Brought to Book: Lordship and land in Anglo-Saxon England
by
David Roffe

I

In 757 King Cynewulf of Wessex was at Meretun visiting his mistress when he was caught unawares by the Ætheling Cyneheard, the brother of the deposed king Sigeberht, and slain. He had only brought a small party of discreet thegns with him and aroused by the disturbance they soon came up and found themselves out-numbered. Cyneheard offered them money and land on their own terms to secure their loyalty, but they refused and fought until all but one were dead. Meanwhile,

This incident, recorded in the Anglo-Saxon Chronicle, is an eloquent illustration of the nature and role of lordship in early Anglo-Saxon society. As a relationship of clientage, the bond between lord and man was articulated through the medium of treasure and land, but it was an absolute which ultimately superceded all considerations of wealth, family, or self. As the word hlaford, 'lord', with the root meaning of 'bread giver', indicates, it was the very stuff of existence: in return for fidelity and service to his lord, a man received protection and sustenance, and even death did not dissolve the mutual obligations which were thereby created, for each pursued the vendetta on the other's behalf.

Lordship remained a datum of society throughout the Old English period, but in the course of time the relationship ceased to be an absolute; by the tenth century it was paralleled by other obligations which theoretically limited the authority of the lord. From the time of the earliest law codes kings had made more or less hopeful claims to some degree of universal competence. In reality, however, as the heroic poetry of the time illustrates, lordship superceded loyalty to the crown in the early period; if a lord were banished for rebellion against the king, however reprehensible his behaviour his men were disgraced and held in scorn if they did not leave their estates and follow him into exile. By contrast, when Edward the Confessor exiled Earl Godwine in 1051, far from expecting the earl's thegns to join him, the king required them to renounce their loyalty to their lord and pledge themselves to him. By then there was a presumption that the king's lordship transcended all other obligations, and they acceded, albeit temporarily, to his demand.

The development of royal power was a protracted and seemingly inexorable process. Kings had always had the right to dues and tributes from their extensive estates, but services were only demanded from those which were alienated from the royal fisc in the mid eighth century in Mercia. By the reign of Offa such were universal throughout the kingdom and its satellites and had been defined as fyrd service, bridge building, and borough fortification. The reservation of these 'common burdens' was quickly adopted elsewhere and development became rapid. In Wessex the obligations of both royal and bookland estates were forged into an integrated system of government in the reign of Alfred with the creation of a standing army and a network of garrisoned burhs. By the time Mercia had been absorbed and the Danelaw conquered under Edward the Elder, boroughs had been established throughout the country, and the new royal centres assumed wider functions such as the minting of coin, the control of trade, and some policing duties within the territories assigned to them. With the reconquest of the five boroughs of the Northern Danelaw in 942 Edmund could demand an oath of loyalty from all his subjects, and by the early years of the reign of Edgar a system of sureties and tithings organized in a proto-shrieval system had come into existence. Henceforward, the king's peace was promulgated through a network of hundred and shire courts and maintained by a territorially-based system of policing and seigneurial sureties.

As such, the mechanisms of royal power are well understood, and their manifest success has brought into question the continuing importance of the relationship of lord and man. Domesday Book seemingly demonstrates its frailty. Numerous references are found to free men who had the right to 'go with their lands to whatever lord they wished', and many historians have followed Maitland and Round in asserting that lordship had became a weak bond by the eleventh century. However, this is at variance with other sources. Tenth-century law codes show that an individual did indeed have the freedom to commend himself to a lord of his choice, but once the bond was forged it was permanent: it could only be dissolved by the death of one of the parties, the consent of the lord, or the gross abuse of the relationship on either side. By the eleventh century, commendation had probably become territorialized in dependent tenures and an individual may therefore have had two or more lords by virtue of the lands he held. Free men, though, were still generally commended to a single superior and they were apparently subject to his authority. To take but one example, before 1066 both Edric and Stanwin, two Suffolk free men, were only at liberty to return to their former lord Edric of Laxfield, who had been outlawed, with the consent of their existing lords. Almost all similar references indicate that freedom of action and movement was circumscribed. It was the lord's right to commendation that was weak rather than his relationship with his man.

Lordship remained a bond which could never be lightly and unilaterally discarded. It was, furthermore, a relationship which was preceived as a necessary and desirable characteristic of a well-ordered society, for by 1066 every free man was required to have a lord. It is thus clear that the growth of royal government was not simply a suppression of lordship. Its dynamics were more complex. The twelfth century saw a similar expansion in central government, but this cannot be explained simply in terms of of the introduction of new offices like that of the coroner and the novel legal procedures of writ and assize. Species of naked royal will was always resisted, but by and large the new instruments of government were accepted with alacrity by the free man only because they provided a quick and efficient means of establishing right independently of his lord's court. Royal power increased with the active connivance of the ruled. Kingly patronage of the church in the Anglo-Saxon period has long been seen as a relationship of mutual and complementary interest of this type. An examination of the interaction of bookland, lordship, and land suggests that the phenomenon was far more widespread. The relationship of lord and man was not monolithic but varied in relation to the different types of land on which it devolved; it was the tension between landlordship, liege lordship, and commendation or surety lordship which constituted the mainspring of developing authority in the Old English state.

II

The nature of bookland is a matter of debate. The landboc, the diploma in which the tenure was granted, was copied from contemporary Italian property dees and emerged in the mid to late seventh century. The distinctive feature of the instrument was that it created perpetual right, and the device appears to have been initially introduced to endow monasteries; it seems that it conferred somewhat more secure title than folkland which in some way was its antithesis. So much is generally agreed. The implications of bookland are more controversial. Stenton, following Maitland, saw it as essentially an immunity which was created by the conveyance of royal farm; it was therefore to be contrasted with folkland which, when not in the hands of the king was only temporarily held in laen. Abels has countered this arguemnt by demonstrating that bookland, at least in the late Anglo-Saxon period, entailed reciprocal duties of service to the king and continued tenure depended upon successful performance. He, however, while recognising that it was largely concerned with fiscal land, has agreed with most commentators that bookland generally conferred full rights to land and was identical with it. Vinogradoff was the first to articulate the problem. He argued that land was normally held by the kin and was folkland by virtue of the folkright which governed its disposal; booking broke these bonds and gave the individual complete freedom of tenure. John, by contrast, has maintained that the land of the family was always precariously held, laenland which was in the gift of the king, and the great attraction of bookland was that it permanently brought land into the family. More recently Wormald has taken issue with all three views and argued that, as an acquired tenement, bookland created unrestricted donation and therefore allowed the individual free disposition outside of the constraints imposed by the family and the crown. However, all have assumed that, whatever the nature and origins of bookright, it was synonymous with lordship. To date the analysis has largely centred on a study of the law codes and survivng pre-Conquest charters. The evidence they provide, however, is limited and ambiguous. Here the starting point is the Domesday survey, for, despite the problems of identifying bookland, it provides a mass of coherent data which have been largely ignored.

Although a survey of estates, Domesday Book reveals remarkably little about the status of land. The term manerium, manor, with its divisions of berewick and sokeland, is ubiquitous but is probably not a unitary concept. In the account of the northern counties of Circuit VI (Derbyshire, Huntingdonshire, Lincolnshire, Nottinghamshire, Rutland, and Yorkshire) its correlates changed in the course of compilation, and it seems to have referred to a number of different types of estate in the rest of the text; it embraced bookland, laenland, thegnages, sokeland, and a host of different relationships between lordship and tenancy. Nevertheless, not all holdings were distinguished as manors or parts of manors; a high proportion throughout the country were merely designated as terra, land. Some of these estates are specifically associated with an overlord or are said to belong to a particular place, but such references are at best sporadic and tenurial status is usually not explicit. The record of tenure, by contrast, is more comprehensive. For almost every parcel of land in England south of the Tees Domesday Book records the name of the individual, or, in the case of groups of free men or sokemen, the number, who held it in 1066 or before. In some circuits, like nos IV (Leicestershire, Warwickshire, Oxfordshire, and Northamptonshire) and VI, nothing more than the name is given, but more usually the tenants are said to hold freely (libere), equally (pariter), in allod (in allodio), or, more occasionally, to have under (sub) or from (de) another lord. However, it is only in the Eastern Counties that a relatively full account of lordship is recorded. It is within this area that the relationship between lordship and land is the most easily perceived.

Throughout the area the lords of earls, bishops, and named predecessors are rarely given, except in Essex where they are occasionally called king's thegns (taini regis). Otherwise relationships of this kind are explicit. In Circuit VII (Norfolk, Suffolk, and Essex) named thegns, free men, and, less often, sokemen are said to be commended (commendatus) to a lord or to hold under (tenuit sub) or from (tenuit de) another. More abstractly. commendation (commendatio) is said to be reserved to an individual or manorial centre. It is in Circuit III (Cambridgeshire, Hertfordshire, Buckinghamshire, Bedfordshire, and Middlesex), however, that the relationship is most completely articulated. Almost every individual, whether a named tenant or sokemen (the local equivalent of the free man) is simply said to be a man of a certain lord, and his rights are consistently delimited. Some sokemen were not able to alienate their land without the lord's permission, and their lands were incorporated into a manor. Most ecclesiastical land was apparently of this type, and the Ely records show that the dependent estates were held in thegnage. The form was by no means confined to the church - Edeva the Fair held the manor of Cherry Hinton (Cambs) with thegnages in Fulbourn, Teversham, and probably many other vills in the vicinity of the estate centre - and such lands were considered to be part of the lord's demesne. The sokeman's tenure was essentially ministerial; he may well have had some presumptive right to inheritance, but his continued seisin was at the will of his lord.

By far the majority of sokemen and free men, though, seem to have had freedom of alienation, for, although men of a superior, they were free to grant or sell their land without the permission of their lord or were free to go with it to another lord. This concept was apparently foreign to the Norman tenants-in-chief, for land was frequently claimed on the basis of 'mere commendation', but as the invasiones section of Circuit VII reveals, the fact was not considered to constitute sound title. The act of commendation did not usually confer rights over the man's estates. Lordship, it seems, as articulated by commendation, was generally independent of land tenure. Free men and sokemen of this type have usually been characterised as either königsfreien, that is free men who, holding allodial land, were only responsible to the king, or booklords who were entirely free. A few were undoubtedly taini regis of the more lowly ministerial kind, who rendered their soke to a royal manor or the hundred. A sokeman in Wickham (Herts), for example, was a man of Godith, but was in the king's soke. To all appearances such were otherwise free of tenurial constraints. However, neither their land nor that of the majority of free men was entirely unencumbered. Throughout the area estates were held by two or more tenants, and parage associated with partible inheritance was probably the norm. A free man called Godric, for example, held land in Colchester and Greenstead (Essex), and on his death the estate was divided between his sons. Land was entailed within the family and was held by socage of one sort or another.

Over many soke is specifically said to be reserved in Domesday Book. Soca in this context can, and often does, refer to simple jurisdiction which commanded suit of court, forfeitures, and sundry minor dues but did not otherwise confer substantial rights over land. More usually the term embraced more extensive bonds; the Inquisitio Eliensis and other Ely sources show that it could also comprise labour dues, customs, services, and farm. It is these renders which are usually noted. Three thegns held East Burnham (Bucks), for example, and could sell, but they paid 5 oras (6s 8d) to the church of Staines in customs. Explicit references of this kind are relatively rare since dues are only recorded in exceptional circumstances, either when there was a dispute over tenure in 1086 or two tenants-in-chief shared the renders. But their value to the overlord is recorded in every entry, for the valuit figures are the annual render that the free men made, and it is therefore clear that soke qua customs was generally paid.

Unlike commendation, the right to custom sanctioned the tenure of what Domesday Book calls land, for, short of a royal writ to the contrary, it invariably conferred title on tenants-in-chief in 1086. The individuals who were entitled to it TRE are not explicitly noticed, but they can be identified since the record of commendation would appear to be directly related to soke structure. Named predecessors and king's thegns who commonly gave title to a single Norman lord, such as Asgar the Staller and Edeva the fair whose lands passed en masse to Geoffrey de Mandeville and Count Alan of Brittany respectively, are hardly ever said to be a man of a lord and no comment is ever made on their right to alienation. It is inconceivable that such lords were not generally free to dispose of their land, and the usage is clearly no accident of diplomatic, for, apart from the exceptional cases of earls and the holders of frankalmoign estates and laenland, all individuals who are treated in this way were succeeded by a single tenant-in-chief. They were evidently those who were entitled to customs and could therefore confer title in 1086. The record of commendation and the right of alienation in other entries, then, must indicate that the free men were not predecessors in their own right of this kind and signals that the renders from their lands were not due to the lord to whom they were commended, and thereby his post-Conquest successor, but were the right of the tenant-in-chief through his predecessor's tenure of their soke qua custom. To a Norman it was probably natural that the land of a vassal should be subject to his lord. This device, however, was a reminder that lordship did not usually imply such a relationship in East Anglia and established the legal right of the tenant in 1086 at the same time as limiting the claims of any possible rival. Thus, the statement that 'X held under Y; he could withdraw' indicates a free man who rendered soke, while 'Y held' alone identifies the recipient as soke lord.

If Domesday Book obliquely notices the holders of custom and those who rendered it in this way, there is less indication of the organization and extent of such sokes in East Anglia. As might be expected from the nature of the dues, there is some evidence that they were territorialized. For example, fifteen sokemen, who were free to sell their land, in six vills in Bedfordshire belonged to the manor of Cople. It was probably only in Bury's ancient liberty in Suffolk that soke dues of this kind, and then only those from its own lands, were considered to be hundredal renders. Even there, however, services were performed at specific demense centres. Ixworth (Suff), for example, belonged to the manor of Pakenham, but the fact is only known from a near contemporary note on estate mangement.

In reality estate structure throughout East Anglia must have been very similar to the highly developed system of manors and sokes of the Northern Danelaw. Domesday Book is explicit in its description of the East Midlands and Yorkshire. Land, held in parage and subject to the custom of partible inheritance and the closely related practice of ultimo-geniture, is specifically called sokeland and is associated with a manorial centre; some tenements were in the tenure of undifferentiated sokemen, while others were thegnland, laenland, or free tenements which were themselves often called manors and were held by named thegns. However, the Descriptio Terrarum of Peterborough Abbey, a Domesday satellite which emanates from an early stage in the Domesday inquiry, describes the house's Lincolnshire lands in much the same terms as the those of the East Anglian circuits, and it would seem that in the compilation of the Exchequer text the articles of the inquiry were simply interpreted more ambitiously than elsewhere in the country. The apparent differences between the Southern and Northern Danelaw are largely Domesday artefacts occasioned by differences in procedure and composition. Outside of Circuit VI, the details of estate structure were omitted because they were perceived as largely irrelevant.

The phenomenon of sokes, however, is by no means solely a characteristic of the Danelaw. Socage-like tenures are found throughout the country. The form is most apparent in Kent where the custom of the shire was gavelkind, that is partible inheritance, but in the West and South-West the tenure is reflected in the terminology of Domesday Book, for land is frequently said to be held pariter, that is equally or in parage. As in East Anglia the dues that the tenure implied are probably contrasted with lordship. Commendation is only sporadically recorded in most areas. Nevertheless, it is clear that the information was collected, for it is regularly found in the Liber Exoniensis, part of the Circuit II (Wiltshire, Dorset, Somerset, Devon, and Cornwall) return or an earlier recension thereof. However, it was omitted from the Exchequer text and replaced with a statement that land was held libere, that is freely. In allodio, 'in freehold', is used in Circuits I (Kent, Sussex, Surrey, Berkshire, and Hampshire) and V (Gloucester, Herefordshire, Shropshire, Cheshire, and Staffordshire) in a similar fashion, and, like commendatio, the two terms are probably employed to draw the distinction between the respective rights of the commendation and soke lords.

The creation of post-Conquest castleries in the south-east and marcher lordships on the Welsh border made the record of pre-Conquest rights largely irrelevant, and the existence of the territorial sokes to which the tenure gave rise is therefore often, if not always, obscured. But they are more apparent in the occasional reference elsewhere in Domesday Book. The inclusion of freely-held land in a manor is sometimes explicit. Lydeard and Leigh (Somerset), for example, had always rendered customs and services to the bishop of Winchester's manor of Taunton, but it was held TRE by a thegn who was free to go with his land to whatever lord he wished, and the bishop's tenure of the estate was only warranted by a specific grant of the king after the Conquest. More obliquely, notices of capital manors in the south and south-west, such as Northwick (Somerset), and Deerhurst (Glouc), hint at the pre-eminence of some estate centres, while the repeated claim in the terre occupate sections of the Liber Exoniensis that land which was held freely did not belong to a predecessor's 'honour' indicates that the form was general. Similar patterns are found in the West Midlands and the North-West. Geoffrey de la Guerche's estates in Warwickshire, for example, were constituted as a single unit, although individual manors were held by tenants, and the whole area between the Ribble and Mersey was divided into six northern shires which consisted of groups of manors largely held by drengs. Sokes would appear to have been all but universal.

The nature of the right they conferred is no where made explicit in Domesday Book, but its main characteristics are apparent. The evidence is clearest in the account of the Northern Danelaw. There the soke lord's interest, sharply contrasted with simple soca qua jurisdiction, was not merely a random and ad hoc collection of dues. Although the free man could sell or otherwise dispose of his estate, neither grantor nor grantee could unilaterally withdraw the custom, for by virtue of the possession of a capital manor it was vested in an individual who was endowed with sake and soke. This liberty, the holders of which are systematically recorded in the Yorkshire, Lincolnshire, Nottinghamshire, and Derbyshire folios, entitled the holder to the receipt of customs and sundry personal services and, more importantly, the king's two pennies, that is the wites and the like which accrued from the processes of justice; apart from suit to the hundred and county courts, the land was held with full rights and forfeitures were only made to the king and the earl. Elsewhere the holders of sake and soke are not systematically listed, but nevertheless, the term, with but few exceptions, generally expressed the same rights. Askell, for example, who gave title to Earl Hugh of Chester, had sake and soke over nine freely held manors in Northamptonshire, and the liberty is regularly associated with diplomatic forms in the Staffordshire folios that indicate the receipt of custom. Sokes throughout the country were clearly of this kind, and the lords who held them were evidently of that class of king's thegn 'who has his soke' of the Laws of Cnut and the late eleventh- or early twelfth-century Instituta Cnuti, and the group of estates over which his soke extended was bookland.

Although distinctive diplomatic forms identify booklords in many circuits, it may never be possible to reconstruct the exact limits of their bookright. Nevertheless, the general characteristics of the tenure on the eve of the Conquest can be perceived. First and foremost, there was a demesne. This was sometimes extensive, especially in ecclesiastical estates, encompassing a number of berewicks and sokelands which were tenanted by peasants who were not free to go with their land. The hall where services were rendered symbolized the lord's right, and in some areas the term manerium or mansio was confined to it. However, throughout much of England by far the greater portion of bookland was soke. It likewise rendered customs and services at the hall, but it was freely held by liberi homines or sokemen as inalienable family or patrimonial land. Neither aspect of bookright entitled the lord to the commendation of his tenants; those who were unfree to leave with their land were generally his men, but those who held in laenage were not necessarily so, while the sokemen were free to seek lords where they would. Bookright, land tenure, and lordship, then, were not intrinsically coextensive.

III

The identification of these characteristics of the tenure in Domesday facilitates an examination of bookland in the Anglo-Saxon period. All three of its salient features of demesne, soke, and freedom of commendation are evidenced in tenth- and eleventh-centuries sources. Pre-Conquest charters frequently emphasize the unequivocal rights in demesne land that were conveyed by booking; estates were granted with all appurtenances in fields, woods, pastures and the like, and the boundary clauses define the exact limits of property and proprietorial rights. Free sokeland is less easily discernible. It is occasionally noticed in charters, and the different nature of the land, and the booklord's interest in it, is often apparent, for it is not always included in the boundary clause. Southwell (Notts), for example, which was granted by King Eadwy to the archbishop of York in 956, received soke dues from twelve of the surrounding vills, but the bounds only define the estate demesne. More usually sokeright is not mentioned, and its silent inclusion in grants is only apparent in those areas in which Domesday provides greater details of land tenure. The book of a certain Thurketel, for example, granted Culford (Suff), which the donor described as his own estate, to Bury along with Wordwell and Ixworth in 1040-2, but it is only Domesday Book that reveals that the first was held in demesne in 1066 and the other two were held by free men, one group of whom are specifically said to have rendered their service in Culford.

Probably reflecting their origin in late Roman property deeds and transactions, books only explicitly convey the tangible rights of the tenure. The sokeman held freely and acquitted the land of the geld levied on it; the booklord therefore only received dues and these were engrossed in the issues of the demesne. Pre-Conquest law codes are more revealing: the dichotomy between demesne and soke, on the one hand, and bookright and lordship, on the other, is explicit. From the reign of Athelstan, if not before, fines and forfeitures incurred in tbe breach of the peace were reserved to soke lords who therefore presumably enjoyed the other issues of sokeland and the free men who held it. Inalienable and concrete, such bookright is consistently contrasted with the rights conferred by commendation lordship. Thus, in II Edgar 3.1 it is stated that if a man did not pay his tithes, regardless of whether he was a king's man or a thegn's, his forfietures were to be divided between the bishop and the landlord, while in I Æthelred 1.5,9 it is provided that if a man was found guilty his wergeld went to his lord, but if he escaped it went to the lord, that is landlord, who was entitled to his fines. From the reign of Athelstan hlaford, lord, usually refers to the commendation lord and landhlaford or landrica, landlord, to the holder of bookland.

The admixture of demesne and sokeland was characteristic of bookland throughout the Anglo-Saxon period. Some compact ecclesiastical estates appear to have survived from an early time as large demesnes. Fladbury (Worcs), for example, which was granted to the church of Worcester in the late seventh century, still embraced the same area in the eleventh, and Domesday suggests that it was a large manor which, although leased out, had always been held in the hands of the bishop. However, some degree of free tenure must be suspected in this type of estate. Early charters often hint at the fact. Selsey (Sussex) and its dependent vills were apparently granted with free men as well as land and slaves by King Æthelwalh to St Wilfred in the seventh century, and sources such as Æthelwulf's first Decimation of 844 show that typical socage services like riding, farm, and witeræde were due from them. Furthermore, it is clear that land was entailed within the families of such free men at this time. The explicit assessment of estates in terms of their land and its close connection with the hide, invariably glossed as the land of one family, suggest that laymen had already established such rights to the soil by the seventh century, and Ine's laws indicate that the ceorl, and possibly the gesith, had the expectation of inheritance in common with his siblings.

Since bookland was in the gift of the king, in origin the demesne was probably always royal land which, populated and worked by geburs, slaves, and sundry dependents, was the full right of the monarch. Other types of land, however, may have become assimilated to this central core in the course of time. Bookright conferred the forfeitures of the soke and rights to land could revert to the booklord on the outlawry or otherwise of tenants. More usually in these circumstances, however, land probably remained in soke. Wiglac and Godric forfeited two parcels of sokeland in Scremby (Lincs) shortly before 1066, but it was not incorporated into the demesne of Ulf Fenisc, the booklord, at Bardney but remained in the tenure of a number of sokemen. It is not clear whether they were free to grant their land, but, as in Nottinghamshire and Derbyshire, the family of the tenant may well have retained residual rights in forfeited land. By custom and established interests, the extent of the demesne probably remained fairly constant. Conversely, it is possible that some sokeland was patrimonial land of the booklord; he is ocasionally said to hold freely in the soke of his own manor in Domesday Book. Most, however, was entailed within free men's families, and the tangible benefits of the tenure were the transfer of the customs or farm of this land from the king to an indivdual.

Demesne and sokeland were constants. By contrast, the relationship between the crown, bookright and lordship was more variable. As in the tenth and eleventh centuries, a lord uneqivocably assumed the protection of his dependent tenants and slaves in the early Anglo-Saxon period, but the free man had resort to his family. It is this characteristic which distinguished him from Cynewulf's thegns whose clientage irrevocably separated them from their own blood. It was his kin which took primary responsibility in all matters that devolved upon soke and the forfeiture of land, that is causes that were emended by the payment of wites; it demanded and paid the wergeld as appropriate and generally stood surety for the performance of whatever obligations were incurred. The lord was not without a role: he aided the family in the pursuit of its interests, whether it be a feud or a legal matter, in return for a portion of the blood price, but his right in this context amounted to little more than soke. In essence the lord was a landhlaford and an individual's allegiance was largely determined by soke nexus.

In origin the relationship was one between family and king and was probably the essence of folkland. With the advent of booking the dynamic changed as lordship of this kind became vested in the booklord. Domesday Book may well misrepresent the incidence of freedom in its account of ancient estates, for most for which pre-Conquest documentation survives were held by churches, and the survey often exaggerates their interests. Nevertheless, such estates were signficantly larger and more demesnal than others, and many seem to have had their origin in royal shires, sokes, or multiple estates. As complete units of administration, the churches that received them also succeeded to the courts through which they were governed: manors of this type regularly had extensive liberties attached to them in later centuries which approximated to hundredal jurisdiction, a relationship which has been termed manerium cum hundredo. Until the late eighth century no services were rendered in return, and it would seem that the free man therefore had few remedies against seigneurial aggression since the booklord's court was the only forum in which disputes were resolved. His independence turned on the strength of his unsupported kin, and the high proportion of unfree tenants and liberi homines who were not free to go with their lands, along with the reservation of commendation, in such estates in the eleventh century may well reflect the outcome of the conflict of interests. Contrasted with folkland, bookland was effectively an immunity at the early period.

The tenure, then, was an absolute loss to the king's fisc and his fund of service, and its institution severed the direct link with free kins. Nevertheless, the advantages of endowing churches in this way outweighed the cost. The most frequently expressed benefit was that of prayer, for every ruler needed the spiritual support that the church provided. However, the alienation of rights to a monastery presented more tangible rewards, for it created a strong royal party with a vested interest in promoting the rights of the crown against local interests; religious corporations lacked the support of local lordship and kin and relied upon the crown for their protection. Throughout the Anglo-Saxon period bookland probably fulfilled this function in both ecclesiastical and secular contexts. Bookright countered the demands of the kin over whose land the immunity was exercised by creating a rival interest.

The church was not always as loyal as a king might deem desirable. but by and large bookland was probably an effective instrument of royal power in this way. The careful balance of interests on which it depended, however, was shattered by the creation of family monasteries as laymen acquired the liberty under the guise of religious endowment. The attractions of bookland were great. The early law codes are silent as to the patrimonial interests of noble families, but it seems likely that they had always held free land like ordinary ceorls. The compounding of personal names with various habitative elements in place-names from an early period may well point to the permanent settlement of gesiths, and their land is referred to in law codes from the late seventh century. But its extent was limited by the rights of other families and kins, and kings had probably always supplemented their men's meagre resources by granting out farm. The interest conferred was no different in kind from that used to endow the church, but it was not alienated permanently, for it was a laen to which the family had no right; a lord had to earn his father's tenement as a reward for faithful service to the king. The creation of a monastery, then, divorced tenure from service and introduced real hereditability. The effect of the deceit was subversive. In 734 Bede wrote to Ecgberht, bishop of York, castigating the pseudo-religious foundations for stripping the realm of the resources to defend itself. This was no more than genuine endowment did, but the real danger probably lay in the union of immunity with kin, for it led to the territorialization of lordship to the detriment of royal power and the independence of the free man.

How widespread this abuse was we cannot say, but a remedy was only found once the common burdens were reserved in the reign of Offa. Although immunities were occasionally granted to favoured communities, thereafter bookland generally took on a new aspect. The context of the change was probably the threat of Danish raids and the need to raise forces to counter them. From the outset it was evidently seen as an arrangement which was of mutual benefit; in imposing the service on ecclesiastical lands in Kent in 792, Offa declared that he requested the churches consent to the measure so that his grant of liberty to them should remain more stable. Henceforward, the booklord was in the king's soke and became his thegn; the constraint on the grant of the land to laymen was therefore removed, and for the first time he could legitimately enjoy ius perpetuum.

The most immediate outcome of the measure was to increase greatly the king's capacity for patronage without permanently jeopardizing his military resources. From the ninth century vast tracts of land were booked out. But the device also opened up opportunities for the extension of royal power. Large compact estates were less often granted, and increasingly there was a tendancy to divide dues. The pattern is most apparent in the North where royal sokes were divided element by element rather than by area. The soke of Sleaford (Lincs), for example, seems to have formerly encompassed the whole of the wapentake of Aswardhurn, but was divided into a number of separate bookland estates probably sometime in the tenth century. By the time of Domesday Book four of the subdivisions survived as manors with almost identical structures to indicate their origin in a single whole. But throughout the country charters attest the grant of single vills and parts of vills. Such dispersed patterns of bookland tenure broke up large concentrations of land, and, by linking continued tenure to loyalty to the king, weakened local power centres, be it the ealdorman or a regional magnate. Although the constitution of bookland estates in 1066 is not always immediately apparent, post-Conquest patterns of tenure frequently perpetuated the structures and reveal great complexes of interlocking interests throughout the country.

An insistence on the common burdens, however, did not just introduce a remote balance of power. It also heralded a reservation of jurisdiction and service which distinguishes it from earlier practice. As before, the landhlaford continued to enjoy the forfeitures which accrued from his men and estates and received a share of the fines which they paid for various misdemeanours and crimes, but the context of the relationship apparently changed. The laws are silent as to the circumstances in which soke dues were received, but from the sixth code of Athelstan the booklord was only entitled to a share; the victim and the hundred were also interested parties, and, since the king's reeve was apparently always associated with these transactions, it would seem that the forum was a hundred or shire court rather than that of the lord. As is clear from Domesday Book and later records, simple sake and soke never conferred hundred jurisdiction on its own, and it would appear that its corrollary of bookland merely entitled its holder to a share of farm, fines, and forfeitures in a central royal court.

In the course of time the specifically tributary dues, such as food rents and labour services, became territorialized and manorial courts developed to administer them. However, the free sokemen, albeit under the supervision of the landhlaford, was still personally burdened with royal services. It was he rather than his lord who paid fyrdwite, a fine for non-performance of army service which was normally reserved to the king, and Domesday Book provides numerous examples of him discharging his duties without reference to the booklord. Likewise, he appears to have always been vested with some responsibility for the maintenance of the peace. In 844 Æthelwulf released one tenth of the land in Wessex from riding (witeráæde), but the rest, whether bookland or folkland, appears to have continued to perform the service, while Domesday Book shows that free men in East Anglia were still burdened with escort duties and cartage for the sheriff as late as 1086 regardless of the bookright exercised over their land. The widespread incidence of the five-hide unit and the practice of counties like Berkshire suggest that the distribution and execution of such services were probably based upon the amount of land held, and it is clear that they were rendered in the hundred and the shire. Except in special circumstances, booking did not sever the direct relationship between king and free man, for royal justice and the mechanism of peace-keeping were usually reserved. Bookright was evidently a franchise which conferred the issues of the king's majesty without its apparatus.

Since the booklord had neither land nor the control of local government at his disposal, it was perforce bookright itself which was the main medium of patronage. Primarily, the franchise enabled the lord to accumulate resources through the receipt of tribute which could be redirected towards clients; throughout much of the period the transfer of treasure remained a fundamental nexus between lord and man. More directly, the right itself was increasingly used. It was probably only in very exceptional circumstances that it was granted outright, for a book forged a direct relationship between a client and the king and thereby effectively isolated the lord. Rather bookland was given for life or a term of lives in return for service. The use of laenage by the great churches in this way is well-recognised. Its more general role is less apparent. By the beginning of the eleventh century, however, it was probably the main means of maintaining a comitatus, for II Cnut 77 stipulates that a man who deserted his lord was to forfeit all the possession and land that he had given him and to lose his life. Domesday Book suggests that some, probably a majority, of the thegns who held manors in 1066 only had a life interest in their estates. The relationship, which can be termed one of liege lordship, was much closer than that between the lord and the free man on his bookland and was probably the main instrument of seigneurial power.

If the grant of bookland by the king conferred the means to create a potent type of liege lordship which was independent of land and the services due from it, the booklord was obliged to execute the common burdens in return. As a king's thegn, the service due was personal. The Worcester Domesday makes it clear that he was individually summoned to the fyrd and was responsible for the execution of the service due. The implications are explicit in the Domesday account of Tewksbury (Gloucs); Brictric son of Algar alone aquitted the manor, consisting of some 150 demesne hides, even though numerous thegns held the land from him. Service of this kind seems to have been specific to the bookland tenement and therefore additional to the service due from the sokemen over whom it was exercised. Thus, 2Å hides in Lydeard and Leigh (Somerset) rendered customs and service to the bishop of Winchester's manor of Taunton TRE, but the thegn who held them was free to go to whatever lord he wished and did not render army service with the bishop's men, that is those who were tenants of the bishop's demesne. Whether the amount rendered was based on the hidation of the bookland is unclear, but frequent reductions in assessment, as in Tewkesbury, suggest that the number of men who had to be found was ultimately a matter of negotiation between the king and his thegn. However, the military units which were thereby formed appear to have had a distinct identity. The bishop of Worcester's scattered booklands in Worcestershire were grouped together as a ship soke and the bishop was the archeductor of the thegns who served in return for their laenlands. Nevertheless, they seem to have been coordinated under the ealdorman or earl at the villa regis, and later in the borough and shire. The record of sake and soke by shire in Domesday Book suggests that the dues from bookland were organized on a county basis, and pre-Conquest evidence suggests that all matters of bookright were adjudicated in local forums.

The characteristics of bookland are thus clear in their general outline. The tenure appears to have conveyed a farm consisting of various dues and tributes along with the centre at which they were rendered. These were essentially public dues and the term folkland was probably applied to them in the earlier period when in the hands of the king or held by a thegn for a life or term of lives in laenage. Booking merely transferred them to an individual. Paradoxically, bookright created landlordship, but it was a fiscal tenement and did not confer full rights over land. As such, apart from a small demesne bookland neither brought land into a family nor took it out, but conferred regalian rights over freely-held sokeland in perpetuity with full freedom of alienation of the dues. An acquired tenement, it was thus free of the family entail and therefore contrasted with patrimonial land. It was this characteristic which made it a powerful resource for the creation and maintenance of a loyal and dependent clientage. Nevertheless, since it was a right over free land, it did not in itself create a bond of liege lordship. Before the common burdens were reserved it was an immunity and seigneurial rights were probably asserted. Subsequently, however, it became a dependent tenure and the free man's relationship with the king through communal courts remained unchanged.

IV

The reservation of the common burdens, and the control of local government to which it gave rise, effectively prevented the territorialization of lordship, and it is in the context of the essential independence of the institution and land that the emergence of an integrated system of local government in the ninth and tenth centuries must be understood. The tenure of folkland, sokeland, and latterly bookland had always suggested service at local centres, and the introduction of a burghal system merely introduced new nexus. The more revolutionary change was the development of a mechanism to enforce the performance of services and the maintenance of the peace. The effectiveness of local government was severely limited by the reluctance of overweening and recalcitrant families and lords to act as sureties and keep the peace for their respective members and clients. Athelstan stipulated that posses should be sent out against both noble and common kins which would not bring their members to justice, and lords were enjoined to surrender delinquents in their charge. Peace guilds and the like, however, were of limited efficacy in this context, and the problem was only resolved when lordship took over surety from the kin. Whether the process was initiated by the crown is unclear, but it is evident that it was in the mutual interest of lord and man and was probably actively espoused by both. The former was invested with the warrantee's wergeld and assumed an authority that bookright denied him, while, equally, in the unsettled circumstances of the time, the latter acquired powerful protection beyond the means of his own kin. The relationship, however, was different from that of either landlord and man or liege lord and man, for it did not devolve on the transfer of resources or land. Lacking any element of personal or military service, surety lordship was essentially confined to frankpledge; patterns of lordship in Domesday Book suggest that free men often consciously avoided potentially dangerous concentrations of interests that could lead to more dependent tenures.

As a duty to maintain pledges and tithings, surety lordship added what was later called leet jurisdiction to the lord's private court, and its forum was inevitably local government. By the second half of the century every man was required to have a lord who would vouch for his good behaviour, and the relationship which was thereby formed was firmly integrated into the structure of local government. By the terms of the Hundred Ordinance and IV Edgar a system of territorialized tithngs had been set up, and henceforward, under the supervision of the lord the vill became the context of peace keeping within the jurisdiction of the hundred. Lordship had become a public matter; it was declared in the hundred or shire court and could not be repudiated without raising suspicion of criminality. Lordship of this kind was rooted in royal power and the mutual interests of both lord and free man thus became grounded in an acceptance of the king's peace.

The bond is represented by the hlaford of the tenth- and eleventh-century law codes who is contrasted with the booklord or landhlaford, and it is to this relationship that the commendatio of Domesday Book and related formulas refer. The network of alliances that is apparent in the text is therefore of a different order to the bond between King Cynewulf and his thegns. Liege lordship is more difficult to identify in the Domesday text - laenland is only noticed where it impinged on questions of title in 1086. However, it was still an important element in eleventh-century society but was not such an untramelled relationship as formerly. In the early eleventh century King Cnut had found it necessary to enjoin lords to control their men, but a client was no longer solely dependent on his lord. Earl Godwin's men were probably bound by mutual obligations which devolved upon the exchange of resources such as laenland. However, they were equally as likely to hold bookland on their own behalf, and their compliance with King Edward's demands in 1051 must have been as much prompted by the fear of forfeiting it, as stipulated in II Cnut 77.1, as loyalty to the crown. Clients' interests were not necessarily identical with those of his lord, and through the nexus of warranty the divergence brought the relationship into local government. By the late tenth century lordship over both free man and client attests less to strong seigneurial power than effective royal government.

In consequence it was the structure of local government which was perceived as the means to power and influence. It was from the late tenth century that the crown employed public courts as an instrument of patronage. Unlike the earlier grants in which soke and land were conterminous, the rights which were conferred were generally exercised over the land of a number of lords. The great liberties of the East Anglian monasteries, for example, encompassed not only ecclesiastical land, but also that of many king's thegns as well as more lowly lords. The courts to which these estates paid suit were still public and were burdened with all the duties of local government. Nevertheless, they were valued not only as a financial asset in their own right, but, more importantly, as local political centres to which the free men of the area were bound. The fact seems to have been generally recognized, for seigneurial discontent and subversion were rarely directed against local government per se. The North rose against Earl Tosti in 1065 because of his excessive use of the machinery of administration rather than its form which seems to have been long accepted, and Ealdorman Eadric Streona probably marshalled local support in the west and north-west against the king in 1015 through the system. By and large local government maintained the peace and even under the stress of irresistible outside pressure in the late tenth and early eleventh century continued to function and raise enormous sums of money for the defence of the realm. It was thus a natural vehicle for seigneurial ambition.

The courts of hundred and shire had assumed primary political significance, and the crown enhanced the functions of the shire reeve to counter the influence of ealdormen, bishops, and king's thegns who presided over them. Territorialized lordship was less of a problem. Seigneurial aggression on, or absorption of, free men's estates was not unknown, but, as Domesday Book shows, the rights of the liberi homines were generally secure within the peace of the king. The reservation of the common burdens and the recruitment of lordship to the maintenance of the peace brought both lord and man to book in a common system of local government and thereby created a monarchy of unparalleled power.

ã David Roffe, 2002.

Back