Family and Community: Structures

We should begin by describing the variety of forms which family and community could take in the middle ages. Not only were there significant regional differences between types of family and communal organization but these types underwent important developments during the medieval period. The nature of familial and communal structures, their degrees of coherence and of independence, were all consequences of a range of influences which varied in their effectiveness. In the high and late middle ages, family structure was not determined solely by biology but by a combination of pressures and inducements, not all of which operated to the same general effect. Similarly, the nature and structure of village communities underwent major transformations in the four hundred years after Domesday Book. Population movements and shifting patterns of mobility, changes in landholding and agrarian practices, the altered role of the laity in the workings of the parish, all had consequences for communal organization and solidity.

It is now an axiom of the historical literature that the majority of peasant families or households in the high and late middle ages were neither large, which is to say more than four or five individuals, nor complex, that is, other than the two generations of parents and their children. In this the English peasantry conformed to the general experience of almost all populations since, for reasons of demography alone, the majority of households in any population are, at any one time, likely to be simple, two-generational units. This is not to say, therefore, that there were no large or complex peasant households in medieval England, but rather that their proportion relative to the majority of households was always small. The extent to which that proportion varied across time and between regions will need to be considered later.

There is only scant evidence for the structure of the peasant family and household and, unsurprisingly, it is not uniform across the periods. Eleventh - and twelfth-century sources have almost nothing to say on the size and structure of the peasant family or household, tending, instead, to identify household heads rather than the members of families. Similar problems prevail in the sources from later centuries but some listings, such as the late thirteenth-century Spalding serf lists, do survive which seem to offer direct information on family form. Records of post mortem transfers of property provided in manorial court rolls and wills also offer glimpses of family members as heirs and beneficiaries, while archaeology presents oblique views of family form through the excavation of medieval houses. However, it is the poll-tax listings of the late fourteenth century that may offer the most important insights into family size and form in medieval England. The extensive national coverage of the poll-tax lists has encouraged historians to propose typicalities of family and household forms. Allied to its wide geographical coverage is the fact that, since the poll tax was not a tax on property but on individuals, it should illustrate family and household form amongst a range of economic groups or sub-groups. More than a half-century ago, J. C. Russell proposed that the listings for the first poll tax of 1377 indicated that peasant families were small. He estimated that the average family contained 3.5 persons. Although his estimates have not been universally accepted and questions raised regarding his methodologies, the listings do suggest a preponderance of small, conjugal units. Comparison of manorial records with poll-tax data also appears to confirm Russell’s average. At Kibworth Harcourt (Leics.) the average size of households in the late fourteenth century was 3.72 persons, a calculation based upon the 1377 and 1379 poll-tax listings with extensive cross-referencing to local records. However, if the average was low, investigation of the range of household sizes suggests a significant array of sizes. At Kibworth Harcourt, the 1377 and 1379 poll-tax lists indicate that the largest households were composed of six or seven individuals, whilst the smallest contained just a single person.1

Although nothing directly comparable to the poll-tax listings exists for earlier periods, a few, scattered pieces of information suggest that the average family sizes, at least for the thirteenth century, were not dramatically different from those to be found in late fourteenth-century Kibworth. Late thirteenth-century listings of serfs on manors belonging to the prior of Spalding indicate that family size may have been slightly larger than in the later fourteenth century. The average size of households on three Spalding manors in south Lincolnshire varied between 4.37 persons and 4.81 persons. Interpretation of the source has been shown to be problematic, however, and it is possible that the calculations based upon the lists overestimate family size on these manors. That said, the figures do accord with other calculations of late thirteenth-century family size. Smith, using manorial court rolls, has estimated that the average family size on the Bury St Edmunds manors of Redgrave and Rickinghall (Suffolk) was 4.7 and 4.9 respectively, while, for central England, Razi has produced the broadly comparable figure of 4.7 for peasant families at Halesowen.2

As always such averages hide a potentially wide range of family and household sizes and, in particular, obscure the presence of a minority of larger households. As Hallam’s survey of the Spalding serf lists suggests, some households in medieval villages could have been quite substantial, with more than six persons and more than two generations co-residing. Although we should not risk treating the serf lists as ‘censuses’, the variance in number of individuals per family does suggest that a proportion of families on the priory’s manors could have been large and complex, if not so large and complex as Hallam’s initial investigations may have suggested. Most telling, however, is the high proportion of families and households that were small, composed of two, three or four individuals. At Moulton (Lines.), Hallam estimated that 58 per cent of households were smaller than his calculated mean household size, which was, itself, less than five.3 It is possible that complex households were relatively more common in areas of low population density, pastoral husbandry and where opportunities for alternative employment, which would have encouraged out-migration, were limited. In certain parts of the country, beyond the boundaries of champion England, the outlying farmsteads on moorland and rough pasture may also have conformed more closely to the complex, multi-generational household types traditionally associated with a European peasantry. The isolated farmsteads of Ashwater on Dartmoor (Devon) perhaps included two or more generations of the same family but would also have been bolstered by live-in servants, known as servants-in-hus-bandry. Archaeological evidence also encourages the view that such farmsteads may have contained complex household units.4 In fifteenth-century England, similar conditions of economic independence and de facto ‘isolation’ were also created by market forces that led to a polarized society of landed peasant entrepreneurs and landless or near-landless labourers. Peasant entrepreneurs or ‘yeomen’ in southern and eastern England constructed substantial houses for themselves and their families and employed live-in servants or servants-in-husbandry. The survival of late medieval ‘wealden’ houses, with their separated living and servant quarters, reminds us that the wealthiest members of the peasantry could afford the complexity of household structures which their poorer peers could not. But in all periods, these appear to have been exceptions and it is now generally acknowledged that complexity was not the norm.

What were the forces which promoted this polarization of family and household forms, encouraging a preponderance of small, nuclear units and a minority of larger, more complex households? G. C. Homans, writing in the mid-twentieth century, had suggested that the complex peasant household of the medieval countryside was the typical household form in the thirteenth-century countryside. He argued a case based upon an assumed relationship between the acquisition of land and the establishment of a household, what is sometimes termed a neolocal household formation system, where marriage and household formation are dependent upon acquisition of necessary resources. Where opportunities for such acquisition are constrained in some way, for example, by rules of inheritance which favour a single son over his siblings, then the opportunities for the non-inheriting siblings to establish households and families of their own are limited. In such cases, Homans contended, they may become dependents in the households of the inheriting son, thus forming complex, here, technically, ‘stem-’, families. For Homans, as for certain other writers in mid-century, family and household formation were dependent upon opportunity and, according to his view of the medieval countryside, opportunity was extremely limited and largely dependent upon demographic processes, in particular, upon the deaths that freed up land for heirs.5

Homans’s model, as more recent research has shown, cannot be applied universally, or, indeed, hardly at all, to the peasantry of medieval England. Although it is generally assumed that some form of neolocality applied in medieval rural household formation systems, it also seems evident that no single household formation regime operated amongst the medieval peasantry. Instead, it seems most sensible to acknowledge two separated forms of neolocality, commonly distinguished as, on the one hand, a ‘peasant’ or ‘niche’ system of household formation, in which individuals intending marriage and household formation had to await the availability of land, often through some demographic moment or crisis, and, on the other, a ‘real wages’ or ‘proletarian’ system, in which the individual labours to accumulate sufficient capital that will permit him or her to marry. Historians in attempting to apply these separate models to the evidence from rural medieval England have tended to give primacy to the ‘peasant’ or ‘niche’ model. This is unsurprising, given that surviving sources concentrate our attention upon issues largely applicable to that model: inheritance, dower, family property, tenant deaths, and so on. This has the added effect that our view of the peasantry is skewed towards the wealthier tenantry who also, of necessity, dominate our sources. Theirs was not, however, the only type of peasant family in medieval England and, indeed, in most parts of the country they are hardly evident at all. We will, however, begin with some observations regarding these wealthier peasants, their household formation systems and their familial ties before proceeding to consider the remainder of the peasantry, those whose lives conformed rather more closely to the ‘proletarian’ or ‘real wages’ model.

Complexity and relative ‘largeness’ of family size were features more typical of the wealthier peasantry. Landed resources provided opportunities for both heirs and the non-inheriting to gain access to land at reasonably early ages. Contrary to the early statements of Homans, failure to inherit did not, as the research of a number of historians has now shown, prevent marriage and household formation. Even if, as is also widely assumed, marriage and household formation were dependent on the prior acquisition of sufficient resources, inheritance was not the only channel through which these resources could be obtained. Opportunities for marriage to heiresses or to widows presented other points of access into landholding, while parents frequently provided non-inheriting offspring with land or financial support in the form of dowry for daughters, sufficient to establish them in married life. Marriage and the onset of child-rearing tended, therefore, to be early while the advantaged position of these wealthier peasants may have been some protection against diseases, especially of infants and mothers. Furthermore, the wealthier peasantry were more likely to employ servants-in-husbandry and to provide accommodation for dependents, in particular for retired parents. All of these factors could have boosted household size and increased complexity. Additionally, the resources of the family, which permitted the establishment of family members within the vicinity of the parental household, ensured that, over time, wide networks of kin would also be established. Familial support was, for the most part, a contingent consequence of the ‘peasant’ or ‘niche’ model; typically, the family made available resources which, in some way, could be employed to establish a separate household.

For others of the peasantry, the piecemeal acquisition of sufficient capital and/or land was achieved independently of familial support or the chance appearance of landholding opportunities. Where household formation was independent of inheritance or bequest by family members, efforts to accumulate sufficient capital were likely to be concentrated in some form of labour or service. This appears to have been especially true for the poorer sections of the peasantry. In this circumstance, which approximates to the ‘real wages’ or ‘proletarian’ model, the various forces in play tended to promote simple, two-generational households. First, mortality and fertility combined, to varied extents, to limit family size. In a neolocal household formation system, any drains on landed or financial resources, essentially the exactions of family, church, state or lordship, might interfere with individual marriage plans and effectively delay them. Attempts by parents to provide for a number of offspring, and the additional burdens of rent, tithe and taxation, may all have helped generate smaller family sizes, simply by delaying marriage. Where resources were scarce, members of the family would have been obliged to abandon the family and household and seek their opportunity elsewhere. Further, and perhaps most importantly, the decision to marry and form a household was not, in this situation, one dependent upon the timing of inheritance or the transference of land or some other resource from the family. Life expectancy amongst the peasantry, although it cannot be accurately calculated in this period, was almost certainly low, probably less than thirty years at birth, and is likely to have been lowest amongst the poorer peasantry. Fertility amongst the poorer peasantry is also likely to have been lower than amongst the wealthiest peasantry as a consequence of later marriage.

There is, therefore, an important distinction to be drawn between those families and households which could maintain such close contacts between their members and those which could not. Broadly speaking, complexity, whether of actual family or household form or of wider networks of kin, was a function of wealth. The more resources which a family could muster, the greater was the potential to establish heirs and non-heirs within the main households or its environs. The relatively well-to-do peasantry, especially the tenants of the larger units of servile land, were best placed to achieve this.

The possibility that the majority of the peasantry could provide family members with sufficient resources to permit them to remain within the main household or to establish households of their own is, however, questionable. E. A. Kosminsky, in his analysis of the late thirteenth-century Hundred Rolls, found that for a block of counties in central England, the vast majority (c.75 per cent) of free and unfree tenants did not hold as much as a standard holding or virgate and that a significant proportion of these held less than five acres (perhaps 30 per cent of the unfree and a greater percentage of the free tenantry).6 Instead, it is likely that, throughout the period for reasons of land-hunger (notably, thirteenth and early fourteenth centuries) or polarization of landed resources (fifteenth century), a large proportion of the peasantry had to look elsewhere than to their own families to accumulate the resources necessary to establish families and households of their own. This is a point to which we will need to return when we consider the function of the family.

Finally, shifts in these variables over time effected changes in family and household form. The period between 1100 and 1500 witnessed dramatic movements in population, a general increase in levels of trade and commercial activity, and the growth of an urban sector. Additionally, there was an overall decline in the strength of lordship and its claims upon a servile tenant population. Alongside the decline of serfdom, there is evidence of increased self-determination in other facets of peasant life, particularly in terms of labour choices, mobility, religious expression and consumption. In reviewing these developments, historians have spoken of a period of transition from feudalism to capitalism and of the rise of the individual as entrepreneur and as free agent within a money economy. Most importantly, despite the abundance of land in the post-plague countryside, such developments prompted a process of polarization in landholding and an effective proletarianization of the countryside which is likely to have increased the proportion of small, nuclear households ill-supported by more distant networks of kin. Not all of these developments occurred with the same force in every part of the country. In particular, the growth of a market economy appears to have had a more significant impact in southern and eastern England than it did in the highland zone of England where long-established forms of landholding and social structure persisted into the early modern period.

To move from discussion of the medieval family to consider the ‘structure’ of the village community is, in part, to turn our attention to the physical layout of the village. The function of the village community, which will be discussed below, was, to a large extent, a response to the geography of the village. But the cohesion of the village community, its energy and coherence, were also the products of other variables, not the least of which were the nature of land-use and of lordship. As with the family, it is, consequently, the variety of village forms, husbandry and lordships that impress, and not the ubiquity of a single form. But also, as with the family, one form in particular has tended to dominate our view.

The ‘traditional’ village of the medieval countryside has standard elements: the village plan is nucleated, with the village houses concentrated upon a central road, with a church and manor house at its core. Beyond the village, two or three great, ‘open-’fields are given over to arable, with one field fallow in a two - or threefold course of rotation. Each of the open-fields is divided into strips and villagers hold their land as composite units of these strips. The lord of the manor, in this standard village type, manages part of the land directly, the demesne, whilst extracting rent, some of which would be in the form of labour, from his tenants. Again, in this traditional model, the manor, the unit of lordship, and the village, the geographical unit, are co-terminous. Finally, the village and the parish are synonymous, with the parish church situated at the village’s centre. Villages such as these did exist in the medieval countryside.

In central, southern, midland and northeastern England, it is possible to find examples of settlements consistent with this model. The manor of Cuxham (Oxon.) provides a closely studied example of a nucleated village, surrounded by open-fields and held by a single lord.7 By the close of the middle ages, Cuxham was held by the Fellows of Merton College, Oxford. From the eleventh century until the late thirteenth, the manor had been in the possession of a succession of minor noble families. By the end of the thirteenth century, when sources permit close investigation, the unfree or servile tenantry were composed of thirteen ‘cottagers’ and thirteen villeins, the latter holding twelve acres or half of a standard holding, or virgate. There were also two holdings held by free tenants of the manor. The cottagers paid most of their rent in the form of money but also performed some minor labour services for the lord. By contrast, the villein tenants paid the bulk of their rent as labour, each villein providing about two days’ work per week for the lord. At Cuxham, the houses of the villagers appear to have followed the course of the main road. P. D. A. Harvey’s reconstructed map of the village suggests that the tenants of the larger villein holdings and the cottagers, with their smaller tenements, occupied distinct parts of the village. At the centre of the village were the rectory, demesne farm and church. The village was surrounded by three open-fields, subject to a three-course rotation of crops. Each of the open-fields was divided into furlongs and the furlongs subdivided into strips. While the lord’s portion of the arable at Cuxham was held as single blocks within the open-fields, tenant land was divided into strips, each of which measured no more than an acre. Unlike the demesne, the lands of the tenants were scattered throughout the three fields, their identity marked by boundary stones. While the tenants do not appear to have grown exactly the same crops as those grown on the lord’s demesne, they do appear to have grown similar produce to each other, with a predominance of rye and vetch. Beyond the arable of the open-fields, to the north of the parish, was meadow, which was also let in strips to the villagers. Most tenants had a few head of livestock, perhaps just an ox or a horse, a cow and a pig.

Such arrangements of village, lordship and land were familiar features of the medieval countryside. The physical ordering of the village, the expectations of lordship and the organization of land-use all speak of some communally organized effort. It is in the very nature of open-field agriculture that individuals cooperate: without collective activity and agreement such a system cannot function. Similarly, lordship has also organized the village in ways that seek to maximize the benefits of this collective enterprise whilst, to a degree, providing some measure of protection and security for the labour force. It has, in fact, been suggested that villages such as Cuxham were the planned products of lordship in the decades either side of the Norman Conquest: an artificial community constructed for collective endeavour and seigniorial benefit. In the next section of this chapter, we will return to the nucleated village to consider more closely the nature of the village community in such settlements. Before we do, we need to consider the variety of rural settlements that did not conform to the model simplicity of the nucleated village.

Even within the boundaries of champion England, the nucleated vill co-terminous with the manor and the parish was not the only village form, nor was open-field agriculture the only type of land-use. Villages within champion England did not all display the neat focus of a Cuxham. In Oxfordshire, in the heart of champion England, the physical layout of a significant number of modern villages suggests that their medieval antecedents were not composed of a single core but were, instead, agglomerations of smaller hamlets and farmsteads. These ‘polyfocal’ villages, such as Hook Norton in the north of the county, were characterized by divided lordships and irregular field-systems with a combination of open-fields and some enclosed arable.8 Furthermore, not all lordships operated in ways similar to that found at Cuxham. There was, on many manors, no demesne to speak of and the tenantry were certainly not everywhere divided between free tenants, cottagers and villeins owing labour services. Neither was there, as appears to have been the case at Cuxham, a neat uniformity of landholding. Topography, early patterns of settlement and land-use, the nature of local lordship, the influence of local and regional markets: all of these helped determine the structures of communities and produced significant variations, even within regions where standard forms appear to have dominated.

When we move beyond the boundaries of champion England, however, the nucleated settlements and, especially, the open-fields tend to disappear from view to be replaced by hamlets, dispersed and isolated settlements, areas of ancient enclosure, pasture, woodland and moorland. In the south-west, for instance, in areas of moorland and pasture, hamlets and farmsteads were most typical. Similarly, in the far northwest, whilst the lower-lying settlements, many of which were established after the Conquest, conformed in certain respects to the villages of the lowland south, a process of colonization of the higher moorland produced isolated farmsteads and dispersed settlements. In the east of the country also, a system of land-use which included anciently enclosed fields and wood-pasture also encouraged a degree of dispersed settlement. In Essex, for example, only the far north-west of the county was identifiably open-field countryside in the middle ages. The rest of the county tended to hedged fields and coppiced woodland. Whilst settlement in the Essex countryside could conform to the nucleated villages found in central England, a significant proportion of the inhabitants lived in much smaller settlements dotted amongst the woodlands and fields. Hamlets and smaller moated settlements are standard features of this countryside. At Cressing, in central Essex, the parish contained a number of distinct settlements, none of which were particularly large. It was densely settled countryside of moated sites and small hamlets formed around greens.9

Within most of these regions, there were significant differences and variations produced by the vagaries of topography and of geology. In particular, distinctive natural features promoted agrarian practices and rural industries which, in their turn, effected developments in the size, layout and nature of the communities in which they were situated. Areas of pastoral husbandry, such as the wool-producing uplands of the Cotswolds, the downlands of the south coast or the cattle-grazing regions in the west of England, prompted distinctive forms of land-use, communal and agrarian organization. The heavily wooded or forested areas of the countryside, such as the Forest of Dean, militated against the close nucleation of the villages of the central midlands. The terrain, as well as the demands of labour and of lordship, ordered the lives of medieval men and women in ways that were also distinctive. The pastoral economies of western England imposed a range of employments upon rural dwellers and regulated their lives differently from those who worked in areas that were predominantly arable. The management of herds and flocks, their husbandry and marketing made specific demands of the labour force. Similarly, the inhabitants of areas which were largely forested and wooded lived their lives according to the dictates of their byemployments: coppicing, charcoal-burning, hunting and so on.

Certain parts of the medieval countryside were also home to industry which led to further diversities of community scale and structure. Lead-mining in Derbyshire and the Pennines and tin-mining in Cornwall engineered distinctions within the local rural communities and their economies. Miners used the profits of their labour to buy into local land and, in so doing, effected changes in the balance of local communities. The same could be said of the cloth industry in the countryside. In the principal areas of the cloth industry in the late medieval countryside, in-migration swelled rural populations while the capital generated by the industry encouraged a stratification of wealth and a significant range of landholdings. In the late fourteenth and fifteenth centuries, for instance, villages such as Castle Coombe (Somerset) or Lavenham (Suffolk) were transformed into significant towns by the presence of the cloth industry and of those who had benefited from participation in it.

As well as topography and geology, geography also dictated difference. Distance from or proximity to particular features of the medieval political and economic landscape influenced the structure and organization of rural communities. Those rural settlements within the hinterland of larger towns or cities also found themselves drawn within the economies of those urban centres, both as providers of produce for their markets and as sources of investment for wealthy townsmen. The gravitational pull of these towns and cities also drew country-dwellers away from their own communities. Villages and other rural settlements within urban hinterlands are likely, therefore, to have been less stable than were more isolated settlements. But it was not only towns that exerted such influences upon the countryside. Communities could be moulded and refashioned by a range of other factors peculiar to certain regions. In those parts of the country which were more vulnerable to political disturbance and warfare, the tenor of community life and the physical ordering of communities were affected accordingly. The countryside of the Scottish borders, an area which saw severe disruption from Scottish raids in the high and late middle ages, was characterized by its fortified settlements and an unstable population. The physicality of community in the more remote reaches of the northern borders must have been something wholly distinctive from that to be found in central and southern England.

As with the family, no single chronology will serve to account for developments in the village community. Local and regional variations in population levels, climate, soil quality, lordship, land-use, the urban sector, trade and politics ensured that, as in the case of developments in family and household form, changes in the form of rural settlement were also not uniform. Historians have tended to account for particular developments and shifts according to their own predilections as to the driving forces of historical change. Population movement and economic shifts feature large in most historical accounts of changes in patterns of settlement. Most obviously, the decline and, on occasion, wholesale abandonment of settlement in the post-plague period have been described in terms of demographic stagnation and migration from the countryside to the towns. But this is not a model that can be consistently applied: whilst certain reaches of the countryside declined in the fifteenth century, others prospered. Relocation of industry into the countryside, itself a response to shifts in international trade and politics, caused the population of villages in advantaged areas to swell. In earlier centuries, sudden dramatic shifts caused fundamental changes in social structure and settlement. The post-Conquest harrying of the north or the Scottish raids of the early fourteenth century affected patterns of settlement and threatened the existence of whole communities in ways that were utterly alien to the experience of communities elsewhere in the country.