The Household

views updated

THE HOUSEHOLD

Richard Wall

This essay assesses the multifaceted character of the household as a residential and social unit. Later sections of the essay consider the role of economic, demographic, and social factors in shaping household forms, and discuss the strengths and weaknesses of the various attempts that have been made to map the variation in household patterns across Europe.

The household is usually defined as a residence unit. The members of a household include all persons, whether or not they are related to each other, who share a clearly defined living space or dwelling. The household is in this way clearly distinguished from the family, whose members are related to each other, however distantly, but do not necessarily coreside. Households can also have other attributes. Members of the household (or some of them) may pool their incomes, eat communally at least once a day, and earn their livelihood from working together to exploit assets rented, leased, or owned by the household, such as a farm or workshop. Other ties may develop from this level of cooperation: a sense of mutual dependence among the members of the household, respect for the authority of the household head, and a desire to preserve the privacy of the physical space occupied by the household. The household may also undertake the socialization of the young and afford shelter to members of the local community (primarily relatives but in some cases also nonrelatives such as foster children and the childless elderly) unable to provide for themselves from their own resources. Vestiges of the religious and judicial functions of the household lingered from medieval times.

CONCEPTUALIZATIONS OF THE HOUSEHOLD

The multifaceted character of the household has been a major source of its appeal to economists, sociologists, and social geographers as well as historians. The household has attracted attention both as the unit which ensures the reproduction of the labor force and the unit through which capitalist economies pressurize individuals as workers. The household has also been viewed as the locus of many of the relationships between men and women, where much work, domestic labor, and child care is undertaken. Different theorists have stressed some functions of the household at the expense of others. For Michel Verdon it is the criteria of residence; for Kathie Friedman it is its role as an income-pooling unit. This leads the latter to redefine the household to include persons who live elsewhere who contribute to its economic well-being, thereby identifying as the significant unit in societies not those persons who live together but the wider group of persons who share (some) of their income. As Diana Wong has pointed out, such an approach assumes that this support network provided generalized support according to need rather than, as was usually the case, limited assistance in exceptional circumstances. Inequalities in access to the resources of the household on the part of husband and wife, parents and children, master and servant, are also ignored.

Nevertheless, many households are not autonomous economic and social units. For example, Martine Segalen has documented for the west of Brittany in the nineteenth century the extent to which networks of neighbors as well as kin, smallholders as well as farmers, cooperated in a range of labor-intensive tasks such as ploughing, harvesting, and threshing. Such networks also channeled information, offering the individual material, social, cultural, and political privileges and introductions to a potential spouse. Euthymios Papataxiarchis has argued that on the Greek island of Lesbos in the nineteenth century, mutual aid, care of children, and even significant interpersonal relationships extended beyond the household. Women's ties were with their families of origin, while the men had their (exclusively male) social groups. Variation can be expected in the strength of the social ties which united members of different households according to area of residence, time, the nature of the particular household economy, and the social standing of the household, but too little research has been completed to provide a wide-ranging comparative perspective.

The situation is considerably better as regards the economic ties between households. Analysis of the time budgets collected by Frédéric Le Play and his followers indicate that in Europe in the middle and later nineteenth century more than half of young couples working as peasants, artisans, tenant farmers, and laborers had received substantial financial support from their parents or parents-in-law either at the time of their marriage or later. In most cases, and in particular in western and northern Europe, this support was received without the necessity to coreside. Parental support was forthcoming less frequently in England, the Low Countries, Spain, Germany, Switzerland, and Paris than in other parts of Europe. Underlying these variations were differences in the nature of the household economy. Couples who were peasants or smallholders were much more likely to receive financial support from their parents than were couples who were factory workers or laborers. Other than parents, more distant relatives, employers and landlords, and other nonrelatives also made significant contributions to the standard of living of persons resident in other households.

In practice it is no easy matter when working with historical documents to identify an unambiguous, let alone a consistent, definition of the household. The lists of inhabitants which furnish the information on family and household patterns in England prior to the official censuses of the nineteenth century almost without exception fail to provide any definition of how individuals were set out in groups, whether separated by lines, spaces, or numbered consecutively. The most detailed of these lists have been identified as listing households on the basis of the information provided on the members of these groups: spouse, sons and daughters, other relatives, and servants were all described in relation to the person listed first in the group. This also has the advantage of providing a measure of consistency with the definition proposed in the British Census of 1851, where the household (referred to as the family) was defined as follows: "The first, most intimate, and perhaps most important community, is the family, not considered as the children of one parent, but as persons under one head; who is the occupier of the house, the householder, master, husband or father; while the other members of the family are, the wife, children, servants, relatives, visitors, and persons constantly or accidentally in the house" (quoted in Wall, 1972, p. 160). Even so, total consistency is not assured, as the British censuses of the nineteenth century were conducted on a de jure basis, recording all persons resident in a household on a given night, whereas the precensus lists of inhabitants registered only the de facto population, the habitual residents of the household.

One of the severest problems encountered by historians seeking a coherent definition of the household is occasioned by the presence of lodgers and other nonrelatives who rented their accommodation from the principal household but budgeted separately. Some relatives and even adult children may occasionally have been in the same situation. On a definition of the household which focuses on income pooling, they should logically be considered as constituting separate households. Conversely, on a definition of the household based on common residence they should be counted as members of the same household unless known to be occupying separate living space. Most nationwide surveys of living arrangements in present-day Europe attempt to give equal weight to the common dwelling and common housekeeping in their definitions of the household, as recommended by the United Nations.

EUROPEAN FAMILY AND HOUSEHOLD SYSTEMS

If the way households function and even the way a household is defined are embedded in a particular culture as well as in a particular economic system, considerable care is required in comparing the structure of households across both time and space. Some scholars have argued that all comparative work is flawed, as it simplifies and distorts social realities by using the criteria of areal coverage and prevalence in conjunction with the selection of supposedly objective standards or norms to determine the significance of a social structure. Classifications of household types, David Sabean has declared, are useless unless they take into account power relationships within and beyond the household and the networks which linked household with household, family with family, and individual with individual—an approach that can only be achieved through the study of a particular locality. The contrary view is that each local or regional study needs a wider comparative survey to place it in context. This is the approach adopted here. The loss of precision and context that this entails is admitted, as is the fact that similarity in household structure of populations from different time periods or regions is not to be taken as evidence that social relations invested in that structure were necessarily identical.

There have been a number of attempts to map the variation in household forms across Europe. In a two-stage process, first the distinguishing features of the household system are arbitrarily determined and then, second, the regional prevalence is measured. For John Hajnal, building on his earlier work on European marriage patterns, the key aspects of the household system of northwest Europe in the seventeenth and eighteenth centuries were a late age at first marriage by men and women (over twenty-six for men and over twenty-three for women), the immediate assumption by the newly married couple of the headship of a household, whether newly formed or a continuation of the parental household (in which case the parents would retire), and the circulation of young people between households as life-cycle servants prior to marriage. Northwest Europe as defined by Hajnal encompassed Scandinavia (including Iceland but excluding Finland), the British Isles, the Low Countries, the German-speaking area, and northern France.

Hajnal's rule about headship runs counter to Peter Laslett's earlier (and later) conceptualizations of the west European family system, which envisaged the formation of an independent household on marriage as one of its key characteristics. In his 1983 paper "Family and Household as Work Group and Kin Group: Areas of Traditional Europe Compared," Laslett also argued that it was possible, using a broader set of criteria as the defining characteristics of each family and household system, to identify four, rather than two, distinct family and household systems that were dominant in, if not entirely exclusive to, particular areas of Europe: northwest, west and central, Mediterranean, and east. These criteria included, in addition to household formation rules, procreational and demographic characteristics and the types of kin present in the household, as well as aspects of the role of the household in the area of work and welfare. The effect, possibly unintentional, was to anchor each household system more firmly within a particular economy and broader social structure, the latter reflecting in particular the extent and nature of community and state support for disadvantaged groups within the population.

According to Laslett, the family and household system of northwest Europe was distinguished not only by the formation of new households at the time of marriage, a late age at first marriage, and the predominance of simple-family households (households consisting of couples with or without unmarried offspring or lone parents and unmarried offspring) but by the rarity with which households functioned as work groups and the presence of households which received a large part of their income in the form of transfer payments from the community. By contrast, the family and household system of central Europe contained a large proportion of stem-family households (where a married son, on his marriage or later, continued the parental household by succeeding his father as the household head), and many households were work groups. In other respects, Laslett argued that the household systems of central and western Europe were similar. The household systems of Mediterranean and eastern Europe shared with central Europe the association between the household and the work group but departed from its other features through higher proportions of complex households (particularly in eastern Europe), early ages at first marriage (for both sexes in eastern Europe; for men only in Mediterranean Europe), and absence of the link between marriage and the formation of a new household.

Hajnal's and Laslett's delineations of European family and household systems have now been challenged from a number of quarters. One concern has been the fluidity of the boundaries between the various regions. Boundaries between "systems" might also shift over time. During the course of the eighteenth and nineteenth centuries the household system of Hungary evolved from the simple-family household system as in northwest Europe toward more complex structures in the face of land scarcity in one part of the country and labor scarcity in another. In the latter part of the nineteenth century, as the economy of Corsica deteriorated, extended- and multiple-family households came to predominate in place of less complex households. Other societies moved in the reverse direction. For example, in the southwest of Finland simple-family households increased at the expense of complex households during the late eighteenth and nineteenth centuries in response to legal reforms which permitted the division of farms and the formation of households by the landless. After 1850 the trend toward more simple household forms was reinforced by innovations in methods of fishing which reduced the amount of capital and labor required. Simple-family households also replaced complex-family households for reasons still uncertain in some parts of Sweden during the eighteenth century. Instances in which, in different parts of Europe between 1750 and 1950, there were fewer complex households than fifty years earlier were almost matched by instances when there were more complex households later on. In some populations (Hruni, Iceland, and Cuenca, Spain) a trend toward more complex households is even evident after 1900, although increasing complexity of household structures was most in evidence in the nineteenth century.

A second challenge to both Hajnal's and Laslett's conceptualization of marriage and family patterns has involved a search for inconsistencies within the defining characteristics of a particular family system: for example, signs of the presence of a late age at first marriage in conjunction with low proportions remaining unmarried and a high proportion of complex households, or, alternatively, of an early age at first marriage coexisting with a preponderance of simple-family households. Such evidence has been duly produced, particularly from Italy, making it difficult to maintain that there was just one Mediterranean family pattern. It is clear that the variability is too great to be accommodated within one household system, even with a generous allowance for the fluidity of boundaries between systems and the presence of marriage and household patterns incompatible with the characteristics of the household system of which they were supposedly part, as argued by Laslett.

An even more fundamental attack on the premises of the conceptualization of the northwest European household system has been mounted by Daniel Scott Smith. According to Smith, two of its key characteristics, a late age at first marriage and a high rate of permanent celibacy, were not intrinsic elements of the family system but the product of external constraints. Whenever there was an open frontier, as in North America, age at marriage and the proportions of never married fell below the levels associated with a northwest European household pattern, leaving only the establishment of a new household on marriage as the defining characteristic of the system. Yet it is possible to show that even this principle might be violated at times, such as when economic circumstances, in the form of a shortage of housing at a suitable price or the need for young married women to seek employment outside the home, enforced the coresidence of relatives outside the immediate nuclear family of parents and unmarried children. Smith also envisages, as does Michel Verdon, a universal preference for small and simple households. Households, they argue, would always adopt this form but for the existence of a variety of constraints which prevent such preferences being implemented.

Demographic, economic and social change, particularly in the twentieth century, has had a profound effect on household forms. One such change was the fall in fertility which substantially reduced the size of the average household during the first half of the twentieth century. Rising living standards, in conjunction with an increased preference for residential independence on the part of both the elderly and their adult children, has also reduced the frequency of multigenerational households since the end of World War II. For the same reason, boarders and lodgers, so common in households in western Europe in the nineteenth century, all but disappeared during the twentieth century. Instead of living with relatives or non-relatives, many more persons at the beginning of the twenty-first century lived on their own, in a one-person household. Some changes, of course, occurred earlier, such as the decline in demand for male farm servants throughout much of England in the late eighteenth century. However, the reduction in demand for female domestic servants, and the willingness of young women to undertake such work, can be dated, at least for England, to the first half of the twentieth century.

THE GEOGRAPHY OF EUROPEAN HOUSEHOLD FORMS

Tables 1 through 3 provide a more detailed perspective on family and household patterns in the European past by setting out the variation in the proportions of extended- and multiple-family households and in membership of the household defined by relationship to the household head. Table 1 measures the variation in the proportions of extended and multiple households in the middle of the nineteenth century, when data are most plentiful. Extended-family households include both a family group (couple with or without unmarried offspring or lone parent with unmarried offspring) and other relatives such as a parent, sibling, or grandchild. If these relatives themselves constitute a family group (couple or parent and child), then the household is classified as multiple. In mid-nineteenth-century England, as table 1 indicates, there were almost five times as many extended-family households as multiple ones (the average frequency [median] for the eleven populations is 14 percent against 3 percent). Just under one in six households were complex: i.e., either extended or multiple. The range in values was also considerable: 1 to 7 percent for multiple family households, 11 to 16 percent for extended households, and 12 to 21 percent for complex households. This, then, is the English experience behind Laslett's suggestion that northwestern Europe in the past had very low proportions of multiple-family households. There is, therefore, some justification for the claim of "very low" proportions of multiple-family households, given a maximum 7 percent of households of the multiple type in mid-nineteenth-century England. However, the fact that up to a fifth of households in some English populations in the middle of the nineteenth century were complex must raise doubts about the claim that the proportions of complex households were "very low."

Nor are these patterns particularly distinctive. Multiple-family households were equally rare in some French populations: as, for example, in Montplaisant, although located in the south of the country, as well as in a number of populations in northern France. There were also Spanish, Italian, Swedish, and Icelandic populations with as few multiple-family households, and even some populations from southwest Finland with no more multiple-family households than in the English populations with the highest frequency of multiple-family households of the eleven English populations. On the other hand, there were other French, Spanish, Italian, Icelandic, Swedish, and Finnish populations with proportions of multiple-family households far in excess of the experience of any of the English populations. What is therefore most distinctive about the English experience is its uniformity, relative to the variation in household forms occurring in other parts of Europe.

The division of Europe into distinct familial regions as proposed by Laslett—northwest, central and middle, Mediterranean, and east—also looks problematic. Most distinctive, and with very high proportions of multiple and complex households, are the populations of eastern Europe. Yet even in this instance, multiple-family households occur almost as frequently in some of the northern Italian populations (and in higher proportions than on the Linden Estate in Kurland, Lithuania), and proportions of complex households present were as frequent in western districts of Finland. Mediterranean populations look particularly diverse, as others have noted. In the middle of the nineteenth century, there were fewer extended-family households in southern Italian populations than in England (although probably more multiple-family households), whereas in northern Italy the proportions of multiple-family households were close to those of eastern Europe. Household patterns in the Nordic countries were also extremely variable, with several populations from Iceland and Sweden not conforming to the tenets of the northwest European household system, although placed by Hajnal within the ambit of this system for the seventeenth and eighteenth centuries. Unfortunately, we cannot proceed further with delineation of the sphere of influence of west (or northwest) European family patterns, as the selection criteria used to produce table 1 resulted in the inclusion of only one population from Germany and one from Switzerland, and none at all from Ireland, the Netherlands, Belgium, and Denmark. Other studies indicate, however, that rural households in England contained as many relatives as did households in Denmark and Flanders and more relatives than households in the Netherlands. We may reasonably infer, therefore, similar proportions of complex households in Denmark, Flanders, and England (lower in the case of the Netherlands). By the same token, households in nineteenth-century Ireland were considerably more complex.

Rural households. Table 2 sets out the membership of the household in a number of rural populations enumerated at a variety of dates between the late sixteenth and early nineteenth centuries, with membership of the household expressed in terms of the number of persons present of each type: heads of household (married and nonmarried), offspring, relatives, servants, and, finally, any other persons not known to be related to the head of the household. The number of persons of all types found within the household varies considerably, and the variation would no doubt be greater if it had been possible to include more populations from southern and central, let alone from eastern, Europe. In the case of servants, for example, in the selected rural populations the range is from 118 per hundred households in Iceland in 1703 to fewer than 20 per hundred households in the countryside around Gouda in the Netherlands in 1622, in certain Swiss communities between the mid-seventeenth and early eighteenth centuries, and in the Spanish province of Cuenca in the eighteenth century. For offspring, the range is from 279 per hundred households in Egislau in Switzerland, in the area around Gouda, and in West Flanders in 1814, to 157 per hundred households in Cuenca in 1724.

Text Not Available

Gauging the significance of the variation is more difficult but can be considered from the following points of view. In the first place there is the question of the smoothness of the distributions when the populations are placed in rank order from those with most offspring, kin, or servants to those with least. Three populations, for instance, stand out as having an above average number of servants (Iceland, Denmark, and West Flanders). Then follow a number of populations with more moderate numbers of servants (Norway, west Nord Brabant, and England) and finally three populations with very few servants (the Swiss communities, Cuenca, and three of the rural areas in the Netherlands). In a similar vein, populations can be identified where very few households were headed by nonmarried persons (Denmark, West Flanders, and the Swiss communities) or contained a large number of unrelated persons (Iceland and to a lesser extent Norway). On the other hand, there are very few occasions

Text Not Available

when any one population is sufficiently distinctive as regards a particular component of the household—for example, the number of its offspring, relatives, or servants—to stand apart from all other populations. Of the various populations examined so far, the most distinctive in view of the large number of servants and other unrelated persons is Iceland, but even Iceland may come to look less distinctive as investigations of other European populations are completed.

Indeed, already the differences between Iceland and the rest of Europe look quite modest when set alongside the structure of the household in some non-European populations—that of India, for example, where there were 122 relatives per hundred households in 1951, and the Russian serfs of the nineteenth century, with their 520 relatives per hundred households. Nevertheless, it is obvious that there has been considerable variation in household structure even within the confines of northern and central Europe. For example, the Danish, Norwegian, and West Flemish populations had many married household heads and many offspring and servants, the number of offspring being boosted in West Flanders by the frequency with which both widowers and widows remarried. The populations of rural Holland stand out on account of the relative rarity of kin and servants in the household. The distinctiveness of Iceland, on the other hand, as has already been mentioned, was due to the large number of servants and other unrelated persons attached to its households. Finally, a fourth household pattern may exist, exemplified by the relative frequency with which nonmarried persons headed households. This pattern is found in England, west Nord Brabant, and Cuenca. Iceland, it will be noticed, also had many nonmarried heads of household.

Urban households. A place may also have to be reserved for a European urban household. Table 3, using the same classification scheme as table 2, shows that, in general, urban households were less likely than rural households to be headed by a married couple and that they contained fewer offspring but more relatives and many more unrelated persons, many of whom of course would be the lodgers and boarders traditionally associated with town life. Overall, urban households, even including lodgers, were generally smaller than rural households. As with the rural households, however, there is also evidence of considerable variation from place to place. The households of the inhabitants of Norwegian towns, for example, were most likely to have married couples as heads. Households in Bruges, Gouda, and Zurich were more likely than those of other towns to contain offspring. Relatives, other than members of the head's own nuclear family, were most often to be found in the households of the inhabitants of Bruges and Fribourg. Servants turn up most frequently in Norwegian and Swiss towns and in Konstanz, and unrelated persons in Rome, Bruges, and Fribourg. Yet despite this variation in the composition of the urban household, the association of specific types of households with particular towns is not an easy task. In part this reflects the very fragmentary nature of the evidence currently available. London, for example, is represented only by one of its central and wealthier parishes, Rome by a handful and variable number of parishes at different points in time, and even Southampton by only half of its parishes. This should increase the measured degree of variation from area to area, yet the reality is that households from different urban populations seem to differ somewhat less in certain key respects (in numbers of households with married and nonmarried heads and in the number of offspring they contain) than do households from different rural populations. Of the towns and sections of towns covered in table 3, only those in Holland really stand out on account of their low numbers of relatives, servants, and other unrelated persons in their exceptionally small households.

Text Not Available

Sex ratios. The emphasis on variability is reinforced if we consider the sex ratio of particular categories of people within the household, nonmarried heads, offspring, relatives, servants, and other unrelated persons in the rural populations of Europe. As might be expected, through women generally being younger than spouses on marriage and generally outliving them, most of the nonmarried persons heading households were women. Even here, however, Iceland provides an exception, while two of the three Swiss communities lie at the other end of the distribution, with more than three times as many nonmarried women as nonmarried men heading households.

Sons and daughters who resided with their parents were usually present in almost equal numbers (a sex ratio of around one hundred). Any marked departure from a sex ratio close to a hundred in a population of any size would indicate either a mortality differential by sex or, most probably in these populations, an earlier exit from the parental home either by sons or by daughters. In rural populations, other factors being equal, any desire to retain male family labor in farming and keep the heir in residence, assuming the heir was, by preference a son, would tend to raise the sex ratio. However, the effect on the sex ratio of the entire offspring group (as opposed to offspring over the age of ten) is likely to be muted since the vast majority of offspring would be of an age when both sons and daughters would normally still be in the parental home. Nevertheless, in the case of England, where local censuses giving ages have been analyzed, it emerges that prior to the late eighteenth century it was sons and not daughters who were first to leave the parental home.

Whether this is the same elsewhere would merit investigation. What is already evident is that the majority of the young rural labor force recruited in the form of servants was male. Many of these servants, of course, were the offspring who were "missing" from the homes of their parents. The surplus of male servants shows up strongly in West Flanders in 1814 and in England before 1750. However, the surplus is less marked in Denmark at the end of the eighteenth century and is reversed in two of the Swiss communities and in Iceland, indicating considerable differences in the way in which these societies used service as a source of labor. Comparable differences occurred in the sex ratio of the groups of related and unrelated persons in the household. In a number of the populations, for example, such as rural Denmark in 1787, female relatives outnumbered male relatives by more than two to one, whereas in West Flanders there were considerably more male relatives than female.

Text Not Available

A glance at the sex ratio of the urban populations suffices to show that females predominated in the majority of towns that it was possible to examine (few data sets, unfortunately, were available). Females were generally in the majority among the nonmarried heads of households and among relatives, servants, and other unrelated persons to a much greater extent than was the case with the rural populations. The effect of this excess of females (in Bruges, for example, there were only six males over fifteen to every ten females) on the economic and social life of certain towns was considerable. Through their preponderance in the population these women made a major contribution to the economic vitality of these towns, and their networks of contacts with other women, both relatives and nonrelatives, were an important feature within the social structure.

Not all city populations, of course, were like this. The City of London parish of St. Mary Woolchurch, for example, had a marked surplus of men among the nonmarried household heads. However, of all the urban populations examined, it is those from Rome which are most distinctive on account of the relative preponderance of males in all constituent parts of the household, at least in the first two of the periods studied (1650s and 1700s). In the case of Rome, there is no reason to doubt the representative nature of these results since year by year through the course of the seventeenth century a marked surplus of males was recorded in the total population of the city.

Text Not Available

THE FORCES SHAPING EUROPEAN FAMILY AND HOUSEHOLD SYSTEMS

To explain the essential elements of the family system at the earliest point at which its workings can be observed, different scholars have pointed to the significance of a broad range of economic, cultural, and demographic factors. Economic forces have probably commanded the greatest attention in the attempts to explain historical family structures. In contrast, contemporary household structures, particularly supposedly new forms such as nonmarital cohabitation and "living apart together," as well as the increased numbers of lone parents and persons living alone, are seen as the result of the exercise of personal choices on the part of those concerned—in other words, as cultural preferences, although with the economic wherewithal to live in the desired way taken as a prerequisite.

The capacity of economic forces to shape family and household patterns is self-evident and can take a multitude of forms. For Pier Paolo Viazzo and Dionigi Albera, environmental factors, particularly the varying labor requirements of different mountain communities of northern Italy, explained the variations in their demographic and family patterns. The significance of the local labor market is also stressed by Michael Mitterauer for Austria, by John Rogers and Lars-Göran Tedebrand for Sweden, and by James Lehning for the Loire region of France, among many others. More generally, economic factors appear to underpin the marriage and household patterns of northwest Europe. According to this scenario, the timing of marriage and the formation of a new household were postponed until a suitable farm became available, sufficient savings had been accumulated, or an appropriate skill gained on the labor market. In eastern Europe, on the other hand, the serf owner, together with the village community, occasioned the formation of complex households.

In societies where land was a key resource, a shortage of land, whether as a result of population growth or landlord restrictions on land use, could occasion the formation of more complex households. However, greater security of tenure could also encourage the formation of more complex households, even when the economic situation of the farming population was, relative to other sections of the economy, in decline, as in Hruni, Iceland, between 1880 and 1930. Hruni provides a particularly interesting example, as early in the nineteenth century the structure of households in Hruni had became less complex during a period of severe economic hardship resulting from disruption to trade and depleted catches of fish, even though the crisis was less severe in its impact in Hruni, an agricultural parish, than in parishes whose inhabitants depended on fishing.

The role of demographic factors as determinants of family patterns is also very evident. High mortality limits the opportunities for parents to coreside with their adult children. A rise in life expectancy at older ages, a rise greater for women than for men, as in the twentieth century, increases the numbers of persons at risk of living on their own. The growth of population may also strain the existing family system, directly or indirectly: directly by forcing parents to export children to the grandparental home, indirectly by promoting the subdivision of landholdings, thereby making complex households less viable. The demographic impact of male migration, rather than a set of inheritance rules, is cited as the factor occasioning the presence of extended households in Lanheses, Portugal.

Much more difficult to identify with any degree of precision are the norms and expectations influencing residential choices. However, most interpretations of historical household structures, even while according preeminence to economic factors, have awarded at least a minor role to cultural forces. For Mitterauer, for example, cultural forces limited the explanatory power of eco-types (local economies that suited the topography) as determinants of household patterns in the extreme east and west of Austria. According to Inez Egerbladh's account of the family patterns of landed peasants in coastal areas of northern Sweden, cultural influences such as a strong regional church helped to shape family patterns within the context established by demographic and economic factors. Cultural influences lie embedded in the ways in which property is transferred between generations and in whether the care of the elderly is assumed almost entirely by the family or is shared with the community.

Admittedly, there have been some dissenting voices. Smith and Verdon have separately argued that there was a natural preference in historical populations for the formation of simple, noncomplex households. Smith saw it as natural because the of the resemblance between the simple family and the basic biological unit, and Verdon because of the "natural" preference for every adult not part of a couple to maximize their individual autonomy. These accounts at first sight leave no room for the forces of cultural change, although it would seem that both Smith and Verdon see culture as the prime determinant of a presumably universal family system. Other factors, primarily economic but also including, at least for Smith, well-established behavioral patterns, feature only in the role of constraints which prevent individuals from following what would otherwise be their natural inclinations to live separately from other adults.

Smith and Verdon assume a major disjuncture between the real (but largely unexpressed) preferences for particular types of living arrangements and the households that are actually formed. In this respect they are in agreement with Laslett, but with the important difference that Laslett saw the familial normative structure as enabling individuals to survive demographic and economic crises and ideological transformations. Verdon and Smith, by contrast, see the natural preferences of populations as completely subverted by economic and other constraints. Others have predicated a more harmonious relationship between economic and demographic realities and the familial system, in which choices are framed taking account of the options available. Such arguments have been advanced to explain the evolution of family patterns in a specific microregion in Croatia after the defeat of the Ottomans, the long-term persistence of simple-family households in the Spanish province of Cuenca, and the continued dominance of the northwest European household system. The significant factors shaping family patterns in northeast Croatia were, according to Jasna Capo Zmegac, the timing of resettlement, the amount of land available, and the family patterns of the first settlers. These factors acted in combination to establish preferences for particular types of household. In a similar vein, David Reher has argued that in Cuenca relatively early and universal marriage, neolocal household formation, and property transfers through inheritance ceased to be demographic, social, and legal acts and became normative cultural behavior. Finally, Mitterauer traces the origins of the northwest European household system back to the early Middle Ages and the combined influences of the Catholic Church and a tighter control over access to land consequent on a deterioration in the land-labor ratio. According to Mitterauer, therefore, the northwest European household system was the joint creation of economic circumstances and a specific institutional structure (which was itself embedded in a number of cultural values). These forces then shaped the cultural preferences of various European populations which kept the system in place thereafter.

BIBLIOGRAPHY

Alderson, Arthur S., and Stephen K. Sanderson. "Historic European Household Structures and the Capitalist World-Economy." Journal of Family History 16 (1991): 419–432.

Andora, Rudolf, and Tamás Farragó. "Pre-Industrial Household Structure in Hungary." In Family Forms in Historic Europe. Edited by Richard Wall, Jean Robin, and Peter Laslett. Cambridge, U.K., 1983. Pages 281–307.

Andorka, Rudolf, and Sandor Balazs-Kovács. "The Social Demography of Hungarian Villages in the Eighteenth and Nineteenth Centuries (with Special Attention to Sárpilis, 1792–1804)." Journal of Family History 11 (1986): 169–192.

Benigno, Francesco. "The Southern Italian Family in the Early Modern Period: A Discussion of Co-Residence Patterns." Continuity and Change 4 (1989): 165–194.

Brettell, Caroline B. "Emigration and Household Structure in a Portuguese Parish, 1850–1920." Journal of Family History 13 (1988): 33–57.

Caftanzoglou, Roxanne. "The Household Formation Pattern of a Vlach Mountain Community of Greece: Syrrako, 1898–1929." Journal of Family History 19 (1994): 79–98.

Capo Zmegac, Jasna. "New Evidence and Old Theories: Multiple Family Households in Northern Croatia." Continuity and Change 11 (1996): 375–398.

Egerbladh, Inez. "From Complex to Simple Family Households: Peasant Households in Northern Coastal Sweden 1700–1900." Journal of Family History 14 (1989): 241–264.

Friedman, Kathie. "Households as Income Pooling Units." In Households and the World-Economy. Edited by Joan Smith et al. Beverly Hills, Calif., 1984. Pages 37–55.

Guinnane, Timothy. "Coming of Age in Rural Ireland at the Turn of the Twentieth Century." Continuity and Change 5 (1990): 443–472.

Gunnlaugsson, GisliÁgúst. Family and Household in Iceland, 1801–1930: Studies in the Relationship between Demographic and Socio-Economic Development, Social Legislation, and Family and Household Structures. Uppsala, Sweden, 1988.

Hajnal, John. "European marriage patterns in perspective." In Population in History: Essays in Historical Demography. Edited by D. V. Glass and D. E. C. Eversley. London, 1965. Pages 101–138.

Hajnal, John. "Two Kinds of Pre-Industrial Household Formation System." In Family Forms in Historic Europe. Edited by Richard Wall, Jean Robin, and Peter Laslett. Cambridge, U.K., 1983.

Henderson, John, and Richard Wall. Poor Women and Children in the European Past. London, 1994.

Kertzer, David I., and Dennis P. Hogan. "Reflections on the European Marriage Pattern: Sharecropping and Proletarianization in Casalecchio, Italy, 1861–1921." Journal of Family History 16 (1991): 31–45.

Laslett, Peter. "The Character of Familial History, its Limitations, and the Conditions for its Proper Pursuit." Journal of Family History 12 (1987): 263–284.

Laslett, Peter. "Family and Household as Work Group and Kin Group: Areas of Traditional Europe Compared." In Family Forms in Historic Europe. Edited by Richard Wall, Jean Robin, and Peter Laslett. Cambridge, U.K., 1983. Pages 513–564.

Laslett, Peter. Family Life and Illicit Love in Earlier Generations: Essays in Historical Sociology. Cambridge, U.K., 1977.

Laslett, Peter. "Introduction: the History of the Family." In Household and Family in Past Time. Edited by Peter Laslett and Richard Wall. Cambridge, U.K., 1972.

Lehning, James R. "Socioeconomic Change, Peasant Household Structure, and Demographic Behavior in a French Department." Journal of Family History 17 (1992): 161–181.

Marchini, Antoine Noble.À Propos de la Casinca, Mediterranéens, Corses, des Gens et des Pays de France, L'Histoire (1770–1968). Individus, Familles, Cours de la Vie dans les Alèas de la Transition. Université de Nice Sophia Antipolis: Thèse de Doctorat d'État, 1996.

Martinez Carrión, José Miguel. "Peasant Household Formation and the Organization of Rural Labor in the Valley of Segura during the Nineteenth Century." Journal of Family History 1 (1988): 91–109.

Mitterauer, Michael. "Medieval Routes of the European Family Development." Unpublished Paper presented at the conference Where does Europe End?. Budapest, 1994.

Mitterauer, Michael. "Peasant and Non-Peasant Family Forms in Relation to the Physical Environment and the Local Economy." Journal of Family History 17 (1992): 139–159.

Mitterauer, Michael, and Reinhard Sieder. The European Family: Patriarchy to Partnership from the Middle Ages to the Present. Translated by Karla Oosterveen and Manfred Hörzinger. Oxford, 1982.

Moring, Beatrice. "Household and Family in Finnish Coastal Societies, 1635–1895." Journal of Family History 18 (1993): 395–414.

Morris, Lydia. The Workings of the Household. Cambridge, U.K., 1990.

Papataxiarchis, Euthymios. "La Valeur du Ménage. Classes Sociales, Stratégies Matrimoniales et Lois Ecclésiastiques à Lesbos au XIXE Siècle." In Espaces et Familles dans l'Europe du Sud à l'Âge Moderne. Edited by Stuart Woolf. Paris, 1993.

Reher, David Sven. "Household and Family on the Castilian Meseta: the Province of Cuenca from 1750–1970." Journal of Family History 13 (1988): 59–74.

Rogers, John, and Lars-Göran Tedebrand. "Living by the Sea: Farming and Fishing in Sweden from the Late Eighteenth to the Early Twentieth Century." Journal of Family History 18 (1993): 369–393.

Sabean, David Warren. Property, Production, and Family in Neckashausen, 1700–1850. Cambridge, U.K., 1990.

Schiavoni, C., and A. Sonnino. "Aspects Généaux de l'Évolution Démographique à Rome: 1598–1826." Annales de Démographie Historique (1982): 91–109.

Segalen, Martine. "Nuclear is not Independent: Organization of the Household in the Pays Bigouden Sud in the Nineteenth and Twentieth Centuries." In Households: Comparative and Historical Studies of the Domestic Group. Edited by Robert McC. Netting et al. Berkeley, Calif., 1984.

Smith, Daniel Scott. "American Family and Demographic Patterns and the North West European Model." Continuity and Change 8 (1993): 389–415.

Smith, Joan, et al. eds. Households and the World-Economy. Beverly Hills, Calif., 1984.

Verdon Michel. Rethinking Households: An Atomistic Perspective on European Living Arrangements. London, 1998.

Viazzo, Pier Paolo, and Dionigi Albera. "The Peasant Family in Northern Italy, 1750–1930: A Reassessment." Journal of Family History 15 (1990): 461–482.

Wall, Richard. "Historical Development of the Household in Europe." In Household Demography and Household Modeling. Edited by Evert van Imhoff, Anton Kuijsten, Pieter Hooimeijer, and Leo van Wissen. New York, 1995.

Wall, Richard. "Leaving Home and the Process of Household Formation in Pre-Industrial England." Continuity and Change 2 (1987): 77–101.

Wall, Richard. "Mean Household Size in England from Printed Sources." In Household and Family in Past Time. Edited by Peter Laslett and Richard Wall. Cambridge, U.K., 1972.

Wall, Richard. "Zum Wandel der Familienstrukturen im Europa der Neuzeit." In Historische Familen-forschung. Edited by Josef Ehmer, Tamara K. Hareven, and Richard Wall. Frankfurt, 1997.

Wall, Richard et al., eds. Family Forms in Historic Europe. Cambridge, U.K., 1983.

Wong, Diana. "The Limits of Using the Household as a Unit of Analysis." In Households and the World-Economy. Edited by Joan Smith et al. Beverly Hills, Calif., 1984.

Encyclopedia of European Social History