Abstract
This article offers a new interpretation of the Domesday survey, drawing upon a collaborative study of its earliest surviving manuscript, Exeter Cathedral Library MS 3500 (Exon). It identifies five principal stages: first, the survey was launched at Gloucester in midwinter 1085; secondly, fiscal information extracted from geld assessment lists was integrated with manorial detail supplied by landholders to create a survey organised on a geographical plan by hundred; thirdly, this hundredal recension was checked by a second group of commissioners at meetings of shire courts, generating a substantial corpus of contested matter; fourthly, the hundredal recension was restructured into circuit returns which grouped together and summarised the holdings of barons who held directly from the king; fifthly, Domesday Book itself was written directly from these circuit returns. Royal assemblies held at Easter, Whitsun and Lammas functioned as deadlines for the second, third and fourth stages respectively; and a major geld levied at the rate of six shillings to the hide was collected and accounted for during this period. The survey generated a range of different outputs, each intended to serve specific fiscal and political purposes: the hundredal recension was designed to facilitate a reassessment of geld liabilities; the lists of contested matter anticipated a later judicial review; the circuit returns, summaries and Domesday Book were designed to make the administration of the royal demesne and the profits of royal lordship more efficient. The latter also supplied barons with what amounted to confirmation charters of their uncontested holdings, for which they performed homage.
The annal for 1085 in the Anglo-Saxon Chronicle says that King William I convened an assembly at Gloucester in midwinter, and there had ‘mycel geþeaht swiðe deope spæce wið his witan ymbe þis land’ (‘much thought and very deep discussion with his council about this land’) before launching the Domesday survey.1 What were they thinking and talking about? These questions matter because the survey’s surviving records are of profound historical importance and the more we know about the circumstances in which they were made, the better placed we will be to interpret them. They have therefore been intensively explored.2 However, a new interpretation has emerged from a recent collaborative study of an important but relatively neglected manuscript, Exeter Cathedral Library MS 3500 (Exon).3 This is invaluable for several reasons. Although it covers only Wiltshire, Dorset, Somerset, Devon and Cornwall and survives in an incomplete form, it remains the earliest extant manuscript of the survey. It also comprises several distinct kinds of text written by a large group of scribes at various stages of the process. The bulk of the manuscript consists of manorial descriptions listed under the names of landholders who held directly from the king (fiefs), which informed the corresponding entries in Great Domesday Book (GDB); but it also contains material excluded from GDB including records of contested property labelled Terrae Occupatae, summaries of fiefs, three pairs of hundred lists and a series of accounts relating to the land tax known as the geld. In addition, since every scribal stint has now been identified, it is possible to reconstruct the scribes’ working patterns, draw inferences about the sources they used and examine how the fiefs were subsequently redacted by the GDB scribe.4 Exon Domesday therefore constitutes a unique opportunity to observe certain stages of the survey directly and to draw inferences about the remainder; and when this evidence is placed alongside the rest of the Domesday corpus, it becomes possible to construct a new interpretation of the entire process and what it was intended to achieve.
The making of Domesday remains controversial, principally because none of the texts in the Domesday corpus or any other contemporary record contains a detailed description of it.5 As we shall see, the annals for 1085 and 1086 in the Anglo-Saxon Chronicle, a short contemporary description of the survey written by Bishop Robert of Hereford, and the prologue to a collection known as Inquisitio Eliensis contain vital clues, but otherwise the survey’s processes and purposes must be inferred from its three surviving contemporary manuscripts and other texts extant in later copies. What follows is a fresh interpretation of this evidence informed by close reading of Exon. It argues that the process by which Domesday was made can be reconciled into five stages, which generated several distinct outputs, each intended to serve specific purposes. During the first stage, the survey was launched and data was collected by royal officials and landholders; during the second, a geographically arranged survey was made by combining material extracted from existing fiscal records with manorial detail supplied by landholders; during the third, this material was checked and amended with contested matter in dramatic, widely attended meetings of shire courts; during the fourth, the material was rearranged and listed under the names of landholders who held directly from the king; and during the fifth, the two manuscripts known separately as Little Domesday Book (LDB) and GDB and collectively as Domesday Book were written. It is suggested that the first four stages were structured around meetings of royal assemblies and were completed before 1 August 1086, and that Domesday Book was written shortly afterwards in 1086–7. It is also suggested that the survey was intended to maximise the king’s income by strengthening the administration of several forms of royal income, each of which required information structured in specific ways. That is just what the survey did. It therefore empowered the king to exploit landholders’ incomes more intensively, but it could not have been made unless landholders co-operated, and they did so because they also received something precious in return.
I
During stage one, plans for the survey were finalised and agreed at Gloucester. The kingdom was divided into seven circuits, each consisting of several contiguous shires, and two groups of commissioners were appointed to run each circuit. A contemporary description of the survey written by Robert, bishop of Hereford (1079−95), supplies important evidence here. It says that ‘Alii inquisitores post alios, et ignoti ad ignotas mittebantur prouincias, ut alii aliorum descriptionem reprehenderent et regi eos reos constituerent’ (‘Other investigators were sent after the others, as strangers into unfamiliar shires, in order that they might find fault with the other men’s survey and report those guilty to the king’).6 This indicates that two groups of inquisitores were assigned to each circuit, one responsible for producing a first draft of the survey, the second for checking it.7 The latter were presumably intended to be ‘strangers’ in the sense that they were not meant to have significant landholdings in the shires they surveyed. That interpretation is strengthened by a document preserved in the late eleventh-century compilation known as ‘Hemming’s cartulary’, which records the identity of the commissioners responsible for the survey in Worcestershire and neighbouring shires: Remigius, bishop of Lincoln, Walter Giffard, Henry de Ferrières and Adam, brother of Eudo the king’s steward, ‘qui ad inquirendas et describendas possessiones et consuetudines, tam regis quam principum suorum, in hac prouincia et in pluribus aliis, ab ipso rege destinati sunt eo tempore quo totam Angliam idem rex describi fecit’ (‘who had been sent by the king himself to investigate and survey the possessions and customs of both the king and his magnates in this shire and several others at the time when the king caused the whole of England to be surveyed’).8 The circuit in question (circuit V) comprised Gloucestershire, Herefordshire, Shropshire, Cheshire and Worcestershire; and Domesday Book reveals that, although these men between them held manors in twenty-four different shires, none of them held land in this circuit except for Henry de Ferrières, who held two manors in Herefordshire.9 This is the only source that specifically names Domesday commissioners, but there is suggestive evidence as to the identity of others. Bishop Robert’s description implies that the first group of commissioners could comprise locally powerful lords and officials, and we have a letter addressed by Archbishop Lanfranc to a certain ‘S.’ who was responsible for conducting an inquiry into certain shires, almost certainly in 1086, and who is plausibly identified as Swein of Essex, quondam sheriff of that shire.10 There are also strong hints that bishops William of Durham, Walkelin of Winchester, Geoffrey of Coutances, Osmund of Salisbury and Maurice of London, and laymen such as Robert d’Ouilly, Hugh de Port and Ivo Taillebois served as commissioners during the checking stage. All of these men were front-rank barons, who were comfortably among the wealthiest hundred landholders in the kingdom; they could therefore call upon the services of extensive networks of clerics, household officials and subtenants to assist them in their duties and represent them in their absence.11
The commissioners were instructed to obtain answers to a questionnaire for each manor, which is preserved in the prologue to Inquisitio Eliensis, a collection of Domesday-related texts preserved in three twelfth-century manuscripts:
Quomodo uocatur mansio, quis tenuit eam tempore regis eduardi, quis modo tenet; quot hide, quot carruce in dominio, quot hominum; quot uillani, quot cotarii, quot serui, quot liberi homines, quot sochemani; quantum siluae, quantum prati, quot pascuorum, quot molini, quot piscine; quantum est additum uel ablatum, quantum ualebat totum simul et quantum modo; quantum ibi quisque liber homo uel sochemanus habuit uel habet. Hoc totum tripliciter scilicet tempore regis Ædwardi et quando Rex Willelmus dedit et quomodo sit modo et si potest plus haberi quam habeatur.
What is the manor called? Who held it in King Edward’s time? Who holds it now? How many hides? How many ploughs in demesne; how many belong to the men? How many villans, how many cottars, how many slaves? How many free men, how many sokemen? How much woodland, how much meadow, how many grazing lands, how many mills, how many fisheries? How much has been added or taken away? How much was it worth altogether and how much now? How much each free man or sokeman had or has there? All this in triplicate; that is, in the time of King Edward, when King William gave it and as it may be now; and if more can be reckoned there than may [currently] be reckoned.12
There are excellent reasons for thinking that this questionnaire (or a very similar one which perhaps included a question about livestock omitted in the surviving copies) formed part of the survey’s terms of reference.13 The fact that Exon Terrae Occupatae, which relate to a circuit distant from Ely, systematically answer the question ‘quantum est additum uel ablatum’ constitutes clinching evidence in this connection.14 However, to judge from Domesday Book and other texts in the corpus, the commissioners and scribes responsible for the first draft of the survey in each circuit (in stage two) addressed the question in a slightly different order using distinctive formulaic conventions; indeed, it is this that permits us to reconstruct the seven circuits.15
The terms of reference were probably issued as writs intended to be read out at meetings of shire courts, for this was the standard mode of communication between the king and the localities.16 Such meetings had probably been scheduled in advance, but if not they could have been arranged at short notice: an early twelfth-century legal compilation records that shire courts could be convened with seven days’ notice.17 It is also probable that deadlines were set to coincide with meetings of royal assemblies. The annal for 1086 records that royal assemblies were held in Winchester at Easter (5 April), in Westminster at Whitsun (24 May), and in Salisbury at Lammas (1 August);18 and we shall see that it was logistically feasible to complete the second, third and fourth stages of the survey between these dates. In this connection, it is also relevant that Carolingian rulers are known to have ordered surveys to be made between meetings of royal assemblies: for example, in his edict of Pîtres (864), King Charles the Bald demanded that a detailed survey of markets be delivered to him ‘ad proximum placitum nostrum’,19 and the annals of St-Bertin record that, after Easter in 869, Charles instructed bishops, abbots, abbesses and counts to produce breues of their honores recording how much land they held, and to bring these to his next placitum in May.20
The launch of the Domesday survey triggered an intensive period of data collection, lasting about a month, during which landholders and their officials gathered the information they needed to answer the survey’s questionnaire with respect to each of their manors, and royal officials extracted information from geld assessment lists—that is, assessment lists used to administer the geld and other dues and services levied on the hide.21 Extracts from geld assessment lists have been identified in later manuscripts,22 and their existence can be confidently inferred from the fact that gelds were levied regularly throughout the eleventh century,23 for that would not have been possible without detailed records of liability. Both the Exon geld accounts and the Northamptonshire geld account—an Old English document which pertains to an otherwise unrecorded geld levied earlier in the reign—presuppose the existence of detailed assessment lists organised by hundred.24 We also have assessment lists which were compiled in the early twelfth century on a cadastral plan, by hundred or wapentake, from Leicestershire,25 Northamptonshire,26 Middlesex,27 and Lindsey.28 Similar records must have been used in 1086 to compile skeletal lists containing provisional answers to some of the survey’s questions: the name of the manor, its geld assessment, who held it in the time of King Edward and who holds it now. The plan was to flesh these lists out with manorial data, though in a few instances where that proved impossible the skeletal lists found their way into Domesday Book.29
II
During stage two of the survey, the information extracted from geld assessment lists was combined with manorial detail supplied by landholders to create fully formed Domesday entries in documents organised by hundred or wapentake: let us call this the hundredal recension. This was achieved by interviewing all landholders who had no intermediary between themselves and the king. By the early twelfth century, such lords were known as tenentes in capite—tenants-in-chief.30 That usage was not current in 1086, but the underlying phenomenon plainly existed, for the structure of the Exon geld accounts and Domesday Book both assume its existence: they refer to such lords as barones and tenentes respectively, so let us call them barons. The commissioners assigned each baron an appointment to have their holdings recorded in documents or breues: LDB refers to the day on which Robert Malet ‘fuit inbreuiatus’.31 The barons were strongly incentivised to attend, partly because failure to account for land at the survey could result in forfeiture,32 and partly because the survey represented an excellent opportunity for barons to secure their holdings.33 The process can be imagined as follows. When a baron arrived, doubtless with an entourage of men bearing manorial details recorded in documents,34 tally sticks,35 or aides-memoires, he or she would be directed into a room where several scribes were at work, each furnished with documents containing the skeletal lists for certain hundreds. The baron would then visit each desk to be ‘inbriefed’. The scribes would check whether the baron held land in the hundreds they were writing up and, where relevant, combine the baron’s manorial detail with the information extracted from geld assessment lists to produce a draft entry for each manor. That process would be repeated until the baron had visited every desk to ensure that all his manors were inbriefed; once every baron had done so, the hundredal recension would be complete.
The text known as Inquisitio Comitatus Cantabrigiensis (ICC), which survives in a single manuscript written in the late twelfth or early thirteenth century, preserves a text of the hundredal recension for Cambridgeshire.36 This closely resembles the Cambridgeshire folios of GDB except that it omits some royal demesne manors and is incomplete, owing to the loss of one or more quires at the end. However, ICC is organised on a different plan to GDB, for whereas the latter is arranged by fief, the former proceeds by hundred and vill in that order. As John Horace Round first showed by collating the two texts, the Cambridgeshire folios of GDB were written later than ICC. Round also argued that ICC represents a unique survival of the first outputs from the survey, which he labelled the ‘original returns’; that such returns lie behind the entire Domesday corpus; and that Domesday Book itself was written directly from them.37 We shall see that the hundredal recension was in fact reconstituted into fiefs (in stage four) before Domesday Book was written, but otherwise Round was right, for it is possible to detect traces of the hundredal recension throughout the Domesday corpus.
The nature of these traces is partly a function of the way the scribes organised their work in stage four.38 To explain: if the hundredal recension was at that stage bound together in a single volume, the scribes would tend to copy entries from each booklet into each fief in the same order, so that each fief would tend to evince not only a high proportion of ‘hundredal grouping’, with material pertaining to specific hundreds consistently grouped together, but also a regular ‘hundredal order’, with hundreds consistently treated in the same sequence. Alternatively, if the hundredal recension consisted of physically separate documents, and if the scribes treated these in roughly the same order but with licence to deviate from the sequence where necessary, that would result in a high proportion of hundredal grouping but a less regular hundredal order. Finally, if the scribes chose to treat the hundredal recension in a random order, controlling their progress with checklists, that would result in a high proportion of hundredal grouping but an irregular hundredal order. In a pioneering study, Peter Sawyer showed that a consistent hundredal order can be detected within the fiefs of several shires in Domesday Book: he analysed a sample of fiefs in twenty-six shires and concluded that seventeen of them showed clear traces of a hundredal order.39 Frank Thorn has subsequently confirmed that a hundredal order can be detected in many Domesday shires,40 and Colin Flight has made a strong case for thinking that Domesday-related texts from Ely and St Augustine’s Canterbury derive from the hundredal booklets.41 However, the frequency of hundredal grouping within fiefs in Domesday Book constitutes the clearest and most decisive evidence that they derive ultimately from hundredally arranged texts, and a computer-assisted study has revealed that 90 per cent of entries in fiefs in Domesday Book fall within hundredal groups (see Appendix I).
This is sufficient to refute one of the central premises of Vivian Galbraith’s analysis of the making of Domesday Book. He accepted that the survey began with a hundredally arranged text in some circuits but denied that this was universal. His treatment of Exon was crucial in this connection, for he argued that the absence of hundredal organisation in Exon fiefs and the corresponding shires in GDB ‘proves to demonstration that the original returns made by the circuit were not a series of hundred rolls’, and that ‘the information was from the very first committed to writing by the circuit commissioners in the form in which it is now found both in Exon and Domesday Book’. This led him to conclude that ‘the primary interest of the commissioners in the south-west was not with the assessment or collection of the geld but with the total wealth and lands of the king and his tenants-in-chief’.42 But this was the point at which Galbraith’s analysis began to go seriously wrong.43 Instead, drawing on Frank Thorn’s meticulous analyses of the place-names and hundreds of the south-western shires, it has now been possible to determine the extent of hundredal order and grouping in Exon with considerable precision. About 40 per cent of Exon fiefs entries manifest a perfect hundredal order, another 55 per cent follow a broadly consistent hundredal order, and about 90 per cent of the entries occur in hundredal groups.44 In addition, as we shall see, the distinctive palaeography of Exon fiefs is best explained by presuming that they were written by groups of scribes who found ways to collaborate while working on the same set of hundredally arranged booklets.45 Indeed, the scribes were so preoccupied by this hundredal order that they recorded it in lists. Exon contains six lists of hundreds, two each for Cornwall, Devon and Somerset, and the second of these for each shire informs the order in which the hundreds were treated within fiefs.46 Taken together, this evidence ‘proves to demonstration’ that the survey began with the production of a hundredal recension in every shire.
Provided the commissioners employed a large enough team of scribes, it would have been possible for them to complete the hundredal recension in about a month and a half. To judge from the surviving portion of Exon fiefs, the text originally occupied about 1,200 pages, generally ruled at twenty lines to the page. The Exon scribes were essentially copying and reorganising material in stage four, so we can safely assume that the hundredal recension booklets they worked from contained roughly the same amount of matter. Michael Gullick has estimated that Romanesque scribes could copy between 150 to 300 lines per day.47 The scribes who compiled the hundredal recension were doing something more complicated, for they were interviewing and integrating material as they wrote. Let us therefore suppose that they wrote on average eighty-five lines per day. On that basis, the first draft would have taken 282 working days to complete. The critical variable was therefore the number of scribes: the greater the number of scribes that could be simultaneously deployed, the more quickly the work could be done. Exon is crucial because it demonstrates that a team of about twenty scribes worked on the south-western circuit return during stage four, which was a less labour-intensive phase of the survey. If the commissioners maintained a team of twelve scribes at work for six days a week the hundredal recension could have been written in twenty-six days. Then there is the question of travel. It is not clear whether the hundredal recension was produced at one centre in each shire or at single centres within each circuit; however, since the former is the most conservative assumption, we should allow twenty-one days for the commissioners and their scribes to travel within their circuit. The general plausibility of this is suggested by a record of the expenses incurred by Thomas Walmesley as a judge of the assize in 1598, which reveals that he rode through a circuit similar to Exon’s in three weeks, presiding over sittings in each county.48 We thus arrive at a total of forty-six days for the whole exercise. If it began on 14 February, the commissioners could have finished the survey with enough time to return to the Easter royal assembly at Winchester by 2 April, the day before Good Friday. Of course, none of this proves that the survey was completed on this timetable: merely that it could have been, using conservative assumptions. However, one further detail is suggestive here. Having described the Domesday survey, the annal for 1085 says that all the records (gewrita) were brought to the king, and the annal for 1086 begins at Easter.49 This suggests that the author of these annals was following the convention of beginning the year at or near Lady Day (the Feast of the Annunciation, 25 March).50 It is therefore probable that the 1085 annal describes the delivery of the first draft of Domesday survey in the spring of 1086.
III
During stage three of the survey, the hundredal recension was checked by the second set of commissioners at extraordinary meetings of shire courts attended by large numbers of participants.51 The Ely prologue says that the king’s barons made the survey ‘per sacramentum uicecomitis scire et omnium baronum et eorum francigenarum et tocius centuriatus presbiteri, prepositi, .vi. uillani uniuscujusque uille’ (‘by the oath of the sheriff of the shire and of all the barons and their Frenchmen and of the whole hundred, the priest, the reeve and six villans from each and every vill’).52 Both Inquisitio Eliensis and ICC contain lists of homines iurauerunt, the teams of jurors who represented certain hundreds and gave sworn testimony at the survey, and these generally comprised equal numbers of Englishmen and Frenchmen drawn from the lower reaches of landed society.53 In addition, Domesday Book records testimony given by various groups including the representatives of shires, towns, ridings, hundreds and wapentakes, as well as the barons and their men.54 If the survey was indeed attended by six or more representatives from each vill, we must confront the possibility that tens of thousands participated, for Domesday Book records more than 5,000 vills.55 This helps to explain why the author of a Domesday-related text compiled at Bury was moved to observe that the survey was attended by ‘pene universi incole’ (‘almost all the inhabitants’) of the land.56
The commissioners presumably arranged for each entry to be read out in French and English before inviting sworn jurors to approve or contest its content. The focus was primarily on tenure at this stage. A few entries describe testimony relating to manorial values,57 geld assessments,58 service,59 and other manorial details,60 but the great bulk of the recorded testimony responds to two questions: ‘who holds it now?’ and ‘has anything been added to or taken from the manor?’ The commissioners did not usually attempt to resolve disputes: they simply ensured their scribes recorded the relevant testimony before moving on. There are a few exceptions. For instance, an entry in Exon records that the abbot of Tavistock was in seisin of a manor at Werrington ‘ea die qua rex Willelmus misit barones suos ad inquirendas terras Angliae’ (‘on the day that King William sent his barons to enquire about the lands of England’), and that he was dispossessed of it by these barons because English jurors testified that it did not belong to the abbey on the day King Edward was alive and dead.61 However, only a handful of entries record or imply that Domesday commissioners took action to deal with contested matter while the survey was in progress,62 and several indicate that participants expected it to be followed up later: for example, an entry in the Berkshire folios records that the jurors ‘judicium non dixerunt, sed ante regem ut judicet dimiserunt’ (‘did not give judgment but referred [a contested matter] to the king to judge’).63 This was partly because if the commissioners had tried to resolve disputes, the survey would have ground to a halt, for property disputes were often time-consuming: for instance, King William is known to have settled one dispute between the abbey of Fécamp and William de Briouze at Lacock in Wiltshire sometime in or after April 1086, and this single placitum is said to have lasted from morning until evening.64
The commissioners evidently interpreted this part of their brief in different ways, for the nature, quantity and structure of Domesday’s contested matter varies considerably between circuits.65 No manuscripts survive from this stage of the survey, but it is a reasonable guess that the commissioners’ scribes recorded material using space left blank in the margins of the hundredal recension, or by listing it separately in appended documents. The bulk of Domesday’s contested matter can be reconciled into three broad categories: infringements on royal property, disputes between landholders other than the king, and unauthorised tenures.66 The latter make a remarkable assumption: that unless barons were able to prove that they held a manor with the king’s express consent, their tenures were considered dubious, even where there were no other claimants. Many entries imply that Domesday jurors expected, as a matter of course, to have seen a liberator bearing the king’s writ or seal to deliver seisin at meetings of shire or hundredal courts.67 This logic is applied to a wide range of transactions involving the full spectrum of landed society. Numerous cases relate to land said to have been taken, seized or appropriated without the king’s permission,68 but the language of illegality is also applied to many seemingly routine transactions: for instance, land acquired through marriage,69 by exchange,70 or by gift.71 There is also abundant evidence that barons were in theory expected to obtain the king’s permission before absorbing the lands of former free men and sokemen into their manors.72
Stage three could have been completed between the Easter and Whitsun royal assemblies, that is, between 5 April and 24 May. To judge from the surviving portion of Exon, there were about 3,200 entries to check in the five south-western shires. However, it was only necessary to read out certain details from each entry: the name of the manor, its geld liability, value, who held it in the time of King Edward, who holds it now. Even in two languages, two minutes per entry would suffice for this, provided the court was kept to order. Since contested matter was potentially time-consuming, the commissioners were instructed only to record the essential details. Exon Terrae Occupatae happen to be one of the fullest records of this material, and they relate to about 8 per cent of entries in the south-western shires. We should allow for the possibility that some disputes went unrecorded, so let us estimate that 10 per cent of entries were contested and allow twenty minutes for each in a ten-hour working day. Based on these assumptions, the hundredal recension for the south-western circuit could have been checked in twenty-two days. We should then allow another twenty-one days for the commissioners and their scribes to travel between shires, but that would still leave sufficient time for them to return to Westminster by Whitsun.73
The Exon geld accounts establish that the survey coincided with the collection of a major geld.74 There is no manuscript justification for the term ‘geld accounts’, for the documents occur under headings such as ‘Isti sunt hundreti de Devensir’,75 and a contemporary but incomplete schedule of the contents of Exon refers to them as records of an Inquisitio Gheldi.76 To judge from the documents themselves, this inquisitio involved an audit of the accounts in the treasury followed by an inquiry which aimed to collect arrears and to address any other problems which had arisen during the collection process.
The accounts were drawn up after the collection of a geld levied at the rate of 6 shillings to the hide. They record for each hundred (i) the total number of hides; (ii) the number of demesne hides within manors held by the king and his barones; (iii) the number of hides which had paid geld, and the amount of money collected from them; (iv) the number of liable hides for which payment had not been made; and (v) other relevant matter, including records of allowances for those involved in the collection process. The logic of the accounts is that (i) minus (ii) should equal (iii) with (iv) plus (v) accounting for any disparity. They also reveal the following system of exemption: the demesne portions of manors held by the king and his barons were exempt from geld, whereas the non-demesne portions of those manors and the entirety of manors held by subtenants were liable.77 In other words, the burden of the geld was shouldered by subtenants and peasant cultivators. Exon contains one set of accounts for most of the hundreds in Cornwall, Devon, Dorset, and Somerset (for which there is a second fragment), plus three sets of accounts for Wiltshire.
The geld was collected in two instalments with payment due at two termini, the second of which fell at Lady Day or Easter. Although it was paid in coin by the peasantry and subtenants, liability rested with barons. The accounts mention manorial reeves in contexts that suggest they were intermediaries between lords and geld collectors. The latter are variously styled collectores, congregatores, fegadri, and hundremanni; they operated in teams of four within each hundred and were paid an allowance, although they could also be fined for negligence. They handed the geld over to geld-bearers (portatores) who took it to Winchester. One entry refers to the moment when the geld-gatherers in Wonford Hundred in Devon handed over their haul to William the usher and Ralph de La Pommeraye ‘qui debebant geldum portare ad thesaurum regis Wintoniae’ (‘who had to carry the geld to the king’s treasury at Winchester’).78 The Somerset fragment supplies further arresting detail:
De Sumerseta habet rex de gildo suo .d. libras et .ix. libras in thesauro suo Vintoniae et illi qui portauerunt has Vintoniam habuerunt .xl. solidos de conregio suo et inter saginarios conducendos et scriptorem et forellos emendos et ceram dederunt .ix. solidos et .viii. denarios et de .l. et .i. solidum et .iii. denariis quos receperunt portatores geldi non habuit rex denarium et non potuerunt compotum reddere. Hos uadiauerunt sese reddituros legatis regis.
From Somerset the king has 500 pounds and 9 pounds of his geld in his treasury at Winchester and those who have carried these to Winchester have had 40 shillings for their allowance, and including hiring packhorses and a scribe and buying forels and wax they have expended 9 shillings and 8 pence; and from 51 shillings and 3 pence, which the geld-carriers have received, the king has not had a penny and they have not been able to provide an account. They have pledged that they will pay these to the king’s legates.79
This appears to be a copy of a receipt given to the geld-bearers by treasury officials. It is striking that the geld-bearers were given allowances for hiring scribes, and that the text refers to a computus—a written account—for this establishes that the process involved the production of documents in the localities as well as the treasury. The geld-bearers presumably employed scribes to write receipts for the geld-gatherers when they received the money from them, and to produce a first draft of the accounts. The Exon accounts thus demonstrate the existence of one type of document used to administer the geld and imply the existence of two more: detailed assessment lists which underpinned the whole system, and receipts produced each time money changed hands throughout the collection process.
When the money reached the treasury, it was counted and checked against the accounts, which were amended as appropriate to inform the collection of arrears. The three sets of Wiltshire accounts were written at different stages of this process. They are labelled Wiltshire C, B, A in the edition and were originally written in that order. Wiltshire A also refers to the activities of men charged with collecting unpaid geld led by ‘Bishop William’80 and a certain ‘Walter’ and his colleagues (socii).81 These men were evidently responsible for the inquisitio gheldi in the south-western shires. There are several possible candidates for the identity of Walter, but Bishop William can only be William de Beaufai, who was appointed bishop of Thetford at the midwinter assembly of 1085, or William de Saint-Calais, bishop of Durham (1080–96). The latter identification is the more probable, for he is elsewhere named in connection with the survey of the south-western shires.82 It is therefore probable that the geld and survey were checked in the period after Easter in the same meetings of shire-courts, convened for both purposes.
That conclusion is strengthened by the fact that there are close codicological and palaeographical connections between the geld accounts and the remainder of the Exon collection. Most of the geld accounts were written in booklets belonging to the same stock of blank quires used for Exon fiefs, cut and folded to the same page size and pricked and ruled to the same pattern of twenty lines per page. They were also written by the same group of scribes. Building on earlier work, Fran Álvarez López has identified every scribal stint in Exon, assigning each scribe a name, most of which are letter names from the Greek alphabet.83 Alpha wrote most of the geld accounts for Somerset, Devon, Cornwall and Dorset, and Beta and Mu wrote the remainder; and they are the most prominent scribes throughout the manuscript. The Wiltshire accounts were written by Ksi, Rho, Sigma, Tau and Mu; Ksi and Tau contributed to the production of the Exon fiefs in Wiltshire and Dorset, of which only fragments survive, and although Rho and Sigma cannot be identified elsewhere in Exon it is possible that they contributed to the lost sections of Dorset and Wiltshire fiefs. The fact that the geld accounts were written by the same scribes as Exon fiefs makes it as good as certain that they relate to a levy taken in 1085–6. They also prove that the geld was administered with stunning efficiency. The accounts deal with just over 10,000 hides; 32 per cent of this was exempt demesne; 85 per cent of the geld was collected from the remaining hides before Easter; and if the Wiltshire accounts are representative, 96 per cent had been collected or accounted for once the inquisitio was complete.
IV
The principal tasks of stage four of the survey were to reconstitute the hundredal recension into separate documents for each baron, to compute summaries for each fief, to list contested matter into separate documents, and to bundle all this together into reports: call them circuit returns. Exon is the only manuscript witness to this stage of the survey. It reveals a group of about twenty scribes working under considerable pressure and therefore collaborating in ways that maximised productivity.
Colin Flight has advanced an ingenious hypothesis to explain how Exon fiefs were written in a way that accounts for their distinctive palaeography.84 He suggests that the hundredal recension booklets were unpacked and laid out on a convenient workspace (call it a table) to form a production line; and that the scribes were instructed to work down the line in sequence, ensuring that every entry was copied into the relevant fief. This procedure made it possible for several scribes to operate simultaneously, but since there were bound to be occasions when more than one scribe needed the same booklet, the scribes devised a modus operandi to overcome this problem. Suppose scribe A went to the table and found a gap in the sequence because the hundredal recension booklet he needed was being used by scribe B: scribe A would then place the fief he had been working on in a pending position to signal how far it had progressed, and then start a new fief from the beginning. Once scribe B had finished using the missing booklet and returned it to the correct place in the sequence, scribe C could use it to resume work on the fief left pending by scribe A—and so on.
This amounts to a significant reinterpretation of Exon fiefs, for Galbraith and David Roffe have both posited that they represent the ‘process of inbreviation’ when manorial descriptions were first compiled.85 However, it is now possible to test Flight’s hypothesis using a relational database to compare the pattern of scribal stints with the occurrence of other boundaries in the manuscript and text. It emerges that the Exon scribes collaborated at every level: about 70 per cent of the fiefs in Exon were written by two or more scribes; 60 per cent of folios containing Exon fiefs were written by two or more scribes—indeed, one was written by five different scribes;86 and 70 per cent of the hundreds in Exon fiefs were written up by two or more scribes. There is also a close correspondence between scribal stints and hundredal boundaries, for more than 80 per cent of the transitions between scribal stints occur at or close to the point where the text passes from one hundred into another. The material was first divided into two groups, one covering Devon–Cornwall–Somerset and the other Wiltshire–Dorset, and fiefs were written in that order, one hundred at a time. This is all consistent with Flight’s hypothesis. It now seems clear that the hundredal recension consisted of a series of separate documents containing of material relating to one, two or three hundreds, and that the scribes then treated them in a regular order, except that they sometimes skipped ahead and treated booklets out of sequence if this enabled them to stay productive.
Once they reached the end of the production line, a summary was computed for each fief.87 Exon contains summaries of two ecclesiastical and six lay fiefs. These vary in format and content, but all supply totals of manors, hides, ploughlands, and value. For example, the summary of Miles Crispin’s fief in Wiltshire reads as follows:
Milo crispinus habet .xv. mansiones in Wiltesira. In his habentur .lxx. hide et .i. uirga et dimidia. Haec terra sufficit .l. et .vii. carrucis et dimidie et est appreciata .lx. libras .x. solidos minus. De his habet Milo in dominio .xx. et .iii. hidas quae sibi ualent .xxiiii. libras. Caetera habent homines sui.
Miles Crispin has 15 manors in Wiltshire. In these there are reckoned to be 70 hides and 1 virgate and a half. This land is enough for 57 ploughs and a half and it is valued at 60 pounds less 10 shillings. Of these Miles has 23 hides in demesne which are worth 24 pounds to him. His men have the rest.88
Most of the summaries deal with one shire at a time, but Gilbert de Bretteville’s treats Wiltshire and Dorset together, then Somerset separately, before combining all three shires into a single entry, and Count Robert of Mortain’s summary treats Wiltshire, Dorset, Devonshire and Cornwall in a single composite entry. This suggests that summaries were compiled one shire at a time before being engrossed. The Exon summaries were written by four scribes, two of whom also contributed to Exon fiefs.89 The fact that the Exon summaries are carefully written and largely free of corrections indicates that they are copies, not original calculations. Only a handful of summaries survive elsewhere in the Domesday corpus: there are four in Inquisitio Eliensis,90 and two in Great Domesday Book.91 However, since these and the Exon summaries between them cover thirteen shires in five circuits, it is probable that they represent chance survivals from a much larger corpus and that the survey was intended to generate summaries for each baron, which covered their holdings throughout the kingdom.
In ten instances, the words consummatum est occur, usually in large letters on blank folios at the end of fiefs.92 This could derive from Latin summa and may have been written after the summaries were made, in which case it could mean ‘it has been added up’. However, consummatum est could also mean ‘it has been finished’, or ‘it is finished’, or just ‘done’, and was probably written simply to indicate that a fief was complete.
The same group of scribes was also responsible for a self-contained section of material relating to contested matter labelled Terrae Occupatae (TO).93 This consists of 408 entries written in sixty-one pages and amounts to roughly 10 per cent of the manuscript. The surviving material pertains only to Devon, Cornwall and Somerset, though TO doubtless originally covered Wiltshire and Dorset too. TO was written by eleven scribes, all of whom contributed to Exon fiefs, and they used the same stock of parchment. This indicates that TO and fiefs were written at about the same time. The material is not organised in the same way as fiefs; indeed, TO lacks a clear organising principle, for it does not consistently group manors by baron or hundred. However, entries relating to the same hundred do often occur in close proximity. Most of TO is duplicated in Exon fiefs, but collation establishes that TO could not have been copied from the corresponding passages in Exon fiefs or vice versa.94 It follows that TO and fiefs were written from a common source at the same time and place: in other words, both drew directly from the hundredal recension booklets. The scribes who wrote fiefs probably enjoyed priority access to those booklets, for this would explain why the pattern of hundredal order and grouping is less pronounced in TO.
The content of TO broadly resembles the contested matter recorded elsewhere in Domesday in that it consists of infringements on royal property, disputes between barons and unauthorised tenures. However, a distinctive feature of TO is that most entries constitute a direct answer to the question quantum est additum uel ablatum, for they frequently use the phrases additus est and ablatus est, thus echoing the survey’s questionnaire.95 TO also has an unusually high proportion of unauthorised tenures, which account for about 60 per cent of the material. The closest comparanda to TO are the lists labelled Inuasiones super regem in LDB and Clamores in Yorkshire, Lincolnshire and Huntingdonshire, but these differ significantly from each other and TO in form, content and arrangement. Inuasiones were written in long stints by the scribes responsible for fiefs, they consist mostly of fully formed Domesday entries, and they organise the material by ‘defendant’—the person accused of each ‘invasion’. They are principally concerned with unauthorised tenures; infringements on royal property and disputes between barons are usually treated in LDB fiefs. Inuasiones therefore seem to be an attempt to ringfence unauthorised tenures pending further judicial processes. Clamores are different: they are not fully formed entries and they assume the existence of related or corresponding entries in the text of fiefs; they are geographically organised, by riding and then by wapentake or hundred; they frequently invoke jurors’ testimony; the material relates to the full gamut of contested tenures; and most of the material is not duplicated in fiefs, though there are many entries in fiefs which record contested tenures for which there are no corresponding clamores entries. Clamores thus seem to represent an attempt to list cases for which panels of sworn jurors had supplied testimony.96 It is possible that lists of contested matter were compiled for other circuits but omitted by the GDB scribe, for he certainly omitted TO. Either way, it seems clear that this part of the process was less uniformly organised than others, presumably because the commissioners and their scribes were given a less specific brief.
The fourth phase of the survey thus consisted of the production of circuit returns, compiled in various ways, which comprised fiefs, summaries, and lists of contested matter. Because it survives in its original form, Exon allows us to reconstruct in remarkable detail how the scribes responsible for one such return organised their work.97 Since they were collaborating closely, the scribes must have worked in close physical proximity, probably in the same hot, sweaty, stressful room: call it the Exon office. A total of twenty scribes contributed to the production of Exon fiefs and eleven of them also wrote TO. They clearly operated within a hierarchy. Mu was the boss. Although he wrote a relatively small proportion of what survives, his contributions were wide-ranging and managerial in nature. He wrote small contributions to the hundred lists and the geld accounts, a handful of manorial descriptions in each shire, about a quarter of all TO entries, half of the fief summaries, all but one of the consummatum est entries, and the incomplete schedule of contents; he was also responsible for about 10 per cent of the corrections in the manuscript. His interventions suggest that he exercised considerable authority over the whole operation and was prepared to engage with its detail, but the scale of his contributions suggests that he was not necessarily physically present throughout the process. Alpha and Beta were his principal deputies and team leaders. They worked on all the shires and were together responsible for about half of the manorial descriptions, half of the corrections, and about 15 per cent of TO; the scale of their input suggest that they worked full-time in the Exon office. They were supported in each shire by teams consisting of three or four junior assistants plus a similar number of occasional helpers. With the exception of one scribe (Eta), who worked on Dorset and Somerset, there was no overlap between the teams. Alpha and Beta must have worked full-time in the Exon office, but their assistants and occasional helpers evidently had spare capacity and may have been seconded from elsewhere. We can be reasonably certain that all the scribes were of continental origin, for the orthographical evidence indicates that they were not native speakers,98 and Julia Crick’s review of palaeographical comparanda concludes that they were all trained in northern France or Flanders.99 We occasionally catch the scribes relieving their tedium by writing pen-trials or doodles,100 bishops’ croziers,101 increasingly elaborate paraphs,102 and even musical neumes;103 and one of them appears to have ended a tedious day spent on Somerset TO looking forward to a glass of wine, for he jotted a biblical tag invoking the wedding feast at Cana where the best wine was saved for last.104
Where was the Exon office? Some evidence points towards regional production. The fact that the inquisitores of stage three were meant to be outsiders does not preclude the possibility that those responsible for stages two or four were locally powerful lords.105 There is not a scrap of information in Exon relating to land outside the five south-western shires, and there is no overlap between the scribes who wrote Exon and LDB. The hands of three of the scribes who contributed to the writing of Exon have been matched with some of those who wrote a large number of books for Salisbury Cathedral in the late eleventh century, and, like the Exon scribes, the Salisbury scribes often collaborated closely, sometimes writing together in short stints.106 Bishop Osmund of Salisbury (1078−99) played an important role in royal government prior to his promotion, for he served as the king’s chancellor in England between c.1070 and 1078,107 and we will shortly see that Salisbury was an important focal point for the survey. It is therefore possible that Exon was written at Salisbury, conceivably under Osmund’s direction. However, the fact that three Exon scribes wrote Salisbury books does not prove that Exon was written there: among other possibilities, it may be that a few clerks in Osmund’s entourage were drafted in to work in the Exon office, wherever it was located.108 There is also evidence which points to central production at Winchester. In general terms, the production of circuit returns at a central location would have been consistent with the principle of neutrality which clearly determined the selection of commissioners for the checking stage of the survey; for, having gone to considerable lengths to ensure that the survey was impartially supervised at this stage, it would have seemed illogical then to entrust the material to men with local vested interests. There is also palaeographical evidence which is suggestive of central production. At least five of the scribes who wrote Exon fiefs also worked on the Exon geld accounts,109 and these were surely copied from documents redacted in the treasury at Winchester to which they specifically refer.110 Since several of the scribes involved in Exon had spare capacity, it is possible that they contributed to the production of other circuit returns. If the circuit returns were compiled centrally in a series of separate offices, that would have made it easier to pool and allocate scribal resources. That makes Mu’s role all the more intriguing. His contributions to Exon amounted to about one-twelfth of the work undertaken by Alpha and Beta, so he could have spent as little as half a day of a six-day week in the Exon office. Did he spend the rest of his time supervising the other six circuits in Winchester in the summer of 1086? Whatever the case, the balance of evidence does now appear to tip in favour of the hypothesis that circuit returns were compiled there.
The timetable for stage four can be estimated from the volume of material written by Beta, Exon’s most prolific scribe. He wrote about 30 per cent of the extant text of Exon fiefs, and if he wrote a similar proportion of the lost sections of Dorset and Wiltshire, at a rate of 150 lines a day, starting a week after Whitsun, he would have finished in the last week of July. That is almost certainly the deadline that Exon scribes were working to. The annal for 1086 says that, having held court at Westminster at Pentecost, King William
ferde abutan swa þet he com to Lammæssan to Searebyrig,
þær him coman to his witan and ealle þa landsittende men þe ahtes wæron ofer eall Engleland, wæron þæs mannes men þe hi wæron,
ealle hi bugon to him
weron his menn
him holdaðas sworon þet hi woldon ongean ealle oðre men him holde beon.
travelled about so that he came to Salisbury for Lammas, and his councillors came to him there, and all the landholders who were of any account over all England, whichever man’s men they were, and they all bowed to him and became his men, and swore loyalty oaths to him that they would be loyal to him against all other men.111
Although the chronicle does not explicitly link the two events, Sir James Holt made a powerful case for thinking that the oath at Salisbury was the climax of the Domesday survey.112 Exon was certainly written before the Salisbury assembly, for one entry records that King William granted Bishop Walkelin of Winchester two manors (at Lydeard St Lawrence and Leigh near Taunton, Somerset), and instructed Bishop William of Durham to amend the breues accordingly;113 and it is clear that the breues were indeed amended, for the corresponding section of Exon is written by a scribe who does not otherwise occur in the manuscript.114
V
A few weeks after the great assembly at Salisbury, King William made what turned out to be his last Channel crossing to Normandy; but work on Domesday continued in his absence and now entered stage five, which involved the writing of Domesday Book itself: that is, LDB (Little Domesday Book), which covers Norfolk, Suffolk and Essex, and GDB (Great Domesday Book) which covers all other shires south of the Tees.115
The arrangement of LDB differs from Exon principally in that it structures material first by shire and within shire by fief, rather than vice versa. LDB also evinces a different pattern of scribal collaboration, for it was written by six scribes who tended to write long uninterrupted stints, and by a seventh who added running titles giving the name of the shire and the baron’s name at the top of each folio, as well as the manuscript’s colophon.116 This suggests that LDB is a fair copy of a feudally arranged recension which, like Exon, derived ultimately from a hundredally arranged text.
It remains unclear whether a fair copy of each circuit return was made. Galbraith and Flight have argued that a fair copy of Exon was produced before GDB was written and that this was the norm.117 Their case turns on two groups of entries in Exon which, they contend, were written after the rest of the manuscript by scribes who were not otherwise involved in its production. The first group consists of the ten consummatum est entries, but, according to Álvarez López, one of these was written by Alpha,118 and with one possible exception the rest were written by Mu.119 The second group consists of three entries in the margins of Exon: ‘Hoc scripsit Ricardus’, written at the foot of folio 316r; ‘hic debet esse hoc quod Jordan scripsit’, written at the foot of 406v; and ‘usque huc scripsit. R.’, written in the right-hand margin of 414r. Álvarez López has suggested that the first of these ‘could be the work of Alpha’,120 but has not matched the other two with scribes elsewhere in Exon. They may therefore be the work of fair-copyists. This must be weighed against the evidence that GDB was written directly from Exon. The GDB scribe certainly had Exon before him at some stage, for he wrote two short stints in the manuscript.121 Caroline and Frank Thorn have drawn attention to several features of GDB that are best explained by assuming that its main scribe worked directly from Exon.122 Chris Lewis has now identified further evidence which points in this direction: for example, he suggests that the letter ‘f’ was written frequently in the margins of the fief of Count Robert of Mortain in Exon when the GDB scribe restructured this complex fief to group the manors held by the count’s subtenants.123 This evidence tips the balance of probability towards the hypothesis that the GDB scribe worked directly from Exon without recourse to a fair copy.
However, the matter does not materially affect the chronology of the process. The colophon of LDB asserts (in capital letters): ‘anno millesimo octogesimo sexto ab incarnatione domini, uigesimo uero regni Willelmi, facta est ista descriptio, non solum per hos tres comitatus sed etiam per alios’ (‘this survey was made in the year one thousand and eighty six from the incarnation of the lord and the twentieth of the reign of William, not only throughout these three counties but also throughout the others’).124 This is not conclusive evidence that LDB was written in 1086, for the word descriptio could refer either to the survey, or to LDB itself, or both. However, the fact that the scribe did not specify which King William he was referring to commends the view that he was writing during the Conqueror’s lifetime. The manuscript’s palaeography certainly indicates that it could have been written in less than two months. The largest contribution was made by Rumble’s ‘scribe 2’, who wrote 276 pages averaging twenty-five lines per page;125 if he wrote 150 lines per day, this would have taken him less than eight weeks. It follows that if work on LDB began in early August, it could have been finished by Michaelmas (29 September) 1086.
A single scribe wrote most of GDB, though he was supported by a second scribe who checked the text and made about five hundred corrections and minor additions.126 The hand of the main GDB scribe has been identified in three manuscripts other than Exon, and Pierre Chaplais adduced evidence for linking these with Durham.127 However, in a wide-ranging review of this evidence, Julia Crick now argues that there are no specific palaeographical grounds for connecting the GDB scribe with Durham.128 By comparing Exon with the corresponding sections of GDB, it is also possible to reconstruct the main scribe’s working methods in detail.129 His editorial procedures were drastic. He jettisoned the geld accounts, TO and the summaries altogether; he compressed Exon fiefs by about 40 per cent, mainly by using much shorter formulae, and by omitting livestock and bynames; he reorganised the terra regis, grouping manors from the king’s various antecessores together; in other fiefs, he brought capital manors forwards and sometimes grouped the manors held by specific subtenants together. He also made the text more visually accessible by writing numbered chapter lists which correspond closely, though not always exactly, with the headings in the text, and by using rubrication and a range of other visual devices to pinpoint manors.130 He probably worked in a logical geographical order, proceeding north to south in great westerly sweeps, treating circuits VI, III, IV, V, I and II in that order. If so, Exon was the last circuit return he worked on, and he treated this in a westerly direction too, writing Wiltshire, Dorset, Somerset, Devon and Cornwall in that order. His intention was doubtless to treat the eastern shires last but he was for some reason prevented from doing so.
There is no decisive evidence to prove when GDB was written. David Douglas thought that it could have been written a decade or more after the survey,131 and David Roffe has recently argued that it was written circa 1089, connecting its production with the rebellion of 1088.132 There are, however, good reasons for thinking that the GDB scribe began in the autumn of 1086 and finished about a year later. The two entries which appear to style William de Warenne ‘earl’ turn out to be red herrings which can be set aside.133 Otherwise, there is no textual hint that GDB was written after the Conqueror’s lifetime. Indeed, as Sally Harvey acutely observes, the only entry in GDB which mentions William Rufus accuses him of spoliation.134 Frank and Caroline Thorn have shown that the main GDB scribe appears to have been working under considerable pressure of time and in the expectation that further detail might still become available to plug the gaps he occasionally left in the text.135 The best available evidence indicates that he could have written GDB in about 240 days.136 Howard Clarke has made a persuasive case for thinking that an abbreviated version of Domesday was made in the early 1090s, and that Orderic Vitalis was referring to this, not Domesday Book, when he described how Ranulph Flambard persuaded King William II to revise the survey of all England.137
The best explanation for the fact that GDB omits the eastern shires is that news of the king’s death (on 9 September 1087) caused a change of plan. We know that the first thing William Rufus did after his coronation at Westminster (on 26 September 1087) was to scrutinise the treasury at Winchester.138 Domesday Book was presumably among the treasures that it contained. However, because the GDB scribe had not by then written up LDB, a treasury official (Rumble’s ‘scribe 7’) added running rubrics to make it resemble GDB, presumably so that the two volumes could be presented to the new king.139 More generally, Exon and GDB evince continuity in form and purpose at a fundamental level, for both arranged material under the names of barons within shires and, as we shall see, this was an essential prerequisite for the administration of the royal demesne and royal lordship. Although GDB enhanced the utility of this material, it did so within a broad framework which must have been conceived before Exon was compiled. Taken in isolation, none of these points is decisive, but their cumulative weight is considerable. Domesday Book was almost certainly written between August 1086 and September 1087 to a plan carefully designed from the outset.
VI
Why was Domesday made? Several explanations have been posited: for example, Round argued that the survey was intended to facilitate the administration of the geld, and Maitland considered this to be the purpose of Domesday Book itself. Galbraith rejected this interpretation and argued that the sole purpose of the survey was to produce Domesday Book, which he thought was intended only to enhance the administration of the royal demesne and the exercise of royal lordship. Roffe contends that the ‘inquest’ was a collective response to the threat of invasion in 1085, and was intended to serve a range of purposes including the administration of the geld and knight service, and that Domesday Book was made for different purposes in the aftermath of the 1088 rebellion. Harvey argues that the book was the Conqueror’s initiative, and was designed to serve various fiscal purposes including the maximisation of income from the royal demesne and the profits of royal lordship, and the reassessment of geld liabilities based on ploughland data; she also contends that the survey was intended to make royal officials more accountable and landholders more secure in their tenures.140 The interpretation developed below shares elements in common with some of these arguments, but proceeds from the conviction that none of them satisfactorily accounts for all of Domesday’s outputs. The suggestion is that the survey was carefully designed to generate several different types of record, each intended to serve a specific administrative function. This argument is informed by the analysis in the foregoing sections of this article, and where possible by comparison.
The pipe roll of King Henry I for the 1129–30 fiscal year is especially valuable, for it constitutes our only opportunity to examine the English revenues of a Norman king in detail.141 It reveals that Henry’s revenues flowed from four main sources: the royal demesne, the geld, the profits of government and justice, and the profits of royal lordship (see Appendix II). The kingdom’s fiscal machinery doubtless evolved between 1086 and 1130, but it is demonstrable that the Conqueror generated revenue from similar sources and that each posed specific administrative challenges. It is also clear that the Domesday survey generated records that were carefully structured to address precisely those challenges: it was, therefore, a fiscal exercise in the broadest sense, intended to maximise royal income from every conceivable source. The 1130 pipe roll also reminds us that there was an important political dimension to the administration of royal finance, for it reveals that Henry extracted nearly half of his income directly from members of the landholding elite through a plethora of placita (pleas) and conuentiones (agreements), mostly relating to baronial acquisitions of property, offices and other privileges. This meant that many barons were indebted to the king at any moment in time, and the king could exploit this for political leverage by deciding whether to call in his debts rapidly, gradually or not at all by granting pardons and exemptions.142 The system therefore enabled the king to exchange financial capital for political capital with ease, and that is also important for understanding the purposes of Domesday, for its fiefs and summaries were structured to inform similar negotiations, and the whole survey embodied a great conuentio between the Conqueror and his barons.
Domesday Book was not intended to facilitate a reassessment of the geld, but the first draft of the survey was. The hundredal recension created the potential to undertake a systematic reassessment of geld liabilities, aligning them more closely with their capacity to pay.143 It structured information in precisely the way that the geld was administered and collected, by hundred, vill and manor; and it recorded details of each manor’s productive assets together with an estimate of how much money it could be expected to generate each year.144 Equipped with this information, treasury officials could readily identify manors that were too heavily or lightly burdened and reassess liabilities using common ratios of hidage to value to make the system more equitable, efficient and profitable.
This is essentially the case which Round and Maitland made more than a century ago and which Galbraith tried to refute. Round drew attention to the many anomalous geld assessments recorded in Domesday and argued that the survey was intended to facilitate a reassessment: one of its ‘main objects’ was to ‘obtain a register of real, as distinct from assessed, value’.145 He also argued that the survey’s ‘original returns’ were intended for this purpose: that is why he considered ICC to be the ‘true key to the Domesday survey, and the system of land assessment it records’.146 Maitland agreed that the geld system ‘was full of anomalies and iniquities’, partly because it was originally imposed using a simple, top-down method of ‘subpartitioned provincial quotas’ which did not result in an even or equitable distribution of the geld’s burden, and partly because kings subsequently used the system as an instrument of patronage, making grants of ‘beneficial hidation’ to their favourites. Consequently, by 1086, the ‘assessment of the geld sadly needed reform’. Maitland conceded that the king’s commissioners may have achieved other ‘minor purposes’ but thought that ‘the quest for geld was their one main object’. Indeed, unlike Round, he concluded that this was the fundamental purpose of Domesday Book itself. ‘Our record is no register of title, it is no feodary, it is no custumal, it is no rent roll; it is a tax book, a geld book.’ Its ‘main theme’ is ‘geldability, actual or potential’.147
Galbraith’s attempt to refute this hypothesis consisted of four main propositions. First, he argued that, since Domesday Book is organised on a feudal plan, it could not have been intended to facilitate the administration of the geld—and he was surely right about this. However, as we have already seen, Galbraith also argued that the survey’s ‘original returns’ were not always hundredally organised like ICC; that Domesday Book was the survey’s only intended output; and that the reassessment hypothesis was ‘just another Victorian anachronism’.148 He was wrong about all of this. It should now be clear that the first draft of the survey was hundredally organised throughout the kingdom and that this was one of several outputs from the survey. It is also demonstrable that several ancient and medieval regimes used cadastral surveys to bring fiscal burdens and landed wealth into closer alignment. For instance, that was the essence of the reforms of the late Roman taxation system introduced during the reign of the emperor Diocletian (284−305).149 It also lay at the heart of the land tax system in Byzantium, which used periodic cadastral surveys to distribute the burden of taxation in proportion to taxpayers’ resources; indeed, we have two treatises roughly coeval with Domesday which describe how this was done, village by village, such that each taxpayer’s liability was a function of his share of the village’s landed resources.150 An exact contemporary of William the Conqueror in Sung China, Wang An-shih, one of the leading government officials under Emperor Shen-tsung (1067−85), conceived and implemented a ‘Land Survey and Equitable Tax Policy’: from the early 1070s, landholders were asked to report the extent and quality of their own land, and taxation was reapportioned on this basis, so that by 1085 some 35 million acres were recorded in the north of the empire—an area roughly five times the size of Domesday England.151 The administration of the land tax in Egypt during the central Middle Ages also drew upon elaborate surveys. These were essential because the flow of the Nile peaked at different levels each year, causing significant variations in the extent of land available for agricultural production: fiscal surveys were therefore compiled in the autumn and spring of each year to record the capacity of every cultivated plot, which tax officials then used to calculate tax liabilities.152 The use of surveys to align fiscal liabilities with economic realities was a common denominator among ancient and medieval land-tax systems because it was common sense.
Further objections to the geld hypothesis have since been made, but none of these persuade. It would indeed ‘have been a herculean task to re-assess hides and carucates completely, to make them a more realistic measurement of the country’s landed wealth’,153 but the Domesday survey reduced this task to manageable proportions by assembling all the data in a coherent format. It also constitutes incontestable evidence that the Conqueror’s regime could successfully implement major administrative exercises. It is true that we have no clear evidence that a reassessment of the geld was attempted after 1086, but it is essential to distinguish between the survey’s intended purposes and how its outputs were eventually used, for the Conqueror’s sudden death prevented him from making systematic use of them, and triggered a lengthy period of political turbulence during which geld reform is unlikely to have been a priority.154 The suggestion that Domesday Book was intended to restructure the system of geld collection on a feudal plan is improbable,155 for the Exon geld accounts prove that the system of geld collection was remarkably effective.156 We also know that assessment lists continued to be drawn up on a cadastral plan in the early twelfth century,157 and that the land tax was still collected on a geographic plan in the 1190s.158 It is also improbable that Domesday was intended to facilitate a reassessment of the geld based on ‘ploughland’ data,159 for there is a simpler and better explanation for this: it indicated where there existed scope to exploit manors more intensively, and was intended to make the management of the royal demesne and royal lordship—not the geld—more efficient.160
The Domesday survey was also designed to maximise income from the royal demesne. The motives for this are not difficult to identify. The total value of revenues due to the king from royal manors and boroughs recorded in Domesday Book amounted to about £17,800 per annum.161 This was a prodigious sum, representing about one-fifth of the landed wealth recorded in Domesday.162 The royal demesne was, however, a complex operation and its revenues were far from guaranteed; indeed, it is instructive that about one-fifth of the revenues from the royal demesne remained outstanding in 1130.163 The Domesday ualet or value represented an estimate of the total income the lord of each manor could expect to generate each year from demesne farming, the rents paid by dependent peasantry, and other renders from customary or jurisdictional dues.164 It was therefore a notional figure, akin to the revenue budget line in modern management accounts. To collect what was due to him, the king relied upon many agents including sheriffs plus a wider network of custodians and reeves responsible for managing royal manors, either singly or in groups;165 and it was not always easy to prevent the king’s money from sticking to the fingers of those who handled it as it passed through the system.
The Domesday survey addressed these challenges in various ways. The survey constituted a searching inquiry into the activities of all the officials responsible for administering the royal demesne, creating a comprehensive list of infringements on royal property and encouraging a culture of accountability.166 It also created a comprehensive description of the royal demesne—an inventory. The 1085 annal says that the survey recorded ‘hwet se cyng him sylf hæfde landes orfes innan þam lande oððe hwilce gerihtæ he ahte to habbanne to .xii. monþum of ðære scire’ (‘what land and livestock the king himself had in the land, or what dues he ought to have in twelve months from the shire’),167 and that is just what Domesday Book does. Most shires begin with a passage, written before the chapter lists, recording the king’s property in boroughs together with a miscellany of specific services and payments due to the king, before listing all the king’s manors under the terra regis.168 This defined the totality of the royal demesne and made it possible to track changes to its composition. It also furnished the treasury with all the information it needed to account for the king’s revenue from each shire: it was indeed ‘the king’s rent book’,169 which enabled treasury officials to calculate precisely what they should demand from the various officials who managed the king’s boroughs and manors. Domesday Book rarely records such totals, but some are known to have been calculated during the survey, for a few of them found their way into GDB,170 and the Exon scribes used a space left blank at the end of the account of the king’s manors in Devon to make (or copy) a few calculations of this kind.171 The Domesday corpus created the potential to recalculate such totals each time the configuration of local personnel and their responsibilities changed.
In addition, the survey enabled treasury officials to determine where there was potential to generate more income from the king’s manors. Here it is relevant that many royal manors were often held by reeves ad firmam, that is for a fixed rent,172 and that this privilege could be auctioned to the highest bidder. According to the annal for 1087:
Se cynge sealde his land swa deore to male swa heo deorost mihte. Þonne com sum oðer
bead mare þonne þe oðer ær sealde,
se cyng hit lett þam menn þe him mare bead. Þonne com se þridde
bead geat mare,
se cyng hit let þam men to handa þe him eallra meast bead
ne rohta na hu swiðe synlice þa gerefan hit begeatan of earme mannon ne hu manige unlaga hi dydon.
The king sold his land on very hard terms—as hard as he could. Then came somebody else, and offered more than the other had given, and the king let it go to the man who had offered him more. Then came the third, and offered still more, and the king gave it into the hands of the man who offered him most of all and did not care how sinfully the reeves had got it from poor men, nor how many unlawful things they did.173
Domesday’s manorial descriptions were designed to inform such bargains, for the details they included combined to answer the last question in the survey’s terms of reference: whether more could be taken from each manor. The ploughland formula, ‘terra est pro n carrucis’ (‘there is land for so-many ploughs’), constituted a particularly well targeted answer to this question, for it represented an estimate of the number of plough teams required to cultivate each manor’s arable land, which could be compared with the existing number of operational teams to pinpoint inefficiencies. Thus, if an entry revealed that the ploughland figure was greater than the recorded number of plough teams, that would indicate scope to generate more revenue by investing in plough traction. King Henry II’s survey of the lands of widows, heiresses and wards ‘in the king’s hand’ helps bring this into focus, for it routinely indicates whether a manor could be worth more (potest plus ualere) if it were more fully stocked with oxen or other livestock.174
There are strong grounds for thinking that the Domesday survey of the royal demesne was partly inspired by Carolingian models.175 Carolingian legislation contains many exhortations to royal officials—missi, counts, bailiffs and so on—to safeguard and manage royal estates and benefices honestly and efficiently, and to document their activities.176 For example, Capitulare de Villis is a normative statement of the responsibilities of an ideal estate manager, which included the regular delivery of written reports to the palace; and the Breuium exempla, which occurs immediately before de Villis in the same manuscript, was clearly intended to provide models for estate records, conventionally known as polyptychs.177 Although most of the extant polyptychs relate to the endowments of religious houses, they survive from many parts of the empire and in numbers that suggest that literate landlordship was both extensive and routine. Polyptychs served various functions. Some served as inventories which defined the overall structure of endowments and thus constituted proof of title, affording religious communities a measure of protection from the political and military risks to which they were exposed.178 However, most contained detailed descriptions of individual estates and their assets and human resources and were surely intended to be used alongside other more ephemeral records to facilitate routine administration of landlordship.179 If so, polyptychs served similar functions to Domesday entries and fiefs. This poses a problem of transmission, for most polyptychs were compiled between the late eighth and early tenth centuries. James Campbell’s solution was to suggest that, although the estate memoranda which survive from late Anglo-Saxon England differ considerably in form and scale from Carolingian polyptychs, they should nevertheless be placed in broadly the same tradition.180 According to this reading, the technologies of Carolingian estate administration were transferred to Anglo-Saxon England and thence transmitted to its conquerors. That hypothesis remains attractive, but it is also possible that the technologies of literate estate management remained current on the Continent and were imported to England from there during the Conqueror’s reign, for Jean-Pierre Devroey has shown that polyptychs continued to be used and updated long after they were produced, sometimes deep into the tenth and eleventh centuries;181 and although there are no surviving polyptychs from Normandy, it is demonstrable that several English bishops and abbots at the time of Domesday had been trained in, or were in contact with, continental religious houses from which polyptychs survive.182 It is therefore probable that many of those who planned and implemented the Domesday survey were familiar with such records, and it follows that if someone at the assembly where Domesday was launched had explained that the survey would result in the production of a polyptych covering the entire the royal demesne, a critical mass of those present would have understood just what this meant.
In addition, Domesday Book both articulates and was designed to facilitate an intensive form of royal lordship, which exploited the fact that all land except for the royal demesne was considered to be held from the king.183 The pipe rolls, treatises and surveys of the later twelfth century reveal how this worked and why it was immensely profitable.184 Whenever a baron died, his or her honor would escheat to the king and would remain under his control until a successor performed homage for it before entering the land. The king could also profit from this in various ways. An honor might escheat permanently to the king if a baron forfeited for political disobedience or a felony, or died without heirs, in which case the king could either keep the honor in hand or sell it to another lord who had made a suitable proffer for that privilege. If a baron were survived by an adult son or heir, the king could charge him a fee, known as a relief, to enter the land. If a male heir was a minor, the king could exercise rights of wardship: that is, he could take custody of the heir’s person and lands until he came of age, enjoy their revenues throughout the intervening period, and then decide whether to charge the heir a relief to enter the land once he reached maturity. If a baron was survived by a widow, she would be entitled to her marriage portion and dower, but this would make her an attractive financial proposition to potential suitors, for if she married again, her new husband could assume control of her lands for his lifetime, and since widows were considered to be ‘in the king’s gift’, the king could take proffers from prospective suitors, or from widows for the right to remain unmarried. The king could similarly profit from heiresses by taking control of their lands if they were minors, or by selling the right to marry them. The revenues from reliefs, wardships and marriages are often labelled ‘the incidents of tenure’ or ‘feudal incidents’ in the literature. The king could also profit from succession to the religious honors, for whenever a bishop died, the king would appoint officials to secure control of the relevant bishopric’s temporalities until a successor was installed in office and enjoy its revenues throughout the intervening period, and similar procedures applied to the temporalities of abbots and abbesses; these practices are referred to as ‘vacancies’ or ‘regalian right’ in the literature.
Similar forms of royal lordship can be traced back close to the time of Domesday.185 Almost every membrane of Henry I’s pipe roll records payment for vacancies, forfeitures, escheats, reliefs, marriages, wardships, and other succession proffers—indeed, these accounted for about a third of Henry’s income in 1130.186 In his coronation charter, issued in 1100, Henry promised to make more restrained use of precisely these rights and implied that his brother and father had abused them.187 Several chronicles confirm that William II exploited vacancies intensively.188 For example, Eadmer supplies a vivid description of the administration of Christ Church Canterbury’s endowment during the vacancy after Lanfranc’s death in 1089, when the king sold jus in ea dominandi to the highest bidder and renegotiated the price frequently, so that grim-faced custodians were regularly seen throwing their weight about in the monastic precincts while exacting the king’s money. Eadmer explains that such practices were a function of a new phenomenon, namely that after the conquest of England, no one was ever made a bishop or abbot ‘qui non primo fuerit homo regis’ (‘who had not first been made the king’s man’). He goes on to say that the Conqueror made bishops, abbots and other principes obey him to the extent that ‘cuncta … diuina simul et humana ejus nutum expectabant’ (‘all things, spiritual and temporal alike, waited upon his nod’); and that his readers could readily infer what the king decreed in saecularibus from what he ordained for the divine.189 There are echoes of all this in the Anglo-Saxon Chronicle. The annal for 1100 says that whenever bishoprics or abbacies became vacant, Rufus sold them for money or kept them in hand and let them out for rent ‘forþan þe he ælces mannes, gehadodes læwedes, yrfenuma beon wolde’ (‘because he intended to be heir of everyone, cleric and lay’).190 We have seen that the annal for 1087 describes how the Conqueror auctioned the custody of land to the highest bidder; it also says that powerful and wretched men alike had to ‘þes cynges wille folgian gif he woldan libban oððe land habban’ (‘follow the king’s will if they wanted to live or have land’).191
Royal lordship was not, of course, a new phenomenon in 1066, but there are good reasons for thinking that it became suddenly more intensive during the Conqueror’s reign because he advanced a novel claim to be the source of all tenure in England. George Garnett has shown that this was a function of his claim to the kingdom itself: Edward the Confessor had made William his heir, but when Harold’s usurpation prevented him from entering his lawful inheritance, all legitimate tenure in England ceased until it was restored by William’s gift. That is why every entry in Domesday Book records not only the current landholder, but also the antecessor who held tempore regis Edwardi, or as Exon puts it with greater precision, ‘ea die qua rex Eduuardus fuit uiuus et mortuus’ (‘on the day King Edward was alive and dead’). This formulation implied that all landholding was legitimated in the same way as the conquest itself, and it followed that all landholders in 1086 were doubly dependent upon the king, since he had both granted them their lands and legitimated their tenures.192 Domesday Book implies tenurial dependence in other ways. Its chapter lists make it clear that all the land in each shire consisted of either the terra regis or the lands of other tenentes, and this structure implies that the whole kingdom was conceived as a gigantic manor in which all land was either royal demesne or held from him by tenants.193 Admittedly, the chapter lists do not usually specifically assert that these tenentes held their land from the king.194 However, two of them do. The short list for Shropshire begins with the heading ‘Hic Annotantur Tenentes Terram De Rege In Sciropescire’ (my italics).195 The tenentes in this list comprise two English bishops, one French abbey, five secular tenants-in-chief and Earl Roger, who held the rest of the shire with his men. The list for Cheshire consists of the following statement: ‘In Cheshire, the bishop … holds de rege what belongs to his bishopric. Earl Hugh with his men holds de rege all the rest of the land of the shire.’196 This proves that the GDB scribe considered all the land in Shropshire and Cheshire to be held de rege without exception, and it is probable that he thought this point needed special emphasis in these two shires because their tenurial structure was exceptional in that the king had no demesne manors there. This suggests that the GDB scribe considered the notion that all land was held de rege to be implicit in other shires. His use of the formula ‘X tenet de rege’ within fiefs is similarly suggestive. In many fiefs, especially those of secular tenentes, the scribe used this formula only once, prominently, in the first line of the first entry. It therefore functioned like a fief heading, or an extension of it, in that it applied to the entire fief: the clear implication is that every manor in the fief was held de rege. The average frequency which this formula occurs in the first entries of fiefs within each circuit is as follows: 5 per cent in circuit VI, 24 per cent in circuit III, 42 per cent in circuit V, 43 per cent in circuit I, 70 per cent in circuit IV, and 68 per cent in circuit II. This was probably the order in which the GDB scribe wrote up the circuits.197 It therefore emerges that he articulated the idea that tenentes held de rege with growing confidence as his work progressed.198
It is also clear that Domesday Book was designed to give that idea greater practical force. The Dialogus de Scaccario, written by the king’s treasurer Richard fitzNigel in the last decade or so of Henry II’s reign, says that whenever an honor fell into the king’s hand, royal officials were immediately sent to inspect it thoroughly, add up its revenues and appoint the sheriff or another custodian to account for it at the exchequer.199 Domesday Book was designed to facilitate precisely these procedures. The chapter lists, the corresponding numbered headings for each fief, the running headings throughout the text and its systems of capitalisation and rubrication, enable the reader to locate each honor and its constituent manors efficiently.200 This is precisely what Richard fitzNigel says: he observes that Domesday Book is arranged so that the lands of those who ‘de rege tenent in capite’ are easily found (‘facilius occurrunt’).201 This meant that treasury officials could swing into action the moment they heard news that a baron had died. However, instead of sending officials to survey the honor, they could take immediate control of it using Domesday Book to list all its manors before instructing sheriffs to take them into the king’s hand. Treasury officials could then determine the value of the honor by adding up its revenues. If the survey generated summaries for every honor,202 that information would also be readily available; if not, it could be calculated from scratch with reasonable dispatch. The treasury would then be able to assist the king with whatever course of action he chose. For instance, if the king wished to install a successor immediately, he could negotiate the relief from a position of strength, knowing the honor’s precise content and value; the same would apply to negotiations with suitors for marriage to widows and heiresses. If the king’s intention was to keep the honor in wardship or in hand throughout a vacancy, his officials would know precisely what to demand from its custodians. In addition, Domesday’s manorial details would help prevent custodians from stripping assets and would allow treasury officials to determine where there was scope to generate more revenue by investing in plough traction or livestock. In short, Domesday was beautifully designed to make the new tenurial principles of conquered England more workable in practice.
We turn finally to Domesday’s judicial functions. Why did the survey record so much contested matter? That problem is itself contested.203 The suggestion here is that the survey was designed to identify contested tenures pending a judicial review, and that the determination of title to property was integral to the whole exercise.
The matter turns partly on one’s estimate of how orderly the process of land taking was between 1066 and 1086. Sir Frank Stenton argued that the ‘tenurial revolution never degenerated into a scramble for land’ because in ‘every part of England the great redistribution was controlled by the king, and carried out by his ministers on lines which the king himself laid down’.204 Robin Fleming has challenged this view, characterising the two decades before Domesday as ‘a period of jarring and violent tenurial discontinuity’, precisely because there had been a scramble for land—or, as she put it, ‘a land rush … which more closely resembled a free for all than a centrally sponsored programme of colonization’. Indeed, Fleming considers ‘the scale of such appropriations … astonishing’, estimating that it accounted for ‘well over half the land transference’ in some regions.205 She therefore interprets the Domesday survey as an attempt to restore order to a process which had got out of control.206 This remains an important corrective. However, Fleming’s estimate of the scale of the problem is out by an order of magnitude. Exon TO supplies the most detailed record of contested tenures in the Domesday corpus, and the total reported value of the land it records was about 7 per cent of the total value of the land in the three shires Exon covers. We should allow for the possibility that some unauthorised land grabbing went unreported, but it remains clear that the proportion of contested property was much lower than Fleming suggests—nearer 10 per cent than 50 or more. The very basis on which tenures were contested during the survey is also instructive, for, as we have seen, literally hundreds of entries assume that no one could legitimately acquire land in England without the king’s consent.207 There are also good reasons for suspecting that the whole process of land redistribution was controlled with reference to geld assessment lists, for the geld system could not have functioned unless such lists were updated every time the king authorised a transaction. Indeed, it is probable that the regime’s geld assessment lists functioned as a land registry system, for there was an intimate connection between the payment of geld and title to property throughout the eleventh century.208 In this connection, it is relevant that the Normans used the remnants of Arabic tax records to inform the process of land taking when they conquered Sicily in the late eleventh century.209
It follows that the Domesday survey was intended to bring closure to a process which had already been carefully documented. Its record of contested matter was also integral to the survey’s wider fiscal objectives. The identification of infringements on the king’s property was part of a drive to increase yields from the royal demesne. The administration of the geld turned on knowing precisely who held each parcel of land and who, therefore, was liable to ‘defend it’ to the geld; the process of identifying and ultimately resolving contested tenures would therefore improve yields from the geld by eliminating ambiguity as to the locus of liability. The administration of royal lordship was also dependent upon accurate records of baronial holdings, and by resolving disputes and unauthorised tenures, the king would remove any vestige of doubt that all land was held de rege. In addition, the settlement of disputes would itself be a profitable exercise. The 1130 pipe roll reveals that Henry I routinely demanded substantial sums from his subjects for ‘help in judicial matters’210—essentially bribes paid to secure the outcome of property disputes. Here one recalls a passage in the Gesta Stephani, which says that no one outside Henry’s group of favourites could gain a hearing relating to any matter until he had oiled the king’s palm.211 The Conqueror doubtless intended to do much the same on a greater scale. Indeed, there is suggestive evidence that he began to do so in the immediate aftermath of the survey, for the 1086 annal says that, before crossing to Normandy, the king obtained a great deal of money from his men on any pretext, whether just or otherwise.212 The fact that the Conqueror died suddenly in 1087 is sufficient to explain why Domesday disputes were never systematically resolved, though there is charter evidence that he and his sons processed some of the litigation it generated.213
The barons’ motives were more straightforward: they simply wanted greater security of title. The survey took place after two decades of tenurial upheaval. Many belonged to the first generation of barons in conquered England who held through acquisition not inheritance.214 Indeed, succession practices in England remained distinctly unsettled, for in theory all legitimate tenures were defined with reference to that slippery phenomenon, the antecessor, and needed to be personally sanctioned by the king. Although the regime had made a serious and sustained attempt to record and communicate each transfer of property, it had inevitably proved impossible to keep track of each one, and although many barons possessed writs or charters relating to some of their holdings, few possessed royal documents that authorised them all. There were, however, precedents for such documents, for pancartes and ‘confirmation charters’ had become an increasingly common way of documenting and protecting the endowments of religious houses in Normandy during the Conqueror’s reign.215 These were surely important among the conceptual models that informed the survey, for as Holt put it, Domesday gave each baron ‘a record of their tenure, in effect a confirmation of their enfeoffment’.216 The Domesday survey was also an intensely dramatic episode which participants were unlikely to forget,217 and that is relevant because memorable events or gestures were often important elements of the property conveyance process, helping to fix details of transactions into the memories of witnesses.218 That is why the whole process culminated in another remarkable ritual, when ‘ealle þa landsittende men þe ahtes wæron’ performed homage to the king at Salisbury for lands assigned to them in the great confirmation charters the survey had generated.219
The Domesday survey thus embodied many of the documentary and ritual elements of land conveyance procedures, and each stage made tenures progressively more secure. Stages one and two grew out of geld assessment lists and proceeded in tandem with a major geld levy, in a legal milieu which still privileged those who ‘defended’ their land to the geld. Stage three ensured that property rights were witnessed in public courts, drawing upon sworn testimony, thus invoking the sanction of divine retribution like the anathema clauses of charters. Stage four reconstituted this matter into feudally arranged lists, which were effectively confirmation charters of uncontested tenures, and inventories which would help barons and subtenants protect their land when it escheated into royal control. The whole process was memorably dramatic and reached a ritual climax with the performance of homage for the holdings they recorded. This amounted to just about the most comprehensive package of security that was conceivable within the framework of dependence that the survey enshrined.
VII
It is now possible to imagine what the king and his witan were thinking and talking about at Gloucester in the midwinter of 1085. Since they were celebrating Christmas, they were doubtless reminded that Caesar Augustus had decreed that all the world should be surveyed at the time of Christ’s birth.220 Because William intended to issue a similar decree within his own orbis, everyone had to be briefed. His kingdom would be divided into seven circuits, each surveyed by two groups of commissioners. The first group would integrate information taken from geld assessment lists with manorial detail supplied by landholders in documents organised by hundred within each shire: the deadline for this was Easter. That descriptio would then be checked, corrected and fleshed out with contested matter in public sessions organised by the second group: the deadline for this was Whitsun. All this would proceed in tandem with the collection of the geld. The survey would then be rearranged into further documents recording the holdings of the king and each baron, and these would be summarised into totals; the deadline for this was Lammas, when everyone would perform homage to the king. After that, the survey would be compressed into a single volume, and the entire collection could be used for its intended purposes. There would be a comprehensive review of geld liabilities, which would be reassessed from the bottom up, linking each manor’s liability with its capacity to pay on a broadly equitable basis throughout the kingdom. The royal demesne would be administered with a tighter grip: any property which had been taken from the king would be restored to him, and treasury officials would explore ways of making more money from the king’s manors. The profits of royal lordship would also be administered more systematically, for the king and his officials would now know the precise composition and value of every honor and would therefore be better placed to manage escheats and negotiate reliefs, marriages, wardships and vacancies. In addition, when circumstances allowed, the king would lead a great judicial enquiry to work through the survey’s contested matter, and once that was done everyone would know exactly who held what, and that it was all held de rege. The atmosphere doubtless became tense as the barons absorbed the enormity of all this, for they were essentially being asked to furnish the king with the material he needed to tighten his fiscal grip upon them. The more courageous and vociferous among them must have demanded to know what they might expect in return. However, the survey’s brief contained a good answer for that too: the whole process would make their holdings more secure, because their fiefs would function like the confirmation charters that were all the rage in France—and that was just enough to secure baronial compliance.
This was an extraordinarily well formulated plan. When and by whom was it conceived? There have been several attempts to identify ‘the man behind the survey’.221 It is, however, improbable that the survey was the brainchild of a single genius. It is also important to differentiate between the plan and its implementation. Chaplais listed excellent reasons for thinking that Bishop William of Durham was placed in charge of the operation once it was launched.222 The palaeographical evidence that he adduced for linking the GDB scribe with Durham must now be set aside,223 but the rest of his case remains compelling: William is named in connection with the geld inquest in the south-western shires;224 he was instructed by the king to amend the Exon breues at Salisbury;225 he witnessed a writ relating to Pyrford in Surrey which was issued ‘post descriptionem totius Angliae’ and caused the GDB scribe to make a marginal amendment;226 and he witnessed seven further extant charters of the Conqueror and William II which deal with matters relating to the survey.227 Bishop William was also a natural candidate to direct a survey which embodied the principle that it should be checked by neutral parties, for the church of Durham and much of its endowment lay beyond the survey’s northernmost limits. There is also cumulatively compelling evidence that he enjoyed a prominent role at the heart of the Conqueror’s government throughout the last few years of the reign.228 He was surely the man who ran the survey. It is equally clear that Ranulph Flambard eventually became the master of the fiscal apparatus it enhanced.229 However, that does not prove that either of them designed the survey alone. Indeed, it is much more likely that the survey was planned by a group of people who ran the Conqueror’s government: men who had staffed the king’s household, writing office and treasury, and had mastered the technologies of English government, but who had nevertheless struggled to cope with the growing scale and complexity of the king’s revenues. Most of them had been trained in the schools of north-western Europe in the hope of finding service in English episcopal households, or better still the royal household, which offered the prospect of lucrative rewards for talented administrators, including the ultimate prize of promotion to bishoprics: men such as Bishop Osmund, who had served as the king’s chancellor before he became bishop of Salisbury in 1078;230 and Maurice, who succeeded Osmund as the king’s chancellor between 1078 and 1085, before being promoted to the see of London at the very assembly where Domesday was launched.231 It is unlikely to be a coincidence that Osmund and Maurice both enfeoffed Ranulph Flambard, an otherwise modest landholder.232 They and other members of this group were surely the men behind the survey, for they had a detailed grasp of the physicalities of fiscality in conquered England, and therefore knew precisely what structured information the king needed to enhance the flow of revenue into his treasury. They had probably been drawing up plans for the survey for many years, hoping that there would be an opportune moment to persuade the king to bring them to fruition.
That moment came in 1085 and was triggered by an external threat. The king had been in Normandy when news came that Cnut IV of Denmark was planning to invade England with Count Robert of Flanders. With characteristic vigour, the Conqueror mobilised a great army from throughout northern France, sailed it to England, arranged for the troops to be billeted among his barons, strengthened fortresses, and wasted parts of the coastline to prevent invaders from easily securing provisions. Then, when further news came that Cnut’s invasion had been hindered, the Conqueror allowed some of his troops to return home, arranged for the remainder to stay in England for the winter and finally convened the Gloucester assembly where Domesday was launched.233 These events surely caused the king to focus on all of his resources, fiscal, military and political; but it also meant that the winter of 1085 was a uniquely propitious time to take the survey, for the simple reason that it was a rare occasion when the king and all the landholders of any account were all in England. The survey was therefore a consequence of the invasion threat, but its origins lay deeper.234 It was also executed with stunning efficiency. Historians have often boggled and sometimes baulked at the notion that the survey was completed in 1086 and that the book was written shortly afterwards. They must now confront what amounts to overwhelming evidence that it was made, checked and rearranged in seven months flat while a heavy geld was collected and accounted for, and that Domesday Book was finished about twelve months later. Little wonder all this provoked ‘mycel geþeaht swiðe deope spæce’: that midwinter assembly was contemplating one of the most ambitious feats of government ever undertaken.
Appendix I
The proportion of hundredally grouped entries in Domesday Book.235
| Shire | Circuit | Percentage of entries grouped in hundreds |
|---|---|---|
| Berkshire | I | 99 |
| Hampshire | I | 100 |
| Kent | I | 88 |
| Surrey | I | 91 |
| Sussex | I | 88 |
| Cornwall | II | 70 |
| Devon | II | 85 |
| Dorset | II | 88 |
| Somerset | II | 86 |
| Wiltshire | II | 82 |
| Bedfordshire | III | 97 |
| Buckinghamshire | III | 100 |
| Cambridgeshire | III | 99 |
| Hertfordshire | III | 100 |
| Middlesex | III | 100 |
| Leicestershire | IV | 91 |
| Northamptonshire | IV | 81 |
| Oxfordshire | IV | 74 |
| Staffordshire | IV | 90 |
| Warwickshire | IV | 91 |
| Cheshire | V | 77 |
| Gloucestershire | V | 97 |
| Herefordshire | V | 92 |
| Shropshire | V | 73 |
| Worcestershire | V | 97 |
| Derbyshire | VI | 95 |
| Huntingdonshire | VI | 100 |
| Lincolnshire | VI | 93 |
| Nottinghamshire | VI | 84 |
| Rutland | VI | 100 |
| Yorkshire | VI | 88 |
| Essex | VII | 90 |
| Norfolk | VII | 88 |
| Suffolk | VII | 92 |
| Mean average | 90 |
| Shire | Circuit | Percentage of entries grouped in hundreds |
|---|---|---|
| Berkshire | I | 99 |
| Hampshire | I | 100 |
| Kent | I | 88 |
| Surrey | I | 91 |
| Sussex | I | 88 |
| Cornwall | II | 70 |
| Devon | II | 85 |
| Dorset | II | 88 |
| Somerset | II | 86 |
| Wiltshire | II | 82 |
| Bedfordshire | III | 97 |
| Buckinghamshire | III | 100 |
| Cambridgeshire | III | 99 |
| Hertfordshire | III | 100 |
| Middlesex | III | 100 |
| Leicestershire | IV | 91 |
| Northamptonshire | IV | 81 |
| Oxfordshire | IV | 74 |
| Staffordshire | IV | 90 |
| Warwickshire | IV | 91 |
| Cheshire | V | 77 |
| Gloucestershire | V | 97 |
| Herefordshire | V | 92 |
| Shropshire | V | 73 |
| Worcestershire | V | 97 |
| Derbyshire | VI | 95 |
| Huntingdonshire | VI | 100 |
| Lincolnshire | VI | 93 |
| Nottinghamshire | VI | 84 |
| Rutland | VI | 100 |
| Yorkshire | VI | 88 |
| Essex | VII | 90 |
| Norfolk | VII | 88 |
| Suffolk | VII | 92 |
| Mean average | 90 |
| Shire | Circuit | Percentage of entries grouped in hundreds |
|---|---|---|
| Berkshire | I | 99 |
| Hampshire | I | 100 |
| Kent | I | 88 |
| Surrey | I | 91 |
| Sussex | I | 88 |
| Cornwall | II | 70 |
| Devon | II | 85 |
| Dorset | II | 88 |
| Somerset | II | 86 |
| Wiltshire | II | 82 |
| Bedfordshire | III | 97 |
| Buckinghamshire | III | 100 |
| Cambridgeshire | III | 99 |
| Hertfordshire | III | 100 |
| Middlesex | III | 100 |
| Leicestershire | IV | 91 |
| Northamptonshire | IV | 81 |
| Oxfordshire | IV | 74 |
| Staffordshire | IV | 90 |
| Warwickshire | IV | 91 |
| Cheshire | V | 77 |
| Gloucestershire | V | 97 |
| Herefordshire | V | 92 |
| Shropshire | V | 73 |
| Worcestershire | V | 97 |
| Derbyshire | VI | 95 |
| Huntingdonshire | VI | 100 |
| Lincolnshire | VI | 93 |
| Nottinghamshire | VI | 84 |
| Rutland | VI | 100 |
| Yorkshire | VI | 88 |
| Essex | VII | 90 |
| Norfolk | VII | 88 |
| Suffolk | VII | 92 |
| Mean average | 90 |
| Shire | Circuit | Percentage of entries grouped in hundreds |
|---|---|---|
| Berkshire | I | 99 |
| Hampshire | I | 100 |
| Kent | I | 88 |
| Surrey | I | 91 |
| Sussex | I | 88 |
| Cornwall | II | 70 |
| Devon | II | 85 |
| Dorset | II | 88 |
| Somerset | II | 86 |
| Wiltshire | II | 82 |
| Bedfordshire | III | 97 |
| Buckinghamshire | III | 100 |
| Cambridgeshire | III | 99 |
| Hertfordshire | III | 100 |
| Middlesex | III | 100 |
| Leicestershire | IV | 91 |
| Northamptonshire | IV | 81 |
| Oxfordshire | IV | 74 |
| Staffordshire | IV | 90 |
| Warwickshire | IV | 91 |
| Cheshire | V | 77 |
| Gloucestershire | V | 97 |
| Herefordshire | V | 92 |
| Shropshire | V | 73 |
| Worcestershire | V | 97 |
| Derbyshire | VI | 95 |
| Huntingdonshire | VI | 100 |
| Lincolnshire | VI | 93 |
| Nottinghamshire | VI | 84 |
| Rutland | VI | 100 |
| Yorkshire | VI | 88 |
| Essex | VII | 90 |
| Norfolk | VII | 88 |
| Suffolk | VII | 92 |
| Mean average | 90 |
Appendix II
Money paid into the royal treasury in 1129–30.236
| £ | % of total237 | |
|---|---|---|
| Royal property | ||
| Borough farms | 248 | 1.1 |
| Forest revenues other than pleas | 121 | 0.5 |
| Shire farm | 8,080 | 35.1 |
| Subtotal | 8,449 | 36.7 |
| Taxation | ||
| Aids of boroughs, cities and counties | 390 | 1.7 |
| Cornage, geld of animals | 306 | 1.3 |
| Danegeld | 2,458 | 10.7 |
| Subtotal | 3,154 | 13.7 |
| Justice and government | ||
| Help in judicial matters | 649 | 2.8 |
| Miscellaneous agreements | 276 | 1.2 |
| Offices | 847 | 3.7 |
| Pleas: forest | 367 | 1.6 |
| Pleas: murder fines | 147 | 0.6 |
| Pleas: other | 2,140 | 9.3 |
| Subtotal | 4,426 | 19.2 |
| Royal lordship | ||
| Aid of knights | 69 | 0.3 |
| Estates in hand (escheats) | 2,882 | 12.5 |
| Marriages | 569 | 2.5 |
| Miscellaneous agreements | 980 | 4.3 |
| Regalian rights (vacancies) | 1,097 | 4.8 |
| Relief | 1,309 | 5.7 |
| Wardship | 89 | 0.4 |
| Subtotal | 6,995 | 30.4 |
| Total | 23,024 | 100.0 |
| £ | % of total237 | |
|---|---|---|
| Royal property | ||
| Borough farms | 248 | 1.1 |
| Forest revenues other than pleas | 121 | 0.5 |
| Shire farm | 8,080 | 35.1 |
| Subtotal | 8,449 | 36.7 |
| Taxation | ||
| Aids of boroughs, cities and counties | 390 | 1.7 |
| Cornage, geld of animals | 306 | 1.3 |
| Danegeld | 2,458 | 10.7 |
| Subtotal | 3,154 | 13.7 |
| Justice and government | ||
| Help in judicial matters | 649 | 2.8 |
| Miscellaneous agreements | 276 | 1.2 |
| Offices | 847 | 3.7 |
| Pleas: forest | 367 | 1.6 |
| Pleas: murder fines | 147 | 0.6 |
| Pleas: other | 2,140 | 9.3 |
| Subtotal | 4,426 | 19.2 |
| Royal lordship | ||
| Aid of knights | 69 | 0.3 |
| Estates in hand (escheats) | 2,882 | 12.5 |
| Marriages | 569 | 2.5 |
| Miscellaneous agreements | 980 | 4.3 |
| Regalian rights (vacancies) | 1,097 | 4.8 |
| Relief | 1,309 | 5.7 |
| Wardship | 89 | 0.4 |
| Subtotal | 6,995 | 30.4 |
| Total | 23,024 | 100.0 |
| £ | % of total237 | |
|---|---|---|
| Royal property | ||
| Borough farms | 248 | 1.1 |
| Forest revenues other than pleas | 121 | 0.5 |
| Shire farm | 8,080 | 35.1 |
| Subtotal | 8,449 | 36.7 |
| Taxation | ||
| Aids of boroughs, cities and counties | 390 | 1.7 |
| Cornage, geld of animals | 306 | 1.3 |
| Danegeld | 2,458 | 10.7 |
| Subtotal | 3,154 | 13.7 |
| Justice and government | ||
| Help in judicial matters | 649 | 2.8 |
| Miscellaneous agreements | 276 | 1.2 |
| Offices | 847 | 3.7 |
| Pleas: forest | 367 | 1.6 |
| Pleas: murder fines | 147 | 0.6 |
| Pleas: other | 2,140 | 9.3 |
| Subtotal | 4,426 | 19.2 |
| Royal lordship | ||
| Aid of knights | 69 | 0.3 |
| Estates in hand (escheats) | 2,882 | 12.5 |
| Marriages | 569 | 2.5 |
| Miscellaneous agreements | 980 | 4.3 |
| Regalian rights (vacancies) | 1,097 | 4.8 |
| Relief | 1,309 | 5.7 |
| Wardship | 89 | 0.4 |
| Subtotal | 6,995 | 30.4 |
| Total | 23,024 | 100.0 |
| £ | % of total237 | |
|---|---|---|
| Royal property | ||
| Borough farms | 248 | 1.1 |
| Forest revenues other than pleas | 121 | 0.5 |
| Shire farm | 8,080 | 35.1 |
| Subtotal | 8,449 | 36.7 |
| Taxation | ||
| Aids of boroughs, cities and counties | 390 | 1.7 |
| Cornage, geld of animals | 306 | 1.3 |
| Danegeld | 2,458 | 10.7 |
| Subtotal | 3,154 | 13.7 |
| Justice and government | ||
| Help in judicial matters | 649 | 2.8 |
| Miscellaneous agreements | 276 | 1.2 |
| Offices | 847 | 3.7 |
| Pleas: forest | 367 | 1.6 |
| Pleas: murder fines | 147 | 0.6 |
| Pleas: other | 2,140 | 9.3 |
| Subtotal | 4,426 | 19.2 |
| Royal lordship | ||
| Aid of knights | 69 | 0.3 |
| Estates in hand (escheats) | 2,882 | 12.5 |
| Marriages | 569 | 2.5 |
| Miscellaneous agreements | 980 | 4.3 |
| Regalian rights (vacancies) | 1,097 | 4.8 |
| Relief | 1,309 | 5.7 |
| Wardship | 89 | 0.4 |
| Subtotal | 6,995 | 30.4 |
| Total | 23,024 | 100.0 |
Footnotes
I wish to thank Julia Crick and Chris Lewis for commenting on a draft of this article and for their intellectual companionship throughout its gestation. I also wish to thank Peter Stokes, Frank Thorn and Fran Álvarez López, together with all those who contributed to the Exon research project (see n. 3 below), which laid the foundations on which this article builds.
The Anglo-Saxon Chronicle: A Revised Translation, ed. D. Whitelock, D.C. Douglas and S.I. Tucker (London, 1961), MS E [hereafter ASC E], s.a. 1085; The Anglo-Saxon Chronicle, A Collaborative Edition, VII: MS. E, A Semi-Diplomatic Edition with Indices and Notes, ed. S. Irvine (Cambridge, 2004), p. 94. Domesday Book is cited by folio number from Great Domesday, ed. A. Williams and R.W.H. Erskine (Alecto Historical Editions; Library Edition, 6 cases, London, 1986–92) and Little Domesday, ed. A. Williams (Alecto Historical Editions; Library Edition, 6 vols, London, 2000) [hereafter GDB and LDB respectively], and in parentheses by shire and entry number from Domesday Book, ed. J. Morris et al. (34 vols, Chichester, 1974–86).
For a survey of the literature from the mid-eighteenth century to date, see S. Baxter, ‘The Domesday Controversy: A Review and a New Interpretation’, Haskins Society Journal, xxix (2018), pp. 225−93.
The present article represents one of the fruits of ‘The Conqueror’s Commissioners: Unlocking the Domesday Survey of South-Western England’, a project funded (between 2014 and 2017) by the Arts and Humanities Research Council (grant number AH/L013975/1, principal investigator J. Crick, co-investigators S. Baxter and P.A. Stokes). Its principal output to date is a website on which facsimile images, text, translation, and a comprehensive codicological description of the manuscript are published: Exon: The Domesday Survey of South-West England, ed. P.A. Stokes, Studies in Exon Domesday, gen. ed. J. Crick (London, 2018), available at http://www.exondomesday.ac.uk. The materials include F. Álvarez López, ‘Codicological Description’, and F. Thorn, ed., ‘Exon Domesday Book: The Latin Text and Translation’ [hereafter cited as ‘Exon’ with reference number]. An analytical volume is in progress: S. Baxter, J. Crick, and C.P. Lewis, Making Domesday: The Conqueror’s Survey and its Context (forthcoming).
The scribal stints are identified in Álvarez López, ‘Codicological Description’, which develops earlier work on the identification of Exon scribes including R.W. Finn, ‘The Evolution of Successive Versions of Domesday Book’, English Historical Review, lxvi (1951), pp. 561–64, at 562–3; id., ‘The Exeter Domesday and its Construction’, Bulletin of the John Rylands Library, xli (1959), pp. 360–87, at 363–8; N.R. Ker, ‘The Beginnings of Salisbury Cathedral Library’, in J.J.G. Alexander and M.T. Gibson, eds., Medieval Learning and Literature: Essays Presented to Richard William Hunt (Oxford, 1976), pp. 23–49, at 35–6; id., Medieval Manuscripts in British Libraries, II: Abbotsford–Keele (Oxford, 1977), pp. 800–807; T. Webber, ‘Salisbury and the Exon Domesday: Some Observations Concerning the Origin of Exeter Cathedral MS 3500’, English Manuscript Studies, 1100–1700, i (1989), pp. 1–18; ead., Scribes and Scholars at Salisbury Cathedral, c.1075–c.1125 (Oxford, 1992), pp. 8–19; and C. Flight, The Survey of the Whole of England: Studies of the Documentation Resulting from the Survey Conducted in 1086 (British Archaeological Reports, British Series, 405; Oxford, 2006) [hereafter Flight, Survey], pp. 56–9. For further analysis of the Exon scribes and their milieux, see J.C. Crick, ‘The Form and Shape of the Exon Manuscript’, in Baxter et al., Making Domesday.
Important recent contributions include D. Roffe, Domesday: The Inquest and the Book (Oxford, 2000), Decoding Domesday (Woodbridge, 2007), and ‘Domesday Now: A View from the Stage’, in id. and K.S.B. Keats-Rohan, eds., Domesday Now: New Approaches to the Inquest and the Book (Woodbridge, 2016), pp. 7–60; Flight, Survey; and S.P.J. Harvey, Domesday: Book of Judgment (Oxford, 2014). Significant similarities and differences between these interpretations and my own are signalled in notes below, and are treated more fully in Baxter, ‘Domesday Controversy’, pp. 251–78. It should be apparent that the thesis developed in the present article differs fundamentally from that advanced by C. Symes, ‘Doing Things beside Domesday Book’, Speculum, xciii (2018), pp. 1048–101, except in so far as the latter integrates material from a report on the Exon research project delivered by Julia Crick, Chris Lewis and myself to the 35th Annual Meeting of the Haskins Society at Carleton College on 5 Nov. 2016 (acknowledged by Symes on p. 1051).
W.H. Stevenson, ‘A Contemporary Description of the Domesday Survey’, English Historical Review, xxii (1907), pp. 72–84, at 74; translation by Frank Thorn.
Flight, Survey, pp. 108–9.
Hemingi Chartularium Ecclesiæ Wigorniensis, ed. Thomas Hearne (2 vols, Oxford, 1723), pp. 287–8; also printed with translation in Domesday Book, XVI: Worcestershire, ed. and tr. F. Thorn and C. Thorn (Chichester, 1982), Appendix V, Worcester F. Hemming’s cartulary contains two further documents which name the same four men in this connection: Hemingi Chartularium, ed. Hearne, i. 75–7 (Domesday Book: Worcestershire, ed. Thorn and Thorn, Appendix V, Worcester H, nos. 4–5).
S. Baxter, ‘The Representation of Lordship and Land Tenure in Domesday Book’, in E. Hallam and D. Bates, eds., Domesday Book (Stroud, 2001), pp. 73–102, 203–8, at 81−2.
F. Barlow, ‘Domesday Book: A Letter of Lanfranc’, English Historical Review, lxxviii (1963), pp. 284–9.
S. Baxter, ‘The Men Behind the Survey’, in Baxter et al., Making Domesday. References to testimony given at the survey by barons’ men are collected by R. Fleming, Domesday Book and the Law (Cambridge, 1998) [hereafter DBL], p. 543.
The text is printed in Inquisitio Comitatus Cantabrigiensis … subjicitur Inquisitio Eliensis, ed. N.E.S.A. Hamilton (London, 1876), p. 97, and, with the translation adopted here (with minor amendments), by F. Thorn, ‘Non Pascua sed Pastura: The Changing Choice of Terms in Domesday’, in Roffe and Keats-Rohan, eds., Domesday Now, pp. 109–36, at 112. For the manuscripts of Inquisitio Eliensis, see Flight, Survey, pp. 81–93.
For a different view, see Roffe, Domesday: The Inquest and the Book, pp. 114–17, 171.
See below, at n. 95.
Baxter, ‘Representation of Lordship and Land Tenure’, pp. 74–80; J.J.N. Palmer, ‘A Digital Latin Domesday’, in Roffe and Keats-Rohan, eds., Domesday Now, pp. 61–80, at 63–71.
R. Sharpe, ‘The Use of Writs in the Eleventh Century’, Anglo-Saxon England, xxxii (2004), pp. 247–91.
Leges Henrici Primi, ed. L.J. Downer (Oxford, 1972), 7.4 (p. 100).
ASC E, s.a. 1086 (ed. Irvine, p. 94).
Capitula Regum Francorum, ed. A. Boretius and V. Krause, Monumenta Germaniae Historica, Legum Sectio II (2 vols, Hanover, 1883–97), no. 273 c. 19 (ii. 317–18).
Annales de Saint-Bertin, ed. F. Grat, J. Vielliard and S. Clémencet with an introduction by L. Levillain (Paris, 1964), pp. 152–3; The Annals of St-Bertin, tr. J.L. Nelson (Manchester, 1991), pp. 153–4.
That the hidage system assessed liability to dues and service (including military service) as well as geld is stressed in a document issued in 1086: Regesta Regum Anglo-Normannorum: The Acta of William I (1066–1087), ed. D. Bates (Oxford, 1998), no. 350, which says that the men of Hampton in Worcestershire ‘debent placitare, et geldum et expeditionem et cetera regis servitia de illis xv hidis secum persolvere’ (accepting the editor’s suggested emendation of regis for legis).
S.P.J. Harvey, ‘Domesday Book and its Predecessors’, English Historical Review, lxxvi (1971), pp. 753–73; ead., Domesday, pp. 67–71; H.B. Clarke, ‘The Domesday Satellites’, in P.H. Sawyer, ed., Domesday Book: A Reassessment (London, 1985), pp. 50–70, esp. 54, 59.
See most recently R.P. Cohen, ‘Danegeld: The Land Tax in England, 991–1162’ (Univ. of Oxford D.Phil. thesis, 2018).
Anglo-Saxon Charters, ed. and tr. A.J. Robertson (Cambridge, 1939), Appendix I, no. 3, pp. 230–36, 481–4.
J.H. Round, Feudal England: Historical Studies on the XIth and XIIth Centuries (London, 1895), pp. 196–214; The Leicestershire Survey, c.1130: A New Edition, ed. and tr. C.F. Slade (Leicester, 1956).
Round, Feudal England, pp. 215–24; K.S.B. Keats-Rohan, Domesday People: A Prosopography of Persons Occurring in English Documents, 1066–1166, I: Domesday Book (Woodbridge, 1999), pp. 98–117.
J.H. Round, The Commune of London and Other Studies (London, 1899), pp. 258−60; J.S. Cockburn, H.P.F. King and K.G.T. McDonnell, eds., A History of the County of Middlesex, I: Physique, Archaeology, Domesday … (Victoria County History; London, 1969), pp. 135−8.
The Lincolnshire Domesday and the Lindsey Survey, ed. and tr. C.W. Foster and T. Longley, with introduction by F.M. Stenton (Gainsborough, 1924), pp. 237–60; Keats-Rohan, Domesday People, pp. 77–97.
Harvey, ‘Domesday Book and its Predecessors’, pp. 761–3; S. Baxter, ‘Title to Property in the Age of the Great Confiscation’, in id. et al., Making Domesday.
R. Sharpe, ‘Tenere in Capite and Tenant in Chief’, currently accessible via The Charters of William II and Henry I: History from the Writs and Charters of Two Norman Kings (2017–), at https://actswilliam2henry1.wordpress.com/2017/05/23/tenere-in-capite/.
LDB 276v (Norfolk 66,61).
The following entries are suggestive in this connection: GDB 22 (Sussex 10,82), 164, 166v, 170 (Gloucestershire 1,63; 28,7; 75,2), 182v (Herefordshire 2,57); LDB 279v (Norfolk 66,99), and 291v (Suffolk 2,8). In addition, the letters ‘nfr’ are entered in the margins next to about a dozen entries, and this was perhaps shorthand for ‘non fecit retornum’, indicating failure to make a return; examples are listed in DBL, p. 548, and discussed by C. Thorn, ‘Marginal Notes and Signs in Domesday Book’, in A. Williams and R.W.H. Erskine, eds., Domesday Book Studies (London, 1987), pp. 113–35, at 130–31.
See below, at nn. 215–20.
Recent work on ‘seigneurial returns’ includes Roffe, Domesday: The Inquest and the Book, pp. 110–12, 140–46; Baxter, ‘Representation of Lordship and Land Tenure’, pp. 81–93; Flight, Survey, pp. 111–13; C. Flight, The Survey of Kent: Documents Relating to the Survey of the County Conducted in 1086 (Oxford, 2010), pp. 33–70; Harvey, Domesday, pp. 58–63; P. Taylor, ‘The Episcopal Returns in Domesday’, in Roffe and Keats-Rohan, eds., Domesday Now, pp. 197–217. Some landholders may have been able to draw on existing manorial records, such as the estate memoranda pertaining to the endowment of Dorchester-on-Thames compiled under the auspices of Bishop Remigius: J. Blair, ‘Estate Memoranda of c.1070 from the See of Dorchester-on-Thames’, English Historical Review, cxvi (2001), pp. 114–23.
For tally sticks, see L. Kuchenbuch, ‘Les Baguettes de taille au Moyen Âge: Un moyen de calcul sans écriture’, in N. Coquery, F. Menant and F. Weber, eds., Écrire, compter, mesure: Vers une histoire des rationalités pratiques (Paris, 2006), pp. 113–42.
Inquisitio Comitatus Cantabrigiensis, ed. Hamilton, pp. 1–96 (text); Flight, Survey, pp. 87−93 (discussion).
Round, Feudal England, pp. 1–146.
See section IV below.
P.H. Sawyer, ‘The “Original Returns” and Domesday Book’, English Historical Review, lxx (1955), pp. 177–97, at 181.
F.R. Thorn, ‘Hundreds and Wapentakes’, in the Introduction and Translation volume for each county other than Lincolnshire in Great Domesday, ed. Williams and Erskine; see also his contributions to the introductions for each county in Electronic Edition of Domesday Book: Translation, Databases and Scholarly Commentary, 1086, ed. J.J.N. Palmer (2nd edn, 2010), currently accessible through the UK Data Service at https://discover.ukdataservice.ac.uk/catalogue?sn=5694.
Flight, Survey, pp. 94–107; Flight, Survey of Kent, pp. 72–88.
V.H. Galbraith, The Making of Domesday Book (Oxford, 1961), pp. 102–22, 156–65, quotations at 115, 117.
For further discussion of the problems associated with Galbraith’s interpretation, see below, section VI.
S. Baxter, ‘Exon Fiefs I: Hundredal Organisation, Scribal Collaboration and the Nature of Domesday’s First Draft’, in id. et al., Making Domesday.
See section IV below.
Exon 63a1–63a33 (Devon list A), 63a34–63a64 ((Devon list B), 63b1–63b7 (Cornwall list A), 63b9–63b15 (Cornwall list B), 63b16–64a3 (Somerset list A), 64a4–64b40 (Somerset list B).
M. Gullick, ‘How Fast Did Scribes Write? Evidence from Romanesque Manuscripts’, in L.L. Brownrigg, ed., Making the Medieval Book: Techniques of Production (London, 1995), pp. 39–58.
Walmesley’s itinerary began at Andover in Hampshire on 20 February and included stops at New Sarum (Wiltshire), Dorchester (Dorset), Chard (Somerset) and Exeter (Devon), before ending at Launceston (Cornwall) on 13 March: W. Durrant Cooper, ‘The Expenses of the Judges of Assize Riding the Western and Oxford Circuits, Temp. Elizabeth, 1596–1601’, in Camden Miscellany IV, Camden Society, lxxiii (1859), pp. 27–30. I owe this reference to Chris Lewis.
ASC E, s.a. 1085 (ed. Irvine, p. 94).
D. Whitelock, ‘On the Commencement of the Year in the Saxon Chronicles’, in Two of the Saxon Chronicles Parallel, ed. J. Earle and C. Plummer, reissued with a bibliographical note by D. Whitelock (2 vols, Oxford, 1952), vol. ii, p. cxliic.
F.W. Maitland, Domesday Book and Beyond: Three Essays in the Early History of England (1897; new edn, Cambridge, 1987), p. 11 n. 1.
Inquisitio Eliensis, ed. Hamilton, p. 97; Thorn, ‘Non Pascua sed Pastura’, p. 112.
C.P. Lewis, ‘The Domesday Jurors’, Haskins Society Journal, v (1993), pp. 17–44.
S. Baxter, ‘The Making of Domesday Book and the Languages of Lordship in Conquered England’, in E.M. Tyler, ed., Conceptualizing Multilingualism in Medieval England, c.800–c.1250 (Turnhout, 2011), pp. 271–308, at 285.
H.C. Darby, Domesday England (Cambridge, 1977), p. 336.
Feudal Documents from the Abbey of Bury St Edmunds, ed. D.C. Douglas (London, 1932), p. 3.
DBL, p. 529.
GDB 47v (Hampshire 35,2), 50 (Hampshire 69,30), 169 (Gloucestershire 59,1), 170 (Gloucestershire 72,3).
GDB 36 (Surrey 25,2).
Galbraith, Making of Domesday Book, p. 80; DBL, pp. 2–3.
Exon 178b2. The manor was restored to the abbey in 1096; see H.P.R. Finberg, ‘The Early History of Werrington’, English Historical Review, lix (1944), pp. 237–51, at 245–8.
GDB 181 (Herefordshire 1,75) is the clearest case. See also GDB 57v (Berkshire 1,39), 39 (Hampshire 1,42), 373 (Yorkshire CN,5) with 300v (Yorkshire 1N,87); and LDB 6–7 (Essex 1,27–8).
GDB 58v (Berkshire 5,1). See also GDB 44v (Hampshire 23,3), 50 (Hampshire 69,16), 191 (Cambridgeshire 5,21), 377v (Lincolnshire CK,50); LDB 377 (Suffolk 16,34), 448v–449 (Suffolk 76,16), 450a (Suffolk 77,4).
Regesta Regum Anglo-Normannorum, ed. Bates, no. 146.
P. Wormald, ‘Domesday Lawsuits: a Provisional List and Preliminary Comment’, in C. Hicks, ed., England in the Eleventh Century (Stamford, 1992), pp. 61–102, at 64–9, 95.
The categories are defined and explored by Baxter, ‘Title to Property’. Wormald, ‘Domesday Lawsuits’, contains a selective list of ‘king’s pleas’ (pp. 98–101) and of disputes between other landholders (pp. 77–94); DBL prints all of the non-standard matter in Domesday Book in translation with invaluable indices.
DBL, p. 534, collects references. For discussion, see Edward A. Freeman, The History of the Norman Conquest of England: Its Causes and Its Results (3rd edn, 6 vols, Oxford, 1870–79), v. 787–98, Note I; G. Garnett, Conquered England: Kingship, Succession, and Tenure, 1066–1166 (Oxford, 2007), pp. 66–72.
DBL, pp. 515–17, 534.
DBL, nos. 158, 575, 626, 3127.
DBL, no. 1988.
DBL, nos. 1502, 3127.
DBL, nos. 401, 417, 790, 924, 983, 1857, 1858, 1860, 1876, 1878, 1882, 1885, 1894–7, 1900, 1930, 1954, 2009, 2025, 2026, 2027, 2032, 2033, 2054, 2229, 2269, 2369, 2508, 2660, 2679, 2693, 2748, 2781, 3199, 3203.
See above, at n. 18.
R. Welldon Finn, Domesday Studies: The Liber Exoniensis (London, 1964), pp. 97–123; C.P. Lewis, ‘The Exon Geld Accounts and the Great Geld of 1085–6’, in Baxter et al., Making Domesday.
Exon, 65a1.
Exon, 532a13.
For the novelty of this system, see D. Pratt, ‘Demesne Exemption from Royal Taxation in Anglo-Saxon and Anglo-Norman England’, English Historical Review, cxxviii (2013), pp. 1–34.
Exon 70b1.
Exon 526b5.
Exon 1a11.
Exon 1b4, 1b6, 2a1, 2b8 (Wiltshire A); also Exon 8a6 (Wiltshire B).
Exon 175a6; below, at nn. 222–9.
See n. 4 above.
Flight, Survey, pp. 45–8.
See n. 42 above; Roffe, Domesday: The Inquest and the Book, p. 173.
Exon 286r is written by Alpha (heading and lines 1–13) and Beta (lines 14–20), and 286v is written by Lambda (lines 1–7), Iota (lines 7–12), Alpha (lines 13–17) and Theta (lines 18–20).
Finn, Liber Exoniensis, pp. 124–30; Flight, Survey, pp. 103–6, 148–50; C.P. Lewis, ‘Fief Summaries’, in Baxter et al., Making Domesday.
Exon 530b3.
Mu wrote summaries for Glastonbury Abbey in Wiltshire, Dorset, Devon and Somerset, treating each shire separately (Exon 527b1–528a4). Pi wrote six lay fief summaries (530b2–6, 531a1–3), plus a single entry in Dorset fiefs. Omega wrote the summary for St Petroc in Cornwall (528b1–3) and ‘GFS’ wrote a summary for Glastonbury Abbey in Somerset (173a2).
Inquisitio Eliensis, ed. Hamilton, pp. 121–3.
GDB 270 (Cheshire R7,1), 381 (Yorkshire SN, CtA,45).
Exon 155a1, 209b1, 370b1, 387b1, 449b1, 451b1, 467b1, 474b1, 490a4, 494b1.
Finn, Liber Exoniensis, pp. 55–96; C.P. Lewis, ‘Terrae Occupatae: Recording Unauthorized Tenures’, in Baxter et al., Making Domesday.
R.S. Hoyt, ‘The Terrae Occupatae of Cornwall and the Exon Domesday’, Traditio, ix (1953), pp. 155–75, at 168.
See above, at n. 14.
For details, see Baxter, ‘Title to Property’.
S. Baxter and C.P. Lewis, ‘Inside the Exon Office’, in Baxter et al., Making Domesday.
P.H. Sawyer, ‘The Place-Names of the Domesday Manuscripts’, Bulletin of the John Rylands Library, xxxviii (1956), pp. 483–505, at 492–5.
Crick, ‘Form and Shape of the Exon Manuscript’.
Exon 512v (lower margin).
Exon 515v (lower margin).
For example, Exon 521v. Paraphs are symbols indicating the beginning of an entry.
Exon 531r (lower and right-hand margins). Neumes are a system of musical notation.
Exon 521v: ‘om[n]is omo primum bonum’, a quotation from John 2:10; see C.P. Lewis, ‘In the Margins of Exon: Eta’s Note’, available via Exon: The Domesday Survey of South-West England, at http://www.exondomesday.ac.uk/blog/in-the-margins-of-exon-etas-note/ (published 1 Nov. 2015).
See above, at n. 10.
Ker, ‘Beginnings of Salisbury Cathedral Library’, pp. 34–8; Webber, ‘Salisbury and the Exon Domesday’; ead., Scribes and Scholars at Salisbury Cathedral, pp. 8–19.
Regesta Regum Anglo-Normannorum, ed. Bates, pp. 98–9.
These questions are explored by Crick, ‘Form and Shape of the Exon Manuscript’.
Alpha, Beta, Ksi, Mu, and Tau (Sigma and Rho wrote geld accounts for Wiltshire).
Exon 70b1, 78b1, 526b5.
ASC E, s.a. 1086 (ed. Irvine, p. 94) (translation adapted with reference to The Anglo-Saxon Chronicles, tr. and ed. M. Swanton [London, 2003], p. 217).
J.C. Holt, ‘1086’, in id., ed., Domesday Studies: Papers read at the Novocentenary Conference of the Royal Historical Society and the Institute for British Geographers, Winchester, 1986 (Woodbridge 1987), pp. 41–64. For different interpretations, see J. Prestwich, ‘Mistranslations and Misinterpretations in Medieval English History’, Peritia, x (1996), pp. 322–40, at 333–6; P. Wormald, ‘Engla Lond: The Making of an Allegiance’, in his Legal Culture in the Early Medieval West: Law as Text, Image and Experience (London, 1999), pp. 359–82, at 367; J. Maddicott, ‘The Oath of Marlborough, 1209: Fear, Government and Popular Allegiance in the Reign of King John’, English Historical Review, cxxvi (2011), pp. 281–318, at 298–9, 304–6; D. Bates, William the Conqueror (London, 2016), pp. 475–7.
Exon 175a6.
Exon 173v–175v.
For the writing of Domesday Book, see H. Jenkinson, Domesday Rebound (London, 1954); A.R. Rumble, ‘The Palaeography of the Domesday Manuscripts’, in Sawyer, ed., Domesday Book: A Reassessment, pp. 28–49; A.R. Rumble, ‘The Domesday Manuscripts: Scribes and Scriptoria’, in Holt, ed., Domesday Studies, pp. 79–99; M. Gullick, ‘The Great and Little Domesday Manuscripts’, in Williams and Erskine, eds., Domesday Book Studies, pp. 93–112; C. Thorn and F. Thorn, ‘The Writing of Great Domesday Book’, in Hallam and Bates, eds., Domesday Book, pp. 37−72; and C.P. Lewis, ‘Writing Great Domesday Book’, in Baxter et al., Making Domesday.
Rumble, ‘Domesday Manuscripts’, pp. 80, 98–99.
Galbraith, Making of Domesday Book, pp. 31–2, 102–22; Flight, Survey, pp. 2–4, 71–80, 126–37.
Exon 209b1.
The entries attributed to Mu are Exon 155a1, 370b1, 387b1, 449b1, 467b1, 474b1, 490a4, 494b1; the exception is 451b1, which Álvarez López, ‘Codicological Description: 451v’, considers to be ‘the work of an unidentified hand’.
Álvarez López, ‘Codicological Description: 316r’.
Exon 153v and 436v.
Thorn and Thorn, ‘Writing of Great Domesday Book’, pp. 48, 56–62, 67–8.
Exon folios 210–264v; Lewis, ‘Writing Great Domesday Book’. Unfortunately, the style of the marginal ‘f’ is so generic that it is impossible to attribute it to any particular scribe.
LDB 450.
Rumble, ‘Domesday Manuscripts’, pp. 98–9.
Thorn and Thorn, ‘Writing of Great Domesday Book’, p. 38. I am very grateful to Frank Thorn for allowing me to see the late Caroline Thorn’s hitherto unpublished list of this scribe’s stints.
P. Chaplais, ‘William of Saint-Calais and the Domesday Survey’, in Holt, ed., Domesday Studies, pp. 65–77, at 73–5.
Crick, ‘Form and Shape of the Exon Manuscript’.
Lewis, ‘Writing Great Domesday Book’.
See below, at n. 200.
D.C. Douglas, ‘The Domesday Survey’, History, xxi (1936), pp. 249–257, at 254–5.
Roffe, Domesday: The Inquest and the Book, pp. ix, 243–4.
Baxter, ‘Domesday Controversy’, p. 287 n. 154.
Harvey, Domesday, p. 100, citing GDB 77 (Dorset 3,6).
Thorn and Thorn, ‘Writing of Great Domesday Book’, pp. 48–55, 71–2.
Jenkinson, Domesday Rebound, p. 34; Gullick, ‘Great and Little Domesday Manuscripts’, p. 105.
The Ecclesiastical History of Orderic Vitalis, ed. M. Chibnall (6 vols, Oxford, 1969–80), iv. 174; H.B. Clarke, ‘Condensing and Abbreviating the Data: Evesham C, Evesham M, and the Breviate’, in Roffe and Keats-Rohan, eds., Domesday Now, pp. 247–75, at 259–62.
ASC E, s.a. 1087 (ed. Irvine, p. 98).
Rumble, ‘Domesday Manuscripts’, pp. 80–81, 99.
For references and fuller treatment, see Baxter, ‘Domesday Controversy’.
The Pipe Roll of 31 Henry I, ed. and tr. J.A. Green, Pipe Roll Society, xcv (2012).
J.A. Green, The Government of England under Henry I (Cambridge, 1986), pp. 51–94, 220–25; S. Baxter, ‘The Significance of Henry I’s Pipe Roll’, in Baxter et al., Making Domesday.
This exercise would also affect landholders’ liabilities for other payments or services levied on the hide, including fyrd service—for which, see C.W. Hollister, Anglo-Saxon Military Institutions on the Eve of the Norman Conquest (Oxford, 1962), pp. 38–102, and id., The Military Organisation of Norman England (Oxford, 1965), pp. 216–60.
For the Domesday ualet, see below, at n. 164.
J.H. Round, ‘Danegeld and the Finance of Domesday’, in P. Edward Dove, ed., Domesday Studies: Being the Papers Read at the Meetings of the Domesday Commemoration 1886 (2 vols, London, 1888–91), i. 77–142 (quotation at p. 116).
Round, Feudal England, p. 3.
Maitland, Domesday Book and Beyond, pp. 3–6, 107–28, 399–489 (quotations at pp. 4, 5, 6, 120, 448).
Galbraith, Making of Domesday Book, pp. 12–44 (quotation at pp. 14–15).
A.H.M. Jones, ‘Census Records of the Later Roman Empire’, Journal of Roman Studies, xliii (1953), pp. 49–64; id., ‘Capitatio and Jugatio’, Journal of Roman Studies, xlvii (1957), pp. 88–94; J. Percival, ‘The Precursors of Domesday: Roman and Carolingian Land Registers’, in Sawyer, ed., Domesday Book: A Reassessment, pp. 5–27, at 5–13; P. Thonemann, ‘Estates and the Land in Late Roman Asia Minor’, Chiron, xxxvii (2007), pp. 435–78.
N. Oikonomidès, Fiscalité et exemption fiscale à Byzance (IXe–XIes.) (Athens, 1996); C.M. Brand, ‘Two Byzantine Treatises on Taxation’, Traditio, xxv (1969), pp. 35−60.
P.J. Golas, ‘The Sung Fiscal Administration’, in J.W. Chaffee and D. Twitchett, eds., The Cambridge History of China, V, pt. II: Sung China, 960−1275 (Cambridge, 2015), pp. 139−213, at 152, 166.
G. Frantz-Murphy, The Agrarian Administration of Egypt from the Arabs to the Ottomans (Cairo, 1986); ead., ‘The Corpus and Context: Agrarian Fiscal Administration and State Formation in Early Islamic Egypt, 694−1035 A.D. (75−427 A.H.)’, in B. Palme, ed., Akten des 23. Internationalen Papyrologenkongresses, Wien, 22−28 July 2001 (Vienna, 2007), pp. 221−31.
J.A. Green, ‘The Last Century of Danegeld’, English Historical Review, xcvi (1981), pp. 241–58, at 245.
F. Barlow, William Rufus (London, 1983), pp. 243–5.
Prestwich, ‘Mistranslations and Misinterpretations’, pp. 330–31.
See the last paragraph of section III above.
See above, at nn. 25–8.
Three Rolls of the King’s Court in the Reign of King Richard the First, A.D. 1194–1195, ed. F.W. Maitland, Pipe Roll Society, xiv (1891), pp. xxii–xxv; Roger of Howden, Chronica Magistri Rogeri de Houedene, ed. William Stubbs, Rolls Series, li (4 vols, 1868–71), iv. 46–7.
S.P.J. Harvey, ‘Taxation and the Ploughland in Domesday Book’, in Sawyer, ed., Domesday Book: A Reassessment, pp. 86–103; Harvey, Domesday, pp. 210–38.
See nn. 173–4 below.
A. Dymond, ‘The Estates of William the Conqueror: Royal Property in England and Ducal Property in Normandy in the Eleventh Century’ (Univ. of Oxford D.Phil. thesis, in progress).
S. Baxter and C.P. Lewis, ‘Domesday Book and the Transformation of English Landed Society’, Anglo-Saxon England, xlvi (2019), pp. 343−403, at 368.
Green, Government of England, p. 223.
Baxter, ‘Domesday Controversy’, pp. 273–4.
R. Lennard, Rural England, 1086–1135: A Study of Social and Agrarian Conditions (Oxford, 1959), pp. 142–59.
In other words, the Conqueror was seeking to foster a culture which is treated for a later period, taking account of deep hinterlands, in a fine study by J. Sabapathy, Officers and Accountability in Medieval England, 1170–1300 (Oxford, 2014), especially chs. 2, 3 and 6, which resonate with many of the suggestions made here.
ASC E, s.a. 1085 (ed. Irvine, p. 94).
For an important discussion of this material, see J. Munby, ‘The Domesday Boroughs Revisited’, Anglo-Norman Studies, xxxiii (2010), pp. 127–49. See also Roffe, Domesday: The Inquest and the Book, pp. 128–40, 230–34, who suggests that a survey of ‘the royal fisc’, which covered boroughs as well as royal manors, was connected with the geld inquest and constituted a distinct early phase of the survey; and Harvey, Domesday, pp. 239–50, who suggests that material relating to boroughs was captured late in the survey and related to an ‘inquest of sheriffs’. My own view is that material relating to the royal demesne and boroughs was first recorded in stage two of the survey and subjected to public scrutiny in stage three, in common with the rest of the survey, though, as Munby shows, the procedure for this varied considerably between circuits, presumably because the terms of reference for boroughs were less clearly defined than those for manors; see S. Baxter, ‘Exon Fiefs III: The Sources of Information’, and ‘Conclusions’ in id. et al., Making Domesday.
Harvey, Domesday, p. 4.
For example, GDB 154v (Oxfordshire 1,12), 230 (Leicestershire C,3), 238 (Warwickshire B,4), 293v (Rutland R,4).
Exon 97b3–5: ‘.ccc. libras et .lxxv. de terris comitum reddit Balduinus regi ad firmam per annum’; ‘Gotselenus reddit per annum regi pro terris Edide reginae .c. libras et .viii.’; ‘Raginaldus reddit per annum ad firmam regis pro terra Ordulfi .xxiiii. libras.’
Lennard, Rural England, pp. 113–28, 146–50.
ASC E, s.a. 1087 (ed. Irvine, p. 95).
Widows, Heirs and Heiresses in the Late Twelfth Century: The Rotuli de Dominabus et Pueris et Puellis, ed. and tr. J. Walmsley (Tempe, AZ, 2006), pp. 2, 8, 12 et passim.
This point is developed by J. Campbell, ‘Observations on English Government from the Tenth to the Twelfth Century’, Transactions of the Royal Historical Society, 5th ser., xxv (1975), pp. 35–54, at 48–51; Percival, ‘Precursors of Domesday’, pp. 13–27; R.H.C. Davis, ‘Domesday Book: Continental Parallels’, in Holt, ed., Domesday Studies, pp. 15–39; J.L. Nelson, ‘Henry Loyn and the Context of Anglo-Saxon England’, Haskins Society Journal, xix (2007), pp. 154–70, at 164–9; and S. Baxter, ‘Estate Management and Written Records in England and the Continent Before Domesday’, in id. et al., Making Domesday. For the application of literacy to the administration of royal property in Carolingian Francia, see F.L. Ganshof, Carolingians and the Frankish Monarchy: Studies in Carolingian History, tr. J. Sondheimer (London, 1971), pp. 130–31; R. McKitterick, The Carolingians and the Written Word (Cambridge, 1989), pp. 160–63; J.L. Nelson, ‘Literacy in Carolingian Government’, in R. McKitterick, ed., The Uses of Literacy in Early Medieval Europe (Cambridge, 1990), pp. 258–96, at 272–8.
For example, Capitula Regum Francorum, ed. Boretius and Krause, vol. i, no. 18 c. 5 (p. 43), no. 23 c. 35 (p. 64), no. 24 c. 6 (p. 65), no. 34 cc. 10–11 (p. 100), no. 46 cc. 6–7 (p. 131), no. 49 c. 4 (p. 136), no. 77 cc. 4, 19 (pp. 171–2), no. 80 cc. 5–7 (p. 177), no. 95 c. 14 (p. 201); vol. ii, no. 191 c. 2 (p. 12), no. 259 c. 6 (p. 268), no. 273 cc. 5, 19, 27 (pp. 313, 318–19, 321–2), no. 292 c. 20 (p. 403).
Capitula Regum Francorum, ed. Boretius and Krause, vol. i, nos. 32 and 128; Capitulare de villis: cod. guelf. 254 Helmst. der Herzog August Bibliothek Wolfenbüttel, ed. C. Brühl (2 vols, Stuttgart, 1971). For discussion, see most recently D. Campbell, ‘The Capitulare de Villis, the Brevium exempla, and the Carolingian Court at Aachen’, Early Medieval Europe, xviii (2010), pp. 243–64.
E. Lesne, Histoire de la propriété ecclésiastique en France (6 vols, Paris, 1910–43), iii. 1–17; R. Fossier, Polyptyques et censiers (Turnhout, 1978), pp. 25–33.
J.-P. Devroey, ‘Élaboration et usage des polyptyques: Quelques éléments de réflexion à partir de l’exemple des descriptions de l’Église de Marseille (VIIIe–IXe siècles)’, in D. Hägermann, W. Haubrichs and J. Jarnut, eds., Akkulturation: Probleme einer germanisch-romanischen Kultursynthese in Spätantike und frühem Mittelalter (Berlin, 2004), pp. 436–72; J.P. Devroey, Puissants et misérables: Système social et monde paysan dans l’Europe des francs (VIe–IXesiècles) (Brussels, 2006), pp. 426–41. For the suggestion that a group of Merovingian estate records from Saint-Martin Tours represent records written literally in the field as the abbey’s official collected renders and dues and checked them against amounts expected recorded in estate surveys (now lost, but inferred), see S. Sato, ‘The Merovingian Accounting Documents of Tours: Form and Function’, Early Medieval Europe, ix (2000), pp. 143–61.
Campbell, ‘Observations on English Government’, p. 50.
J.-P. Devroey, ‘Au-delà des polyptyques: Sédimentation, copie et renouvellement des documents de gestion seigneuriaux entre Seine et Rhin (ixe–xiiie siècle)’, in X. Hermand, J.-F. Nieus and É. Renard, eds., Décrire, inventorier, enregistrer entre Seine et Rhin au Moyen Âge (Paris, 2009), pp. 53–86.
Baxter, ‘Estate Management and Written Records’.
This was the essence of Galbraith’s case; see V.H. Galbraith, Studies in the Public Records (London, 1948), pp. 104–5; id., Making of Domesday Book, pp. 1–44, esp. 15–17, 28–30; id., Domesday Book: Its Place in Administrative History (Oxford, 1974), pp. 13–15, 163. That case is developed by Holt, ‘1086’, pp. 50–55; G. Garnett, The Norman Conquest: A Very Short Introduction (Oxford, 2009), pp. 63–4, 70–71, 75, 80; and Harvey, Domesday, pp. 5–6, 119–20, 208–9, 232–8, 269.
See, for example, The Treatise on the Laws and Customs of the Realm of England Commonly Called Glanvill, ed. G.D.G. Hall (Oxford, 1965), pp. 82–4, 106–10; Richard fitzNigel, Dialogus de Scaccario: The Dialogue of the Exchequer, ed. and tr. E. Amt (Oxford, 2007), pp. 140–44, 180, 186; Rotuli de Dominabus et Pueris et Puellis, ed. and tr. Walmsley, passim. Recent surveys include D. Carpenter, ‘The Second Century of English Feudalism’, Past and Present, no. 168 (2000), pp. 30–71; J.G.H. Hudson, The Oxford History of the Laws of England, II: 871–1216 (Oxford, 2012), pp. 333–75, 435–54, 627–77, 776–811; S. Baxter, ‘The Profits of Royal Lordship’, in id. et al., Making Domesday.
Garnett, Conquered England, pp. 45–135.
See Appendix II.
Henry I’s coronation charter (Hn cor), ed. and tr. R. Sharpe, caps. 1–4, 6, currently available via Early English Laws, ed. B. O’Brien and J. Winters (Univ. of London), at http://www.earlyenglishlaws.ac.uk/laws/texts/hn-cor/view/#edition/translation; Garnett, Conquered England, pp. 105–20.
M. Howell, Regalian Right in Medieval England (London, 1962), pp. 5–48.
Eadmer of Canterbury, Historia Novorum in Anglia, ed. Martin Rule, Rolls Series, lxxxi (1884), pp. 2, 9–10, 26.
ASC E, s.a. 1100 (ed. Irvine, p. 110).
ASC E, s.a. 1087 (ed. Irvine, p. 98).
Garnett, Conquered England, pp. 1–44.
Freeman, History of the Norman Conquest, v. 8.
S. Reynolds, ‘Tenure and Property in Medieval England’, Historical Research, lxxxviii (2015), pp. 563–76, at 565.
GDB 252.
GDB 262v.
See above, after n. 131.
Baxter, ‘Profits of Royal Lordship’.
FitzNigel, Dialogus de Scaccario, ed. Amt, p. 186.
Holt, ‘1086’, pp. 50–55.
FitzNigel, Dialogus de Scaccario, ed. Amt, p. 96.
See above, at nn. 87–91.
Maitland, Domesday Book and Beyond, p. 5; D.C. Douglas, ‘Odo, Lanfranc, and the Domesday Survey’, in J.G. Edwards, V.H. Galbraith and E.F. Jacob, eds., Historical Essays in Honour of James Tait (Manchester, 1933), pp. 47–57; Galbraith, Making of Domesday Book, pp. 77–8; R. Welldon Finn, The Domesday Inquest and the Making of Domesday Book (London, 1961), pp. 92–111; Holt, ‘1086’, pp. 54–64; P. Hyams, ‘“No Register of Title”: The Domesday Inquest and Land Adjudication’, Anglo-Norman Studies, ix (1987), pp. 127–41; Wormald, ‘Domesday Lawsuits’; DBL, pp. 1–85; Roffe, Domesday: The Inquest and the Book, pp. 165–8; Harvey, Domesday, pp. 239–70.
F.M. Stenton, Anglo-Saxon England (3rd edn, Oxford 1971), p. 626.
R. Fleming, Kings and Lords in Conquest England (Cambridge, 1991), pp. 183–214 (quotations at pp. 184–5, 193, 194).
DBL, pp. 1–86, esp. pp. 17, 35.
See above, at nn. 67–71.
M.K. Lawson, ‘The Collection of the Danegeld and Heregeld in the Reigns of Æthelred II and Cnut’, English Historical Review, xciv (1984), pp. 721–38, at 723–7.
J. Johns, Arabic Administration in Norman Sicily: The Royal Dīwān (Cambridge, 2002), pp. 1–62.
See above, at nn. 141–2, and Appendix II.
Gesta Stephani, ed. and tr. K.R. Potter with an introduction and notes by R.H.C. Davis (Oxford, 1976), pp. 26−7.
ASC E, s.a. 1086 (ed. Irvine, p. 94).
D. Bates, ‘Two Ramsey Abbey Writs and the Domesday Survey’, Bulletin of the Institute of Historical Research, lxiii (1990), pp. 337–9; Wormald, ‘Domesday Lawsuits’, pp. 72−4; below, n. 227.
J.C. Holt, ‘Politics and Property in Early Medieval England’, in his Colonial England, 1066–1215 (London, 1997), pp. 113–59.
D. Bates, ‘Les chartres de confirmation et les pancartes normandes du règne de Guillaume le Conquérant’, in M. Parisse, P. Pégeot and B.-M. Tock, eds., Pancartes monastiques des XIeet XIIesiècles (Turnhout, 1998), pp. 95–109; Regesta Regum Anglo-Normannorum, ed. Bates, pp. 22–30. There is also good evidence that general confirmation charters were drawn up in England late in the Confessor’s reign: S. Keynes, ‘Giso, Bishop of Wells (1061–88)’, Anglo-Norman Studies, xix (1997), pp. 203–71, at 232–9.
Holt, ‘1086’, p. 56; see also Bates, ‘Les chartres de confirmation et les pancartes’, p. 105, and id., William the Conqueror, pp. 471–2.
Harvey, Domesday, pp. 276–91; Bates, William the Conqueror, pp. 476–7.
M.T. Clanchy, From Memory to Written Record: England, 1066–1307 (3rd edn, Oxford, 2013), pp. 38–41.
See above, at nn. 111–12.
Luke 2:1: ‘factum est autem in diebus illis exiit edictum a Caesare Augusto ut describeretur universus orbis’.
See, most recently, Harvey, Domesday, pp. 107–33.
Chaplais, ‘William of Saint-Calais’, pp. 65–78.
Crick, ‘Form and Shape of the Exon Manuscript’.
Exon 1a11, 7b8.
Exon 175a6.
Regesta Regum Anglo-Normannorum, ed. Bates, no. 326; GDB 32 (Surrey 6,5).
Regesta Regum Anglo-Normannorum, ed. Bates, nos. 137, 146, 333, 352; Regesta Regum Anglo-Normannorum, 1066–1154, I: Regesta Willelmi Conquestoris et Willelmi Rufi, 1066–1100, ed. H.W.C. Davis with R.J. Whitwell (Oxford, 1913), nos. 373, 383 (with Bates, ‘Two Ramsey Abbey Writs’, pp. 337–9); Regesta Regum Anglo-Normannorum, 1066–1154, II: Regesta Henrici Primi, 1100–1135, ed. C. Johnson and H.A. Cronne (Oxford, 1956), p. 404 (no. 386a). See also Chaplais, ‘William of Saint-Calais’, p. 77 n. 32.
W.M. Aird, ‘An Absent Friend: The Career of Bishop William of St Calais’, in D. Rollason, M. Harvey and M. Prestwich, eds., Anglo-Norman Durham, 1093–1193 (Woodbridge, 1994), pp. 283–97.
R.W. Southern, Medieval Humanism and Other Studies (Oxford, 1970), pp. 183–205; J.O. Prestwich, ‘The Career of Ranulf Flambard’, in Rollason, Harvey and Prestwich, eds., Anglo-Norman Durham, pp. 299–310. For the suggestion that Ranulph was the ‘mastermind’ behind the survey, see Harvey, Domesday, pp. 115–24.
T. Webber, ‘Osmund [St Osmund] (d. 1099), Bishop of Salisbury’, ODNB.
F. Neininger, ‘Maurice (d. 1107), Bishop of London’, ODNB; L. Lane, ‘Clerks and Commissioners: The Role of Bishops in the Government of England, c.1050–1087’ (King’s College London Ph.D. thesis, 2017), pp. 275–85.
For these men’s careers and the evidence relating to their involvement in the survey, see Baxter, ‘Men Behind the Survey’.
ASC E, s.a. 1085; J. Maddicott, ‘Responses to the Threat of Invasion, 1085’, English Historical Review, cxxii (2007), pp. 986–97; Bates, William the Conqueror, pp. 452–9.
For different views, see Maddicott, ‘Responses to the Threat of Invasion’; and N.J. Higham, ‘The Domesday Survey: Context and Purpose’, History, lxxviii (1993), pp. 7–19.
As defined above, at nn. 38–41. The statistics have been compiled from the database of Domesday landholdings which lies behind PASE Domesday, ed. S. Baxter, second revised edition (2016), published online at PASE (King’s College London, 2010–), available at http://www.domesday.pase.ac.uk.
The table represents an extract from a new analysis based on Pipe Roll of 31 Henry I, ed. Green; see Baxter, ‘The Significance of the Pipe Roll of King Henry I’, in id. et al., Making Domesday.
The totals and subtotals do not always cast precisely due to rounding.
