18

Domesday

The Anglo-Saxon Chronicle begins its account of the year 1082 with a startling thunderbolt. ‘In this year,’ it says, ‘the king arrested Bishop Odo’. Having startled us, however, the anonymous author, our only contemporary informant, leaves us entirely in the dark, adding only ‘and in this year there was great famine’ before moving on briskly to deal with the events of 1083.

Thus, in order to discover why William arrested Odo, we are forced to rely on the reports of chroniclers writing forty or fifty years later. Some of these writers declared that the king had discovered his half-brother was planning to usurp the throne. One pointed to Odo’s oppressive record as England’s regent, while others believed that the bishop was arrested because he was plotting to make himself pope.

On the face of it, this third story might seem to be the most farfetched, given what we know of Odo’s character and the high standards set by the reformed papacy. But by this date the reform movement had run into grave difficulties. The death of Pope Alexander II in 1073 had led to the succession of Gregory VII, a firebrand reformer who took an uncompromising view of the relationship between lay rulers and the Church. His insistence on the supremacy of papal authority had led quickly to a bitter conflict with the king of Germany, Henry IV: Gregory had twice excommunicated Henry and declared him deposed; the king had responded in kind, declaring the pope unfit to hold office and nominating a rival to rule in his place. Their quarrel had divided opinion right across Europe.

Set against this background, the story that Odo was aiming at the papacy starts to seem more credible. It also draws strength from being told by three chroniclers, whose differing accounts show that they were written independently. According to Orderic Vitalis, the bishop purchased a palace in Rome, decorated it sumptuously, and secured the support of several leading Roman families by scattering lavish gifts. William of Malmesbury claimed that Odo advanced his cause by stuffing the wallets of pilgrims with letters and coin. Whatever the precise means, it is not hard to believe that Odo, a man possessed of enormous wealth and power as well as apparently limitless ambition, may well have tried to advance himself as a compromise candidate in the struggle for St Peter’s throne.¹

But what may have seemed a logical career move to Odo appeared otherwise to his elder half-brother. William was far from being a staunch supporter of Gregory VII. Relations between them had started well enough, for in 1066 Gregory, then a cardinal, had been one of the principal cheerleaders for the Norman invasion of England. But latterly their friendship had been dented by the pope’s attempt to call in this debt by asserting that England was a papal fief for which William owed homage – a claim that the king had naturally rejected. Gregory had also fallen out with Lanfranc over the latter’s repeated refusal to visit him in Rome, even to the extent of threatening to suspend the archbishop from office. Yet despite these tensions there had been no serious rift. In a letter to two French bishops in 1081, Gregory had praised William as a pious ruler who supported the Church and governed his subjects with peace and justice. (‘Although in certain matters he does not comport himself as devoutly as we might hope,’ the pope admitted, ‘he has shown himself more worthy of approbation and honour than other kings.’) Similarly, when Gregory was eventually ousted from office by Henry IV in 1084, Lanfranc still refused to denigrate him, and replied angrily to a partisan letter from one of Henry’s supporters which did so.²

It is highly unlikely, therefore, that either William or Lanfranc would have given their backing to a plan to unseat Gregory in 1082 and replace him with Bishop Odo. Whatever their views on Gregory, both were firm supporters of the reform movement and could hardly have believed that Odo would make a suitable supreme pontiff. But papal politics and questions of suitability aside, there was another more important reason for objecting to the plan. According to both Orderic and William of Malmesbury, Odo had been determined to back up his bid by leaving England with a large force of distinguished knights – Orderic names Earl Hugh of Chester among their number. ‘They resolved to abandon the great estates they possessed in western parts and took an oath to accompany the bishop to lands beyond the Po.’ The Conqueror, when he learned of this intention, naturally took a very dim view. The knights he had planted on English soil were his knights – his security against further insurrection or foreign invasion, whose removal might very well imperil his grip on the kingdom. And so, says Orderic, the king, who was in Normandy during the first part of 1082, quickly crossed to England and took his brother by surprise. Odo, who was making ready to sail from the Isle of Wight, was seized and put on trial.

Orderic’s account here cannot be taken entirely at face value. William did cross to England in the autumn of 1082, but he cannot have arrested his brother immediately, for the two men are found in each other’s company later that year, witnessing a charter in Wiltshire. Nor can we place too much trust in the long speech that Orderic attributes to William during the trial, in which Odo is denounced primarily for his oppression of the English. Nevertheless, that something like this did take place after William’s return is proved by its brief but emphatic mention in the Anglo-Saxon Chronicle. Orderic reports the conclusion of the trial in dramatic terms, telling us that none of the magnates assembled in the king’s hall dared to lay hands on the accused, forcing William to carry out the arrest himself. When Odo protested (ironically, giving his alleged ambition) that as a bishop he could be tried only by the pope, the king reportedly replied that he was not condemning a bishop but arresting an earl (a line that finds its way into other accounts – Malmesbury attributes it to Lanfranc). However it happened, Odo was taken across the Channel and imprisoned in the ducal castle at Rouen.³

Odo’s betrayal – as William must have regarded it – was the first of several family disasters to overtake the king in a short space of time. Soon after his return to Normandy in 1083, his wife, Queen Matilda, fell sick, and eventually died on 2 November. This can have been nothing other than a devastating blow for William, for their marriage had clearly been based on love and trust – there is no other way to account for his consistent reliance on her to act as his regent in both England and Normandy. Unlike almost every other eleventh-century ruler William had no bastard children and no reported infidelities – at least no credible ones. William of Malmesbury recorded the scandalous rumour that, after his accession as king, the Conqueror had ‘wallowed in the embraces of a priest’s daughter’, but dismissed the story as lunacy: after all, it concluded with Matilda having her rival hamstrung and William vengefully beating the queen to death with a horse’s bridle. In reality, says Malmesbury, the royal couple had a minor falling-out in their later years over the support she had secretly supplied to Robert Curthose during his rebellion. ‘But that this occasioned no lessening of their affection as man and wife he himself made clear; for when she died … he gave her a most splendid funeral, and showed by many days of the deepest mourning how much he missed the love of her whom he had lost.’ Matilda was buried in Caen, in the abbey of Holy Trinity she had founded over twenty years earlier. Her tombstone, complete with its original carved epitaph, can still be seen in front of the high altar.⁴

Not long after the loss of his wife, William was hit by a third personal crisis when Robert Curthose once again went into exile. The exact cause of their renewed estrangement is unknown, but its occurrence so soon after Matilda’s death is surely significant. Orderic says that despite their earlier rapprochement the king had continued to pour abuse on his son, frequently reproaching him in public for his perceived failings; elsewhere he says that Robert was angry with his father ‘for some silly reasons’. Whatever the cause, it did not have the same disastrous effect as their earlier quarrel, since on this occasion Robert was followed into exile by only a few companions. William of Malmesbury says that he went off to Italy, intending to marry the countess of Tuscany, and hoping to secure support against his father. It may be that he also toured other regions to the same end: Orderic at one point mentions visits to sympathetic lords in Germany, Aquitaine and Gascony. But if on this occasion Robert posed no immediate military threat, his departure must still have occasioned great uncertainty, not least with regard to the succession in both Normandy and England. William made no public moves to disinherit his eldest son or to promote his younger brothers; but from 1084 William Rufus assumes Robert’s place in the witness-lists to royal and ducal charters.⁵

The Conqueror thus suffered three successive blows that struck very close to home: the imprisonment of Odo, the death of Matilda and the renewed rift with Robert. These losses and betrayals must have taken their toll on him, and left him perhaps feeling increasingly isolated. Apart from anything else, these had been the figures he had relied upon to act as his regents. At the same time, there is no sign that William’s family problems from 1082 to 1084 triggered any wider crisis across his dominions. Robert may have hoped to stage a comeback, but chronicle reports depict him as a wandering exile, lacking friends and funds. His former allies, for example, the king of France and the count of Anjou, appear to have lent him no material support; the peace agreed with France in 1079 seems to have held good, while in 1082 William had struck a similar truce with Anjou that reportedly lasted until the end of his life. The only problem the king faced in the wake of Robert’s departure was in Maine, after the viscount of Le Mans rebelled and seized the castle of St Suzanne on the county’s southern border. In 1084 William set about besieging it with customary speed and vigour.⁶

Some way into the siege, however, the king delegated its prosecution to others, and returned to Normandy to deal with what Orderic calls ‘great matters’. What these were the chronicler does not specify, but we may suppose they involved the intelligence William received in 1085 about a planned invasion of England by the king of Denmark.⁷

This was, of course, hardly a new development, for the prospect of Scandinavian invasion had hung over William’s reign from the very beginning, and Danish fleets had twice crossed to England, only to return home with very little to show for their efforts. In 1085, however, the threat seemed altogether more serious. Five years earlier Denmark had witnessed the accession of a new King Cnut, a man seemingly determined to repeat the victories of his illustrious namesake. One of the many sons of the late Swein Estrithson, Cnut IV had been among the captains his father had sent to England in 1069 and 1075, leading fleets that contemporaries estimated at 300 and 200 ships. But once he wore the Danish crown himself, the new king became bent on something even bigger. ‘With his former failures in mind’, says William of Malmesbury, ‘Cnut prepared a fleet of a thousand ships or more (so I’ve heard) for the invasion of England.’ Even if we dismiss this as hearsay, there is no difficulty in believing that on this occasion the forces being assembled were much larger than before, for the Danish king was not acting alone. ‘In this year’, said the Anglo-Saxon Chronicle, ‘people said and declared for a fact that Cnut, king of Denmark, son of King Swein, was setting out in this direction and meant to conquer this country with the help of Robert, count of Flanders’. Some years earlier Cnut had married Robert’s daughter, Adela. According to William of Malmesbury, the count had assembled a further 600 ships in readiness for the planned invasion.⁸

The seriousness of the threat is further attested by the reaction of King William. In the autumn of 1085 the Conqueror rushed across the Channel to England, bringing with him what the Chronicle calls ‘a larger force of mounted men and infantry from France and Brittany than had ever come to this country’. Whether or not that assessment included the invasion force of 1066, the king’s army in 1085 was unquestionably massive: John of Worcester writes of ‘many thousands of paid troops, footsoldiers and archers’, while Malmesbury calls it ‘a great multitude of knights serving for pay, from every province this side of the Alps’. William, he says, had even engaged the services of the French king’s brother, Hugh of Vermandois, along with all his knights.⁹

We can well believe that Malmesbury was right, therefore, when he said ‘the king was very scared’. William’s first action on arriving in England was to summon an emergency council of magnates to debate how to deal with the crisis. The first and most pressing consideration was what to do with the mercenary army. Unlike every other force William had brought to England, their job was not to harry the kingdom but to defend it, and as such they could not resort to the usual medieval practice of living off the land at the expense of its inhabitants. ‘People wondered how this country could maintain all that army’, says the Chronicle. The solution arrived at in council – proposed, says Malmesbury, by Archbishop Lanfranc – was to disperse the troops all over the kingdom. ‘All agreed that the households of the magnates should be reinforced by the presence of knights, so that if need be everyone could unite to defend the public weal and private fortunes against the barbarian.’ As Malmesbury’s account makes clear, this made the magnates themselves responsible for the mercenaries’ every need. Bishop Wulfstan of Worcester, for example, ‘began to maintain a sizeable force, keeping them happy with high pay and filling them up with choice food’.¹⁰

It was not just the poor magnates who suffered. ‘People had much oppression that year’, recalls the Anglo-Saxon Chronicle, ‘and the king had the land near the sea laid waste, so that if his enemies landed, they should have nothing to seize.’ According to an early twelfth-century source, William also ordered the coastline to be closely guarded; castles were strengthened and town walls repaired and manned. The towns themselves were required, like the magnates, to billet large numbers of French mercenaries, to the extent that there was hardly enough room left for their English inhabitants.

So, as the days grew shorter and autumn turned to winter, England entered a state of high tension. As in 1066, the whole kingdom held its breath and waited for news. Then, shortly before Christmas, news came. ‘The king found out for a fact that his enemies had been hindered’, says the Anglo-Saxon Chronicle, ‘and could not carry out their expedition.’ It was only a temporary reprieve – Cnut had apparently decided to delay his invasion until the new year – but it must have eased the tension a little. William responded by sending some of his mercenary troops back to the Continent. Others, however, he kept in England throughout the winter, indicating that the fear of invasion remained.

Christinas 1085, which the king once again kept at Gloucester, must therefore have been a stressed affair, with the thought of the ongoing Danish threat never far from the minds of the assembled magnates. William revealed his own anxiety at this time by dismissing from office the abbots of Crowland and Thorney, a pair of neighbouring monasteries in the Fens. This, of course, was a region where the Danes had landed in the recent past and received considerable support from the local population – not least the local monks. Not wanting another Ely on his hands, and evidently doubting the loyalty of Crowland’s and Thorney’s existing abbots, the king replaced them with two monks from the Norman abbey of St Wandrille.¹¹

The Christmas court at Gloucester lasted five days, says the Anglo-Saxon Chronicle, the Church council that followed a further three. The Chronicle then goes on to tell us what happened next:

After this, the king had much thought and deep discussion with his council about the country – how it was occupied and with what sort of people. Then he sent his men over all England into every shire and had them find out how many hundred hides there were in the shire, or what land and cattle the king himself had in the country, or what dues he ought to have in twelve months from the shire. Also he had a record made of how much land his archbishops had, and his bishops and abbots, and his earls, and – though I relate it at too great length – what or how much everybody had who was occupying land in England, in land or cattle, and how much money it was worth. So very narrowly did he have it investigated, that there was no single hide nor yard of land, nor indeed (it is a shame to relate but it seemed no shame to him to do) one ox nor one cow nor one pig was there left out, and not put down in his record.

Such is the Chronicle’s description of the Domesday Survey, which produced the Domesday Book, one of the most famous documents in English history (in terms of celebrity only Magna Carta trumps it), and certainly the most voluminous. It is, in fact, two volumes: a large one, called Great Domesday, and a smaller one, known as Little Domesday. Between them they contain 832 folios, filled on both sides with closely written, abbreviated text, and containing somewhere in the region of two million words.¹²Because of its vast size Domesday is, and always has been, unique. A handful of subsidiary documents relating to the survey have survived – sectional drafts which scholars have dubbed ‘satellites’ – but the book itself, in two volumes, survives only as a single copy. It is, without question, the most important document in English history, for, as the Anglo-Saxon Chronicle indicates, it contains the results of an inquest of unparalleled scope and magnitude: a kingdom-wide investigation of rights, dues, land and economic assets, which produced the single greatest description of a pre-industrial society anywhere in the world.

Hence, of course, its ominous name. During the twelfth century, Domesday was kept in the royal treasury at Winchester, and so initially government officials referred to it as ‘the book of the Exchequer’, ‘the great book of Winchester’, or simply ‘the king’s book’. But, as one of those officials noted in the 1170s, ‘this book is called by the natives “Domesday” – that is, by metaphor, the Day of Judgement’. One suspects that this English nickname had existed from the very beginning, because ample other evidence exists to indicate that William’s inquiry left contemporaries awestruck. Part of the reason for their astonishment must have been the speed with which the project was carried out. Begun soon after Christmas 1085, it was apparently completed before 1 August the following year. Producing the final draft of Great Domesday took somewhat longer, but the survey itself was accomplished in a little over six months.¹³

Why the Conqueror launched this great investigation in the winter of 1085 is a question that has long been debated by historians. Given its timing, few have been content to believe that it was merely an act of royal curiosity, and most would link its origins to the ongoing threat of invasion. Exactly how the two events were linked, however, remains a matter of dispute. Despite over a century of rigorous scholarship, there is still no consensus as to what the Domesday Book was for.

By looking at the book itself, the satellites, and the comments of contemporary chroniclers, we can at least understand something of how the survey was made. As the Anglo-Saxon Chronicle indicates, it began with William sending commissioners into every shire, and from the Domesday documents themselves we can see that the shires were grouped into at least seven circuits (mostly consisting of five counties each, but in one case containing only three). For one particular circuit we can see that there were four commissioners, so overall there were probably no more than thirty or so men directly responsible for compiling the Domesday data.

Providentially, thanks to a document known as the Ely Inquest, we also appear to have a list of the questions to which these commissioners were required to find answers. From these we see that the principal unit that Domesday was concerned with was the manor – a term that appears to have been coined after the Conquest, which signified something like ‘individual lordly estate’. The commissioners began with the basics: what was the manor called? Who held it in the time of Edward the Confessor, and who holds it now? They then moved on to specifics: how many hides of land are there? How many ploughs? How many slaves? How many freemen? They also asked about natural resources: how much woodland was there? How much meadow? How much pasture? Lastly they asked about value: how much was it worth then? How much is it worth now?¹⁴

This is just a selection from the total list of twenty questions given in the Ely Inquest, all of which had to be posed in the case of each individual manor. Domesday contains many, many entries: at a rough estimate it mentions about 13,000 places and around30,000 manors.¹⁵ The thirty or so Domesday commissioners, we assume, must have had many assistants. But, even so, how could they possibly have compiled all this information in just six months?

Part of the answer was that they relied on that perennial government time-saver, self-assessment. Tenants-in-chief – those who held their land directly from the king – seem to have been required to supply the commissioners with written returns; in some instances we can see their fossilized form in the pages of the Domesday Book itself. The commissioners could also have obtained a lot of information quickly from existing written records. Eleventh-century England, it bears repeating, was a much-governed country, a medieval bureaucratic state. By turning to earlier surveys and tax rolls, many answers could be uncovered, especially those relating to the time of Edward the Confessor.¹⁶

Had Domesday involved no more than this, it would probably have generated little comment – indeed, it would have been just another royal survey in the Anglo-Saxon tradition. It would also have had little value, because self-assessment by the landowners would have inevitably delivered a skewed result. But Domesday involved a great deal more, and is a great deal more valuable, because all the written evidence assembled, whether from self-assessment or government archives, was subjected to public scrutiny. During the spring of 1086, in every county, extraordinary sessions of the shire court were convened, to which an extraordinary number of witnesses were summoned. Principally these were jurors selected from every hundred, or wapentake. In Cambridgeshire, for example, each of the county’s fifteen hundreds sent eight jurors, including in every case the head of the hundred (the reeve) and a local priest: a total of 120 men. This total is likely to have been low compared to the national average: other counties contained more hundreds, and the juries they sent probably numbered the more typical twelve.¹⁷

But the hundred jurors, while seemingly the most important, were not the only ones required to give evidence. The landowners were also summoned in order to defend their written assessments against the jurors’ oral testimony, and there were also other types of juries, some representing the shire as a whole, some its most important towns, some its monastic communities. According to the Ely Inquest, every settlement (vill) was required to send an eight-man jury, and this statement receives support from the villeins who appear, on two occasions, as witnesses in the Domesday Book. At the very least, therefore – counting only the jurors drawn from the hundreds, shires and boroughs – the survey involved the participation of around 8,000 people; if it really involved juries from every settlement, the number rises to somewhere in excess of 60,000. Each county assembly therefore involved hundreds, possibly thousands, of people, making these meetings far larger than the regular sessions of the shire court, and larger even than the great land pleas (like Penenden Heath) held earlier in William’s reign. One imagines it was these giant gatherings that had the greatest impact on the popular consciousness, and gave rise to the name ‘Domesday’.¹⁸

The jurors were almost certainly required to answer all the commissioners’ questions, including those about the economic assets of a particular manor: from time to time Domesday records their objection to a landlord’s valuation.¹⁹ The crucial questions put to them, however, related to ownership: who held the land during the time of Edward the Confessor, and who holds it now? As we’ve seen, the Norman colonization of England had been a sporadic, piecemeal and confusing process, with land being granted out on two different and potentially conflicting principles, and plenty of property simply seized by those with the sharpest elbows.

In the Domesday sessions of the shire court, this process now came under the spotlight (which is, of course, how we come to know so much about it). In some areas there had been comparatively little disruption, because estates had been transferred unbroken from their former English owner to a Norman newcomer. But Domesday reveals that only a minority of properties had been reallocated in this way – something like ten per cent of the total. A much larger amount of land – over a third of all the land in England – had been granted out on a territorial basis, paying no regard whatsoever to previous patterns of ownership. In the case of the remaining lands, there is no clear pattern of redistribution, suggesting that in these areas the Normans had been helping themselves.²⁰

Hence the thousands of doubtful or contested claims in Domesday. In some instances the jurors validated the right of a particular landowner, testifying that they had seen the original royal writ that ordered the owner to be put in possession. But very often they said they had seen no such writ, and had no idea by what right the defendant came to be holding the estate in question. Sometimes rival claimants came forward to contest a property. In Hampshire, for example, that notorious predator, Picot the Sheriff (‘a hungry lion, a roving wolf, a crafty fox …’) had his title to a small parcel of land in Charford challenged by a fellow Norman named William de Chernet. William, it is recorded, ‘brought his testimony for this from the better and old men from all the county and hundred. Picot contradicted this with his testimony from the villeins, common people, and reeves.’ Sadly, but typically, we do not know who won, but in general Picot was very successful in hanging on to his acquisitions.²¹

Part of the purpose of the Domesday Survey, therefore, was to sort out precisely this sort of tenurial confusion – the net result of twenty years of Norman colonization which had been at best opportunistic and at worst downright rapacious. Richard fitz Nigel, the twelfth-century official who first recorded the name ‘Domesday’, thought that the book had been made ‘so that each man, being content with his own rights, should not with impunity usurp the rights of another’, and there is much to be said for this. Every effort had been made to ensure that the final judgement was as fair as possible. Not only were the jurors summoned to provide a check on the landlords, the names of the jurors, where known, reveal a fifty–fifty split between Englishmen and Frenchmen (thus in Cambridgeshire, each eight-man hundred jury has four English and four French jurors apiece). This ethnic balance meant, of course, that neither nation could use Domesday as an opportunity to settle old scores, but it also implies that ways to ensure the fairness of the survey had been worked out at the very highest level. Similarly revealing is a short, contemporary description of the Domesday Survey by the bishop of Hereford, which states that a second set of commissioners was sent around the country to check the work of the first, and, if necessary, to denounce its authors as guilty to the king. The second team, we are told, were sent into regions where they themselves held no land – another obvious anti-corruption measure – and this, so far as we can judge, was true of their forerunners too.²²

The question that inevitably arises, though, is why this problem could not have waited. Sorting out property disputes was unquestionably important, for left unresolved they might ultimately lead to violence between rival landlords; clearly, some degree of orderhad to be imposed. But why did this take priority at a time when the country was threatened with invasion? After all, if the Danes were successful, any record of who owned what in England would be out of date almost before the ink had dried. That the threat of invasion still hung over England at the time of the inquest is implicit in William’s decision to retain some of his French mercenaries throughout the winter. Indeed, in one or two instances we seem to see these troops in the Domesday Book, garrisoned in towns such as Southampton and Bury St Edmunds.²³

The mercenaries themselves provide a further clue as to the survey’s purpose. Soldiers of fortune, by their very nature, rarely come cheap. By dispersing his great army around the country in 1085, William had passed some of its cost on to his magnates, but not all. Many troops had also been quartered on royal estates and towns, and the same situation must have obtained into 1086, requiring the king to find large sums of money to retain the men who remained in his service.

Earlier English kings had obtained such large sums by taxing their subjects. The geld, as we’ve seen, had been collected since the late tenth century, first to buy off Danish invaders with tribute, and later to maintain a mercenary fleet in order to guard against future attacks. Edward the Confessor had ostentatiously done away with this fleet in 1051 and ceased to collect the money to pay for it, but it seems unlikely that he abolished the geld entirely. The Anglo-Saxon Chronicle’s comment that the tax for the fleet ‘always came before other taxes’ implies that other forms of geld existed, and these almost certainly continued to be levied on an annual basis for the rest of the Confessor’s reign. The sudden reappearance of geld in the Anglo-Saxon Chronicle within days of William’s coronation is thus to be explained by its heavy rate rather than its sudden revival.²⁴

By 1066, however, the geld was no longer the great cash cow it had once been. This was partly because successive kings had run down the system of assessment. Geld was calculated on the basis of the number of hides a particular estate or community contained, so by reducing this number a king could show political favour. By the reign of Edward the Confessor many hundreds – which notionally ought to have contained one hundred hides – were assessed at considerably less. A surviving geld roll for Northamptonshire, drawn up in the 1070s but recording the situation in Edward’s day, shows some hundreds containing eighty hides, others sixty-two hides, and in one case as few as forty. Thus in 1066 the total number of hides in Northamptonshire’s thirty-two hundreds was not 3,200, but 2,663. This erosion, moreover, had continued after the Conquest. Domesday reveals that by 1086 the hidation of Northamptonshire had fallen to just 1,250.²⁵

Nor was that the limit of the problem, because only a fraction of the hides in this much-diminished rump actually paid geld. In every hundred in Northamptonshire, the same geld roll shows that large numbers of hides – in some cases up to half the total – had been written off as ‘waste’. This word, which also occurs over and over again in the Domesday Book, has been interpreted in recent times merely as a term of administrative convenience, used indiscriminately by royal officials to describe land which, for whatever reason, paid no geld. But most historians today would argue that waste (vasta) had a specific meaning, and indicated land that had been ruined – devastated – as a result of the Conquest. A thoroughgoing analysis of the word’s use in Domesday shows a quite precise correlation with those areas visited by the Conqueror’s armies in the years immediately after 1066. Thus we find high concentrations of waste in Sussex, particularly in the rape of Hastings, as well as in the counties along the Welsh border, harried in the period 1069–70. We see it in the coastal areas that William had strategically ravaged in 1085, and we also see it in cities. In Lincoln, Domesday records that there were 970 inhabited dwellings before the Conquest, but afterwards 240 of them were waste: 166 had been destroyed ‘on account of the castle’, the remainder ‘because of misfortune and poverty and the ravages of fire’. Similar accounts equating waste with the construction of castles are found for Norwich, Shrewsbury, Stamford, Wallingford and Warwick. When Earl Hugh received Chester, says Domesday, almost half its houses had been destroyed, and the city had lost a third of its value, ‘for it had been greatly wasted’.²⁶

None of this, however, compares to the amount of waste recorded for Yorkshire, which accounts for over eighty per cent of Domesday’s total for the whole country. This goes a long way to substantiating the chronicle accounts of the Harrying of the North and the scale of the destruction caused by the Conqueror’s armies in the winter of 1069–70. According to Simeon of Durham, the region between Durham and York lay uncultivated for the next nine years, with every settlement uninhabited; William of Malmesbury, writing in the 1120s, declared that the soil there was still bare in his own day. Shockingly, Domesday reveals that in 1086 the population of Yorkshire had dropped to just a quarter of what it had been in 1066, meaning that around 150,000 people had vanished from the record. For once, it seems, the six-figure numbers given by the chroniclers corresponded all too closely with reality.²⁷

Some areas of the country, it is true, seem to have recovered in the twenty years between 1066 and 1086. Whereas, for example, the average value of manors in Yorkshire had plummeted by over sixty-five per cent, in Norfolk it had risen by an impressive thirty-eight per cent. It would be wrong, however, to read too rosy a picture into such rises, for the value of a Domesday manor was the value to its lord in rents, and there is ample evidence to indicate that new Norman lords had racked up rents to intolerably high levels. Domesday abounds with complaints about rents being oppressive, exceeding the actual value of the land (a famous entry records that Marsh Gibbon in Buckinghamshire was held by its English farmer, Æthelric, ‘in heaviness and misery’). According to the Anglo-Saxon Chronicle, such oppressive practices originated at the very top:

The king gave his land as dearly for rent as he possibly could; then came another man and offered more than the first, and the king let it go to the man who had offered more; then came a third and offered still more, and the king gave it up to the man who had offered most of all. And he did not care at all how very wrongfully the reeves got it from poor men, nor how many illegal acts they did.

The Normans appear to have been uniformly rapacious in pursuit of profit: elsewhere the Chronicle laments that ‘the king and his leading men were fond, yea, too fond, of avarice: they coveted gold and silver, and did not care how sinfully it was obtained’. The final question of the Domesday commissioners, as preserved in the Ely Inquest, was whether more could be taken from an estate than was currently being taken.²⁸

For this reason, where the Domesday Book shows a fall in manorial values in areas not associated with widespread ravaging, we can be fairly certain that this was not due to leniency on the part of the landlords. In certain counties it has been shown that the sharpest falls in value coincide quite precisely with brand-new Norman lordships – the kind carved out from scratch, with no reference to previous landholding patterns. One view is that this reorganization process was so disruptive that it caused a drop in economic output. A more likely reason is that new manors had been constructed on a new model, more favourable to the lords and more oppressive for the peasantry. For it is also in these counties that we witness dramatic falls in the number of free peasants. In Cambridgeshire, for example, the number of freemen plummeted from 900 to 177; in Bedfordshire from 700 to 90, and in Hertfordshire from 240 to 43. At the same time, we discover that the number of servile peasants has rocketed. Frequently in Domesday we find the phrase ‘he is now a villein’.²⁹

Values were only lower in these areas, it seems, because the before – and-after figures record different realities. Those for the pre-Conquest period had been calculated by combining the incomes of all the various English freeholders on a particular estate; the figures for 1086, by contrast, simply represented the income the new Norman lord derived in rent, having forced these former freemen into financial servitude. This was a bad bargain for the small landowners, but after the Conquest their bargaining power was not strong. The Anglo-Saxon Chronicle says at one point that the Normans ‘imposed unjust tolls and did many injustices which are hard to reckon up’. The tenurial revolution, in short, had prompted a social revolution. English society, in certain areas at least, was a lot less free after the Conquest than it had been before.³⁰

Yet even as lords were driving up profits, they were paying less and less money to the king. At some stage, probably during the reign of the Conqueror himself, land held in demesne had been made exempt from paying geld. We see this, for instance, in the Northamptonshire geld roll, where hides held in demesne, like those listed as waste, are treated separately from those that actually paid. As the reign continued, the concessions appear to have increased, so much so that by the 1080s some tenants-in-chief were paying almost no tax at all. Indeed, in some cases, they may even have collected the geld from their demesnes and kept it for themselves.³¹

Small wonder, then, that revenue from the geld was not all it had once been. Concessions to individuals and communities had led to a massive reduction in the number of hides, and many more had been written off as waste as a result of war and destruction; the Conqueror himself had exempted the demesne land of his tenants-in-chief, and established a further tax-free zone in the form of the Forest. All of these factors had punched great holes in the money-getting system created by England’s pre-Conquest kings. Added to this there was also outright refusal to pay:‘From 6½ hides at Norton,’ records the Northamptonshire geld roll, ‘not a penny has been received – Osmund the king’s secretary owns that estate.’³²

An obvious way to compensate for the system’s shortcomings was to try to get more geld from hides which were still liable. In 1084, William had ordered a geld at three times the normal rate (six shillings per hide rather than the usual two), probably intending to use the money it raised to finance his ongoing war in Maine. The Anglo-Saxon Chronicle recorded this hike with horror, describing the tax as ‘heavy and severe’, and no doubt it was for the limited number of people who had to pay it. At the same time, we have no way of knowing how lucrative it was. Given the manifold inadequacies of the system by this date, and coming hard on the heels of the 1082 famine, the yield may have been disappointing, even though the rate was exceedingly harsh.³³

The final straw may have come in the autumn of 1085, with the decision to break up the massive mercenary army and billet it around the kingdom. ‘The king had the host dispersed all over the country among his vassals’, says the Anglo-Saxon Chronicle, ‘and they provisioned the army each in proportion to his land.’ If the government used the geld lists to try to arrive at this proportional distribution – and it is hard to see how they could have used anything else – then the billeting arrangements would have been no more equitable than a levy of the geld. By the time of the Christmas court, therefore, there may have been many voices being raised in protest, and a realization on William’s part that the system needed to be fixed, in order that future gelds brought in more money, and that any future billeting would be fairer. And part of the Domesday process was an inquiry into the workings of the geld – we can see as much from the circuit return for the south-western shires known as Exon Domesday. An overhaul of the geld system would explain why the survey was so interested in lordly resources: the king was trying to discover precisely where the profits of lordship were going, perhaps with a view to reversing the exemption of demesne. Lastly, a fiscal motive for Domesday would also explain why William felt it was necessary to establish exactly who owned what, for – as those responsible for its administration must surely have attested – it is difficult to collect a land tax when landholding itself is in dispute.³⁴

Such was the conclusion of early Domesday scholars: the Conqueror’s survey was a tax inquiry, intended to remedy the manifold defects of the geld system. As the greatest of all these scholars, Frederick William Maitland, wrote in his Domesday Book and Beyond (1897): ‘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.’ Even today, more than a century after Maitland’s death, many experts would argue that fiscal reform was the primary purpose of the Domesday Survey.³⁵

And yet it cannot have been the purpose of the Domesday Book. As anyone who has ever tried to do so will readily attest, it is all but impossible to use the book as a tool for assessing geld, because of the way its contents are organized. Geld was national, public tax, administered using public institutions: collectors moved from settlement to settlement within each hundred, and from hundred to hundred within each shire, in order to gather in the money. The Domesday commissioners, as we have seen, had used the same public institutions to gather information, summoning juries from hundreds and townships in order to check the written returns provided by individual landowners. But, crucially, when this information came to be compiled and written up for each of the survey’s seven circuits, it was not arranged by hundred and vill. Instead, it was laboriously rearranged – by landowner. Both the surviving circuit returns (Exon Domesday and Little Domesday) are arranged in this fashion, as is the final redaction of the data in Great Domesday. It is an arrangement that makes it fantastically difficult to calculate geld payments; to work out the liability of a particular landowner requires hours of calculations.³⁶ In conclusion, therefore, there is no doubt that a geld inquiry was launched in 1086, and little doubt that a reform of the geld was high on the agenda. Some of the information collected by the Domesday commissioners may have been intended for such a reform. But the selection and arrangement of data in the Domesday Book indicates that it must have been made for a different reason.

While the commissioners had been gathering their data, assembling jurors and landowners in their hundreds and thousands, King William had been travelling around his kingdom (or at least its southern part). At Easter (5 April), by which time the sessions of the shire courts must either have been well advanced or already over, he wore his crown at Winchester. By Whitsun (24 May), when he was at Westminster for the knighting of his youngest son, Henry, the king must have had a clear idea of when the survey would be completed. As John of Worcester explains, ‘shortly afterwards, he ordered his archbishops, bishops, abbots, earls, barons, sheriffs and their knights to meet him on 1 August at Salisbury’. This was to be the last and greatest meeting of 1086 – an assembly for the whole kingdom, the culmination of the Domesday process.³⁷

Salisbury was not in the same place in 1086 as it is today (the present city is a new foundation of the thirteenth century). In the Conqueror’s time it lay two miles to the north, on the site now known as Old Sarum. The Normans had been drawn to this location from an early date. William had established a castle there, probably before 1070, and the bishop of Sherborne had subsequently moved his cathedral to stand alongside it, thereby becoming the bishop of Salisbury. The site’s attraction in 1086 may have had more to do with its prehistoric past, for Old Sarum is one of the most impressive Iron Age hill forts – both castle and cathedral were planted in the centre of a massive enclosure, 400 metres across, surrounded by an earthen rampart that runs for over a kilometre. It was the exactly the right kind of location, in other words, for the sort of large-scale open-air assembly that the Conqueror had in mind.³⁸

For the meeting at Salisbury was massive. ‘There his counsellors came to him,’ says the Anglo-Saxon Chronicle, ‘and all the people occupying land who were of any account all over England.’ To give us some idea of what that might mean in terms of numbers, we can turn to Domesday. The book names around 1,000 individuals who held their land directly from the king – his tenants-in-chief. Their direct connection with the king, and (in most cases) their superior wealth and status meant that most if not all of them would have attended. Domesday also records a further 8,000 or so landowners at the next stage down – i.e. the tenants of the tenants-in-chief, or the king’s subtenants. How many of these men might have come to Salisbury is open to debate. Some of them were as wealthy, or even wealthier, than the lesser tenants-in-chief, and so would correspond to the Chronicle’s notion of ‘people of account’. But many others were considerably less wealthy, with half of them owning land worth less than £1 a year, and so might be reckoned to lie outside of that description. Depending on how strictly the king’s summons was interpreted, therefore, we should imagine an attendance figure well into four figures, and just possibly nudging towards five.³⁹

The impression of an exceptionally large meeting is reinforced by the composition of William’s court. Sadly, we have no record from the day of the Salisbury assembly itself, but we do have a document issued at another location in Wiltshire, datable to the middle of 1086, with a witness-list that suggests it was drawn up at a time very close to the event. Alongside the Conqueror stand his two younger sons – the newly knighted Henry and his older brother, William Rufus. Then come the higher clergy: Archbishop Lanfranc of Canterbury, Archbishop Thomas of York, and with them the bishops of Durham, Winchester, Lincoln, Chester, Hereford, Salisbury and London. The great lay magnates come next, a list of eighteen names. Odo, still confined to his cell in Rouen, is notably not among their number; but William’s other half-brother, Robert, is present, along with the king’s lifelong friend, Roger of Montgomery. Also among the throng of laymen we see Richard fitz Gilbert, lord of Tonbridge, the greatest landowner in south-eastern England; Henry de Ferrers, a great baron from northern England, and Robert of Rhuddlan, the conqueror of north Wales. Another name that leaps out from the list is Abbot Thurstan of Glastonbury, whose massacre of his own monks just three years earlier was evidently no bar to his attendance at court.⁴⁰

It is an impressive roll call – one of the greatest we have seen since the coronation of Queen Matilda in May 1068. Yet what a change has been wrought in the intervening eighteen years. Royal charters issued at the time of Matilda’s coronation reveal a mix of English and French names, with the majority being English. But in 1086 the English are gone, and the list is exclusively Norman (or at least, in the case of the bishops, Continental). In 1068 ten of England’s fifteen bishoprics had been held by Englishmen, three of the remaining five having been given to Germans by Edward the Confessor and only two recently filled by the Conqueror. But by the time of the Domesday Survey only one English bishop – the wily and venerable Wulfstan of Worcester – remained in office. As for the English aristocracy, the eclipse is total. The three native earls who witnessed in 1068 were all long gone by 1086: Eadwine murdered, Waltheof executed, and Morcar still languishing in prison. Also gone are the lesser English nobles present at the queen’s coronation – men with names like Æthelhead, Tovi, Dinni, Ælfgeard, Bondig, Wulfweard, Herding, Brixi and Brihtric. When we turn to the witnesses of 1086, there is not a single English name among them.⁴¹

What is true of the court, moreover, is also true of the country. Of Domesday’s 1,000 tenants-in-chief, a mere thirteen are English; only four have lands worth more than £100 and the wealthiest, Edward of Salisbury, despite his English name, may well have been half-Norman. The king’s thegns – the ninety or so lords who had each owned more than forty hides of land – were all gone. Even when we descend to the next tenurial level, where we start to find natives in more significant numbers, they remain very much in the minority. Of the 8,000 or so subtenants recorded in the survey, only around ten per cent are English, and, as with the tenants-in-chief, the survivors are small fry – men like Æthelric of Marsh Gibbon, holding ‘in heaviness and misery’. England’s middling thegns, who had numbered around 4,000–5,000, have been swept clean away.⁴²

Domesday therefore reveals cataclysmic change to the composition of England’s ruling class, with Normans replacing native lords in almost every village and hamlet. It also, moreover, reveals dramatic changes within that class in the distribution of material wealth. Put simply, there were more super-rich men in Domesday England than there had been twenty years earlier. Whereas in 1066 there had been several thousand middling English thegns, by 1086 half the land in England was held by just 200 Norman barons.⁴³Having in most cases obtained the estates of multiple English predecessors, these newcomers were many times more wealthy. In Edward the Confessor’s day, for example, only thirty-seven individuals had held lands with an annual value of more than £100; by the time of the Domesday survey the number of such men had more than doubled to eighty-one.⁴⁴ At the very top, the spoils of Conquest had been colossal. Half the country was in the hands of 200 barons, but half that half – i.e. a quarter of all the land in England – was held by just ten new magnates. Their names are by now a predictable roll call of William’s friends and family: Odo and Robert, Roger of Montgomery, Richard fitz Gilbert, Hugh of Chester … Orderic Vitalis had not exaggerated when he said that the Conqueror had raised his dependants to high rank and heaped great honours upon them. ‘He was a great lover of the world and of Worldly pomp,’ said Orderic of Earl Hugh, ‘lavish to the point of prodigality, a lover of games and luxuries, actors, horses and dogs.’ With 300 manors in Domesday valued at £800 a year, he could well afford to be.⁴⁵

And yet, obscenely rich as Hugh and his ilk had become, their wealth and power paled in comparison with that of the English earls before the Conquest. At the start of Edward the Confessor’s reign, Godwine of Wessex, Leofric of Mercia and Siward of Northumbria had towered above the rest of the English nobility: collectively the land held by these three men and their families was equal in value to that of the king himself.⁴⁶ Towards the end of the Confessor’s reign, of course, the Godwinesons had expanded their power at the expense of the others, obtaining a monopoly of earldoms that made them quite unassailable. But even the house of Leofric, marginalized as they became, had remained immensely rich and powerful. Earl Ælfgar had been able to defy the king successfully on two separate occasions during the 1050s, fighting his way back from exile by raising mercenary fleets.

England’s post-Conquest earls were not nearly so mighty. Even the richest of them – say, for the sake of argument, Odo of Bayeux – could not hold a candle to the likes of Earl Ælfgar. And Odo, of course, had fallen, in part because he was perceived by his half-brother to be too tall a poppy. So too had Roger of Hereford and Ralph of East Anglia, whose rebellion seems to have been provoked by the perceived diminution of their power. By 1086 there were only two earls left in England – Roger of Montgomery and Hugh of Chester – and they, despite their gigantic wealth, were pygmies compared to their English predecessors. After the Conquest, no coalition of magnates, however large, could match the resources of the king. Even if we take the top ten magnates of Domesday England and combine their incomes, the total falls far short of William’s own – a staggering £12,600 – for the king had twice as much land as all of them put together.⁴⁷

Supremely powerful as he was, the assembly at Salisbury was set to make the king more powerful still. William had summoned England’s landholding class to participate in a great ceremony. ‘They all submitted to him,’ says the Anglo-Saxon Chronicle, ‘and became his men, and swore oaths of allegiance to him, that they would be loyal to him against all other men.’ This, it has been persuasively argued, was the climactic moment of the Domesday process, intimately connected with the true purpose of the book.

When the Chronicle says that the landowners became William’s men, it is describing an act of homage – a personal submission to a lord, in return for which the lord usually recognized the man’s right to hold particular lands. Many of William’s men, of course, had already been bound to him in this way for decades. But the security they received would have been for their ancestral estates in Normandy, not their acquisitions in England. Plenty of land in England, as we’ve seen, had been obtained by royal grant, but much had also been obtained by intimidation, encroachment and violence. Because the process of acquisition had been so protracted, chaotic and in places illegal, few if any Norman lords at the start of the Domesday process could have produced written evidence of title to all their estates.

But Domesday, once complete, provided precisely that written evidence. The Anglo-Saxon Chronicle rounds off its famous description of the survey by saying ‘all the writings were subsequently brought to him’, and this is almost certainly its laconic way of describing the presentation to William of each of the seven circuit returns at the Salisbury assembly (one of the returns, Exon Domesday, contains an entry that was emended as a result of a decision taken during the course of the king’s visit). The landowners, therefore – or at least the tenants-in-chief – did homage to William at Salisbury, and he, in return, was able to present them with Domesday’s circuit returns as a written record of their landholdings. That the twelfth-century chronicler Henry of Huntingdon, translating this passage of the Anglo-Saxon Chronicle into Latin, chose to render the word ‘writings’ as ‘charters’ is entirely apt. Together these documents were a kind of giant charter of confirmation, giving the Norman newcomers the security of title they needed to guarantee their personal conquests.⁴⁸

But if the barons gained much from the bargain, the king gained more. It is a matter of huge moment that William regarded every landowner in England as holding their land from him, either directly as a tenant-in-chief or indirectly as the tenant of an intermediate lord, for the kings of pre-Conquest England had enjoyed no such monopoly. Before 1066, as we’ve seen, lordship and landholding did not automatically go together; the king was landlord to some men but not to others, and some land was held of no lord at all. The same was also true, albeit to a lesser extent, of Normandy. Despite strenuous efforts by William himself to extend his seigneurial rights in the duchy prior to 1066, he enjoyed nothing like total superiority. It was the circumstances of the Conquest itself that meant that William was able in England to construct a new system from scratch. By treating all land as forfeit at the time of his accession, the king had been able to create an aristocracy bound in every case by strict terms of service, and similarly strict terms were imposed on the lands of the English Church.

The most visible type of service that the king expected was, of course, military service – that his tenants would supply him with the requisite number of knights when he demanded. But the relationship of lord and man gave William other rights too; rights which might be regarded as equally if not more valuable. When, for example, a tenant-in-chief died, his heir was obliged to pay the king a substantial sum of money (known as a relief) in order to enter into his inheritance. If the heir happened to be underage, as was often the case, the king took both him and the estate into his guardianship (wardship), enjoying the profits of the estate for the duration. If the heir happened to be female, the king had the right to arrange her marriage, and the same was true when a tenant-in-chief left a widow. Similar rights existed in the case of the Church: on the death of an abbot or bishop, the estates and revenues of his abbacy or bishopric were seized into the king’s hands until a replacement was appointed (which, since the king also controlled the appointment process, might be a long time).

These rights, known to historians as ‘feudal incidents’, gave the person who exercised them enormous power. Not only did they hand him sizeable sums of money on a regular basis (nothing is so reliable as a tax on death); they also allowed him to shape the inheritance patterns of his tenants, by selling the wardships or marriages of heirs to men he wished to promote. Such rights had existed on the Continent prior to the Conquest, but their value varied according to two factors: the extent to which a lord’s authority was acknowledged, and, within that, the extent of a lord’s knowledge about the size and distribution of his tenant’s estates.

This knowledge was precisely what Domesday gave to William. Thanks to the Conquest he had become the ultimate lord of every man in England; thanks to the survey he knew exactly who owned what and where it was located. The principle that all tenure began and ended with the king defines the very structure of the Domesday Book. Every county has its own chapter, and every chapter begins by listing the king’s own lands in the shire. It then proceeds to list the lands of others – bishops, earls, barons, abbots – all of whom are described as holding their lands from the king. If a landowner died or, God help him, rebelled, the king’s ministers could quickly indentify his estates, no matter how scattered they might be, and send in the sheriffs to seize them.⁴⁹

This, then, seems to have been the purpose of the Domesday Book. It was a charter of confirmation for the landowners, giving them the security of title to their estates acquired by the rough and ready processes of conquest. It was also, simultaneously, a directory for royal administrators, enabling them to see at a glance who owned what, and giving them the ability to seize and deliver lands, and to charge accordingly. It was a powerful tool – a weapon, even – for an already powerful king, allowing him to exploit what has been called ‘the most powerful royal lordship in medieval Europe’. Even the most determined critics of the concept of feudalism have to concede that England after the Conquest constitutes an exception to their arguments, coming close to matching the idealized model.⁵⁰

Of course, the Domesday Book did not exist at the time of the Salisbury ceremony. The commissioners had worked flat out to get the circuit returns completed – each surviving example is the work of many hands. Great Domesday, by contrast, is the work of a single scribe. Probably working at Winchester, this anonymous individual performed the Herculean task of collating and condensing the data from the returns, with the aim of reducing it to a single volume. Most likely he started soon after the August assembly had ended: despite recent controversy on the subject of its date, there can be no doubt that the Conqueror himself commanded the book’s creation. A twelfth-century chronicler at Worcester, adding to the account of the Anglo-Saxon Chronicle, concluded’ the king ordered that all should be written in one volume, and that volume should be placed in his treasury at Winchester’.⁵¹

It is important to emphasize that the Domesday Book, for all its importance to both kings and scholars of the future, was probably just one output of the Domesday Survey. It gave William and his successors the means to manage their aristocracy more effectively than any other rulers in Europe, and to profit handsomely from this arrangement, but it did nothing to solve the pressing financial crisis that confronted the Conqueror in 1086. William needed money immediately, in sums of a magnitude that could be raised only by levying a national tax. It seems very likely that the other function of the Domesday Survey was to gather information with a view to reforming the geld. The king’s desperate need for funds is shown by the fact that, while the Domesday Survey was in progress, royal tax collectors were also touring the country, exacting yet another punitive geld at the rate of 6 shillings per hide. Rounding off his description of the inquiry, the bishop of Hereford recalled how ‘the land was vexed with much violence arising from the collection of royal taxes’.⁵²

Violence; mass assemblies; the continuing threat of foreign invasion; mercenary troops billeted in every town and city; a nation waiting nervously on the brink: 1086 was in so many respects a terrible and portentous year, it is no wonder that people associated the king’s great survey with God’s Day of Judgement. ‘This same year was very disastrous’, concludes the Anglo-Saxon Chronicle,

and a very vexatious and anxious year throughout England, because of a pestilence among the livestock; and corn and fruits were at a standstill. It is difficult for anyone to realize what great misfortune was caused by the weather; so violent was the thunder and lightning that many were killed. Things steadily went from bad to worse for everybody. May God Almighty remedy it when it shall be His will!

And yet, by the time of the great assembly at Salisbury, the storm clouds had providentially cleared. Shortly before the ceremony on 1 August, news must have reached England that Cnut IV was dead. The Danish king had died on 10 July, murdered in church by his own rebellious nobles. His sudden passing meant that the threat of invasion was finally lifted. England was safe.⁵³

But Normandy was not. During William’s stay in England there had been some encouraging news from the duchy – notably in the case of the viscount of Le Mans, who had abandoned his stand at St Suzanne, crossed the Channel and made his peace. Developments elsewhere, however, were deeply disturbing. Cnut was dead but his ally, the count of Flanders, remained very much alive and unlikely to abandon his hostility. Robert Curthose had returned from his wanderings and was once again at the court of the king of France, ready to make trouble. Suddenly freed from his English crisis, the king moved swiftly. After taking as much additional money from his subjects as he could, says the Chronicle, he left Salisbury for the Isle of Wight, and from there sailed back to Normandy.

Meanwhile the single scribe responsible for making the Domesday Book sat down to his monumental task, condensing the information from the circuit returns to make the master volume. From the amount of work involved, experts have calculated that he must have persevered for at least a year. But as the end of his labours approached, with six circuits done and only one remaining, he suddenly put down his pen, leaving the final return unredacted. Hence nine centuries later we have Great Domesday Book in its unfinished state, and the original circuit return for the eastern counties – Little Domesday Book.⁵⁴

Why the scribe stopped is anyone’s guess: one plausible theory is that he did so because he heard the king was dead.