bit of Wales. The relationship between the two parts does not concern us today. Suffice it to say that the account of Stamford occurs on folio 336d of Great Domesday. It appears at the head of the Lincolnshire folios after the account of Lincoln and is followed by Torksey and then the body of the text, the account of the lands of the tenants-in-chief, substantially the barons of the shire. It is written in Latin, but in a highly contracted form. You will notice that almost every word in the text exhibits diacriticals, that is distinctive marks that indicate the omission of letters (figure 1). This system of shorthand speeded up writing, but sometimes it is difficult to reconstruct exactly what the scribe wanted to say. Fortunately, our chap was on his best behaviour when he wrote up Stamford, for there are relatively few contractions and suspensions. We are not always so lucky, but here we can be confident of what he intended (figure 1).

The Domesday Survey

Before we dive in and examine what he wrote we must first get some idea of why the Domesday survey, and Domesday Book that was compiled from its returns, was commissioned in the first place. Otherwise, we shall not be able to interpret what is recorded. You may have a vague idea from schooldays of the Domesday survey as a census. Well, it wasn't. To start with, at least half of the population is missing: there are next-to-no women mentioned in the text. More credible is the notion that Domesday was compiled as an inventory of William the Conqueror's new realm. Even so, there are still problems. William defeated King Harold at the Battle of Hastings in 1066. Why, then, did he leave it until 1086 to undertake the survey?

True the Conquest was not complete until the fall of Ely in 1071, but the Domesday survey was still 15 years after. In reality the survey addressed urgent problems in 1085. England was threatened with invasion.

King Cnut of Denmark had formed an alliance with Count Robert of Flanders and had mustered a fleet to conquer England. His claim to the throne was arguably stronger than William's and he must have thought that large parts of England were in one way or another sympathetic to his cause.

William clearly saw Cnut as an existential threat to his rule. He was in Normandy when he heard of the imminent invasion, and acted with characteristic decisiveness. The Anglo-Saxon Chronicle tells us that he crossed the channel with the largest force of mercenaries that had ever been seen in England. Meanwhile, Archbishop Lanfranc of Canterbury had convened a conference and it was decided to billet the troops on landholders throughout the country.

In the event for various reasons the invasion was put off until the spring of 1086 and was then abandoned after the death of Cnut. William, though, was not to know this in late 1085. He laid off some of the mercenaries over the winter, but nevertheless continued with preparations for a renewed onslaught. He ordered the coast to be wasted to deprive any invading army of sustenance. All military enterprises depended on foraging for supplies in the eleventh cent ury. Further, at his Christmas court in Gloucester he appointed trusted agents to East Anglia which, with its Anglo-Scandinavian population, was seen as a particularly exposed Achilles heel. Wulfketel at Crowland was replaced by Ingulf and Gunter was appointed to Thorney and Maurice to the see of London (with extensive estates in Essex and East Anglia, the bishopric was a key player in the east of England). All three were trusted royal priests whom William could rely upon to represent his interests and act decisively.

This was the context in which the Domesday survey was commissioned. At the same Christmas court, as recorded in the Anglo-Saxon Chronicle, William ordered a survey of the king's income in land and taxes and an equally exhaustive inquiry into the income and services of his tenants- in-chief, that is the barons. It is clear that the immediate aim was to raise cash to pay the mercenaries, but there was also a determination to plan for the future. The survey began in January 1086 and seems to have been complete by late June or early July. At a meeting, 'of all those who held land in England', at Salisbury in August 1086 taxation was extended to the lord's demesne and, it seems, new quotas of knights were determined. Never again was the realm to be dependent on mercenaries for its defence or at least that was the aim.

The writing of the Book

All of the data that were collected were then written up in what we now know as Domesday Book. When is a matter of debate. Many insist that it was compiled as part of the 1086 programme. But it is odd that it was not a complete record of the survey: it omits much of the data. I have therefore argued that it was a later enterprise ordered by William's son, William Rufus, in 1090. In 1088 there had been a rebellion against the new king which had occasioned much tenurial chaos. Domesday Book may have been compiled to inform a settlement. Writing in 1125 or so, the Anglo-Norman historian Orderic Vitalis probably makes an oblique reference to the enterprise. However, there is no explicit mention of the Book until the early years of the reign of Henry I, so it may have been compiled anytime up to 1102.

The structure of the Stamford account

Whatever its date, Domesday Book was clearly a royal feodary, that is a work of reference drawn up to inform government of the king's interests and dues. As such, it was an abstract – most noticeably the livestock so carefully recorded in the survey was jettisoned – but it drew all its data from the survey. The sort of information we get, then, is determined by the concerns of 1086. It is thus not surprising to find that the account of Stamford is divided into two broad sections, that is the income of the king and those of his tenants-in-chief. Both are signalled by capital letters and, intermittently, by paragraphoi, that is gallows marks, for the latter (figure 1).

Now, Domesday talks in terms of land, terra in Latin, but we must take care not to be misled thereby. What is mostly involved is rights over land rather than in what we would call real estate. In an urban context those rights were various. Foremost was landgabel or gafol. This was a rent, usually around a shilling a year, due from the community of the borough. This sum, however, if prominent, was not the sum total of what 'land' signified. There were also renders that Domesday generically calls consuetudines, 'customs', payments in coin and kind that represent a former food rent, and occasional labour services such as mowing on the king or lord's demesne. Then there was soca, 'soke', that included dues for the right to bake and brew as well as the amercements, fines in modern terms, that were paid in the borough court. Finally, there were tolls which were taxes on the buying, selling, and the passage of goods. Land itself, freehold to be anachronistic, generally resided in the townsmen.

The king's income

Having cleared the ground, we can now get down to the nitty-gritty. We first get a statement of the tax the town paid to the king (figure 2); it was assessed at 12½ hundreds. In the Danelaw the hundred was not the institution of the same name of hidated England. That was represented by the wapentake. The northern hundred was more like a vill, the unit of local government at its lowest level – think the civil parish of today – and it was rated at 12 carucates. So Stamford was assessed at 150 carucates for the geld, army and navy service, and Danegeld. The town was divided into six wards. We have no further information, but they were probably each associated with the gates of the town. That was certainly the pattern elsewhere. The ward was a unit of policing – hence watch and ward – and an association with routes into a town is common. In the later Middle Ages there were five gates north of the river in Stamford and probably a sixth in Stamford Baron. All the dues that resulted accrued to the king. It should be noted that the ward south of the river was an integral part of the town even though in Northamptonshire. There is no evidence that it was ever constituted as a separate borough.

Then we move onto the number of properties that owed their dues to the king – 141 and half a mill in 1066 and the same in 1086 less 5 that had been destroyed in the building of the castle (figure 3). The dues of 4 others had been taken away by Brand, whom we can identify as abbot of Peterborough. A further two belonged to a Wulfketel son of Merewine who was almost certainly the former abbot of Crowland.

Finally, there were 77 properties that belonged to privileged sokemen over whom the king had more limited rights, namely heriot, forfeiture, and toll alone, that is death duty, the right to land consequent to felony and the like, and, as we have seen, taxes on traded goods (figure 4).

Sokemen are not usually found in towns. Indeed, this reference is all but unique in an urban context. But in status they were akin to burgesses. Both, for example, had similar inheritance customs. In Stamford throughout the Middle Ages the property of an intestate burgess passed to his youngest son rather than his eldest. Many rural sokemen were also subject to this custom which became known in English law as Borough English. Eustace of Huntingdon who took away the mill was sheriff of Huntingdonshire. He, like sheriffs in general, was a notorious despoiler of land. How he came to put his oar into Stamford is unrecorded.

The land and income of senior townsmen and tenants-in-chief

As we have seen, the second section deals with dues that did not go to the king. Lawmen come first (figure 5). Twelve in 1066 and 9 in 1086 and holding in all 51½ properties, they held with sake and soke, that is all customs apart from the ones explicitly excluded here. They were judges of the law, receptacles of local custom rather than the bewigged panjandrums of today, but it is clear from a twelfth-century text called the Leges Edwardi Confessoris that they also acted as royal agents. They were responsible for recording crimes – in Stamford in 1226 they are called coroners – and they made distraints, that is took goods as bail, for offenders to appear in court. These are functions that were later granted by charter and so we are effectively looking at a town council. Unlike in Lincoln, the lawmen are not named. We do, however, have a valuable record of the properties they held. They were substantial townsmen and in all likelihood English. Fastolf too was English, holding a single church in 1086 which was quit, that is no dues were paid to the king (figure 6). Unfortunately, there is no record of who held the church in 1086.

Then we have a series of barons who had interests in Stamford. We can best start with the 9 properties held by Leofwine in 1066 and Alfred of Lincoln in 1086, We learn from Alfred's chapter in the body of the text that these belonged to his manor of Uffington (figure 7). We have no explicit evidence of this kind for other holdings, but the later history of some indicate other connections with rural manors. We learn from the 1212 survey of Stamford that the 23 properties held by Earnwine the priest and Eadsige and then Eudo Dapifer belonged to the manor of Wakerley in Northamptonshire (figure 8). The same source indicates that the 7 properties of Azor and Gunfrid de Chocques belonged to the manor of Casewick. Edward Cild's 14 properties belonged to Witham, Sproxton, or Tickencote. Finally, there were 10 properties in Lincolnshire, that is north of the river, that belonged to Peterborough abbey. I have been unable to place them in any particular manor.

Queen Edith's fee

Contributory manors, as they are known, of this kind are a common characteristic of English towns. They have been used to indicate the territory that was assigned to boroughs. Here, though, they are few in number and so little can be made of the pattern they describe. The 70 properties held by Queen Edith, the wife of King Edward the Confessor, which are described next may superficially appear to be of the same order. In reality, however, they are of a completely different type of holding (figure 9). Domesday indicates that the fee belonged to Roteland, that is the northern part of what became Rutland in the next century. This was an ancient dowage estate of the queens of England that consisted of a number of separate manors. The value at the end of the Stamford entry clearly indicates that the 70 properties had been one of them. Thus, in 1066 Queen Edith's fee was adjacent to the borough but not part of it. As odd as it may seem, this was by no means a unique characteristic of Stamford. Most English boroughs had royal and/or comital (earl's) estates closely associated with them. What is usual here, however, it that the estate was administered by the sheriff of Nottingham as Domesday indicates at the beginning of the account of Roteland.

By 1086 the 70 properties had become incorporated into Lincolnshire and Stanford. The fee of Albert of Lorraine, a priest of the royal household, hints at a reason. Albert held the church of St Peter's in Stamford which is said to have belonged to the church of Hambledon (figure 10). The same point is made in a reference to Albert's land in the account of Roteland itself (figure 11). However, we know from a charter of William Rufus that Albert had been granted all the churches of Roteland. We can, then, understand that St Peter's was the church of Queen Edith's manor here in Stamford. You are all familiar with the site of the church. It was adjacent to the castle and so it is likely the castle site too was originally in Roteland. If so, the whole area was moved into Lincolnshire to bring it under the control of the sheriff of Lincoln. In the later Middle Ages the combined lands were known as 'the castle and manor of Stamford'.

The boundary between Roteland and Lincolnshire was moved out of the town to Ermine Street to the west. Domesday Book recognizes the change by describing the Roteland portion of Stamford in a postscriptal entry in the Northamptonshire folios where it is called Portland (figure 12). The churches of St Peter's and All Saints held land there under the king. In the later Middle Ages the area, the West Field of Stamford, was known as Sundersoken which means 'estate apart'.

Stamford in the late eleventh century

Well, what does all of this tell us about the town in 1086? We can start with size. The assessment at 12½ hundreds, ie 150 carucates, tells us Stamford was by far the largest settlement in south Lincolnshire in terms of the tax it paid. The number of properties might be expected to give some indication of just how big it was (figure 13). In total 405½ are recorded. But the Latin term, mansio, plural mansiones, is an unknown quantity. It is usually translated as something like messuage or tenement; I have used the more neutral 'property' advisedly. What it was definitely not was a house: in Nottingham we have a reference to 3 mansiones in which there were 11 houses (domus). So, there were probably considerably more houses in Stamford in 1086. Moreover, the account is demonstrably incomplete. Although Stamford Baron was already in existence, no details are given. The earliest survey comes from 1125 when there were 74 properties of various kinds. Many of them must have been in existence in 1086. We must, then, be careful how we interpret the figures. It has often been stated, for example, that the castle was built on the fringe of settlement since only five mansiones were destroyed. In reality, we don't know how any houses this implies and anyway we are told only about those that had rendered dues to the king.

Our numbers are uncertain, but we can say something more about the physical extent of the town north of the river. We have four explicit markers. First, there is the bridge. It was on the same site as today judging from the twelfth-century structure at its south end. Second, the castle, the earliest part of the extant structure being the motte, the present bus station. Third, St Peter's church, now marked by a mound on St Peter's Hill. Finally, All Saints church – later known as All Saints in the Market – which still stands in Red Lion Square.

Although somewhat better than for some towns, this is seemingly a poor yield from such a detailed survey. Fortunately, though, other markers can be deduced from the later history of the fees described. In the early thirteenth century William de Lanvalie granted the church of St Clement's to St Michael's nunnery in Stamford. William was the lord of Wakerley and the successor to Eudo Dapifer. It seems likely, then, that one of the churches that Eudo held in 1086 was St Clements. It was situated somewhere close to Barn Hill House in Scotgate – skeletons have come to light from the garden – and so it seems that the area was already developed in 1086. William is not recorded as granting any other church, so the second one held by Eudo cannot be identified. Fastolf and his church present a more difficult problem. We have no record of who held in 1086, so we cannot trace its later descent. Its status, however, may provide a clue. It was quit, so it was held in chief of the king. This profile fits Holy Trinity outside St Paul's gate. In the thirteenth century it was the only church in Stamford that had its own court, as did other lords in Stamford who held in chief of the king. Finally, there is the land held by Crowland. Abbot Ulfketil is not recorded as holding a church. By the mid twelfth century, however, the abbey held St Michael's and this church is likely to mark at least one of Ulfketil's properties.

We end up with a somewhat detailed topography for such an early period. By 1086 Stamford seems to have reached the full extent of the thirteenth-century town (figure 14). There were certainly no walls at this time – they are of later construction – but it is possible that even the line of the later defences, in part or whole, was already established. It might be doubtful, though, that the town was infilled to the same degree as later.

Despite such riches, ranking Stamford among the 70 or so boroughs of England in 1086 remains a difficult task, not the least because different units are used from town to town. Had all of them been measured in terms of mansiones, comparison would be possible. But they aren't. Beside mansiones we also find burgenses, domus, hagae, acrae, or a combination of two or more. So it is that value to the king is the best metric of size, as imperfect as it is (figure 15). Stamford rendered in total £78 in 1086. This does not put it in the Premier League of the English boroughs recorded in Domesday Book. Lincoln, York, Norwich and, of course, London although not fully described, were altogether much bigger and richer boroughs. But Stamford was well up in the First Division.

Conclusion

Tonight I have confined myself to the Domesday evidence. There is, of course, a lot more evidence out there (figure 16). I have not touched on the archaeology, topography, nor numismatics. However, it is Domesday that provides the framework for understanding the early history of Stamford. The most important conclusion that Chris and I reached in 1982 was that the castle perpetuated a centre of authority of long standing. Adjacent to St Peter's church, it almost certainly replaced the manor house of Queen Edith's fee. That association in its turn provided context to late ninth-century defences that we found beneath the bailey. If we are looking at one area that we can identify as the origin of the town it is here. How it relates to the rest of the town is another story and another lecture.

©David Roffe 2018

A print version can be downloaded HERE