5. SOKE, TITLE, AND THE ORIGINS OF MANORS

 

There is little in the Nottinghamshire Domesday that illuminates the problem of the nature of tenure before the Conquest. It is implied in the text that the lord's hall, the natural focus of the estate, was the basic nexus through which seigneurial dues were collected, and that sokeland in some way belonged to it.[1] Nevertheless, the manor was not the ultimate organising principle. One demesne could belong to another and groups of estates were probably commonplace before the Conquest.[2] However, it is nowhere made absolutely clear what types of relationship were involved. In circuit 6 soca and saca et soca, here translated as 'soke', and 'sake and soke',[3] were the terms used to express almost all relationships, whether political, administrative or tenurial. Soke in itself does not seem to have had any specific meaning beyond the customary transfer of dues. It was used of the king's rights to the forfeitures of felons in the wapentake, on the one hand, to the render of a few pennies quit rent from a mill or wood, on the other.[4]  In between, there were all kinds of possible relationship that the term articulated, but the word soke itself did not define them. Thus, in Lincolnshire the North Riding of Lindsey could declare that the land of Eiric in Tealby hundred was in the soke of Count Alan's predecessor, but they did not know what type of soke he enjoyed.[5] By way of contrast, sake and soke had a more specific application to the right to title  and the dues that that implied. The same categories are almost certainly relevant to Nottinghamshire. In common with Yorkshire and Lincolnshire, the Domesday account records the names of the holders of sake and soke, toll and team in the county, and, as we have already seen,[6] one such immunist appears to have given title to Roger de Bully in his land in Hodsock and Blyth. However, since there is so little evidence of how such categories were articulated - soke appears passim, but sake and soke is found in only four entries[7]- the following discussion inevitably draws heavily upon material from outside the county. It is not the intention to review the whole subject of soke. That is completely beyond the scope of the present work.[8] Nor is the status of sokemen and thanes examined in any detail. It is merely intended to examine the relationships which are apparent, and the terms which are used to express them, in the transfer of title to soke, manors, and groups of manors between 1066 and 1086.

            In his seminal essay Types of Manorial Structure in the Northern Danelaw, and various other works on the East Midlands, F. M. Stenton elegantly argued that the fundamental distinction of pre-Conquest tenure in the North lay between demesne and land in soke. In the one, the lord had full proprietorial rights and could dispose of his land at will. In the other, he only had certain limited interests and dues. The most important of these, and the basic characteristic of sokeland, was jurisdiction. The sokeman was personally free and could usually alienate his land without the consent of his lord, but he  was  obliged  to  make suit to his court.[9] There can be no doubt that there was a very marked divide between the two concepts - land in dominio and in soca is contrasted throughout circuit 6. In terms of title to, and tenure of, estates, however, it is only of secondary importance. The distinction between terra, 'land', and soke was of far greater significance in the transfer of land from Englishman to Norman tenant-in-chief. The dichotomy is most clearly apparent in 'forinsec soke entries'.[10] In Nottinghamshire, there are some sixteen sokeland entries in which the land was held by one tenant-in-chief and the soke by another.[11] In Hodsock, for example, Roger de Bully, had two parcels of land which were soke of the king's manors of Mansfield and Bothamsall.[12] The king had the jurisdiction, but the land itself was held by the tenant-in-chief who clearly derived dues, other than the profits of justice, from it.[13] In general, this interest normally conferred full title,  for  subsequently the soke relationship was forgotten, or relegated to a minor role, and the land said to be held in chief.[14] The lords of Mansfield and Bothamsall, for example, do not seem to have retained any interests of significance in Hodsock.[15] In the same way, both manors and berewicks could be held by one tenant-in-chief and the soke by another. Ralf de Mortemer held a manor in Harmston, Lincolnshire, in succession to Copsi, but Earl Hugh had the soke in Waddington.[16] Indeed, throughout the Clamores of circuit 6, terra and soca are consistently contrasted. Nigel Fossard, for example, held five parcels of land in Yorkshire in succession to three named individuals, but the soke belonged to Conisborough which had been held by Earl Harold in 1066.[17] It is evident, then, that the possession of soke did not in itself confer title to land. It was merely a render, by no means the most important, which was due from an estate. Tenure was derived from a more basic interest in the land itself.[18]

            This interest lies at the heart of the relationship between the hall and sokeland and was evidently the basis of the identity of the manor. It is clear that the soke of many, if not most, manors was not held by the tenant recorded in Domesday Book. We have already seen that Roger de Bully's predecessor probably had sake and soke, toll and team over Hodsock and Blyth, although Wulfsi his tenant was in no way so privileged.[19] Elsewhere the reservation of the liberties to  an  over-lord is more apparent. In Lincolnshire, for example, Fyach, Harold the Staller, and Azer son of Sualeua had sake and soke, toll and team,[20] but they apparently held no land in demesne in the county for their names do not appear in the text. It seems likely, however, that their rights extended over the manors held by undertenants. Thus, Robert Malet claimed sokeland in Ingoldsby (Lincs.) against Gilbert de Ghent through his predecessor Azer.[21] His three manors in Lincolnshire, however, were held by a certain Godwin.[22] Only in his Nottinghamshire breve does Azer son of Sualeua appear,[23] but it was to him that soke was re-served throughout his estates. Nevertheless, although the tenant of the manor did not always enjoy soke qua jurisdiction, there is evidence to suggest that the manor and its appurtenances still had a distinct identity as an integrated whole. First, the grant of the manorial caput, the lord's hall, usually implied a grant of the sokeland associated with it. Thus, in the Lincolnshire wapentake of Graffoe, St. Peter of Westminster claimed various parcels of sokeland against Baldwin the Fleming on the grounds that the abbey had been granted the chief manor by the king. The wapentake agreed on the fact of the grant, and the plea was allowed.[24] In so far as the manor as economic unit was different from manor as soke centre,[25] even forinsec sokeland was probably an integral part of an estate: Roger de Bully's land in Hodsock was evidently part of the manor he held in succession to Wulfsi in the same vill,[26] while six bovates of land were soke of Saundby but belonged to the manor of Bole.[27] Second, the form of many manors suggests an identity somewhat greater than the sum of the parts. The manor of Ruskington in Lincolnshire, for example, consisted of most of the wapentake of Flaxwell, and cannot have derived its form from the chance commendation of 197 sokemen.[28] Finally, the manor was given a value for both 1066 and 1086. Almost without exception, these sums are highly conventional figures, being multiples of standard units of account like the Danish ora of sixteen pence.[29] They clearly cannot be derived from the addition of separate dues rendered by a population free to dispose of its lands at will. The lord expected, and presumably received, a standard sum which implies an established and unilaterally inalienable right in the manorial appurtenances.[30] Indeed, the sokeman's right to the free alienation of his land has probably been exaggerated. It is true that much evidence has been adduced to demonstrate his independence of seigneurial control,[31] but the typical services that he rendered are very similar to those which were due from the precarious thanelands of the abbey of Ely in Cambridgeshire,[32] and recent research has demonstrated that  he  was  tenurially  dependent  in  the  twelfth  and thirteenth centuries.[33] Tenure in socage, then, was clearly not as free as it seems.[34] Although the sokeman was personally free, his manorial lord evidently had an interest in his land which was other than jurisdiction and which could not be unilaterally withdrawn.

            It is unlikely, however, that the tenant of the manor named in Domesday Book had unequivocal rights to terra. As we have seen,[35] the tenant-in-chief's title to an estate was derived from the overlord. Thus, Count Alan had title to land in Billingborough through his predecessor Ralf the Staller rather than the tenant Carle. His interest was evidently more than soke for, where a tenant-in-chief only had jurisdiction, the land was held by a second tenant-in-chief or was entered in the land of the king's thanes. Soke no more conferred right to manors than it did to land. William Peverel, for example, had the soke of two manors in Sutton (Passeys) which were held by Aelfric and Brown, but the holdings were not enrolled in his breve.[36] Count Alan's estate, by way of contrast, was an integral part of his fee.

            The overlord, then, appears to have had residual rights to land in the manors of his tenants which were other than soke. In some instances, his title may have been confined to terra. The king, for example, retained unspecified soke over the archbishop of York's manor of Laneham until 1060×1065.[37] Many, probably most, overlords, however, held both land and soke, and their liberties were usually expressed by the term sake and soke. This privilege, with the root meaning of 'cause' and 'seeking' is generally held to be synonymous with soke, and normally refers to the rights of an individual to the profits of justice within his land.[38] Thus, in both Nottinghamshire and Yorkshire, the holders of sake and soke, toll and team,[39] had the king's custom of two pennies, that is the king's share of the forfeitures of their men.[40] In effect, they were entitled to their own courts, although the king retained soke that was normally reserved to the crown,[41] and their estates were thereby withdrawn from the wapentake or, in hidated England, the hundred.[42] Hence, in the Huntingdonshire Clamores, we read that King Edward gave Swineshead to Earl Siward with sake and soke, and so Earl Harold held it, except that (its men) gelded in the hundred and went with them against the enemy.[43]  Sake  and  soke,  then,  has specific legal referrents. The connotations of the liberty, however, were such that the term was used in a wider context which contrasts the idea with simple soke. In excluding the king, the phrase naturally lent itself to the expression of full rights over a property which preclude all other interests and dispute. Guy de Craon, for example, claimed six bovates of land in Gosberton in the Lincolnshire wapentake of Kirton which had been held by his predecessor Adestan, but he was unsuccessful because Count Alan's predecessor had had sake and soke over it.[44] Throughout the Lincolnshire and Yorkshire Clamores, sake and soke is usually employed in this extended sense, for, with only two possible exceptions,[45] it is carefully distinguished from the more nebulous concept of soke. It is stated by the wapentake of Calcewath in Lincolnshire, for example, that 'In Huttoft hundred Alfred claims two bovates of land. And the men of the Riding say that he ought to have one with sake and soke: the other is his in like wise, but Earl Hugh has the soke in Greetham'.[46] The one expresses the unequivocal tenure of land and all rights over it, while the other refers to the mere receipt of soke dues.

            Since sake and soke expressed such unequivocal rights, it is not surprising that it was the ultimate datum of legal title in 1086. Tenure was frequently established by reference to it where there was a dispute. Thus, Gilbert de Ghent held a manor in succession to Tonna in Willoughby (-in-the-Marsh) (Lincs.), with berewicks in Mumby, Hasthorpe, Sloothby, and Willoughby, and soke in Welton and Boothby. Apart from the relatively large number of berewicks, there is little remarkable about the account of the estate except for a comment that the three bovates of inland in Willoughby were held with sake and soke.[47] By implication, this somewhat anomalous statement suggests that Tonna, the tenant of the estate,[48] did not possess the same liberties in the rest of the manor and, indeed, it appears that they were enjoyed by Gilbert's predecessor, Ulf Fenisc. Thus, Count Alan claimed the berewick of Mumby, but the jury declared that Gilbert's predecessor had held it with sake and soke.[49] Although in the possession of Tonna in 1066, the tenant-in-chief had undisputed right to the land and soke of the berewick in 1086 because Ulf had enjoyed sake and soke over it before the Conquest.[50] Since the overlord had full rights over extended groups of estates, it is evident that, as within the manor, dues could not be unilaterally withdrawn by the tenant. The genesis  of many manors, then, was clearly related to seigneurial initiative. Vestiges of the process can be identified in Domesday Book. It is often clear that large dispersed sokes, a common feature of the tenurial landscape of the North,[51] were formerly of greater extent, for their structure is reflected in the constitution of surrounding estates. The appurtenances of the manor of Laxton, for example, were situated in the same vills as the sokeland of the king's estate of Grimston  [52]which was associated with the manor of Mansfield (figure 13). The two estates had clearly constituted a single unit at an earlier period, and it seems likely that Toki son of Outi's liberties of sake  and soke, toll and team in Laxton were ultimately derived from the larger whole.[53] Some groups of manors may have been related to sokes in the same way. The soke of Oswaldbeck extended into fifteen vills in the north of Nottinghamshire, and Roger de Bully held manors in seven of them (figure 10). His estates almost certainly formed  an extended group - five are multiple manor entries, while the incidental notice of sake and soke in Fenton and Clarborough implies that those dues were normally reserved to an overlord[54] -  and  were evidently related to the soke. It would appear that a single estate had been divided element by element to form two separate interests.[55] A memory of the process survived as late as 1275 for a jury declared that all the fees in the wapentake of Oswaldbeck had formerly belonged to the soke.[56] Since the form of the whole was retained in the parts, the division must clearly have been made by the lord. Indeed, a writ of 1060×1065 issued by Edward the Confessor, the lord of Oswaldbeck, probably effected the final removal of the archbishop of York's manor of Laneham from the soke.[57] It seems likely that division could not be made without the possession of sake and soke, and the lord's interest in land was clearly passed on to the new lord.[58]

   

Figure 13: the interlocking estates of Grimston and Laxton.

 

MANSFIELD

 

LAXTON

1.

Warsop

 

 

2.

Clown

 

 

3.

Carburton

 

 

4.

Clumber

 

 

5.

Budby

 

 

6.

Thoresby

 

 

7.

Scofton

 

 

8.

Perlethorpe

 

 

9.

Rayton

 

 

10.

Edwinstowe

 

 

11

Grimston

 

 

12.

Eakring

4.

Eakring

13.

Maplebeck

 

 

14.

Besthorpe

8.

Besthorpe

15.

Carlton(-on-Trent)

9.

Carlton(-on-Trent)

16.

Kirton

1.

Kirton

17.

Willoughby

2.

Willoughby

18.

Walesby

3.

Walesby

19.

Ompton

5.

Ompton

20.

Carlton(-in-Lindrick)

 

 

 

 

6.

Knapthorpe

 

 

7.

Caunton

NOTE: the numbers represent the order in which the place-names occur in Domesday Book. Knapthorpe and Caunton appear in the soke of Laxton in the position of Maplebeck. All three vills are adjacent to each other and may formerly have constituted a single element in the estate of Mansfield.

The process is ultimately one of booking which, in origin and theory, is a royal prerogative.[59] The tenure of sake and soke - the right to land removed from the king's soke - is a natural consequence and corrollary. The term is thus indicative of bookland and, as such, its identification with title is therefore comprehensible.[60] It expresses the delegation of regalian interests which were organised in large sokes, shires, lathes, or multiple estates,[61] to an individual. The king was thereby excluded from the property, under certain con-ditions, and the lord was free to dispose of it as he saw fit.[62] The interests alienated may have included parcels of demesne, the soil of which had belonged to the king. Most, however, were tributes of various kinds. First, there was jurisdiction. The lord had the right to all the profits of justice, including forfeitures, that arose from the pleas of his land and men, except those reserved to the crown.[63] Second, the lord enjoyed labour dues. Those owed by the villeins were extensive and must have provided much of the manpower for the lord's demesne.[64] But the  the sokemen also rendered labour services. The foundation charter of Blyth, which is almost contemporary with Domesday Book, suggests that they included ploughing, carriage service, reaping, mowing, hay-making, and repair of the local mill.[65] The charter gives no information on the amount of service owed, but the burden may have been heavier than normal for in 1086 the sokeland of Blyth was occupied by four villeins and four bordars.[66] However, it is clear from the c.1125 survey of the estates of Peterborough Abbey and the early twelfth-century foundation charter of Revesby Abbey in Lincolnshire, that these services were of the same kind as those due from sokeland.[67] Day work is but rarely found, and most tenants only had to perform seasonal boon works.[68] Other sources suggest that services such as repair of the lord's hall were also due.[69] Third, there was a financial tribute and/or a render in kind. This due was usually expressed as consuetudines, customs, and was of considerable value.[70] In the soke of Oswaldbeck it seems to have accounted for all the monetary issues of the land, for the 20 shillings in customs that 22 sokemen in Leverton rendered in 1066 seem to represent the value that is appended to all the other parcels of land in the soke.[71] In origin, the payment was almost certainly a commuted food rent or farm. A probably authentic late tenth- or early eleventh-century grant  of  land  to  Ramsey  Abbey  in  Hickling  and  Kinoulton specifies a heavy render in kind, but at the time of Domesday there is no suggestion of anything other than a money rent.[72] Finally, some sort of military service may have been expected from sokeland.[73] The evidence, however, is far from unambiguous. Alsi and Adestan held a manor in Swaton, but Alfric their brother had soke over them in Haceby only in the king's service.[74] Soke in this context may have been used in a non-specific sense, for it probably only refers to an arrangement between co-parceners to acquit their tenement of a personal duty. Seemingly less equivocal is the reference to soke of Offram's manor of Keisby which rendered nothing, but used to help in the king's host (adiuabat in exercitu regis) on land and at sea.[75] Adelid, the tenant, however, was a woman, and it is therefore difficult to understand how the service was acquitted.

            Sake and soke seems to have expressed rights to all of these services and exactions. Not all, however, were rendered to the same hall for the lord could give certain tributary dues to his men. It is probably this process that resulted in the formation of individual manors held by subordinate tenants within the lordship of a superior. It is clear from explicit notice and the basic arrangement of the Domesday text that sokeland belonged to the tenant's manorial hall, but he did not necessarily have the soke, qua jurisdiction, of the land. As we have seen, that was frequently, probably always, vested in the overlord. Thus, the marginal S and soca of this type of entry clearly does not imply that the caput had the profits of justice. The device merely indicates that the land was of the type which rendered  dues.  A  simple generic contrast with demesne is all that is intended. It is evident, then, that only part of the tributary dues were diverted to the tenant - this almost certainly included the labour dues and probably also the customs or farm, the essential characteristics of terra. Such services, however, were but the delegated right of the overlord, and the tenant rendered a farm for the privilege of enjoying them. The value of the manor as recorded in Domesday Book was almost certainly the sum he paid. As we have seen, the figure is glossed as consuetudines in one entry in the Nottinghamshire folios, and the value of one estate is frequently said to belong to another in circuit 6.[76] In Derbyshire, indeed, the value of the manor of Osmaston was rendered to both the king and Henry de Ferrers, in the proportion of two to one, just like any other royal farm which was shared between the king and the earl.[77] The Domesday manor in its essentials was evidently not an economic unit. It was merely a convenient device for the interception of delegated tributary dues. Nevertheless, demesne, which was tilled by a villein population, was frequently attached to it. This was the nucleus from which economic manorialisation could grow.[78]

            The process outlined above explains the origins of many manors and groups of manors in Nottinghamshire, The widespread distribution of large sokes and interlocking estates structures  [79]suggests that it was of common occurrence. It is not the only possible mechanism of estate formation, however. Professor Sawyer has argued that the distribution of some Danish place-names with a personal name as a first element  in  the vicinity of large estate centres suggests that some independent manors in the East Midlands owe their existence to the piecemeal disintegration of extensive sokes under the impact of Danish colonisation.[80] Further, as attractive as the idea may be, we have no grounds for assuming that the whole of Nottinghamshire, or the North come to that, was originally composed of a small number of multiple estates.[81] In densely wooded areas there is at least the distinct possibility that many settlements and estates owe their existence to more or less independent assarting of forest and waste. Some of the small manors of Nottinghamshire may well have been shaped by such a process. This does not in itself imply a great deal of freedom, such as is found in the mediaeval fenlands of eastern England.[82] But it may suggest a sizeable class of small independent lords who were free to commend themselves to whomsoever they wished. A fusion model of estate development is therefore not precluded by the existence of a mechanism of fission. The process may, indeed, be responsible for the formation of some pre-Conquest groups of manors. As is evident from many pre-Conquest charters and wills in which more than one estate was granted, once land was booked, it could be divided or amalgamated with other estates at will. In Scarsdale Wapentake in Derbyshire, for example, Wulfric Spot bequeathed a large number of small discrete estates to Morcar in 1002×1004, and in 1066 they appear to have formed a dispersed group of manors held by Leofnoth, the predecessor of Ralf son of Hubert.[83] The individual elements of an extended group in 1066, then, may have come into being in completely different tenurial contexts. Any assessment of the relative importance of fission and fusion in the genesis of eleventh-century estates must, however, await further research on settlement patterns and estate structures.

            Regardless of the origins of his estates, the tenant-in-chief's title to his land in 1086 was founded upon a well established legal theory. It was derived from a predecessor who generally held his estates and their appurtenances by book. The term sake and soke, with its connotations of full rights, effectively expresses the widespread powers which he enjoyed in his lands. It is not surprising, then, that it was the datum of title - the proven tenure of the franchise dispelled all counter claims - and excluded the king and his officers from the-day-to-day running of the estate. The list of immunists appended to the Nottinghamshire Domesday, although demonstrably incomplete,[84] there-fore provides something of a guide to the predecessors of the Norman tenants-in-chief.[85] Nevertheless, this was probably not its primary purpose. As part of the shire customal, as in similar lists in the Lincolnshire and Yorkshire folios,[86] it was clearly also  intended  to provide a record of royal dues. The holder of bookland had duties, such as military service, as well as privileges, and his continued tenure depended upon successful performance.[87] If he failed, he forfeited his land directly to the king and the earl, although in Nottinghamshire his wife and heirs were entitled to a moiety.[88] The list is evidently a record of the king's rights to such forfeitures rather than a memorandum of liberties per se.[89] In the next chapter, we shall examine how these  dues relate to the common system of royal administration and local government in the shire.



[1] Thus, the value of the hall was that of the whole manor with its appurtenances (TMS, 32-4, 57-9).

[2] See chapter 4.

[3] 'Jurisdiction' and 'full jurisdiction' have been employed in Notts. DB, but, as will become clear in the following analysis, the phrases are positively misleading. See G. Black, D. R. Roffe, The Nottinghamshire Domesday: a Reader's Guide, Nottingham 1986, 24-5.

[4] Lincs. DB, 71/9; 69/20; C. A. Joy, Sokeright, unpublished thesis, Leeds University 1974. 70-77.

[5] Lincs. DB, 70/18.

[6] See chapter 4.

[7] Notts. DB, 9,113; 128. 20,6. 30,39.

[8] The most recent discussions can be found in Joy, Sokeright, C. Stephenson, 'Commendation and Related Problems in Domesday Book', EHR 59, (1944), and A. K. G. Kristensen, 'Danelaw Institutions and Danish Society in the Viking Age', Medieval Scandinavia 8, (1975), 74-85.

[9] TMS, 1-55; Lincs. DB, xxvi-xxviii; Documents Illustrative of the Social and Economic History of the Danelaw, London 1920; The Free Peasantry of the Northern Danelaw, Oxford 1969; 'The Danes in England', Preparatory to Anglo-Saxon England, ed. D. M. Stenton, Oxford 1970, 136-65.

[10] The term 'forinsec soke' is coined by analogy with forinsec service in a feudal context, but no identity of relationship is intended. Forinseca soca is used in mediaeval sources of part of the soke of Chesterfield in Derbyshire (PR 1197, 150). However, it probably refers to the land outside of the town. A foreign jury, used in this sense, existed at Gainsborough into the modern period (A. Stark, The History and Antiquities of Gainsborough, London 1843, 91, 532. See also Black, Roffe, Nottinghamshire Domesday, 19).

[11] Notts. DB, 9,47; 48; 118. 11,15-17; 21. 12,22. 13,5; 13. 17,13; 14. 30,26; 44; 49; 55.

[12] Notts. DB, 9,47; 48.

[13] High values are often associated with such entries. Thus, two bovates in Farnsfield were worth eight shillings to Walter de Aincurt in 1086, even though the soke belonged to the archbishop of York in Southwell (Notts. DB, 11, 17). Where a tenant-in-chief only had title to the soke, the land was normally enrolled with the land of the king's thanes. Canute, for example, held two bovates of land in Misson which were soke of Kirton (in Lindsey), and they are described with his manor in the same vill (Notts. DB, 30,44). Likewise, the two manors of Aelfric and Brown appear in the same breve, even though the soke belonged to William Peverel's manor of Wollaton (Notts. DB, 30,55).

[14] Apart from land attached to the large sokes, the relationship does not generally appear in the records at all. Soke, qua jurisdiction, was evidently a minor due.

[15] The land passed to Blyth Priory.

[16] Lincs. DB, 36/4.

[17] Yorks. DB, CW11-14; see also Lincs. DB, 69/11,14,29,35; 70/3,8,12, 15,21,29; 71/6; 72/6,7,17,60; Hunts. DB, D4; 15-7; 29.

[18] Joy, Sokeright, 78-9.

[19] See chapter 4.

[20] Lincs. DB, p13.

[21] Lincs. DB, 72/35.

[22] Lincs. DB, 58/1-8.

[23] Notts. DB, 25,1.

[24] Lincs. DB, 72/27.

[25] See chapter 9.

[26] Notts. DB, 9,46-8.

[27] Notts. DB, 9,118.

[28] TMS, 43; Lincs. DB, 64/1-14.

[29] TMS, 32-4; D. R. Roffe, The Derbyshire Domesday, Darley Dale 1986, 22; Black, Roffe, Nottinghamshire Domesday, 28.

[30] Sokeland, nevertheless, was sometimes held by an individual other than the lord of the manor. Grim held two bovates of land in Watnall which were apparently soke of Siward's manor in the same vill (Notts. DB, 10,45). Just how common this was, cannot be determined. Subtenancies are often inadvertently noted in the Clamores, while there is no indication in the text that the land was not ordinary, untenanted, sokeland. Thus, Bertor, Summerd, Godric, and Siward held land in Mablethorpe (Lincs.) of Earl Harold's soke of Greetham, but the fact is only explicit because the bishop of Durham and William Blund made claims to the estate (Lincs. DB, 13/7; 69/15). The Domesday commissioners were clearly not interested in the phenomenon for it probably had no bearing on title.

[31] Stenton, Free Peasantry of the Northern Danelaw.

[32] Stephenson, 'Commendation', 297n., 308n.

[33] G. Platts, Land and People in Medieval Lincolnshire, Lincoln 1985, 59-66.

[34] Some at least of the twelfth-century grants of free peasants, cited by Stenton in support of his thesis, were subsequently confirmed by their lords, See, for example, Danelaw Charters, nos 538, 540, and Free Peasantry, no. 118. Grant by charter is no reliable indication of freedom; see Carte Nativorum: a Peterborough Abbey Cartulary of the Fourteenth Century, eds C. N. L. Brooke, M. M. Postan, Northampton 1960.

[35] See chapter 4.

[36] Notts. DB, 10,35. 30,55. In 1198 the two manors are probably represented by the fee of Robert le Passeis which was held in chief by sergeancy (BF, 8).

[37] F. E. Harmer, Anglo-Saxon Writs, Manchester 1952, no. 119. For the identification of the estate, see below.

[38] DBB, 84; Lincs. DB, xxxvii.

[39] These privileges were probably no different in kind from sake and soke. Thus, in Lincoln all the lawmen had sake and soke over their lands, but Ulf son of Suertebrand had in addition toll and team (Lincs. DB, p3/1). It seems that the extended phrase includes the regalian dues of holding a markets and collecting tolls within a fee, liberties which the crown had tried to centralise in the borough and wapentake (Lincs. DB, xxix; see chapters 6 and 8).

[40] Notts. DB, S5; Yorks. DB, C36; Lincs. DB, xxxix.

[41] By the very nature of kingship, no one was outside of the king's soke. The so-called trinoda necessitas was always reserved (E. John, Land Tenure in Early England, Leicester 1960, 64, 73), and forfeitures were made to the king and earl. Hence, it is stated in the Nottinghamshire folios that 'If a thane who has sake and soke should forfeit his land, the king and the earl between them have half his land and his money; his lawful wife, with his lawful heirs, if any, have the other half' (Notts. DB, S4). Such forfeitures were probably made in the wapentake for, although his land might be withdrawn, the lord himself was still obliged to pay suit on his own behalf. See chapters 6 and 8.

[42] See chapter 6.

[43] Hunts. DB, D14.

[44] Lincs. DB, 12/76; 73/5.

[45] Lincs. DB, 69/28; 71/14. In both cases, a Norman tenant-in-chief merely claimed soke over land, even though his predecessor had enjoyed sake and soke in the estate. It was on the basis of these two solitary entries that Stenton postulated the identity of the terms sake and soke, and soke (Lincs. DB, xxxvii). As they stand, both are highly exceptional and cannot be easily reconciled with the usage found throughout the rest of circuit 6. However, it may be supposed that there was some unrecorded transaction which conferred rights to land between 1066 and 1086, or that the scribe was simply in error. Given the similarity of the terms, and the fact that the compiler must have had far more information in front of him concerning the liberties than appears in the text, confusion would not be surprising. Indeed, it can sometimes be directly observed. In a Lincolnshire Clamores entry relating to Osbournby, the curious term soca et soca appears (Lincs. DB, 72/53). This is clearly nonsense, and the editors have emended the text to saca et soca. How-ever, it is clearly soca alone which is intended for the liberty entitled Ralf Pagenel to a horse from the land when he went to war. Such rights are always expressed in terms of soke (see, for example, Lincs. DB, 26/45). Likewise, the statement that Countess Godiva had sake and soke over Newark Wapentake is not only illogical, but also patently untrue since the Abbey of Peterborough had the same liberties in Collingham within the same wapentake (Notts. DB, S5; see chapter 6).

[46] Lincs. DB, 69/16.

[47] Lincs. DB, 24/54-60.

[48] Tonna was also a tenant of Ulf Fenisc, Gilbert de Ghent's pre-decessor, in Baumber and Edlington (Lincs. DB, 24/16,20; 69/23,33). Although clearly an undertenant, he must have been of some considerable importance for he held land in at least nine counties before the Conquest. The power and wealth of 'mesne' tenancies were probably not an exclusively Anglo-Norman phenomenon.

[49] Lincs. DB, 69/18.

[50] Such explicit examples are rare, but see Lincs. DB, 38/3-7. Robert the Steward held a manor in Scrivelsby, with berewicks in Coningsby and Wilksby, and soke in Mareham (-upon-the-Hill). It had been held by Siward in 1066, who also had nine acres of arable  and  eight  acres  of woodland in (Wood) Enderby with sake and soke. This parcel of land was apparently part of the estate, but the exceptional notice of sake and soke suggests that Siward was not so privileged in the rest of the manor. Indeed, Robert claimed title to soke in Coningsby through Achi, his predecessor, who held his estates with sake and soke, toll and team (Lincs. DB, p13; 69/34).

[51] For the most recent discussion of the institution, see W. E. Kapelle, The Norman Conquest of the North, London 1979, 50-85.

[52] Grimston  is at one point called a berewick of Mansfield, but in a duplicate entry is termed a manor (Notts. DB, 1,17; 27). The anomaly is discussed in chapter 9.

[53] Notts. DB, S5. 12,1-10.

[54] Notts. DB, 112; 127. Sparrowhawk held his land with sake and soke, but without a hall.

[55] Although seemingly attached to the manor of Mansfield, the soke of Oswaldbeck probably constituted a separate royal manor in 1086. Its Domesday form may merely reflect its management in the late eleventh century as one of a groups of royal estate which were farmed by a single reeve. The management of the king's estates in Hamenstan Wapentake in Derbyshire is a direct parallel (Derbys. DB, 1,15; 29). See chapter 9.

[56] RH ii, 25. Reference is also made to the soke of Bassetlaw.

[57] Harmer, Anglo-Saxon Writs, no. 119. Edward the Confessor quit all the land of the archbishop of York in Yorkshire and Nottinghamshire which had belonged to the king's soke. Since Southwell and Sutton were held by book (ECNE, 111-2), it seems likely that the archbishop's estates in Oswaldbeck are, inter alia, referred to. For the nature of the soke in this context, see chapter 8.

[58] The process can be directly observed in the Lincolnshire Domesday. The estate of Godwin, which had the same structure as the soke of Bolingbroke, was divided between his sons Siwate, Alnod, Fenchel, and Aschil (Lincs. DB, 69/38; 70/30). Siwate 'was the king's man', and his land  passed to Eudo son of Spirewic with sake and soke. The bishop of Durham succeeded to the rest, likewise with sake and soke. Godwin almost certainly held the same liberties, although the fact is not explicit in the text, and it seems likely that the division of the estate in such a way that it passed to two Norman tenants-in-chief was only possible because of these privileges. There was probably some arrangement like the division of Siwate's estate in the wapentake of Horncastle: his demesne land was shared by his three sons Harold, Godevert, and Alfric, but the soke was only divided between two of  them. Subsequently, the estate passed to two tenants-in-chief (Lincs. DB, 3/10; 29/1; 69/20-1). By way of contrast, the four manors of Ingemund and his brothers were probably not held with sake and soke - enrolled in one entry, they were presumably the right of an overlord (see chapter 4) - and therefore all went to Count Alan (Lincs. DB, 12/31; 70/26).

[59] John, Early Land Tenure, 42-3. If tenure by book conferred something akin to 'freehold', as John suggests, then a lord could presumably transfer his interests without further sanction. However, a grantee probably felt it in his interest to obtain his own book to guard against the claims of the grantor's family. Exchanges of land between predecessors and tenants-in-chief were apparently an official matter. See, for example, Lincs. DB, 72/13, 19.

[60] DBB, 282.

[61] Various terms are used, but the type of estate is found throughout the country. See J. E. A. Jolliffe, 'A Survey of Fiscal Tenements', EcHR, 1st ser. 6, (1935-6), 157-71; Jolliffe, 'Northumbrian Institutions', EHR 41, (1926), 1-42; G. R. J. Jones, 'Multiple Estates and Early Settlement', Medieval Settlement: Continuity and Change, ed. P. H. Sawyer, London 1976, 15-40; Kapelle, Norman Conquest of the North, 50-86. See also chapter 9.

[62] See above.

[63] TMS, 21-2.

[64] TMS, 22-8.

[65] TMS, 22-4, 92-4.

[66] Notts. DB, 9,49.

[67] Chronicon Petroburgense, ed. T. Stapleton, London 1849, 157-83; Mon. Ang. v, 434; TMS, 25-7.

[68] Sokemen of the abbot of Peterborough's manor of Scotter in Lincolnshire performed day work (Chronicon, 164-5). This, however, was not the norm in the abbey's estates. In Derbyshire, sokemen, and tenants of similar status, are consistently omitted from Domesday Book (Roffe, Derbyshire Domesday, 18-9), but their services, as recorded in the early twelfth-century surveys of the estates of Burton Abbey, are almost identical with those of the more normal sokemen (C. G. O. Bridgeman, 'The Burton Abbey Twelfth-Century Surveys', Collections for a History of Staffordshire, William Salt Archaeological Society 1916, 212-47).

[69] Jolliffe, 'Northumbrian Institutions', 6.

[70] TMS, 35-7.

[71] Notts. DB, 1,32; TMS, 35-7.

[72] ECNE, 112-3; TMS, 37-8.

[73] TMS, 28-30.

[74] Lincs. DB, 26/45; 51/12.

[75] Lincs. DB, 57/43.

[76] See above and chapter 4.

[77] Derbys. DB, 6,88. Henry was apparently not the earl at this time, but he succeeded to many of the pre-Conquest comital estates and interests (Roffe, Derbyshire Domesday, 12).

[78] See chapter 9.

[79] See chapters 8 and 9.

[80] P. H. Sawyer, 'Some Sources for the History of Viking North-umbria', Viking Age York, ed. R. A. Hall, London 1978, 7. It may be doubted, however, that all the cases that he cites are in fact piecemeal. Personal names attached to settlement elements may imply some kind of unprecedented lordship over a parcel of land. However, the extensive evidence for ordered division suggests that there was still an overlord who retained rights in the new estate.

[81] M. W. Bishop, 'Multiple Estates in Late Anglo-Saxon Nottingham-shire', TTS 85, (1981), 42-7; Jolliffe, 'Fiscal Tenements' 157-71.

[82] H. E. Hallam, Settlement and Society: a Study of the Early Agrarian History of South Lincolnshire, Cambridge1965, 200-5.

[83] ECNE, 109; Derbys. DB, 10,1-10; Roffe, Derbyshire Domesday, 12.

[84] There are four instances where a pre-Conquest lord is said to have sake and soke, but does not appear in the list (Notts. DB, 9,113; 128. 20,6. 30,39). Morcar, probably the earl of Mercia, had toll in Gunthorpe, although, again, his liberties are not apparent from the shire customs. Many important lords were seemingly disenfranchised in the same way. In fact, however, the Domesday scribes were probably not particularly interested in the information. It only appears in the Yorkshire, Lincolnshire, and Nottinghamshire and Derbyshire folios, and appears to have been enrolled without great care (Lincs. DB, xxxix). Many lords, then, were probably omitted.

[85] Notts. DB, S5.

[86] Lincs. DB, p13; Yorks. DB, C36.

[87] R. Abels, 'Bookland and Fyrd Service in Late Saxon England', Anglo-Norman Studies VII: Proceedings of the Battle Conference 1884, ed. R. A. Brown, Woodbridge 1985, 1-15.

[88] Notts. DB, S4.

[89] Lincs. DB, xxxix. The information is probably derived from a survey of royal lands and rights alone (see chapter 3), and was probably enrolled without very much purpose.