INTRODUCTION

 

Background

In South Lincolnshire, defined as the former county divisions of Holland and Kesteven, there are a large number of earthwork sites of medieval or assumed medieval origin, excluding the more widespread type of feature such as protective earth banks, saltern mounds and field systems. In 1977, a period when the input of the Manpower Services Commission to archaeology was at its height, the then South Lincolnshire Archaeological Unit was fortunate in being able to employ Vic Ancliffe, a surveyor recently redundant from local government service. The Unit had undertaken to assess the archaeology of the whole of the south of the county and was in the process of collecting data on the various classes of upstanding monuments for the Sites and Monuments Record. A list of the more obvious earthworks then known was drawn up with the aim of producing a series of plans of as many sites as possible. In the north of the county, the former Lindsey division, Paul Everson had begun a survey for the Royal Commission on Historical Monuments but it was already understood that there was little likelihood of this survey being continued into the remainder of Lincolnshire.

           Out of the preliminary list of 60 sites, only 18 were at the time scheduled ancient monuments and it was felt that completion of a series of plans and sections was a matter of some urgency. It was not anticipated that all the known sites could be dealt with in the time available and this limitation, together with the inevitable occasional problem over access, meant that, at the close of the programme, a number of earthworks remained unrecorded. In this initial scheme there was no brief to investigate additional potential sites or to consider recorded sites, such as villages or moats, that were already under plough. A further project was set up in 1981-2 which enabled the compilation of a gazetteer of fenland and fen edge medieval earthworks, and included the systematic examination both of air photographs and documentary references to sites no longer visible.

          It is certain that more remains to be discovered. The old pasture and wooded areas of Kesteven are likely to be particularly rewarding. The work of Paul Everson in the north of Lincolnshire and that of the Ropsley Survey Group in the south have demonstrated the potential of parkland and the wealth of information concealed in one small area of woodland respectively. On the other hand the Holland division, which is intensely arable with little tree cover, has already been studied in some detail, and is unlikely to furnish many new earthwork sites other than field systems. The only new information here is from documents which occasionally identify sites no longer visible.

          In the last decade since the survey was carried out, one site, the Manor House moat at Spanby, a scheduled monument, has been levelled. However, new names have been added to the lists and a there is now a total of... known earthwork sitesOf these there are still only 21 scheduled ancient monuments. A gazetteer of all known sites in the various categories described below has been prepared (Appendix) although not all of these necessarily remain as earthworks.

          The MSC scheme was restricted to the practical work of surveying and research on individual sites was only carried out subsequently as 'post-excavation' work.

 

Method and sources

Many different types of mediaeval earthwork, from the vestiges of ploughing through moated sites to whole settlements, can be seen in the landscape today. Historians and archaeologists have been aware of the extent of the remains for many years, but, with the general exception of the physical manifestations of field systems and some remarkable local studies, little systematic work has been undertaken on the recording, identification, and interpretation of such sites. This study cannot claim to be a comprehensive gazetteer, nor  is it in any meaningful sense a considered sample. Nevertheless, it does aim to illuminate the various types of earthwork that have been surveyed by describing the social and economic trends which they illustrate. A wide range of sources from the public records has been consulted (see bibliography), but only a limited selection of private muniments have been searched. Most sites would merit extensive detailed investigation, but the pressures of time have necessitated a more cursory approach.

          In common with the vast majority of archaeological sites, the initial problem is one of identity. Even major local institutions like the castle at Wrangle can escape notice, for the material culture and physical background of society is not the everyday material of mediaeval records. The acuteness of the problem varies with the type of site under consideration. Settlements tend to be fairly easy to name since much of local government was mediated through a network of vills which, in thirteenth-century Lincolnshire at least, approximates to the structure of settlements. Difficulties, however, still arise. Ogarth in the vill of Ropsley appears in fourteenth-century sources, but cannot be located on the ground. Likewise, religious houses are usually easy to identify because many of their muniments tend to survive. Granges, however, have not always left records, and some like Laughton cannot be assigned to a specific foundation with absolute confidence. Castles and moated sites in lay hands present more complex problems. Inquisitions post mortem and estate surveys often mention the existence of capital messuages, but only rarely is a detailed description, such as that of the castle at Welbourn, given. In these circumstances it is the details of estate structure and exploitation that must provide the clues to identity. Mediaeval South Lincolnshire is characterised by great tenurial complexity which derives from the network of pre-Conquest soke relationships. Typically manors consisted of a demesne in a central vill and the right to services and dues from various parcels of land in surrounding villages. In the post-Conquest period, then, most settlements were divided between a number of lords. In some cases several manors can therefore be found, and the tenurial context of a moat can only be suggested by its proximity to existing manor houses or known appurtenances of a fee like the church. More often, however, the out-lying portions of estates continued to render rents and dues to the eleventh-century manorial caput throughout the mediaeval period - the Domesday estate structure, although largely invisible in feudal records, survived with a remarkable degree of integrity for several centuries - and the number of possible manorial curie is reduced to one or two. Only rarely is the use of a site explicit in historical sources, and its function can only be deduced from the location and typology of the surviving earthworks. Finally, features like fishponds and gardens are usually unidentifiable except from their form.

          Only two of the earthwork complexes in the present study have been investigated archaeologically, and the dating of sites is therefore often nebulous and dependent upon analysis of historical background. The depopulation of settlements can usually be charted when they constituted vills. The desertion of hamlets like Humby and Boughton, however, in common with the contraction of villages is often difficult to document in the absence of manorial court rolls. Religious houses are usually the most securely dated sites where foundation charters survive, but even an explicit date must be treated as only approximate since the establishment of communities was always a protracted process in which the grant of a charter was one in a number of stages. For moated sites, both ecclesiastical and lay, only date ranges can usually be provided which are determined by the grant of land or enfeoffment of a knight and the first notice of the site or a reference that implies its existence. Other earthworks remain largely undateable.

          In the following analysis the sites have been divided into six categories on the basis of shared characteristics. Not all, however, are mutually exclusive. Shrunken and deserted settlements constitute coherent groups, but are not always readily distinguishable one from the other. Moated granges have been discussed with the two religious houses surveyed as more readily comprehensible within that historical context. In form, however, they differ little from moated sites in lay hands. Castles also overlap with moats, but the category has been defined by mediaeval usage or the survival of substantial defensible earthworks. The remaining earthworks are miscellaneous sites about which little is known.

 

Deserted settlements

                             

 

CLASS

TENURE

CAUSE/DATE

Boughton

2

d

? C14-C16

Brauncewell

2

e,r

granges C13ff

Dunsby

2

e,r

granges C13ff

Silkby

2

e

? C15-C16

CLASS: MAFF land classification

TENURE: d = demesne, e = enfeoffed, r = grant to religious house(s)

 

Four of the earthworks surveyed relate to deserted mediaeval villages. The type is not always readily distinguishable from shrunken villages, for the land is always exploited in one way or another and therefore habitation has usually been found until the advent of factory farming in the present century. Farms, for example, significantly representing the mediaeval manor houses, are still found at Boughton, Brauncewell, and Silkby. But, unlike sites such as the all but deserted Lenton, all have lost civil, ecclesiastical, or economic functions as parishes or hamlets and have therefore ceased to be separate communities. In most cases such major changes can usually be documented, and it is tempting to look for a single reason at the time of desertion. Decline in population, exhaustion of marginal land, climatic changes, and seigneurial greed have all been adduced as causes, and in some cases prove a sufficient argument. The earthworks at Welbourn, for example, may be the site of the lost village of Sapperton and appear to represent a contraction of settlement into the parent vill in the face of a fall in population in the fourteenth and fifteenth centuries. But many settlements experienced similar pressures and did not succumb, and simple explanations of this kind are therefore often inadequate. Rather it is to the interaction of a number of processes and actions which predisposed settlements to radical changes in fortune that we must look.

          Clearly general conclusion cannot be drawn from the small sample of sites studied. Nevertheless, it is evident that the history of each conditioned their eventual fate. It is true that all four are on marginal or relatively marginal land, but this does not imply that they were necessarily vulnerable to desertion. In the eleventh century both Dunsby and Brauncewell, for example, were small settlements situated on the barren heath and appear to be destined to extinction. In fact they were economically and socially well adapted to the meagre resources of their environment. Much of the land was held by sokemen who rendered tribute to a lord in Ruskington, and settlement was probably of a predominantly dispersed pattern. The intensity of exploitation in this form would have no doubt varied as it underwent climatic and demographic pressures such as those experienced in the fourteenth century, but complete desertion would have been unlikely in what was essentially a simple subsistence economy and society.

          Boughton and Silkby, along with many others like Howell and Humby which survive today, shared this character. But the principal experience which led to their different fate was a process of manorialisation which concentrated the control of economic resources in the hands of a lord. Already in the tenth and eleventh centuries lordship was moving from a tributary relationship to a more directly exploitative nexus as large multiple estates like Sleaford, Ruskington, and Folkingham began to disintegrate into a number of discrete manors. In 1086 the process was incomplete, for the render of customs and food rents were probably still the main issues of such estates. In the twelfth century, however, more rigorous services, notably daywork, were often imposed upon the peasantry in the vicinity of the lord's demesne - outlying sokeland was sometimes enfeoffed and new demesnes created, but more usually remained a source of assized rents - and a more exploitative economy and society emerged. Settlement nucleation and a growth in population probably quickly followed, but at once village communities became more vulnerable to economic pressures. Twelfth-century grants of land to religious houses in Burton Pedwardine and Blankney, for example, led to the creation of intensively cultivated farms which seem to have all but extinguished the small settlements of Mareham and Cotes by the thirteenth century. Dunsby and Brauncewell may have contracted at the same time under similar pressures. More usually communities on marginal land, like Silkby and Boughton, survived until the late fourteenth century when the population decline and consequent increase in the price of labour led to the lord turning over his land to the more profitable business of sheep farming.

          The desertion of settlements in Lincolnshire, then, was not just a simple response to catastrophes like the Black Death or changing economic forces. Such factors of course played their part, but the social structure of each settlement was as equally important as its economic potential and demographic resources.

 

Shrunken mediaeval settlements

                                                           

 

CLASS

TENURE

DATE

Boothby

2

m

?14th-

Howell 2

2

s

?14th-

Humby

2

s

?14th-

Welbourn

2

m

14-15th

CLASS: MAFF land classification

TENURE: m = manorialised, s = socage tenure

 

A further four earthwork complexes attest to the desertion of tofts and crofts within existing villages. The phenomenon is common in Lincolnshire. It is found in at least six of the other settlements studied, and superficially it may seem that the process involved is identical to that of complete desertion. However, it is probably a mistake to view most of such sites as 'failed DMVs'. Some of the villages in which they are found, like Howell and Humby, have an identical physical environment to deserted settlements, but retained the sokage tenure which was appropriate to the context. Most, however, tend to be on better land and exhibit a greater degree of manorialisation in the eleventh century when they first appear in the historical record. All, of course, were subject to the same types of pressure that the lost settlements experienced. But as more prosperous communities recovery was quicker and, as at Langtoft, lords could maximise profitability by leasing land to yeoman farmers rather than introducing sheep. In general terms, then, such earthworks must be regarded as evidence of the normal ebb and flow that is characteristic of all settlement in its response to changing circumstances as opposed to testimony only to incipient decline.

          By the very nature of mediaeval records such circumstances cannot always be documented adequately. However, a fall in population in the late thirteenth and early fourteenth centuries resulting in a contraction of settlement, as at Boothby and Howell, is just one of a number of possible causes. At Lavington settlement drift may have been responsible, for the present area of dense occupation, a complex to the east of the church, may represent a shift of nucleus; while in Scredington the abandonment of no less than five moated sites, whether by amalgamation of estates or other means is not clear, seems to have led to desertion within the village. Elsewhere any number of other factors may have been responsible.

 

Religious houses and granges                                        

                                                           

 

LOCATION

Catley

 fen

Swineshead

 marsh

Branston

 heath

Gosberton 1

 fen

Laughton

 

Linwood

 fen-edge

Mareham

 upland

Pinchbeck

 fen

 

Only two of the sites relate to religious houses. Both Catley and Swineshead were mid twelfth-century foundations of local knights and exemplify a new trend in religious benefaction. In the first fifty or so years after the Conquest most houses were founded by tenants-in-chief, often as cells of Norman houses, and were endowed with large tracts of land, frequently amounting to whole vills. The foundations were not only family cult centres, but also an integral part of the community of the honour, and the founder's vassals naturally discharged their religious obligations by granting land to them. Increasingly in the twelfth century, however, wealthy and important subtenants, like Simon de Kyme who founded the priories of Bullington and Kyme, began to found religious houses of their own in imitation of their social superiors. Catley and Swineshead, established by Peter de Billinghay and Robert de Gresley respectively, were typical of the new foundations that emerged. They were communities which adhered to orders which had been recently formed to accommodate a desire for a more personal  and intimate religious life; their endowments, reflecting the resources of their patrons, were relatively modest; and, although equally family cult centres and mausoleums, they were immensely popular locally and attracted grants from all levels of society.

           The houses, however, not only satisfied religious aspirations and conferred status upon their founders. They also played a part in the exploitation of the lord's manors. Both Catley and Swineshead were situated in remote wastes on the periphery of their founder's principal estate, and the improvements that they introduced must have been to the general benefit of the community and the profit of the lord. This aspect of endowment is most clearly seen in the foundation of granges. Seven of the sites surveyed are of this type. All were situated on heath or adjacent to fen and, dating from the mid to late twelfth century, appear to have taken an active part in the process of reclamation of the waste. The moats that surround the sites are the most eloquent surviving testimony to the process. Records from Catley indicate that the digging of boundary ditches was often specified in charters granting land, and at Rigbolt, Newhall, Wyberton, and Linwood where drainage of the site was a prerequisite of exploitation, the construction of a moat must have been the first activity on the site. The creation of upland and heath granges as at Mareham, Branston, and Laughton were probably of less immediate economic advantage to the lord. Moats in this type of environment ensured security and probably often functioned as pounds for animals as attested in Kirkstead sources. Wool production did little but remove common pasture from the estate, but the grant of such rights was a way of endowing religious houses without alienating resources in short supply. Unfortunately no records have survived to illustrate the internal organisation and estate management of the granges studied in any greater detail, but excavation has uncovered the arrangement of buildings at Laughton.

 

Castles                                                                                                    

                             

 

CLASS

FORM

TENANCY

DATE

Burton Pedwardine

-

m

t

?E12th

Heydour

cm

rw/m

e

?

Moulton

c

moat

t/e

?E12th

Sleaford

c

rw

t

E12th

Stainby

cm

m & b

?t

?E12th

Swineshead

c

m & b

e

E12th

Welbourn

c

rw

t

M12th

Wrangle

-

m & b

t/e

?E12th

CLASS: c = castellum, cm = capitale messuagium

FORM: m = motte, b = bailey, rw = ringwork

TENANCY: t = tenant-in-chief, t/e = enfeoffed tenant-in-chief, e = enfeoffed

 

Eight of the earthwork sites surveyed were called 'castles' in the Middle Ages or can be classified as such on the basis of mottes or ramparts which make them somewhat more defensible than the ordinary moated site. A further two, Corby and Wyberton, may have had similar characteristics, but were probably little more than manor houses. The strength of defences varies. Only Sleaford was a major castle, and, with its impressive water defences, it was favourably compared with Newark in the twelfth century. The remaining Kesteven sites were more modest establishments, but, like Welbourn (unsurveyed) may have had stone keeps and curtains. The castles in Holland, by contrast, although demonstrably important, appear to have been rather slight. Moulton, for example, is merely an oval enclosure, showing no indication of a counterscarp, and it must be supposed that the fastnesses of the fens were sufficient protection in this area. Ringworks and motte and baileys are equally represented, and no significant correlations to form have been observed.

          The date of construction is predominantly the early to mid twelfth century and characteristically all, with the exception of Heydour and Swineshead, were held by tenants-in-chief. Sleaford and possibly Welbourn and Stainby were honourial castles where castle guard was rendered. Moulton and Wrangle, however, were subtenancies, but appear to have been the major residences of their lords in an area where they wished to maintain a presence but held no estates in demesne. Which, if any, were adulterine is not clear.

 

Moated sites

                                                           

SITE

DB

TENANT

SERVICE

Barrowby

M d

e C12

m

Corby

M e

 

m

Dowsby

M e

 

m

Gosberton 2

M d

d(r)

s

Gosberton 3

M d

d

alms

Hacconby

M d

d(r)

s

Haceby

M e

 

m

Haddington

B & S

?

?

Heckington

B & S

?d

f

Hougham

M e

 

m

Kyme

M d

e C12

m

Langtoft

M d

d

alms

Newton

M

 

 

Scredington

 

 

?f

Spanby

S

e C12

m, but much f

Swaton

M d

d

d

Wyberton

S

?f

?f

DB: M = manor, d = demesne, e = tenant, B = berewick, S = sokeland

TENANT: d = demesne, e = enfeoffed + date if known, f = freemen

SERVICE: m = military, s = sergeancy, f = freeholding or socage

 

Seventeen sites are simple moated enclosures which appear to have been in lay hands or were held by religious houses as part of ordinary estates from the Conquest. Their nature and function are rarely explicit in the sources consulted. Some sites like Langtoft 2 are related to livestock and the two moats at Newhall Grange in Pinchbeck were used as fisheries in the early sixteenth century. Most, however, like Barrowby, Heckington Winkhill, and Hougham, where mediaeval structures survive or are known to have existed, must have enclosed manorial residences. As such, it is possible to perceive some of their general characteristics from the nature of the estate structure with which they are associated. Only Langtoft, Gosberton 3, and possibly Swaton belonged to demesne estates of tenants-in-chief, and of the remainder nine were held by military service, two by sergeancy, and three by freemen or in socage.

          Moated sites, then, were predominantly, if not exclusively, associated with residences of tenants holding by knight's service, and would appear to be analogous to more aristocratic establishments. Indeed, many had similar functions within their more circumscribed context, for they were estate centres, and some, like Corby, may have even been defensible at a pinch. Castles, however, where not related to regional politics, usually represent institutional continuity from the eleventh century. Sleaford, Heydour, and Stainby, for example, were built on manors which were held in demesne in 1086 and had belonged to king's thanes holding by book in 1066. Moated sites, by contrast, mark a radical departure, for the phenomenon is evidently related to the manorialisation of society. This process was characterised not so much by an extension of existing seigneurial power as the creation of a new class with a novel power base. In 1066 most of the manor studied were already held by tenants, and, although only four of the seventeen holdings record the fact, many were probably held by subordinates from tenants-in-chief in 1086. Thus, at Laughton two TRE lords appear to be represented by an equal number of subtenancies in the twelfth and thirteenth centuries. However, despite the undoubted existence of halls, whether as physical structures or legal fictions, such tenants were probably predominantly ministerial, that is, were essentially bailiffs, and therefore had no hereditary claim to their lands or indeed full control of their issues - many DB tenants merely had a few bovates of demesne. It was only in the first half of the twelfth century with the territorialisation of military obligations that tenants were generally enfeoffed in hereditary fee. It is in this context that knights became resident within estates that were subject to their direct control and exploitation. None of the sites studied has been securely dated, but elsewhere it would seem that they begin to occur from the mid twelfth century onwards, and such a date would chime well with the emergence of the knight's fee.

          The moat, then, the distinctive surviving feature of many such sites, is probably as much symbolic as functional. As at Lavington, it was not a universal feature, but where physical conditions were suitable, it is often found and marks the social differentiation of the lord from the peasants who rendered services to his estate. This phenomenon was most pronounced in the twelfth and thirteenth centuries with the genesis of military tenures, but it may also be associated with later developments which signalled similar social differentiations. A characteristic of fourteenth-century society is the emergence of large manors, like Heckington Winkhill, Wyberton, and possibly Scredington, which did not owe knight's service. Typically they were composed of many small parcels of land which were held by money rents and often belonged to new upwardly-mobile freeman families (nuffies). The Astys, for example, built up an extensive estate based upon a socage tenement in Heckington Winkhill, and the Wells manor in Wyberton seems to have had a similar origin. In neither case can the moat surrounding the manor house be definitely related to the new estate, but the association seems likely, and it can be supposed that it symbolises the aspirations of the respective landlords to social standing.

 

Other earthwork sites

                                                           

 

TYPE

Burton Coggles

?garden

Frieston

?field boundary

Howell 1

?garden

Langtoft 2

rabbit warren

Lavington

manorial complex

Osbournby

fishpond

Ropsley

dam and fishpond

Somerby

road

 

The eight remaining earthworks form a miscellaneous group of garden and field related features and an apparently unmoated manorial centre. By necessity, most  have been identified on typological grounds alone since no references have been found to them in extant historical sources. However, the identity of a few, such as Langtoft 2, has been suggested by field names. It has not proved possible to date any of them.