Family

I COMPARATIVE STRUCTURERaymond T. Smith

BIBLIOGRAPHY

II DISORGANIZATION AND DISSOLUTIONClifford Kirkpatrick

BIBLIOGRAPHY

The articles under this heading discuss the central aspects of the family as a universal institution in society. Closely related are the entries Kinship; Marriage. The reproduction function of the family is discussed in Fertility; Fertility Control; the socializing function of the family is reviewed in SOCIALIZATION as well as in Culture And Personality; Developmental Psychology; Infancy; Personality, article on Personality Development; Moral Development. Stages in family life are reviewed in Adolescence; Aging; Death; Life Cycle; as well as in Family, article on Disorganization And Dissolution. Specialized aspects of the family are reviewed in Adoption; Affection; Incest; Nuptiality; Paternalism; Sexual Behavior. Many anthropologists, psychologists, and sociologists have contributed to an understanding of the family; the following biographies are of particular relevance: Bachofen; Burgess; Engels; Frazer; Freud; Gesell; Hall; L,E Play; Mclennan; Maine; Malinowski; Morgan, Lewis Henry; Radcliffe-Brown; Tylor; Waller; Westermarck.

I COMPARATIVE STRUCTURE

One of the few widely accepted generalizations of social science asserts that “the family” is an institution found in all human societies. There is controversy over just what constitutes the family and what its functions might be, but the generalization itself is widely accepted. In modern Euro-American societies it is accepted that the normal family is a coresidential group which consists of a married couple and their own children and which lives apart from other kin. This has become the reference type for comparative study, and it is argued that in modern, highly differentiated societies such as those of North America and Europe the family has become a highly specialized agency exhibiting the essential and irreducible characteristics which are significant for cross-cultural comparison (Parsons & Bales 1955, p. 10).

In all human societies provision must be made for biological and social reproduction if the society is to continue, and it is generally assumed that the family performs at least these functions. Mating is never simply random, and children require a long period of care by a limited number of individuals with whom they develop relations of intimacy if they are to grow up as normal human beings capable of playing adult roles. This conclusion is based partly upon experimental evidence (Bowlby 1951) and partly upon inference from the fact that in all known societies children are raised in small kinship-based groups and there are customary modes of regulation between children and their socially recognized parents and between the parents themselves. Looked at in this way there is a very close relationship between the family and the domestic group, and both constitute systems of relationship which vary over time. The form of these systems is tied to the physical processes of individual birth, maturation, and death, so that families, like individuals, pass through developmental stages (Fortes 1958). The processes of individual maturation and death are social as well as physical, since individuals are participants in social systems as well as in biological processes. Societies live longer than either individuals or families, but societies exist only through the patterned interaction of individuals who share common understandings. Mating and procreation provide for the continuity of the biological species, while socialization relates it to the equally important continuity of social structure and cultural pattern. It has been argued on both theoretical and empirical grounds that the primary function of the family is socialization, so that an intimate relation is established between biological and social processes, a relation which is reflected in the dependence of demographic trends upon social custom.

The generalizations set out above seem to hold true in a large number of different societies, but it is not certain to what extent the observations and descriptions used have been influenced by the definitions adopted by field investigators. Some writers have expressed doubt that the family is a universal human institution, but both the assertion and its refutation depend upon some agreed identification of what is being discussed.

Definitions

Domestic groups . A domestic group may be defined as a group of people who habitually share a common dwelling and a common food supply. These minimal activities of domestic groups may be greatly extended, and domestic groups may vary in size and stability. The word “family” has its origin in a Latin word which could be roughly equated with “domestic group,” but for sociological purposes the two must be sharply distinguished. Domestic groups may be made up of individuals between whom no kinship ties exist, and, conversely, members of one family may be distributed over two or more domestic groups. The term “household” may be used interchangeably with “domestic group.”

Biological family . In Euro-American societies the basic model of kinship and family ties is that of biological relatedness and sexual intercourse, so that kin and familial relations are thought of in terms of physical descent or sexual relations. This is not so in all societies, even though it appears that family relations are almost invariably intimately associated with sexual regulation and reproduction. Social relations of family and kinship can develop independently of genetic links or sexual relations, so that adoption or other forms of fictive kinship are just as real as blood ties. Even where actual genetic links exist they do not constitute social relations; these must be learned and developed separately. Similarly, marriage is always more than sexual mating. Schneider (see Aspects of the Analysis … 1965) has expressed in a particularly pointed and convincing way the objections to confusing an analytical definition of kinship with definitions stressing biological relatedness.

The relation between mother and child appears to be in a rather special category, since the human infant is part of its mother’s physical being before birth and continues to be dependent upon her or a substitute for a considerable period afterward. The development of artificial feeding has modified this close dependence, but experience in rearing motherless infants has shown that adverse effects result from deprivation of close continuous contact with a mother figure in early infancy. The “mother figure” can be any suitable person, and experiments with monkeys have shown that, for them, even contraptions of wire and cloth can serve as “mothers” in some sense. This raises the question of just what we mean when we use words such as “mother” and should lead to a careful distinction between the activities involved in mothering on the one hand and the biological relationship of the mother to the child on the other. In the case of the father-child relationship, the more frequent disjunction between biological and social paternity has led to a distinction between pater and genitor. The former term is used to refer to a child’s legally recognized father and the latter to his supposed biological father. In some societies a child’s father (pater) may be a woman or a dead man. Among the Nuer of east Africa, the kin of a man killed in war before he had married would sometimes marry a woman to “his name.” After payment of the bride price the woman cohabited with a lover, but any children born to her were the legal offspring of the dead “husband,” inheriting from him and sacrificing to him as their ancestor. Cases such as this make clear the differences between biological and social links, and although the family is frequently, or even usually, coincident with a particular constellation of biological links, it cannot be defined in terms of them.

Nuclear family . The term “nuclear family” (or “elementary,” “simple,” or “basic”) is most frequently used to refer to a group consisting of a man, a woman, and their socially recognized children. This is a straightforward use of the term; it refers to concrete groups, and the qualification “nuclear” suggests that this is the unit out of which more extensive family groups are built or grow. The group need not be coresidential provided regular relationships are maintained among its members. It need not exist as a separate and isolated entity but may be contained within more extensive groups provided it is given some recognition. It has been suggested that the nuclear family is the universal form of family relations, always fulfilling “distinctive and vital functions–sexual, economic, reproductive, and educational . . .” (Murdock 1949, p. 3). “The nuclear family is a universal human social grouping. Either as the sole prevailing form of the family or as the basic unit from which more complex familial forms are compounded, it exists as a distinct and strongly functional group in every known society” (p. 2). The evidence for this statement is far from conclusive, and much of it that does seem to support the generalization may be biased simply because of the way in which the data have been collected and presented. Cases such as the Nayar and the Ashanti (see below) tend to disprove the assertion, and this has now been recognized by Murdock.

Parsons has argued that the nuclear family exhibits characteristics which seem to be necessary (on theoretical grounds) for the socialization of children and the stabilization of adult personalities (Parsons & Bales 1955). He has argued elsewhere that the incest taboo is also universal in human societies for similar reasons and that taboo results in the perpetual creation of new nuclear-family groups through marriage (Parsons 1954). Even if one accepts the ideas that sexual intercourse must be regulated and that children must be cared for and socialized within small groups with the characteristics exhibited by the nuclear family, this does not necessarily mean that the nuclear family, as defined by Murdock, is universal. Levy and Fallers (1959) have suggested that a distinction should be made between the nuclear family as a concrete group and the “nuclear-family relationship complex.” The latter would consist of the relationships of husband-wife, mother-son, mother-daughter, father-son, father-daughter, brother-sister, brother-brother, and sister-sister considered as a system of interaction between roles. By looking at the nuclear family as a system of roles rather than as a concrete grouping of individuals, it is possible to see that role behavior appropriate to this complex might be distributed among a number of individuals, groups, or agencies that do not themselves constitute a single group. Malinowski showed that certain aspects of what he considered to be the normal “father” role are played by the mother’s brother in some societies with matrilineal inheritance and descent. Similarly, in some societies the “mother” role is played in whole or in part by mother surrogates, such as mother’s sisters, mother’s mothers, paid nurses, or teachers. In order to make statements such as these it is necessary to assume that we know what the “real” or “normal” nuclear-family role complex is, and here it is evident that there is a tendency to take the Euro-American family pattern as the type case. Despite this drawback, the concentration of attention upon a role complex rather than concrete groups is a step toward freeing the concept of nuclear family from the rigidity imposed upon it by Murdock. If one asserts that the nuclear-family relationship complex is institutionalized in all human societies–instead of speaking of the universal occurrence of nuclear-family groups–then it is possible to account for the normal development of children brought up in groups that are not nuclear families in the restricted sense. Where one parent is missing or no siblings are present, children may still be affected by the institutionalized nuclear-family role complex, different elements of which are activated by other kinsmen, neighbors, teachers, or even more remote individuals. We may also relate the necessary biological and social functions to this role complex rather than to concrete nuclear families and leave it to empirical investigation to show just how the roles are embodied in a particular society. While this goes some way toward freeing investigation from the analytical restrictions imposed by the concept of the universality of the nuclear family, there are good grounds for going even further and asking just what is meant by the roles involved in the supposedly universal nuclear-family role complex.

Compound family . The term “compound family” is used to refer to a concrete group formed through the amalgamation of nuclear-family units or parts of them. A polygynous household consisting of one man, his three wives, and their respective children would constitute a compound family, as would a family group constituted by remarried widows or divorcees with children from a previous marriage. A compound family need not constitute a coresidential group.

Joint family . According to the handbook Notes and Queries on Anthropology (see British Association … 1874), a joint family exists when “two or more lineally related kinsfolk of the same sex, their spouses and offspring, occupy a single homestead and are jointly subject to the same authority or single head.” An example would be a group consisting of a man and his wife with their married sons and their wives and children. It would be wrong to think of such a group as being a mere mechanical aggregate of nuclear families. Joint families generally arise, exist, and persist because they carry out activities more extensive than would be possible for a nuclear-family group. Joint families grow as younger members bring in spouses rather than setting up independent households. As the younger married couples beget children, it may be possible to detect the existence of a number of nuclear-family cells within the structure of the joint family. The younger married couples will often have their own living quarters and may establish their own cooking facilities. If they have their own budgetary arrangements, then the joint family has really split–even if the young couples do not actually move out of the house. On the other hand, the joint family may be such a cohesive unit that it is difficult to see nuclear families within it as separate groups in any meaningful sense. The men may form one solidary group and the women another; the children may regard all the women of the house as their “mothers.” Eventually, as the original family grows too big, the joint family will either split or some persons will leave to form separate groups.

Extended family . An extended family is a dispersed version of the joint family. That is, the members of the constituent groups of an extended family do not all live together in one dwelling. They usually live close together and engage in common activities.

The concept of extended family really exhausts the usefulness of the word “family” for this kind of empirical classification, since an extended family is already a short lineage and can be discussed in terms of lineage theory, or it is a kindred and may be analyzed as such. Since birth is the criterion of membership in lineages and kindreds, they tend to develop considerable interest in the birth, training, and ultimate loyalties of their recruits, so that there is a close fit between these formations and the “families” which provide and train the recruits.

Comparative family studies

Because they are a part of all human experience and such a powerful agency in the formation of adult attitudes, familial relations have always been a subject of interest and discussion. Their changing and varied forms have been studied by such diverse and ancient disciplines as theology, classics, law, and philosophy. This interest quickened during the period known in Europe as the Enlightenment, but it was the late nineteenth-century interest in evolution that stimulated extensive comparative studies of family forms. Fantastic theories were proposed in an attempt to reconstruct the history of mankind and to account for the existence of seemingly pointless customs and strange kinship terminologies among non-European peoples. Bachofen’s postulated development of society from promiscuity through a great period of mother right and female dominance to father right and patriarchy was an attempt to account for the recognized importance of matrilineal descent systems and the universal importance of motherhood. Other writers of the period argued in much the same way, although disagreeing on developmental priorities. Such writers as Sir Henry Maine, Edward Tylor, J. F. McLennan, Lewis H. Morgan, Friedrich Engels, Sir James Frazer, E. Westermarck, and Robert Briffault carried forward the debate on origins, and, although their theories were highly speculative, they encompassed a wide range of cases and paid particular attention to historical sources. Despite the shortcomings of both theories and sources, these writers laid the foundations for modern studies by systematizing existing information on the family, and they are still worth reading for their insight into problems that have been neglected.

It is surprising that students of the family have not paid more attention to a series of basic sociological works which concern themselves with the classification of social relationships. Maine, Tonnies, and Durkheim, in particular, set up dichotomous categories which they believed to represent fundamentally different types of social relationship: relations of status as opposed to contract; relations of Gemeinschaft as opposed to Gesellschaft; mechanical versus organic solidarity. In all these systems one pole was most clearly represented by relations of a familial or kinship type. Since it was generally assumed that we know what familial relations are, this reference was intended to be illustrative, although Tönnies, in his Gemeinschaft und Gesellschaft, attempted to show how natural will develops within the complex of relations that go to make up the family. The crux of the matter is that what we term familial relations are relations exhibiting a particular kind of solidarity that involves persons in total and permanent social relations. This type of relationship is contrasted with the partial and specialized character of Gesellschaft or contractual relations. Just where the boundaries of familial relations are drawn varies from one society or subgroup to another and is, of course, affected by a large number of empirical factors.

Intensive field-work methods developed during the early twentieth century by Boas, Kroeber, Lowie, and others in the United States and by Malinowski and Radcliffe-Brown in Britain produced a new body of detailed information on kinship, marriage, and the family. Malinowski’s work on the family in Australia and Melanesia was a study of primitive matriliny and was intended to show (in part at least) that the nuclear family emerges naturally even where there is ignorance of the male role in the physiological process of reproduction. The Trobriand Islanders studied by Malinowski asserted that children are conceived when a spirit enters the body of a woman and that sexual intercourse is simply a matter of mutual satisfaction to the persons involved. Malinowski was impressed by the fact that men developed important social relationships with their wives’ children even though they did not consider that there was any biological connection involved. Furthermore, men did not exercise the kind of authority over their children that is often associated with “paternity”; the mother’s brother was more important in that respect. He concluded from these observations that the nuclear family is a universal institution but does not always assume the emotional configuration suggested by Freud’s theory of the Oedipus complex; it is rather a “functional formation dependent upon the structure and upon the culture of a society” (Malinowski [1927] 1953, pp. 142-143). The implication of this idea was that the nuclear-family complex could be activated by various persons and agencies–in the case of societies with matrilineal descent groups it may be that repressed hatred between uncle and nephew replaces that found between father and son in patriarchal societies.

Radcliffe-Brown’s brilliant and lucid treatment of the jural significance of kinship ties enabled him to look at the family in terms of the formal duties, rights, and obligations between the members and between them and external groups. Whereas Malinowski regarded wider kinship ties as “mere extensions” of the nuclear-family complex and the sentiments generated within it, Radcliffe-Brown spoke of the varying modes of incorporation, for jural purposes, of the unit of mother and children. He saw the relationship between a mother and her children as being somehow “basic” because of the close emotional and nurturant ties involved, while other kinship ties could be stressed or neglected to varying degrees depending upon the structure of the whole social and kinship system. In other words, Radcliffe-Brown’s concern was not with the family as a locus of emotional solidarity, or socialization functions, or sexual satisfactions; he was more concerned to explore the ways in which the empirical reference points of birth, sexual relations, and child rearing could be variously stressed for legal purposes. Apart from the apparent invariability of the mother-child relationship, he was prepared to accept the fact that all the other activities and functions could be distributed in widely varying ways.

In the generation since these pioneers of field investigation, detailed work has been carried out in many parts of the world, yielding new information and continuous refinement of theory. The development of the Yale cross-cultural survey is an attempt to organize material from a wide range of sources in order to make inductive generalizations about human societies or to test particular hypotheses. Other writers make similar generalizations based upon a more selective use of sources. In either case it is evident that the generalizations are no better than the observations on which they are based. This is well demonstrated by Murdock’s generalizations based on the cross-cultural survey files. Observations reported in monographic studies depend to a considerable extent upon the investigator’s preconceived ideas and categories or upon his interests. In the field of family studies all these may be severely limited.

Parallel to the development of a wide-ranging comparative study of kinship and family structure, there has been a continuing tradition of interest in the family life of Europe and America. In the nineteenth century, Le Play constructed a typology of European family structures ranging from the patriarchal systems of the semifeudal eastern European areas, which resulted in the development of extended families, through the famille-souche, or stem-family, of the semi-industrial areas to what he considered to be the unstable and disorganized systems of the urban industrial areas. Le Play’s typology, or something very like it, seems to have regained favor as a universal system of classification (see Levy in Aspects of the Analysis …1965). The stem-family type is of particular interest because it seems to combine an emphasis upon the ideal of family continuity with an actual situation where most of the domestic units are small and self-liquidating.

As a result of their work on the development of capitalism and its institutional structure, Marx and Engels became interested in the development of the family, an interest which was quickened by the publication of Lewis H. Morgan’s Ancient Society (1877). Marx read this work and made extensive notes on it, but it was Engels who published an extensive commentary on it after Marx’s death, under the title The Origin of the Family, Private Property and the State. In this work Marx’s general theories of social and economic history are combined with Morgan’s speculative history of kinship institutions. The most general conclusion of the book is that a stable monogamous family system, dominated by male authority and prescribed and supported by law, has really developed as a device for the perpetuation of the private ownership of property. This conclusion has been widely misinterpreted to mean that Engels was opposed to any sort of family relationships and that the family should be abolished in socialist societies. On the contrary, Engels went out of his way to point out that the emotional and sexual elements involved in family relations can only be paramount when considerations of property do not enter into the picture. Thus, he argued that the proletariat make and break marital unions only upon the basis of mutual attraction, whereas in bourgeois marriage the conjugal pair are indissolubly united even when they have no love for each other and perhaps engage in adultery with other available partners.

The influence of Freud’s work upon comparative family studies can only be mentioned in passing. It has produced some of the most penetrating hypotheses concerning the content of family relationships, particularly in European and American societies where psychoanalysis’ has been most widely practiced as a therapeutic technique. Studies of the relation between culture and personality focus upon the familial milieu as the context of personality development, and students of this field have produced some of the most detailed and intimate documentation of family life in non-Western cultures. It is also evident that Freud’s thought has influenced the formation of hypotheses about the nature and functions of family structure, at least since the work of Malinowski.

Theoretical and research problems

Despite the considerable time and effort devoted to family studies and despite the refinement of definitions and the improvement of concepts, our knowledge of the family remains rudimentary. We use the term “family” in the loosest possible way, perhaps because of the very nature of the data themselves. For comparative study it is necessary to exercise great care in the construction of theoretical models in order to avoid distortions in observation and description. The bulk of family theory is derived from the Euro-American cultural tradition and from the study of European and American societies. The development and extension of psychoanalysis has produced both a large collection of empirical data and a growing sophistication in its interpretation. But this has been accompanied by a tendency to attribute functional necessity to observed patterns, so that there is a danger of assuming that the functions performed by small family groups in Euro–American societies must be performed by similar groups, similarly defined and constituted, in all societies. Parsons (see Parsons & Bales 1955) has tried to show the systematic relation between nuclear-family structure on the one hand and socialization for participation in wider social systems on the other. This analysis is very convincing in its establishment of the general conditions which seem to be necessary for adequate human socialization, but those conditions can be expressed in very general terms and could be met by a wide variety of actual persons and groups. If the nuclear family is to be thought of as that group that carries out primary socialization, then it is necessary to investigate the empirical composition of such groups in a wide range of societies.

Should we continue to use the model of nuclear family or even of nuclear-family relationship complex as a measure for cross-cultural comparison?

Is this anything more than a device for making all observations fit our conception of what should exist? Or is it mainly a problem of language? Will more flexible terms that are not derived directly from a particular cultural tradition enable us to overcome these difficulties? Similar problems have arisen in debates over the question of whether all societies have “law” or “a political system.” In spite of some recent defense of the idea of the universality of the nuclear family from linguistic analysts (Lounsbury 1965), it would seem to be desirable to keep an open mind on the question.

Much of the work already done in the study of comparative family structure has been based upon the collection of information about norms, that is, the ideal modes of behavior in different societies; and, as we have suggested above, these have usually been translated into terms of the norms of Euro-American society. Many anthropologists have supplemented interview data and informant’s statements with case material, genealogical investigation, sample surveys, and statistical analyses. The aim has been either to try to determine what the system “really is” or to discover the manner in which norms work out in practice. Pioneering work in this regard was done by Audrey Richards, John Barnes, and Meyer Fortes. Fortes has always stressed the importance of regarding the family as a process in time and of looking at the cyclical development of family structure. As the individuals who compose family groups grow older and as new members are born and others die, the structure of the group changes as well as the relationships between the individuals concerned. New possibilities for this type of analysis are presented by the development of computers. With the aid of these machines it is possible to simulate the experience of a whole community of families over a long time period by constructing models of family structure or household composition and then applying to them such variable elements as birth and death rates, age at marriage, duration of unions, and fertility patterns. In this way it is possible to determine what varieties of family and household types would be produced under varying circumstances over varying time periods.

A major controversy that has developed within anthropology has considerable bearing on the whole subject of comparative family studies. This controversy starts with Levi-Strauss’ idea that the building block of kinship systems is not the biological or nuclear family but a constellation of roles which includes the siblings of a marrying couple as well as their children. In disagreeing with Radcliffe-Brown’s contention that it is the relations between members of the “elementary family” that constitute the unit of structure from which all kinship systems are built, he says:

Of course, the biological family is ubiquitous in human society. But what confers upon kinship its socio-cul-tural character is not what it retains from nature, but, rather, the essential way in which it diverges from nature. A kinship system does not consist in the objective ties of descent or consanguinity between individuals. … The essence of human kinship is to require the establishment of relations among what Radcliffe Brown calls “elementary families.” Thus, it is not the families (isolated terms) which are truly “elementary,” but, rather, the relations between those terms. (Lèvi-Strauss [1958] 1963, pp. 50-51)

From this starting point Lèvi-Strauss proceeds to a complex analysis of preferential and prescriptive marriage systems, but the perspective which he introduced leads to a new assessment of intrafa-milial relations.

Briefly, the question is whether the relations of nuclear family and of filiation always constitute a basic “natural” system which tends to counterbalance particular kinship ties stressed for jural purposes, or whether we should regard nuclear-family relations as being nothing more than a fortuitous complex of highly variable relations some of which may be dispensed with altogether. For example, if filiation is always bilateral (because every child must have a “father” and a “mother”), then where patrilineal descent is stressed for jural purposes the strong jural tie between father and child will be balanced by the tie between mother and child and, by extension, between the child and its mother’s kin–a tie of complementary filiation. Conversely, it could be argued that where matri-lineal descent is stressed, a compensating bond will exist between father and children and between the child and the father’s kin. The point to be noted in this argument is that the ties between parents and children may be accorded great legal significance or may be simply affective, but they derive ultimately from the bonds of filiation and descent which arise within the nuclear-family relationship complex. These relations are socially rather than biologically defined, but they focus upon activities which are familial.

The contrary view is that relationships between a child and its parents may derive from the fact that the parents are married to each other rather than from any bond of filiation. Thus, the Trobri-and Islanders assert that there is no genetic link between fathers and their children. A man is a “father” because he is the husband of the child’s mother. Of course this ideological view of the matter seems to be related to the political and legal sphere rather than to the field of domestic relations, where fatherhood may be rooted in a set of activities rather than a set of abstract principles.

It is not necessary to document the details of this controversy here; adequate references will be found in Leach (1961) and Schneider (see Association of Social Anthropologists … 1966). The point of interest here is Leach’s contention that relations between children and parents may be as variable as any other social relations and that we should be prepared to find societies in which even the mother-child relationship is conceived of as one wholly derivative from the fact of the mother’s marriage to the father, who is thought of as the person with whom the child has a real consanguineal relationship. In such a situation the mother is an affine; she is not “mother” necessarily at all, but “father’s wife.”

The controversy outlined here does not focus upon the internal relations or the structure of the family as such; it is concerned with an understanding of the structure of total social systems. But it does bring into quite sharp focus the need to abandon a dogmatic view of the universality of nuclear-family relations and the need to differentiate clearly between the activities involved in domestic relations, the facts of biological interconnectedness, and the interlocking of familial and wider kinship relations. It seems likely that future research will operate with a bigger collection of more differentiated concepts derived from a splitting up of such notions as descent, filiation, sibling bonds, alliance, and, of course, the general notion of nuclear family.

Throughout the discussion so far it has been assumed that comparative family studies refer to comparisons between different societies and that “structure” means societal norms. Some of the most interesting modern research is concerned with the study of differences in family structure within societies; studies which throw light upon the nature of society itself. It has long been recognized that the structure of family relations appears to vary with class, ethnic group, and other factors. This raises the question of whether the variation lies at the level of culture (that is, in the norms themselves) or whether the actual distribution of empirically ordered “types” of family is the result of constraints operating to prevent the full realization of cultural ideals. For example, it is generally accepted that Hindus value a joint family system in which marriages are contracted at an early age and are indissoluble and in which brides are brought to live in their husband’s father’s house, where they come under the authority of the senior woman and the men form a corporate group. In fact, it would seem that such ideal patterns are met only by the wealthier families, where a common interest in property serves as a binding factor, while the poorer families tend to be small and to consist of only a couple with their own minor children. It has also been suggested that the size of households would be limited under preindustrial conditions everywhere by the early death of parents and high infant mortality rates. Such arguments ignore the fact that purely demographic constraints can be circumvented, since family and kinship relations are not biological relations and since adoption or other forms of culturally defined kinship ties can be used as the basis of household formation.

Important though adoption and the collapsing of relationships may be as a means of keeping functioning groups intact, it must be realized that the exigencies of birth-and-death rates and other demographic variables do have an effect upon household composition, just as the biological processes of maturation have an effect. Sociologically, a more important consideration is the extent to which norms vary from group to group and result in the precipitation of different configurations of family organization. In the United States there has been discussion of “the culture of poverty” in relation to the incidence of fatherless families among lower-status and lower-income groups. In such cases it is difficult to determine whether the pattern of family living among these lower-income groups is determined by a simple inability to command the means of achieving valued patterns of behavior or whether different values are operative. In such cases the study of family life becomes a very sensitive index to the whole mode of integration of the over-all social system. This question is discussed further below.

Although knowledge of the genesis of emotional disturbance and social deviance is much more extensive than it was 25 years ago, it is still inadequate. The complexity of the relationship system of family groups is understood in only the most general terms despite the considerable body of detailed information pertaining to Euro-American families. The profound hatreds and bitter rivalries which are the counterpart of the relations of familial solidarity are well known to anthropologists, both as the source of conflicts with political and economic significance and as the basis of witchcraft beliefs and magical and religious practices. These conflicts are the product of the interaction systems themselves within small “multibonded” groups and may have considerable significance for wider system organization. It has often been remarked that the family is the unit within which primary socialization takes place and within which the growing child is taught the ultimate value orientations of the whole society. This is not necessarily done formally, but the values may be implicit in the general structure of the familial relationship system itself. If this is so, then it is important to know how these relationships are structured in different societies and just how the values are internalized by the growing child.

The range of variability

In the large Ashanti villages of central Ghana one can often see children carrying pots of food from their mother’s house to the house of their father. The children may eat with their father and then go home to sleep at the house where their mother lives, while the mother may visit the father for the night. This pattern of divided residence and visiting back and forth seems to exist because the Ashanti traditional social system stresses matrilin-eal descent for important social purposes, such as inheritance of land, succession to office, and political status. Women often value the tie to their brother as highly as, or higher than, the relationship to their husband, since it is from the brother that their children will inherit. Because the child’s place in society is mainly determined through its relationship to its mother and her matrilineal kin, the breaking of a marital tie is of little legal consequence either for the spouses or for the children. It is true that fathers have always had responsibility for the upbringing of their own children, including the provision of food and the teaching of crafts and skills, but these duties could be carried out even when the child did not reside with the father and even after the parents had broken their conjugal tie. According to Fortes (1949) another factor inducing women to remain in their own homes even after marriage is the very close relationship between mother and children, and particularly between mothers and daughters. This case demonstrates the way in which interests and relationships can cut right across the solidarity of the nuclear family, and it would be very difficult to demonstrate that the nuclear family is either a normal or a necessary coresidential unit among the Ashanti. The Ashanti concept of family is quite different from the European concept; Fortes mentions that persons who speak English often say, “Your mother is your family, your father is not.”

The most extreme and best known example of the effect of matrilineal descent upon familial relationships was to be found among the Nayar caste of southern India in the period before the full effects of British rule were felt. Among the Nayar, it appears that the marital relationship was reduced to a merely symbolic level, being contracted at around the time that the girl reached puberty and shortly thereafter being ritually broken. Afterward, the women were allowed to have informal love affairs with men who visited them at night. The households consisted of a group of brothers and their sisters and the sisters’ children, and any children born to the woman of the household became members of the matrilineal joint family. The children were discouraged from developing any strong attachments to either their pater or genitor, and while it would be impossible to argue that the role of father was completely absent from this social system, it was obviously diminished in favor of the solidarity of the property-owning matrilineal group. It seems likely that many of the functions that are normally fulfilled by fathers in relation to socialization and personality development were carried out by male members of the matrilineage (Gough 1952).

Cases such as these have led some writers, including Radcliffe-Brown, to suggest that the basic structural unit of kinship systems is the unit of mother and children, in other words, to separate the nuclear family into a number of paired relationships, or dyads, and to see how, in differing societies, they fit in relation to each other and to other groups. This is certainly a step in the right direction toward loosening the categories involved in the nuclear-family complex and exploring more fully their independent connections. Even in societies with pronounced patriarchal authority we find that the mother-children unit often forms an important subgroup, given special recognition and marked by close emotional ties among its members.

Among the Swazi as described by Hilda Kuper (see Radcliffe-Brown & Forde 1950), descent, inheritance of economically valuable possessions, and succession to office is patrilineal, but women are given a position of great prominence in the organization of domestic life. Homesteads are built in the form of a semicircular group of huts surrounding a cattle pen, each hut being occupied by a mother and her children. In the center of the arc is the “great hut” occupied by the mother of the male head of the compound (or one of his senior wives if his mother is dead). When a young man or a group of full brothers sets up a new homestead, their mother is installed in the place of honor, for it was she who reared them, watched over their rights, guarded their inheritance, fed and nurtured them. The father is respected as the head of the group, revered, and eventually worshiped as an ancestor, but the mother is loved and cherished. This pattern is common in patriarchal societies where the father is an authoritarian figure and the holder of property that the heir requires in order to attain full maturity–thus introducing an underlying tension into the formally respectful relationship. Parsons has suggested that the father role is always an “instrumental” one as opposed to the more emotional or “expressive” quality of the mother role (Parsons & Bales 1955, pp. 157-158). In a penetrating study comparing two adjacent groups in west Africa which share a basically similar culture, Goody has shown that in one of them the matrilin-eal inheritance of certain types of property alters the interpersonal relationships between close kin and even alters the mode of ancestor worship (1962). Since intrafamilial relationship patterns seem to be so responsive to the effect of outside influences, it is premature to assert too many conclusions about the structure of the nuclear family as such. The specialization of males and females in instrumental and expressive roles may be an aspect of sex-role differentiation in this case rather than of role specialization within a small group structure. Certainly women take on instrumental roles in the absence of men, but they retain the predominantly expressive quality appropriate to females.

In tribal societies where kinship is the basis of recruitment to all or most important social roles, the family must articulate directly with descent groups and becomes a mechanism for the continuous generation of new kinship ties. As Fortes says, “the workshop, so to speak, of social reproduction, is the domestic group” (Goody 1958, p. 2). In more differentiated social systems with an increased division of labor, a literary tradition, and a well-developed class or caste system, it is still true that families may be large, multifunctional units. In China and India, for example, the whole weight of tradition, values, and religious sentiment favored the growth of large patrilineal joint families and patriarchal authority. The uneven distribution of wealth and status seems to affect the realization of these ideals, so that the majority of households are small and contain only parents and children. This is not simply a matter of demographic determinism; the absence of a sufficient property base leads to the early secession of mature sons, who try to make out for themselves. Large joint families are found mainly among landowning and merchant groups, where sons have a continuing material interest in the patrimony to reinforce and sustain filial piety. Such large “families” are really lineages or descent groups which constitute perpetual corporations managing a common enterprise.

Le Play thought of the famille-souche, or stem-family, as preserving familial continuity while adapting to the demands of industrialism. While many of the younger members of the family migrated or went off to work in industrial centers, the main family, or stem, continued and was usually located on its own land or in its own house. A regular mode of inheritance designed to prevent fragmentation was imposed to ensure both the continuity and the integrity of the familial estate. The existence of such a system in twentieth-century rural Ireland has been described by Arensberg and Kimball (1940), and its presence in parts of the United States has been depicted (Zimmerman & Frampton 1935). On the small farms of rural Ireland it is customary for one son to inherit the family farm, while the other children marry out or migrate to the towns. The heir usually postpones his marriage until his father is nearly ready to hand over control, thus ensuring that the farm will not have to support more persons than it can carry and ensuring the continuity of one male line on an undivided farm. The stem-family is not a product of industrial society. Recent work by Laslett indicates that the preindustrial family system in England did not encourage the growth of large extended families occupying single dwellings; on the contrary, the nuclear-family household seems to have been the normal type of dwelling group, and when large household groups are found, they appear to have been made up by servants or craft laborers rather than by kin. There was also a continuous process of downward mobility in which the children of wealthy or high-status families were sloughed off into lower positions, while the main heir took over an intact patrimony (Laslett 1965).

The urban family system of modern Europe and America developed out of the system found in agrarian society (or “the world we have lost,” as Laslett terms it). It has been a constant preoccupation of writers on the modern family to think of it as being somehow stripped of functions and reduced to a pale shadow of its former ample self. The ideal is that of a nuclear-family group living in its own house independently of other kin and subsisting upon the wages or salary of the husband-father. Falling birth rates, an increase in the incidence of divorce, a decline in home food processing, clothes making, and so on–all these have been held to signify a decline in family living and a shift toward increasing individualism and material values. Parsons and others have argued that the family system of the urban United States is not a denuded form of a more “normal” or “natural” family system but is itself a highly specialized form that articulates most satisfactorily with a highly differentiated economic and political system and with institutionalized values that stress achievement rather than inheritance. It is argued that the smallness and relative isolation of the family from other kinship ties is an adaptation that makes possible the spatial and status mobility of its members. The unit of mother and children remains the basic affective group within which there is close emotional identification, and the husband-father is closely integrated with it both emotionally and in terms of his status-conferring and economic-support functions. It is upon his position in the occupational system that the status and the style of life of the whole family group depends. This functional interpretation of the importance of the nuclear family in industrial society may be considerably weakened by the work of Laslett, referred to above. If the nuclear family has been the major form of European family structure for many centuries, predating the industrial revolution, and if this family form was the product of social values rather than of demographic or other constraints, then the modern family forms may be simply the continuation of a culturally preferred form.

Studies of the structure of kinship and family relations in modern industrial societies are rare in spite of the considerable literature on special problems of family living. Recent work in Britain has shown that lower-class families often live in close proximity to near kin and maintain intimate contact with them despite the “isolation” of the nuclear family (Young & Willmott 1957). The cooperation between mother and married daughters is particularly noticeable, and most writers have remarked upon the great importance of women in their role as mothers of grown-up children. Fathers are important, of course, as wage earners and supporters of their families, but in the lower class the status of the family tends to be fixed less by the occupation of the husband-father than by birth into a particular neighborhood and class. The network of relations between female kin is reinforced by daily contact, mutual help, and support in times of economic hardship, and is balanced by a tendency for men to spend their leisure time together rather than with the members of their own nuclear family.

A more extreme form of matrifocal family structure occurs in low-status ethnic groups where there appears to be some correlation with insecurity in male employment. In the plantation areas of the southern United States, the West Indies, and Latin America and in the urban areas of developing societies in Africa and Asia, one often finds high illegitimacy rates and unstable mating patterns. An array of types of domestic group is produced, the most distinctive being those in which a mother is surrounded by adult children and some of her daughter’s children. Money income flows into such a household from a variety of sources, such as children’s earnings, payment to daughters for the support of illegitimate children, the older woman’s own labor, and perhaps government relief funds. This income is reallocated around the essential tasks of child rearing and nurturance, and it appears that a household group of this kind is a useful adaptation to conditions of economic insecurity. This form of domestic group appears to be always regarded as a deviant form, partly because of the instability of male roles in relation to it. Unlike the matrilineal situation, men do not have important legal ties to their sisters; it is the mother who is the focus of the familial solidarity, and her position derives from affective rather than jural relations. It would be wrong to suggest that this is the typical household or familial form in the societies in which it occurs. Even at the class levels where it is most frequently found it constitutes only a small proportion of actual household groups, but its significance is as an expression of the strength of mother-child relations and the relative weakness of maritalties.

This brief sketch has outlined only some of the major features of family structure. Variations within single societies have hardly been touched upon, and nothing has been said about pathological forms. Nor has attention been directed toward the attempts which have been made from time to time to do away with family solidarity in favor of allegiance to other groups, although such attempts provide valuable experimental data. The Israeli cooperative communities are probably the best recent examples of this, although no definite conclusions about the necessity or otherwise of nuclear-family relationships can be drawn (Spiro 1958). What these cases do perhaps indicate, along with some of the matrilineal and matrifocal cases, is that marriage is not the only means of securing social “legitimacy” for children. Spiro shows that children brought up in central nurseries and schools are in a sense children of the whole community, in spite of the specific relationships they may develop with their “own” parents. This is possible because of the relatively small size of the communities.

However the family may be defined or structured, it always constitutes an area of diffuse and permanent solidarity between a limited number of individuals, and this is probably its most important distinguishing characteristic. This is a view at least suggested by W. H. Rivers in his Social Organization when he deliberately identifies the family of parents and children as the basic unit in European society but points out that in other societies the family may be much bigger, that biological connection is no guide to kinship, and that genealogies express kinship ties but may be fabricated to fit social reality. Rivers is very wary about drawing any precise line between family relationships and other kinship ties. Such lines can be drawn in particular cases but are very difficult to define a priori.

Raymond T. Smith

[Directly related are the entriesKinshipandMarriage; see alsoConflict, article onanthropological aspects.]

BIBLIOGRAPHY

Arensberg, Conrad M.; and KIMBALL, SOLON T. (1940) 1961 Family and Community in Ireland. Gloucester, Mass.: Smith.

Aspects of the Analysis of Family Structure, by Ansley J. Coale et al. 1965 Princeton Univ. Press.

Associationof Social Anthropologists

ofthe Commonwealth, Conferenceon New Approachesin Social Anthropology, Cambridge, 19631966 The Relevance of Models for Social Anthropology. Edited by Michael Banton. London: Tavistock. → See especially “Some Muddles in the Models” by David M. Schneider.

Bachofen, Johann J. (1861) 1948 Das Mutterrecht.2 vols. 3d ed. Edited by Karl Meuli. Basel: Schwabe.

Bell, Norman W.; and VOGEL, EZRA F. (editors) 1960 A Modern Introduction to the Family. London: Rout-ledge; Glencoe, Ill.: Free Press.

Bott, Elizabeth 1957 Family and Social Network: Roles, Norms, and External Relationships in Ordinary Urban Families. London: Tavistock.

Bowlby, John 1951 Maternal Care and Mental Health. Geneva: World Health Organization.

British Associationfor the Advancementof Science (1874) 1954 Notes and Queries on Anthropology.6th ed., rev. London: Routledge.

Evans-Pritchard, E. E. 1951 Kinship and Marriage Among the Nuer. Oxford Univ. Press.

Fortes, Meyer (1949) 1963 Time and Social Structure: An Ashanti Case Study; Radcliffe-Brown’s Thesis. Pages 54-84 in Meyer Fortes (editor), Social Structure: Studies Presented to A. R. Radcliffe-Brown. New York: Russell.

Fortes, Meyer 1958 Introduction. Pages 1-14 in Jack R. Goody (editor), The Developmental Cycle in Domestic Groups. Cambridge Papers in Social Anthropology, No. 1. Cambridge Univ. Press.

Frazier, E. Franklin (1939) 1957 The Negro in the United States. Rev. ed. New York: Macmillan. → First published as The Negro Family in the United States.

Goody, Jack R. (editor) 1958 The Developmental Cycle in Domestic Groups. Cambridge Papers in Social Anthropology, No. 1. Cambridge Univ. Press.

Goody, Jack R. 1962 Death, Property and the Ancestors: A Study of the Mortuary Customs of the Lodagaa of West Africa. Stanford Univ. Press.

Gough, Kathleen 1952 Changing Kinship Usages in the Setting of Political and Economic Change Among the Nayars of Malabar. Journal of the Royal Anthropological Institute of Great Britain and Ireland82: 71-87.

Halpern, Joel Martin 1958 A Serbian Village. New York: Columbia Univ. Press.

Laslett, Peter 1965 The World We Have Lost. London: Methuen.

Leach, Edmund R. 1961 Rethinking Anthropology. London School of Economics and Political Science Monographs on Social Anthropology, No. 22. London: Athlone.

LÉvi-Strauss, Claude (1958) 1963 Structural Anthropology. New York: Basic Books. → First published in French.

LÉvy, Marion J. 1949 The Family Revolution in Modern China. Cambridge, Mass.: Harvard Univ. Press.

LÉvy, Marion J.; and Fallers, L. A. 1959 The Family: Some Comparative Considerations. American Anthropologist New Series 61 : 647–651.

Lounsbury, Floyd G. 1965 Another View of the Tro-briand Kinship Categories. American Anthropologist New Series 67, part 2: 142–185.

Malinowski, Bronislaw (1927)1953 Sex and Repression in Savage Society. London: Routledge; New York: Harcourt. → A paperback edition was published in 1955 by Meridian.

Mead, Margaret 1949 Male and Female: A Study of the Sexes in a Changing World. London: Gollancz; New York: Morrow.

Morgan, Lewis Henry 1871 Systems of Consanguinity and Affinity of the Human Family. Smithsonian Contributions to Knowledge, Vol. 17, Publication No. 218. Washington: Smithsonian Institution.

Morgan, Lewis Henry (1877) 1964 Ancient Society. Edited by Leslie A. White. Cambridge, Mass.: Belknap.

Murdock, George P. 1949 Social Structure. New York: Macmillan.

Parsons, Talcott (1949)1959 The Social Structure of the Family. Pages 241-274 in Ruth N. Anshen (editor), The Family: Its Function and Destiny. Rev. ed. New York: Harper.

Parsons, Talcott 1954 The Incest Taboo in Relation to Social Structure and the Socialization of the Child. British Journal of Sociology5 : 101–117.

Parsons, Talcott; and Bales, Robert F. 1955 Family, Socialization and Interaction Process. Glencoe, Ill.: Free Press.

Radcliffe-Bhown, A. R.; and FORDE, C. DARYLL (editors) 1950 African Systems of Kinship and Marriage. Published for the International African Institute. Oxford Univ. Press.

Schneider, David M.; and GOUGH, KATHLEEN (editors) 1961 Matrilineal Kinship. Berkeley: Univ. of California Press.

Spiro, Melford E. 1958 Children of the Kibbutz. Cambridge, Mass.: Harvard Univ. Press.

Thomas, William I.; and Znaniecki, Florian (1918-1920) 1958 The Polish Peasant in Europe and America.2 vols., 2d ed. New York: Dover.

Young, Michael; and Willmott, Peter 1957 Family and Kinship in East London. Institute of Community Studies, Report No. 1. London: Routledge; Glencoe, Ill.: Free Press. → A paperback edition was published in 1963 by Penguin.

Zimmerman, Carle C. 1947 Family and Civilization. New York: Harper.

Zimmerman, Carle C.; and FRAMPTON, MERLE E. 1935 Family and Society: A Study of the Sociology of Reconstruction. London: Williams & Norgate; Princeton, N.J.: Van Nostrand.

II DISORGANIZATION AND DISSOLUTION

Aside from some slight semantic disagreements, most social scientists agree that some form of a familial group is universal among human beings. Likewise in every society some of the familial groups are imperfect in their functioning and deserving of the label “disorganized.” Generally there is some relationship between disorganization and a more climactic, terminal situation called “dissolution.” Family groups are eventually dissolved by death or some form of separation, insofar as these groups are organized around the matings of men and women. Of course, a family process, in the sense of overlapping and interacting generations, is immortal, providing there is reproductive replacement before the familial group of a particular generation disintegrates (Kirkpatrick [1955] 1963, pp. 192-195).

Nature and significance

The significance of familial disorganization and dissolution depends upon the institutional arrangements for family life in a particular culture. If stress is laid on the nuclear family, then kinfolk are less available to aid members of the smaller familial group. Some families are vulnerable in the sense that their members expect permanence, have made deep emotional commitments, and do not readily redirect affection and loyalty. On the other hand, people may be less hurt by familial instability in cultures where instability is an established norm and where kinfolk are helpful and meaningful.

In most cultures, the familial group helps prepare children both for membership in a society and for effective functioning as parents. Disorganization and dissolution are significant in that they tend to interfere with this socialization process. Furthermore, disorganization and dissolution are often cumulative and recurrent in the family process. There is a convergence of evidence in the United States that people with happy marriages tend to come from happy homes. Divorce also runs in families; divorced persons are divorce-prone, and children suffering from bereavement or parental divorce may be involved with delinquency, illegitimacy, and unwise marriage. The consequences of disorganization and dissolution experienced by the parents are commonly visited upon the offspring, perhaps for several generations. Some children, however, are warned by the fate of their parents and in marriage make an effort to reverse a trend toward disorganization.

Terminology . It is impossible to make a useful analysis of the nature and significance of family disorganization and dissolution without precise terminology. In current usage “disorganization” seems to imply maladjustment, malfunctioning, psychological decay, and the existence of family problems. Goode defines family disorganization as “the breakup of a family unit, the dissolution or fracture of a structure of social roles when one or more members fail to perform adequately their role obligation.” He lists five types of family disorganization, namely, illegitimacy, dissolution, whether as a result of divorce, annulment, separation, or desertion; “empty shell family”; unwilled absence of one spouse; and “unwilled” major role failures (Goode 1964, pp. 91-92).

Clearly, such taxonomy makes no distinction between disorganization and dissolution. It seems from the inclusion of illegitimacy that a family can be “disorganized” by role failure even before it is organized. Caplow defines an organization as “a social system that has an unequivocal collective identity, an exact roster of members, a program of activity, and procedures for replacing members” (1964, p. 1). Following this line of thought, it would seem more proper to regard the “illegitimate family” as unorganized rather than as a type of family disorganization.

Distinction between disorganization and dissolution. In the present discussion a distinction will be made between family disorganization and family dissolution, and marital disorganization will be regarded as a special case of family disorganization. Family dissolution means disintegration, disruption, or chronic instability; it is characterized by “dependence, agency intrusion, member extraction, premature atypical departure and family violence” (Kirkpatrick [1955] 1963, p. 565).

Marital dissolution, as a special form of family dissolution, may take the form of suicide, murder of spouse, annulment, separation (ranging from psychological withdrawal to legal separation), and, finally, divorce. The withdrawal from spouse may be accompanied by substitution for wife of a co-wife, concubine, mistress, or prostitute. Sometimes dissolution is by stages, for example, in a sequence of avoidance, suit for separate maintenance, legal separation, an interlocutory divorce decree, and finally an absolute divorce with the right of remarriage.

Family dissolution is an imperfect index of family disorganization. Divorce is an imperfect index of marital disorganization because there may be disorganization without divorce. For example, divorce was very difficult to obtain in England prior to liberalization of the laws in 1857 and again in 1937. When the divorce door is closed, marital disorganization may be inconspicuously endured. On the other hand, a couple with high expectations of marital bliss but relatively little marital disorganization may seek divorce as an exit from marriage. That marital disorganization is often not extreme is shown by the frequency of reconciliation, dropping of divorce suits, and remarriage to the former spouse. Thus, Harvey J. Locke (1951, pp. 54-55) found that about a fourth of the divorced couples he studied were in the same marital adjustment category as the less happy fourth of the married couples.

Religious and moralistic viewpoints . Family disorganization and dissolution are viewed by many people in moralistic and religious terms. The family at one time performed functions so important that disharmony and disruption were viewed with alarm. One assumption of the patriarchal system in western European culture was that paternal authority could insure efficiency and stability in family life. It is probably still the assumption of middle-class people in the United States that “nice” people get on together and as spouses stay together.

The Roman Catholic viewpoint concerning family disorganization and dissolution has remained rather clear and constant since the Council of Trent, 1545-1563. The ideal is patriarchal harmony established by a marriage sacrament directed toward the goals of parenthood and marital love. The Catholic doctrine assumes various impediments, such as disparity of religion, lack of consent, or intention to seek a divorce. Given the impediment of disparity of religion, the marriage of a Catholic to a non-Catholic, especially one un-baptized, can be nullified, thus permitting the Catholic party to obtain a civil divorce and marry a Catholic partner “in favor of the Faith.” Even a marriage between Catholics may be dissolved by papal dispensation if the marriage is unconsum-mated and grounds exist for the marital disruption. In addition to nullification and dissolution by dispensation, the church grants separation from bed and board, which does not include the right of remarriage. The consummated and valid marriage between two Catholics is indissoluble and permits no remarriage unless terminated by death. Certain Protestant denominations hold equally religious and moralistic views (Kirkpatrick [1955] 1963, pp. 573-575).

The legalistic viewpoint . The legalistic view, which is prevalent in the United States, had its origins in canon law and was influenced by the common law of England. It is complicated by the lack of uniformity among the various statutes and decisions of the 50 states; the practice of seeking divorce in states other than the state of residence has brought controversial cases to the U.S. Supreme Court (Harper & Skolnick [1952] 1962, pp. 451-458).

Basic to the legalistic doctrine of divorce in the United States is adversary procedure. It is assumed that a relatively innocent party may be granted a divorce because of the relative guilt of the other party with reference to “grounds” specified in the statutes of the various states. There are four legal doctrines associated with the basic adversary procedure. If spouses cooperate in providing evidence calculated to obtain a divorce on certain grounds, that is collusion, and the divorce should be denied. If one party forgives an offense such as adultery and continues to live with the spouse, condonation is assumed, and the derivative inference is made that the damage to the innocent party was not great enough to justify a divorce. If one party shows that the other party is equally guilty, successful recrimination is established and the divorce is not to be granted. If one party plots the guilty behavior of the other party, that is connivance, and the divorce would be denied to the allegedly innocent party. Thus the legal operation of divorce is withheld from marriages so sick as to menace the well-being of both adults and children in the familial group (Kirkpatrick [1955] 1963, p. 595).

Social pathology . Disorganization and dissolution are subject to a third viewpoint, that of social pathology. This viewpoint assumes that well-functioning, organized families are “normal”; maladjustment and disruptions are regarded as deviant behavior. The families with problems should be brought back into line by monetary aid, psychological manipulation, and a therapy designed to help people achieve conformity to the social norms. With little reflection upon alternative frames of reference, there is concern for the cure of pathological situations.

Sociological viewpoint . A fourth viewpoint is that of scientific sociology. In the choice of this label there is no intended implication that sociology is a distinct, mature, and sophisticated science. Perhaps at the present time it is merely an attempt at clear, honest, abstract thinking about social phenomena, increasingly aided by systematic observation and analysis. From such an intellectual orientation, which is relatively free from concealed value assumptions and distorting preconceptions, family dissolution is sometimes seen as a relief of family disorganization. For certain infections, amputation is the choice of a lesser evil as compared with further spread of infection. So it may be with divorce.

The sociological viewpoint is generally deterministic and assumes that social changes make inevitable certain reactions in terms of disorganization and reorganization. According to this view, familial institutions, groups, and relationships are part of a larger social structure in which every aspect of the total configuration is related to every other aspect. Family disorganization may be one facet of a larger social disorganization.

Methods and trends in research

Family crisis . Research concerning crises of the family has been extensive. The great depression in the United States brought forth books on unemployment as a crisis for the family, and World War II brought an interest in bereavement and other trials visited upon the family by wartime conditions. More recently scholars have speculated as to the devastating influence that an atomic war would have on the family.

The confusion of terminology in regard to disorganization and dissolution is compounded by the concept of family crisis. Death, like divorce, can cause the dissolution of family units, and often bereavement is placed in the category of family crisis. Hill writes of the event (A) which interacts with the ability of the family to meet a crisis (B); both (A) and (B) interact with the definition which the family makes of the event (C), and thus the situation is an (X)–the crisis for a particular family. One family may take in its stride an impact which would crush another family that is lacking in resources and prone to regard a trifle as a crisis (Waller [1938] 1951, p. 460).

Endogenous and exogenous disorganization. Kirkpatrick drew a distinction between endogenous and exogenous disorganization, the former having an internal origin and the latter a relatively external origin. Distinctions were also made between the normal and atypical, the timely and the premature, and the objectively severe as compared with the objectively trivial. It was suggested that the term “crisis” be restricted to exogenous disorganization due to an objectively severe atypical or premature event (Kirkpatrick 1955, p. 503). Insofar as family dissolution results from exogenous disorganization, as defined above, it would seem proper to speak of exogenous family dissolution; an example would be the wiping out of a family by flood or fire.

Bereavement. The assassination of President Kennedy in 1963 was a conspicuous atypical, extreme experience which brought premature bereavement. The disorganization in the Kennedy family, in the sense of shock and sorrow, was exogenous disorganization; and the dissolution, in the sense of the loss of the head of the family, could be regarded as exogenous dissolution. Perhaps all bereavement, save that due to violence within the family group, could be regarded as exogenous, and if premature or atypical, it would be a family crisis. An expected death from natural causes of old age would be less clearly a crisis by the criteria which have been set forth, although some degree of disorganization and dissolution might result. There is no claim of exclusive categories in the suggested terminology, and the distinctions are relative.

It is important to make a distinction between father-deceased bereavement and mother-deceased bereavement in view of their differential implications for the family in terms of economic consequences and implications for the offspring in terms of sex typing, identification, and preparatory af-fectional relationships with parents of the opposite sex. Many aspects of bereavement experience have been probed by Howard Becker, Willard Waller, Thomas D. Eliot, David M. Fulcomer, and others. Perhaps more attention should be paid to the loss of common memories. If a beloved person is no longer available to confirm a happening, there may be doubt about the reality of prior experience and uncertainty about personal identity. If the identity of A is essentially the identity of A in relation to B, the loss of B drastically alters the subjective identity of A. Improved knowledge of familial disorganization and dissolution calls not only for greater precision of terminology but also for further exploration of bereavement as a family crisis that may lead to subtle forms of exogenous disorganization.

Demographic investigation . Investigators have explored the distribution of familial disorganization and dissolution over space, by social categories, and in time. For example, relief cases of a certain type may be more prevalent in certain urban areas and within social categories such as “nonwhites.” Furthermore, there may be fluctuations in the case load of social welfare agencies by time periods. Unemployment associated with the business cycle has generally meant an increase in the number of families dependent upon external financial assistance.

The most extensive investigation has been that of divorce regarded as a measure of familial dissolution and, more specifically, of marital dissolution. Numerous studies have noted variations in divorce rates by geographical region, sometimes considered in combination with various social categories. Paul Click has studied “marriage instability” as measured by separation ratios, divorce ratios, and proportion of children not living with both parents. His hypothesis that marital disruption in the United States tends to be more closely related to urbanization than to regions of residence was only partially supported (1963, pp. 47-54).

Various writers such as William J. Goode, Robert Winch, and Christine H. Hillman have stressed the higher divorce ratios among persons of lower socioeconomic status (Goode 1964, pp. 88-90; Winch [1952] 1963, pp. 706-708). However, this evidence contains some inconsistencies when men and women are considered separately. One finds a rather high incidence of divorce among college women as compared both with less educated women and with men at the same college level (Kirkpatrick [1955] 1963, p. 589). It should be stressed that the proportion divorced in a particular group or category of people is affected by remarriage and the rapidity of remarriage. If every person receiving a divorce decree remarried on the same day, there would be no residue of unmarried “divorced persons” in the population.

Desertion, another mode of marital dissolution, may not be accurately reflected in the census categories “separated,” “spouse absent,” and “divorced,” because of remarriage. During 1955 a Philadelphia court recorded 4,224 desertions handled, in contrast to only 2,812 divorces (Kephart 1961, p. 548). Desertion may be a substitute for divorce or preliminary to the legal action of divorce.

The demographic approach stresses the changing prevalence of divorce in various countries. The trend of the divorce rate after World War II was upward in many European countries, such as France, England, and Sweden, as well as in the United States and Canada (Jacobson 1959, table 42, p. 90; table 47, p. 98; Kirkpatrick [1955] 1963, p. 586). For the United States, rates are available from 1920 through 1960, expressed as divorces per year per one thousand married females 15 years of age and over. A curve based on such rates shows a drop during the depression in the 1930s, a huge peak in 1946 following the war, a decline to a plateau, and a figure of 9.2 on that plateau in 1960 as compared with 8.0 in 1920 (Carter & Plateris 1963). From a sociological point of view, this increase is far from spectacular.

The clinical approach . For many decades investigators have utilized methods less objective but perhaps more penetrating than those of the demographers. The various theories of Freud concerning emotion-laden interaction in the family group were integrated by J. C. Flïïgel. Waller, in his brilliant pioneer study of divorced couples (1930), drew upon psychoanalytic concepts in his analysis of alienation and postdivorce adjustment. He expressed original intuitions concerning motivation, as, for example, the suspicion that some ex-wives seek alimony not out of greed but because of jealous disinclination to see an ex-husband financially capable of marrying another woman. Some psychiatrists probably go to an extreme in regarding the seeking of divorce as an expression of neu-roticism.

Historical and comparative studies . The older and well-known accounts of the family, such as those by Edward Westermarck, George Elliott Howard, and Arthur W. Calhoun, contain much historical and comparative information concerning disorganization and dissolution. A book by Lichten-berger, published in 1931, gives a thoughtful and comprehensive analysis of divorce.

More recently, Murdock (1950) made a systematic survey of divorce practices in 40 nonin-dustrial societies. In many of these societies he found divorce to be more frequent than in the United States, but he also found more preventives and functional adaptations. From this anthropological perspective, the American family does not seem to be in a state of hopeless decay. Following Murdock’s lead, various anthropologists have investigated determinants of divorce rates in various cultures. Ackerman (1963), from a study of 62 societies, found certain forms of homogamy to be associated with low divorce rates. Christensen (1963) has given an elegant demonstration of the fact that the association between premarital pregnancy and divorce varies with a culture’s sexual permissiveness.

Measurement and prediction . Beginning in the 1930s, researchers attempted to measure and predict marital disorganization. The investigators often collected information concerning background factors at the same time that information was requested concerning marital adjustment. Some of the correlations between the so-called prediction scales (based upon various items, for the most part premarital) and the adjustment scales probably involved a “halo effect” (see Kirkpatrick [1955] 1963, chapters 15, 18, and appendix). Well-adjusted respondents tended to report favorably on both their marriages and their childhood backgrounds. The reverse was probably true for maladjusted couples.

In the longitudinal or forecasting studies of such authors as Ernest W. Burgess and Paul Wallin, Truman L. Kelley and Louis M. Terman, information was first obtained on a sample of individuals, and then after a period of years further information was gathered concerning their marital adjustment. [See the biography of BURGESS.] In general, correlations between scores on a prediction instrument and some categories of outcome tended to be lower when the halo effect was controlled by the passage of time.

In connection with this body of research, a great deal of information was collected concerning factors associated with “marital adjustment” or other criteria of success. The findings of many of these studies were brought together, with the revelation that there was rather a striking lack of convergence in evidence (Kirkpatrick 1947). Cautious verbal generalizations about the convergence of findings concerning factors related to marital adjustment and a guide to the sources from which these findings were derived are available (see Kirkpatrick [1955] 1963, chapter 15 and appendix A).

Improvement of scientific knowledge

More precise terminology, pertinent distinctions, and rigorous theory would all be helpful in furthering scientific knowledge in the area under discussion. Since disorganization and dissolution are aspects of family behavior, any advance in knowledge of family systems in general would indirectly contribute to knowledge in this specific area.

Theories of mating . Winch has systematically carried over his functional theory of the family to the area of family disorganization and dissolution. If he is correct in his theory of complementary mating, a foundation would be laid for generalizations concerning the kind of matings which would avoid family disorganization and dissolution ([1952] 1963, pp. 567-655).

Kirkpatrick derived a theory of mating from a basic postulate concerning the influence of prior satisfactions and dissatisfactions in the course of family experience; this theory was called a theory of “selective needs” based on prior satisfactions and dissatisfactions. In some respects it is different from the Winch theory, and in other respects it is convergent and supplementary (Kirkpatrick [1955] 1963, pp. 300-307). According to the Kirkpatrick theory, an extreme of marital disorganization and perhaps consequent familial disorganization would exist when the spouse violates “minus-plus needs” by providing more of the particular kind of dissatisfaction which was distressing during a prior phase of family experience. For example, the wife who fails to compensate for the dissatisfactions experienced by her husband with his mother and aggravates these same dissatisfactions by behavior similar to that of the mother, risks marital disorganization and possibly divorce.

Criteria of scientific worth . Improvement of scientific knowledge in this as in other areas depends upon meeting five different criteria of scientific worth (Kirkpatrick 1939; 1959). One criterion is that of verifiability. In one sense verifiability implies repeatability of procedure, as in the case of experiments in the natural science field. There is some assurance that research is verifiable by others when it is presented as already verified by the investigators in the sense of proper sampling, validity and reliability of instruments, and correctness of logical inference. It is striking that so many findings concerning factors favorable to success in marriage have been nonverifiable or unverified.

A second criterion is transmissibility. A marriage counselor may achieve striking success and yet be unable to transmit his art or technique to a pupil. Often a description of divorce rates in a certain country cannot be transmitted to another scholar because symbols lack identical meanings for each person.

A third criterion of scientific worth is consistency. The contribution having this quality is logically sound, and the terms are carefully defined and consistently used with reference to their assigned meanings. Theory, hypotheses, research design, findings, and conclusion are interrelated. Any convergence or discrepancy in evidence is systematically presented.

The fourth criterion is scope. Family disorganization and disruption may be analyzed in a single brief case study, or a massive three-volume work may describe this aspect of family systems with global coverage. For example, Goode made an admirable study of divorce as affecting a sample of women in Detroit (1956) and in another book analyzed the family patterns as affected by industrialism in varied countries of the world (1963).

A fifth and somewhat multiple criterion is that of implication. Research having the quality of implication has one or more of four characteristics–namely, prediction, practical application, methodological innovation, and creative integration. It would be useful to construct a questionnaire that would predict divorce with 100 per cent accuracy. A study would have implications when, if applied to family counseling, it prevented 90 per cent of the needless divorces. It would be a methodological contribution to integrate the various measures of marital dissolution into an improved index of marital disorganization. Finally, the subcriterion of creative integration would be illustrated by a new integration of existing theories concerning family disorganization into a generalized theory that generated hypotheses subject to definitive verification by empirical research.

Statistics and indices . The improvement of scientific knowledge concerning disorganization and dissolution depends upon improved statistics. In recent years some improvements have been made in marriage and divorce statistics in the United States, which are often inferior to those of less wealthy countries. In 1958 a reporting system known as the “divorce registration area” was established. For inclusion in the divorce registration area a state must have central files of divorce records; include on its record forms certain standard items; receive regularly reports from local areas such as counties; and cooperate with the National Vital Statistics Division in regard to testing data for completeness and accuracy (Carter 1960). Further support to the movement for improved statistics was given by a committee on marriage and divorce statistics of the American Sociological Association, which published a report in June 1958. As part of an action program initiated by the federal agencies concerned with marriage and divorce statistics, a national sample of marriage and divorce transcripts was collected for the year 1960, and the official improved statistics for 1960 are based on this new source, which extends coverage beyond the 18 states in the divorce registration area.

There are various ways in which scientific knowledge would be advanced by improvement of marriage and divorce statistics. Certain needs and research possibilities will be mentioned.

The duration of marriage prior to divorce is a very important item of information. While it is claimed that the duration of marriage to the time of the decree is known for 96.2 per cent of all divorces granted in the United States during 1960, this information has not been fully used to compute the probability of divorce for couples marrying at a particular time. Such true probabilities for earlier years have been computed by Monahan (1959) and Jacobson (1959), but there is a tendency to regard mistakenly the ratio of divorces to marriages of the same year as expressing the risk of divorce.

Improved divorce statistics would reveal on a national basis the details of prior experience with marriage and divorce prior to a particular divorce that is being recorded and tabulated.

It would also be helpful to have national statistics on the total number of persons of various types who have been affected by divorce. People pass out of categories which imply a record of divorce, but they may carry the subjective experience of divorce with them into a remarriage or a relationship with a stepparent.

Statistics concerning divorce should be more closely related to improved statistics concerning separation and desertion. It may be that concern about the divorce of a particular date is inappropriate in view of the fact that the granting of a decree is merely a legal event far removed in time from significant experiences of the individual concerned (Monahan 1962).

It is highly important to have national statistics which are cross tabulated, showing the characteristics of one divorced spouse as matched with those of the other spouse. For research purposes it would be useful to have such cross tabulations by previous marital experience, religion, age at divorce, education, occupation and social status, and children resulting from prior marriages of the divorced spouses.

An interrelation of divorce, marriage, and birth statistics on a national basis would be helpful. Christensen (1963) has already demonstrated that such comparisons are useful in showing the association of premarital pregnancy with divorce. It may be that computer storage and retrieval of data will make the method of “record linkage” increasingly applicable.

Statistics concerning family dissolution which permit valid international comparisons would be desirable. Obviously, a country which regards divorce as a private family matter and lacks a governmental agency concerned with divorce may seem to have a low divorce rate because of underreporting. A logical extension of trends in the United States would be an international divorce registration area, having as members countries with acceptable statistics. Other countries might seek then to improve their indices of family dissolution.

Given the process of changing marital status, which is conspicuous in the United States and in other industrialized countries, there is need for statistics that contribute to an index of what might be called the velocity of family dissolution. The speed of separation, divorce, remarriage and redi-vorce, could be known only with the aid of information concerning prior family experience, which would be reported at each transition. It is known from unusually good statistics in Iowa for 1953 that the duration of marriage prior to a divorce ranges from an average of 6.3 years for primary marriages, to 3.2 when both parties have been married before, to 1.8 for couples breaking up their third marriage (Monahan 1959). Kirkpatrick attempted a crude measure of velocity of family dissolution by dividing the number of divorces per year by the number of persons who in the year following were in the status of divorced (1955, p. 534). A good index of velocity of family dissolution would require improved statistics and would take into account the speed with which children lose natural parents to acquire a stepparent and perhaps a series of such stepparents.

Knowledge of causes and consequences . The distinction between disorganization and dissolution must again be stressed. As compared with couples equally disorganized but undivorced (either contemporary or living in a previous century), certain modern couples seek divorce for special reasons. Often the woman is able to get a job and hence can insist on a high standard of marital happiness. Thus, dissolution may be a preventive of disorganization. Some people are less tied to marriage than others; family functions, religious taboos, or life-long commitment to parental roles may simply have less meaning for them (Landis 1963). The door of divorce is more open to certain couples than to others because of lenient laws in their locality, relative freedom to migrate elsewhere for an easy divorce, encouragement of relatives, opportunities for remarriage, and in general, the prospect of freedom without the economic, religious, and psychological penalties which have threatened other couples in various times and places. It can even be argued that divorce itself is a cause of divorce, in that frequent divorce weakens the norm of marital stability. Divorce frees bad matrimonial risks for remarriage and subsequent divorce; offspring imitate divorced parents; and the divorce of numerous friends makes the choice of divorce more normal and respectable (Kirkpatrick [1955] 1963, pp. 577-582).

Persons facing divorce during the latter half of the twentieth century in the United States have varied problems. Variations in the divorce laws of the 50 states contribute to the possibility of legal tangles. The divorce procedure is neither consistently routinized nor consistently established as an individualized therapeutic procedure; certainly, as compared with the bereaved person, the divorced person lacks a clearly defined role to play. The more meaningful any marriage is, the more painful will be its dissolution. But the confusion involved in divorce as compared with bereavement is often almost as disturbing as the previous frustration and deprivation. Seriously maladjusted couples are confused and torn between the choices of continued sickness in their marriage and the amputation brought about by the choice of divorce. The grim challenge is to find the choice which brings the lesser evil.

Effect on children. It can be argued that research concerning the effect of disorganization and disintegration on children is important, even if it is imperfect by the criteria previously mentioned. Children are the helpless victims of the mistakes of their elders; they have little control over their destinies and usually cannot select the stepparent who comes to them in the course of the divorce–remarriage process.

Much of the research has failed to compare the children of disorganized but undissolved homes with those from disorganized and broken homes of specific types. Landis and Nye found that children from broken homes fared as well or better, according to certain criteria, than children from intact but unhappy homes (Nye 1957). Burchinal (1964) considered homes reintegrated by remarriage, as well as broken and unbroken homes, and found no differential effect on adolescents.

In regard to the convergence of valid findings, much depends on the methods used, the type of effect on children, and the age of offspring at which possible effects are investigated. Landis (1963) found rather impressive evidence that happy parental marriages were favorably related to the dating experience of offspring. Robert R. Bell and James DeBurger, both using the sibling-cooperator method (Kirkpatrick & Cotton 1951, p. 82), found that parental divorce showed greater influence than bereavement on the marital adjustment of grown offspring (see Kirkpatrick [1955] 1963, pp. 665-674).

Applications of knowledge

Effective application of improved knowledge of family disorganization and dissolution depends on the emerging pattern of interprofessional cooperation. Lawyers, doctors, psychologists, social workers, educators, home economists, clergymen, and demographers, as well as social scientists, draw upon the growing body of knowledge in this area. That thriving interdisciplinary organization in the United States–the National Council on Family Relations–draws members from all of the above-mentioned professions. Furthermore, the emerging international organizations concerned with the family tend to be interprofessional in character.

Legal changes . Legal changes, both actual and proposed, tend to reflect a sharper international perspective. Varied explanations of the divorce trends in England include the influence of changes in the divorce laws. However, Max Rheinstein (1960) found little influence of legal changes on divorce rates in Germany. In the United States, a tabulation of the divorce laws in the various states as of 1952 was cross tabulated by the crude divorce rates of 1950 and the percentages of respondents in the various states answering “No” to a question on a poll of 1936 which read, “Should divorces be easier to obtain in your state?” There was a striking lack of correspondence between severity of the laws, behavior as to obtaining divorces, and public opinion concerning legal restrictions (Kirkpatrick 1955, p. 538). Of course, extreme differences in divorce law among various countries or extreme changes within an area, for instance, in South Carolina, do correspond in some degree with divorce rates.

David and Vera Mace note a more “folksy” and less legalistic type of divorce procedure in Russia than in the United States (Mace & Mace 1963, pp. 203-226). Olson (1961) makes similar observations with regard to Japan. In view of the long and, thus far, futile campaign in the United States to achieve either a uniform divorce law in the various states or federal control by a change in the constitution, it is interesting to observe that Australia established such a uniform federal law in 1959. This may well have some effect on the pattern of Australian divorce rates.

Family counseling . Counseling and therapy offer one means of applying improved knowledge concerning family disorganization and dissolution. At times, legal reforms converge with provisions for family counseling. In Australia the Matrimonial Causes Act (mentioned above) restricts dissolution before three years of marriage and provides for marriage counseling and conciliation; it authorizes federal payment for such services to approved marriage guidance organizations (Harvey 1964). The provision for counseling in connection with the divorce procedure in Australia and Japan may give encouragement to persons in the United States who, like Judge Paul W. Alexander, plead for a therapeutic approach, abolition of adversary procedure, and family courts that provide for counseling prior to the granting of a divorce (Kirkpatrick [1955] 1963, pp. 631-632).

There is some agreement, furthered by experience with counseling and therapy, that prevention of family trouble should be sought by a focus on education and counseling at an early period in the life span. Since family dissolution results in no small measure from the coming together of the wrong combinations of people, there is an argument for premarital counseling and guidance in the process of selecting mates. Many issues, however, remain controversial. For example, the small number of trained counselors in relation to potential demand provides an incentive to sacrifice quality to quantity by approving the activity of many relatively untrained persons as counselors.

Perhaps the dilemma of quality versus quantity may be solved to some extent by group counseling and therapy instead of individual treatment. The mutual support found in such groups is impressive. Certain instruments constructed for measuring attitudinal agreement, marital adjustment, and empathy in dyadic relationships might be adapted for self-analysis on the part of individuals and couples with minimal outside aid from experts (Kirkpatrick & Hobart 1954, pp. 10-19).

International trends . The international outlook with regard to family disorganization and dissolution offers an intellectual challenge that goes beyond the parochial applications of existing knowledge. However, some promising beginnings have been made; Goode, for instance, has noted the influence of industrialization upon differential family instability among the different social classes of various countries. These classes may vary, of course, as to degree of urbanization, type of kinship system, and feministic orientation (de Vis-scher 1956). The most valid generalization seems to be that with the spread of Western technology there tends to be increasing similarity in the family systems of the world, including their provision for marital dissolution. The diffusion of feminism may temporarily increase family dissolution, as women demand equality in the privilege of divorce and become economically independent of both husbands and kinfolk (Sysiharju 1960).

Family systems tend to have a certain resiliency under the impact of drastic social change. The free divorce of the 1920s in communist Russia has been replaced by a conservative and restrictive procedure. Hitler attempted reactionary change of the German family (Kirkpatrick 1938), but after the political and military crisis had passed, the general feministic and egalitarian trends characteristic of Western civilization reappeared (Schelsky 1953). As the dominance of the independent nuclear family is proclaimed, evidence appears of the survival of functional kinship systems. Carle C. Zimmerman, who expressed concern about the return of the unstable “atomistic family,” found substantial evidence that “friend-families” tend to be supportive and homogeneous in regard to low incidence of divorce (Zimmerman & Cervantes 1956, pp. 91–117; see also Shanas & Streib 1965).

Improved facilities for communication are characteristic of the modern world, on the international, the national, and the familial level. Cultural interaction may increase both the similarity of familial institutions and agreement concerning human goals such as peace, freedom, and happiness–ends to which social institutions are means. If improved communication reduces repression, hollow ritual, and pretense at every level, the pursuit of happiness and freedom may lead to an increase in divorce rates or other indices of familial dissolution in conservative areas of society. Nevertheless, if dissolution be distinguished from disorganization, certain increases in dissolution may mean merely a spread of higher aspirations and a reluctance to endure familial disorganization with its associated misery.

Clifford Kirkpatrick

[See also Marriage, article onfamily formation; and the biographies of Waller; Westermarck.]

BIBLIOGRAPHY

Ackerman, Charles 1963 Affiliations: Structural Determinants of Differential Divorce Rates. American Journal of Sociology69 : 13–20.

Burchinal, Lee G. 1964 Characteristics of Adolescents From Unbroken, Broken and Reconstituted Families. Journal of Marriage and the Family26:44–51.

Caplow, Theodore 1964 Principles of Organization. New York: Harcourt.

Carter, Hugh 1960 Plans for Improved Statistics on Family Formation and Dissolution in the United States. Social Forces39 : 163–169.

Carter, Hugh; and Plateris, Alexander 1963 Trends in Divorce and Family Disruption. U.S. Dept. of Health, Education, and Welfare, Health, Education, and Welfare Indicators[1963], September:5-14.

Christensen, Harold T. 1963 Timing of First Pregnancy as a Factor in Divorce: A Cross-cultural Analysis. Eugenics Quarterly10 : 119–130.

Click, Paul C. 1963 Marriage Instability: Variations by Size of Place and Region. Milbank Memorial Fund Quarterly41 : 43–55.

Goode, William J. 1956 After Divorce. Glencoe, Ill.: Free Press.

Goode, William J. 1963 World Revolution and Family Patterns. New York: Free Press.

Goode, William J. 1964 The Family. Englewood Cliffs, N.J.: Prentice-Hall.

Harper, Fowler V.; and SKOLNICK, JEROME H. (1952) 1962 Problems of the Family. Rev. ed. Indianapolis, Ind.: Bobbs-Merrill.

Harvey, Leslie V. 1964 Marriage Counseling and the Federal Divorce Law in Australia. Journal of Marriage and the Family26 : 83–86.

Jacobson, Paul H. 1959 American Marriage and Divorce. New York: Holt.

Kephart, William M. 1961 The Family, Society, and the Individual. Boston: Houghton Mifflin.

Kirkpatrick, Clifford 1938 Nazi Germany: Its Women and Family Life. Indianapolis, Ind.: Bobbs-Merrill.

Kirkpatrick, Clifford 1939 A Methodological Analysis of Feminism in Relation to Marital Adjustment. American Sociological Review4:325-334.

Kirkpatrick, Clifford 1947 What Science Says About Happiness in Marriage. Minneapolis, Minn.: Burgess.

Kirkpatrick, Clifford (1955) 1963 The Family as Process and Institution.2d ed. New York: Ronald Press.

Kirkpatrick, Clifford 1959 Sociological Significance of This Research. Pages 289-295 in Winston Ehrmann, Premarital Dating Behavior. New York: Holt.

Kirkpatrick, Clifford 1965 The Family. Pages 298-305 in Encyclopedia of Social Work. Edited by Harry L. Lurie. New York: National Association of Social Workers.

Kirkpatrick, Clifford; and Cotton, John 1951 Physical Attractiveness, Age and Marital Adjustment. American Sociological Review16 : 81–86.

Kirkpatrick, Clifford; and Hobart, Charles 1954 Disagreement, Disagreement Estimate, and Non-empathetic Imputations for Intimacy Groups Varying From Favorite Date to Married. American Sociological Review19 : 10–19.

Landis, Judson T. 1963 Social Correlates of Divorce or Non-divorce Among the Unhappily Married. Marriage and Family Living25 : 178–180.

Lichtenberger, James P. 1931 Divorce: A Social Interpretation. New York: McGraw-Hill.

Locke, Harvey J. 1951 Predicting Adjustment in Marriage: A Comparison of a Divorced and a Happily Married Group. New York: Holt.

Mace, David; and Mace, Vera 1963 The Soviet Family. Garden City, N.Y.: Doubleday.

Monahan, Thomas P. 1959 The Duration of Marriage to Divorce: Second Marriages and Migratory Types. Marriage and Family Living21 : 134–138.

Monahan, Thomas P. 1962 When Married Couples Part: Statistical Trends and Relationships in Divorce. American Sociological Review27 : 625–633.

Murdock, George P. 1950 Family Stability in Non-European Cultures. American Academy of Political and Social Science, Annals272 : 195–201.

Nye, F. Ivan 1957 Child Adjustment in Broken and in Unhappy Unbroken Homes. Marriage and Family Living19 : 356–361.

Olson, Lawrence 1961 How the Japanese Divorce. American Universities Field Staff, East Asia Series9, no. 8.

Rheinstein, Max 1960 Divorce and the Law in Germany: A Review. American Journal of Sociology65: 489-498.

Schelsky, Helmut (1953)1955 Wandlungen der deut-schen Familie in der Gegenwart.3d ed., enl. Stuttgart (Germany): Enke.

Shanas, Ethel; and Stbeib, Gordon F. (editors) 1965 Social Structure and the Family: Generational Relations. Englewood Cliffs, N.J.: Prentice-Hall.

Sysiharju, Anna-Liisa 1960 Equality, Home, and Work: A Socio-psychological Study on Finnish Student Women’s Attitudes Towards the Woman’s Role in Society. Helsinki: Tekijä.

U.S. National Center

for Health Statistics 1964 Marriage and Divorce: A Selected Bibliography of Statistically Oriented Studies. Washington: Government Printing Office.

Visscher, Pierre De 1956 Attitudes antiféministes et milieux intellectuels. Louvain (Belgium): Institut de Recherches Économiques et Sociales.

Waller, Willard W. 1930 The Old Love and the New: Divorce and Readjustment. New York: Liveright.

Waller, Willard W. (1938) 1951 The Family: A Dynamic Interpretation. Revised by Reuben Hill. New York: Dryden.

Winch, Robert F. (1952) 1963 The Modern Family. Rev. ed. New York: Holt.

Zimmerman, Carle C.; and CERVANTES, Lucius F. 1956 Marriage and the Family: A Text for Moderns. Chicago: Regnery.

FAMILY

FAMILY. There is no natural form of family, just as there has never been historical agreement about the meaning of the word itself. Throughout most of the early modern period "family" usually referred to all the members of one's household, including nonrelatives, such as servants and lodgers. At the same time some authors of the time clearly conceived of "family" in the sense of an extended kinship network. By the beginning of the nineteenth century both of these meanings had been largely supplanted by the modern sense of a cohabiting nuclear group of parents and children, although neither of the older meanings died out completely. Just as importantly there was no common family and household living arrangement during the early modern era and accordingly no clear transition from "traditional" to "modern" families but rather a plurality of household forms that continued to occur well into the nineteenth century.

KINSHIP AND THE HOUSEHOLD BEFORE THE EARLY MODERN ERA

The early modern household represented the basic unit of residence, production, and reproduction in the city and country alike. Its common interchangeability with "family" throughout the period had deep and ancient roots. Almost all modern European languages trace their words for "family" back to the Latin famulus, 'slave', thus signifying the dependence of relatives and servants on the paterfamilias or head of the household. Government officials from antiquity through the early modern period shared this patriarchal view of all authority and accordingly considered the hearth, or household, the key unit in census and tax calculations. Thus for practical as well as ideological reasons, the premodern household was the family.

This residential sense of family, however, coexisted with a broader definition as an extended group of kin or all of the people related by blood. Ancient Greeks and Romans tended to stress the unilineal agnatic kinship group (Latin gens, plural gentes ), tracing blood relations only through the father's ancestry, but still recognized the importance of relations through the mother's side (known as cognates). Latin, for instance, had one word for a paternal uncle (patruus) and another for a maternal uncle (avunculus). During the early and High Middle Ages the Germanic understanding of kinship as bilateral, involving blood relations from both parents' families, dominated. The marriage of two individuals from different clans (German Sippe; French race; Spanish raza; Italian razione ) thus had repercussions far beyond the couple itself. Now each had new parents, siblings, and cousins who were to be treated as blood relations, even after the death of one of the spouses, at least according to canon or church law. Until at least the twelfth century this type of broadly defined family by extended kinship constituted the strongest social, economic, and political bond throughout Europe.

About this time many aristocrats, knights, and wealthy merchants returned to the ancient practice of a patrilineal definition of kinship, inventing permanent family surnames and coats of arms that were passed down from fathers to sons. By the end of the Middle Ages the patrilineal movement had spread throughout Europe. Admittedly some members of the lower orders did not adopt the practice of a first and a last name until governments compelled them to do so at the beginning of the nineteenth century, but these were rare exceptions to the rule. In most cases the family surname was inherited from the father though some islands of bilateral lineage survived, such as Castile, where men often took their aristocratic wives' surnames. Sometimes, as in parts of France, a surname went with an estate and thus could potentially come from a nonrelative.

The success of the patrilineal form of genealogy, however, did not eliminate the importance of all maternal as well as paternal relatives in matters of inheritance, guardianship, and of course affection. Canon law also made no distinctions between the two branches of family in its restriction of marriage to those individuals outside the fourth degree of kinship. Thus family in both its broadest definition of kinship and the narrower sense of household survived into the early modern era.

THE HOUSEHOLD VIEWED FROM THE OUTSIDE: FORMATION AND COMPOSITION

The great diversity of marriage and inheritance practices in early modern Europe make generalizations about the formation and composition of the household extremely difficult if not impossible. There are, however, some notable characteristics, a few of them quite distinctive in history. In 1965 the historical demographer John Hajnal famously identified what he called "the European marriage pattern" of the early modern era. This apparently unique phenomenon, which he later revised to "the northwestern European pattern," originally described marriage practices in all the lands west of an imaginary line drawn between Saint Petersburg and Trieste. The two most striking aspects of this model were relatively late marriage (late twenties for men, early to midtwenties for women) and a high percentage (10 to 25 percent) of never married or widowed individuals. Hajnal theorized that various economic factors—such as extended journeyman status in a craft and a population surplus of marriageable girls and women—as well as religious factors (notably the Catholic celibate ideal) contributed to this pattern, which showed no signs of change before the nineteenth century.

The implications for household formation were significant. First, there was the size of the household itself. Late marriage tended to reduce the number of pregnancies and live births, the latter to about five to seven per woman by the age of forty. Given the high infant and child mortality rates, this meant a nuclear family under five per household most of the time—considerably smaller than was commonly assumed. At the same time the common practice of sending boys and girls by age twelve to apprentice-ships or domestic service meant that many youths would be considered part of a household other than their parents' for at least five and as much as fifteen years, depending on their securing permanent employment, property, or in the case of young women, a suitable dowry. Finally, the large number of single young people and widowed old people resulted in an unusually high proportion (by premodern standards) of lodgers or single households, especially in cities.

During the last forty years of the twentieth century, Hajnal's thesis underwent testing and considerable refinement. Reconstituting families by examining marriage contracts, wills, baptismal registers, tax records, and other legal documents, many historians have confirmed the frequency of late marriage and single households during the era, but they have also made clear that there was no one marriage pattern or household type among the diverse peoples of early modern Europe, east or west. Rather, certain ideal models tend to appear more in certain places and times, but even so there is a risk of distorting the variability of these household forms and their larger social implications in practice.

It is widely accepted that the most common type of household was the nuclear family, with evidence pointing well back to the Middle Ages, at least in England. Composed only of a married couple with or without children and possibly a servant, the nuclear family appears everywhere and at all times during the early modern era. If a son remained at home with his parents into adulthood and postponed setting up his own household, the nuclear unit became what is known as a stem family, also an apparently frequent model throughout Europe. When three generations of a family lived together or when any relatives other than the nuclear unit cohabited, the household became an extended family. This type of household, typically including one elderly grandparent, was most common in southern France and parts of German-speaking Europe, especially in the countryside. Households without a conjugal unit at the center, known as nonfamily households, were more common in cities. Households composed of several conjugal units—known to demographers as joint, multiple, or complex households and to early modern contemporaries as fraternas (Italian) or frérèches (French)—were most common in southern and eastern Europe. Again though, these geographical generalizations are merely crude approximations that are particularly difficult to support in the so-called transition zones of Hungary, Bohemia, Moravia, western Poland, and Estonia.

The erroneous presumption that most early modern families were large appears to have originated in the nineteenth century primarily as a reaction against the perceived dangers to the family (by then meaning the nuclear family) from industrialization and other modern developments. Numerous subsequent studies have instead confirmed Hajnal's conjecture that large or multiple households predominated only in eastern Europe, while small households remained the norm in the west. In Russia, for instance, three-quarters of all serf households through the eighteenth century were multiple, while in Scandinavia, England, and northern Germany the majority of households were nuclear, with only 2 percent of English households containing more than twelve people. In fact except for serfs or sharecroppers, whose landlords structurally encouraged large households, the poorer the family, the smaller the household.

The painstaking work of David Herlihy and Christine Klapisch-Zuber on a fifteenth-century Florentine tax census found that only one out of ten households had more than ten people and that most averaged four to five persons. Significantly those few large households in Florence, England, and elsewhere tended to be proportionate to personal wealth. Royal courts, for example, could encompass hundreds of retainers, and even the household of a seventeenth-century English lord might contain forty or more people (and many aristocrats maintained more than one residence). An ordinary English gentleman, by comparison, might have eight people in his "family," while the average household in the kingdom contained four to five individuals. Even among those households with more than three children, the relatively long span between births in many families combined with the early age of leaving for work often meant than no more than three children might be resident at a given time.

In addition to the various types of blood relatives included in different household models, three types of nonkin might also be in residence and therefore considered members of the family. The most common were domestic servants. In cities these tended to be girls from the country, sometimes relatives but usually strangers contracted for three or more years. The objective for the girls, typically starting work at age twelve or thirteen, was to earn enough money for a decent dowry with which they hoped to secure a good marriage. Salaries were one-half to one-third of what the maids could make in the fields, so there was great mobility during harvest time despite the obligations of their contracts and laws threatening punishment. The one exception to the preference for female domestics was the wealthy household, where many male retainers reflected on the social status and virility of the paterfamilias. Domestic servitude as a life phase was extremely common in early modern Europe, so at least one-third of households at any given time included servants. This also held for apprentices, under contract for up to seven years in the hope of learning a craft. In practice apprentices, who were usually teenage boys, ended up doing many odd jobs and other work not related to a craft and were in that sense on the same level as servants.

The third type of nonrelative who might be considered part of the household was lodgers. These individuals tended to be single young men, often journeymen, who stayed for as little as a few weeks or as long as several years. Unable to set up their own households yet, they were forced to rent and often travel, staying mostly in cities and providing a welcome source of income for many families. Often they worked on the family farm or in some kind of cottage industry.

THE HOUSEHOLD VIEWED FROM THE INSIDE: AFFECTIONS AND MATERIAL INTERESTS

Another modern myth about the premodern family has its origins in the work of the sociologist Philippe Ariès more than forty years ago. Though later scholars have somewhat caricaturized his argument, their criticism has nonetheless thoroughly demolished Ariès's most controversial assertion regarding the family, namely that the sentimental affections seen as normal between parents and their children were alien to Europeans before the eighteenth century. During the 1970s some historians, such as Lawrence Stone and Jean-Louis Flandrin, attempted to modify the thesis with their own theories of an early modern transition, but by the late 1980s the scholarly consensus had clearly swung in favor of greater continuity on the question of familial and parental love and affection. Obviously such matters are highly subjective and thus impervious to quantification. The sources available for an overwhelmingly illiterate society are extremely limited as well. Still one need go no further than the sensationalist press's accounts of infanticide and parricide or the plays of Shakespeare, Molière, or Richard Sheridan to grasp the extreme sensitivity of early modern Europeans to the unnatural treatment of blood relations, particularly acts of betrayal and murder among parents, children, siblings, and other close relatives. There was clearly something special about the relationship between parents and children that went beyond mutual duties and obligations, something that occasionally led to great extremes of passions and above all something that deeply affected one's own sense of identity.

This does not mean that early modern families were immune to material interests or the tensions that property often caused. The common source of such conflicts was of course the question of inheritance, and here early modern Europe possessed as bewildering a set of local variations as can be imagined. Some historians have attempted to tie type of inheritance practice to household type, yet while there is some rough correlation between impartibility and multiple households or between partibility and nuclear households, there are far too many regional exceptions to support any such generalization. Broadly speaking the two most basic distinctions were those between partible and impartible inheritance. In partible inheritance the patrimony, or total estate, was evenly divided among children, usually the sons. Occasionally daughters would inherit (especially if there were no sons), but their property would be administered by their husbands or male relatives. Though intended to minimize conflicts among siblings over inheritance, partibility often had the opposite effect, since the heirs had to inventory anything they received at any time. Dividing up the patrimony every generation could also lead to impoverishment, hence a gradual move toward impartibility developed by the beginning of the early modern era.

During the sixteenth and seventeenth centuries partibility began to be replaced in some parts of central and southern Europe by impartibility, which had already established a foothold in much of southern France and Spain. In most versions of impartibility, one heir (usually the oldest son, a practice known as primogeniture) inherited the entire patrimony, thus acting as a steward to the family property and preserving its integrity. The new practice was especially popular among royal dynasties, large landowners, and other wealthy families. In most places noninheriting brothers and sisters were to receive a "portion," that is, a cash settlement, and in many cases went on to work for the older brother. Among nobles on the Continent, the designated heir was often expected to provide an annual stipend for each of his brothers and a dowry to each of his unmarried sisters. English aristocrats, by contrast, were rarely required to make any such concession and only did so voluntarily.

Despite such counterbalances, many early modern people, especially among the Protestant clergy, considered impartibility inherently unfair and even unchristian. Moreover the multitude of diverse customs throughout Europe could make the rights and prerogatives bestowed upon the head of the household appear quite arbitrary. Within the kingdom of France, for instance, a father in Provence had absolute power over the choice of his heir, an exclusionary tactic that led to the designation of the fortunate beneficiary as l'enfant, 'child' (implying that his excluded siblings were not even children in a real sense). Meanwhile a father in Brittany had no power whatsoever to discriminate among his children. Occasionally a father might attempt to circumvent the local legal tradition and put an entailment in his will (known as a substitution in France; mayorazgo in Spain; majorat in Denmark; fideicommissum in Italy, Germany, Sweden, and Poland; and strict settlement in Britain) that prohibited sale, gift, or division of the land for several generations, but this was not always possible.

The question of inheritance could play a key role in relations within the family, including degrees of affection. Even small amounts of property could trigger bitter rivalries among designated heirs and their brothers and sisters, particularly in the case of children from two different marriages, where each individual or group of siblings attempted to exclude the other from inheritance. Many fathers clearly agonized over their attempts to preserve the patrimony and also provide for all of their children. The results could vary considerably. Impartibility tended to encourage emigration, early marriage, and nuclear households among noninheriting children, especially as travel to the New World became cheaper and more frequent, but many also remained near home. It also usually resulted in an extended family household for the adult heir and parent(s), though this too could vary, depending on the nature of the inheritance, local custom, and preferences of the father. Partibility carried its own set of conflicts and survived in many areas (for example, much of central and western Germany) well into the nineteenth century.

The head of the household's control over the family's patrimony obviously gave him a great deal of authority over his children's lives, particularly over the choice of spouses and the timing of marriages. As a rule the more property involved, the greater the involvement of fathers or male guardians in such matters, although forced marriages remained unusual and in most places illegal. Instead, fathers and children often sought a convergence of economic interests and personal attraction, apparently succeeding a good deal of the time. The prerogatives of inheritance also usually ensured that fathers or mothers would be provided for in old age, either in an extended household or with guaranteed income after passing the family property on. Heads of poor households, on the other hand, especially landless laborers, exercised no such economic authority over their grown children and thus enjoyed much less financial security in old age. Well into the twentieth century small households of single or married people over sixty were the poorest anywhere in Europe or North America, surpassed in that distinction possibly only by female-headed households with children.

The nonrelatives of an early modern household usually had no stake in inheritance strategies but played essential roles in family dynamics nevertheless. Writers of the era were fond of portraying servants and apprentices as scheming enemies of the master and the mistress of the house: lying, stealing, lecherous, and above all lazy. Indeed by the eighteenth century words such as "varlet" and "knave" had taken on almost exclusively derogatory meanings in the English language. Legal and other records convey a more nuanced and complicated tangle of relationships between servants and the relatives of the house. Some servants were treated like surrogate children, while others were clearly neglected or abused, receiving worse or less food, mean accommodations, and no affection. Even that common source of illegitimate babies, the illicit affair between the paterfamilias and a young maid, defies easy generalization. Every case had its own mixture of coercion (even rape) and willingness, of naïveté and cynical manipulation, of secrecy and flagrancy. The authority of the paterfamilias over servants and apprentices moreover was never absolute, though it did generally enjoy the legal benefit of the doubt. Religious authorities held the paterfamilias responsible for the spiritual instruction of the entire household, and immoral acts by children and servants alike reflected directly on his reputation. Protestant consistories and other church bodies frequently censored a father or widow for "bad housekeeping," though secular authorities rarely followed up with any punishments of their own.

THE HOUSEHOLD AS AN ECONOMIC UNIT

Before the wide-scale industrialization of the nineteenth century, the household was the key unit of production in European economies. Urban workshops and rural farms alike relied on the labor of household members, including parents, children, servants, apprentices, and sometimes wage-earning lodgers. The degree to which this "whole house" (das ganze Haus) economic model successfully functioned remains a matter of some dispute among family historians. The key for the early modern paterfamilias lay in meeting his family's immediate and future needs while minimizing the number of mouths he had to feed. In this respect rural households west of the Hajnal line showed much greater flexibility, largely because of the high availability of servants and lodgers to balance the size of the household with the size of the holding. The size of eastern European households, usually complex in structure, was less flexible and therefore fixed the amount of labor available, regardless of the size of the property to be worked. Over all, many factors—type and quantity of product, local agrarian system, interregional and international markets, kin and other social networks, property laws, demographics, and so forth—helped determine which labor strategy a head of household adopted.

The division of labor in the early modern household had not always been gender and age specific. During the Middle Ages wives apparently often shared in their husbands' craftwork, even into widowhood. By the sixteenth and seventeenth centuries, though, this type of labor was increasingly restricted for women, who nevertheless often continued to bring in income through sewing or spinning work (which came to be defined as part of "housekeeping" and thus not really "work"). The division of labor along gender lines was long familiar in rural settings, where men were assigned to tasks requiring greater physical strength, such as plowing and planting, while women (and children) worked on food preparation, cleaning and mending of clothing, housecleaning, water carrying, and so forth. During harvest all members of the household played various parts in getting crops from field to market.

Two major economic events during the early modern era had especially significant consequences for the household economy. The first was the rise of wage labor following the late fifteenth century. At the very time when new impartible inheritance laws were sweeping across Europe, an increasing number of wage-earning jobs in the West provided noninheriting children with an alternative basis for setting up their own households. This loss of labor forced some farmers to hire their own help, further increasing the number of wage earners in the economy and nonrelatives in the household.

The second development began in the mid–seventeenth century and likewise fed the growth of a wage economy in many parts of western Europe. Proto-industrialization, particularly in the textile industry, had the greatest impact on small landholders, whose members (women, children, and sometimes men) increasingly turned to some phase of cloth production for their households' incomes. Unlike the traditional guild system, in which all phases of production were under the direct supervision of a master in the craft, the "putting-out system" (German Verlagssystem ) assigned an intermediate task to an outsider, who was paid at an agreed-upon piece rate. The growth of these so-called cottage industries was gradual and came as a consequence of changes in agrarian markets as well as higher demands and therefore increased production. Before the mechanization and factories of the industrial revolution, for instance, it took four to ten spinners to produce the amount of yarn woven by a single weaver in one day. Like large landowners who had increasingly enclosed and converted their farmland to sheep-grazing fields, these cottagers found income from the cloth industry both more profitable and more reliable than farming.

The immediate impact of increased wage labor and proto-industrialization on household forms was not always clear-cut. Clearly wage labor tended to encourage earlier marriage and more nuclear households. No longer forced to wait for inheritance, noninheriting children had little reason to remain part of an extended or complex household rather than start their own. Yet many such individuals, for reasons of security or family solidarity, did stay on well into adulthood despite the conditions. Similarly proto-industrialization sometimes encouraged more nuclear households and other times had the opposite effect. The formation of new households depended on many factors, including the skills required for production (an important limitation in such industries as arms manufacture in Belgium), the initial outlay of capital required for equipment (for example, expensive looms), and the size of the household.

IDEAS ABOUT THE HOUSEHOLD AND FAMILY

Since antiquity family and household have served as powerful metaphors as well as social realities. Aristotle considered the household (Greek oikos ) the chief building block of society as well as the model for a successful state. "It was out of the association formed by men with . . . women and slaves," he wrote in his Politics, "that the household was formed. . . . The next step is the village. . . . The final association, formed of several villages, is the city or state." Aristotle considered both the household and the state "naturally" hierarchical and patriarchal, with the head of the household and head of state each possessing certain duties as well as prerogatives. Their paternal authority demanded filial obedience but also obliged household heads to display loving concern. Later the Roman paterfamilias and emperor enjoyed expanded rights, often to a shockingly autocratic degree by modern standards. During the sixteenth and seventeenth centuries European writers revived the classical tradition of the household as metaphor for the state, arguing for a strengthening of the patriarchal leader of each. In Germany a number of popular Protestant pamphlets, the so-called Hausvaterliteratur, lamented that the authority of the head of household, or Hausvater, was under assault from an unholy alliance of shrewish wives, wild children, papist agitators, and a variety of devils bent on destruction of the family. A father should be as a king in his own household, they argued, providing biblical, classical, and anecdotal evidence in support. Publications elsewhere in Europe similarly invoked the sovereignty of the paterfamilias, whom Thomas Hobbes (1588–1679) likened to the head of a small monarchy.

The other side of the patriarchal metaphor consisted of a paternalization of political authority. This imagery too was ancient and continued to inspire many medieval paeans to fatherly kings, such as Saint Louis (Louis IX; ruled 1226–1270) of France. During the sixteenth and seventeenth centuries, though, proponents of a stronger "absolutist" monarchy, such as Jean Bodin (1530–1596), applied paternal claims of sovereignty to argue for greater political authority for their kings. Only a father's strong hand, Bodin maintained in his Six Books of a Commonwealth (1576), could prevent the anarchy that had overtaken France during its religious wars.

The comparison of a kingdom to a household probably reached its zenith with the writings of the Englishman Robert Filmer (c. 1588–1653). Beginning in the 1630s Filmer published several tracts that made an argument similar to Bodin's but with even more extensive use of patriarchal imagery. Unlike the proposed absolute monarch of Thomas Hobbes, Filmer's ruler—like a true father—ruled more by moral suasion and education than by force. His ultimate authority, however, was beyond dispute, dating back to Adam and later the dispersion of Babel: "The Nations were distinct Families, which had Fathers for Rulers over them; . . . God was careful to preserve the Fatherly Authority, by distributing the Diversity of Languages, according to the Diversity of Families." After the English Civil War and Commonwealth, Filmer's writings received new attention, culminating in the printing of his previously unpublished masterwork, Patriarcha (The natural power of kings) in 1680. By then patriarchal political imagery and language were so pervasive that even John Locke (1632–1704), who dismissed Filmer's absolutist arguments as "glib nonsense," was forced to acknowledge the paternal nature of government.

The patriarchal revival also had its religious dimension. Both kings and fathers, as Filmer noted, received their authority directly from God, most explicitly in the fourth commandment to love and obey one's parents. This "divine right" to rule became closely identified with the absolutist monarchies of England's Charles I (ruled 1625–1649) and France's Louis XIV (ruled 1643–1715), but the sentiment, if not the political implications, was widespread in early modern Europe. At the same time both fathers and political authorities had clearcut responsibilities for the religious welfare of their respective "families." Martin Luther (1483–1546) even referred to both figures as bishops expected to lead by example and discipline every member of their respective realms. This highly idealized patriarchal hierarchy was best expressed by German authors, who wrote of a Hausvater (head of household) subject to a Landesvater (prince or "father of his country"), with both overseen by the Gottesvater (God the father) himself.

Throughout the early modern period many religious sects and political groups also employed the language and imagery of family. All of the Reformation's leaders spoke to their congregations as "families," and some groups, notably the Anabaptists and their successors, called each other "brother" and "sister." Later movements, such as the Society of Friends (that is, Quakers) and Moravian Brethren, similarly turned to the language of family for coherence and identity. Finally, secular political groups from the Masons to the Jacobins of the French Revolution openly proclaimed brotherhood (French fraternité ) as one of their foundational tenets.

BIRTH OF THE MODERN FAMILY?

Just as no single "traditional" family model ever existed, no single modern family model succeeded it. Nuclear households admittedly became more common during the nineteenth century, but complex and alternate household forms continued to thrive in some places, particularly southern and eastern Europe. "Modernization," in the guise of either industrialization or increased individualism, also did not spell the end of the importance of kinship; some historians argue that family relations became even more important as a result of such larger social transformations. On the other hand, the household itself did experience some significant changes. Most notably, historians detect a discernible increase after the sixteenth century in the desire for privacy, resulting in somewhat larger and more compartmentalized residences among the middle and upper classes. The idea of "home" itself took on a form of separation from society, a haven in a tumultuous world. By the eighteenth century a new "cult of domesticity" was growing, and by the following century it spread to lower-middle-class and working-class cultures. Like the patriarchal model that preceded it, the new idealization of the household often remained a common point of reference rather than a social reality. Still it corresponded nicely with the continuing rise of nuclear households, a convergence one might call the congealment of the modern family if not its birth.

See also Authority, Concept of ; Childhood and Childrearing ; Divine Right Kingship ; Divorce ; Gender ; Inheritance and Wills ; Marriage ; Proto-Industry ; Women .

BIBLIOGRAPHY

Anderson, Michael. Approaches to the History of the Western Family, 1500–1914. 2nd expanded ed. New York, 1995.

Ariès, Philippe, and Georges Duby, eds. A History of Private Life. Vol. 3: Passions of the Renaissance, edited by Roger Chartier. Cambridge, Mass., 1989.

Burguière, André, et al., eds. A History of the Family. Vol. 2, The Impact of Modernity. Cambridge, Mass., 1996.

Flandrin, Jean-Louis. Families in Former Times: Kinship, Household, and Sexuality. Translated by Richard Southern. Cambridge, U.K., 1979.

Goody, Jack. The Development of the Family and Marriage in Europe. Cambridge, U.K., 1983.

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, pp. 101–138. London, 1965.

Herlihy, David, and Christiane Klapisch-Zuber. Tuscans and Their Families: A Study of the Florentine Catasto of 1427. New Haven, 1985.

Houlbrooke, Ralph A. The English Family, 1450–1700. London, 1984.

Hunt, Lynn. The Family Romance of the French Revolution. Berkeley, 1992.

Kertzer, David I., and Marzio Barbagli, eds. The History of the European Family. Vol. 1: Family Life in Early Modern Times, 1500–1789. New Haven, 2001.

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

Medick, Hans, and David Warren Sabean, eds. Interest and Emotion: Essays on the Study of Family and Kinship. Cambridge, U.K., 1984.

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.

Ozment, Steven. When Fathers Ruled: Family Life in Reformation Europe. Cambridge, Mass., 1983.

Plakans, Andrejs. Kinship in the Past: An Anthropology of European Family Life, 1500–1900. Oxford, 1984.

Roper, Lyndal. The Holy Household: Women and Morals in Reformation Augsburg. Oxford, 1989.

Sabean, David Warren. Property, Production, and Family in Neckarhausen, 1700–1870. Cambridge, U.K., 1990.

Stone, Lawrence. The Family, Sex, and Marriage in England, 1500–1800. London, 1977.

Traer, James F. Marriage and the Family in Eighteenth-Century France. Ithaca, N.Y., 1980.

Wall, Richard, Jean Robin, and Peter Laslett, eds. Family Forms in Historic Europe. Cambridge, U.K., 1983.

Joel F. Harrington

FAMILY

FAMILY is vitally important to most religious traditions in two closely interconnected ways: Various ritual processes enacted by, to, and for the family help to create and sustain it as well as give it meaning, and it functions as an important symbol of deity. Historically and cross-culturally, family in various forms has (until the late twentieth century in postindustrialized cultures) been so basic to human existence as to be a universal symbol of ultimacy.

Definition

Exactly what constitutes family is not always clear. Some scholars equate family with household, another imprecise construct that variously includes all permanent members such as servants or else excludes unrelated householders. Further confusion results because most anthropologists posit two basic kinds of family: the nuclear family, consisting of mother, father, and unmarried children, and the extended family, typically including mother, father, all unmarried children, and one or more sons with their wives and children. Numerous complicated variations exist, including different polygynous arrangements in which two or more co-wives live under the same roof. A few domestic groupings, such as those of the Nayar of India, whose men never live with their wives, defy all categories. Nonetheless, family, in some variant, is considered universal.

Also confusing is the fact that all married people simultaneously belong to two different families. Family as seat of origination stresses ties of blood, whereas family as affiliation emphasizes bonds of marriage. To keep separate these two different kinds of family, some anthropologists designate the first as "kin" and the second as "family." Kin are those who share common ancestors, as do mother and child (in contrast to mother and father, who do not). Strictly speaking, although family incorporates kin, the reverse is not true because, excepting incestuous marriages, spouses usually are not blood relatives.

Consequently, family is basically a reconciliation of many different opposites: female and male, life and death, ascendants and descendants, kin and affines (relatives by marriage), biology and culture, freedom and servitude, corporation and individuality. The differing ways in which family contains these opposites represent diverse systems of order in which family roles are valued according to accepted local religiocultural belief and custom. Valuation of all members is almost never equal; therefore, family as a whole embodies and symbolizes order of a particular sort—hierarchy.

In its entirety this "natural" order of human relationships, presumed to have evolved out of earlier hominid bands of approximately thirty, has frequently been deified, with members typically reflecting family as experienced in a particular culture. Thus Kwoiam, the warrior hero of Mabuiag, an island off New Guinea, lives with his mother, her brother and sister, and his sister's son in a matrilineal "family" (technically, a kin system) that omits the father. Very different is the Homeric extended patriarchal family of Zeus and Hera, which includes variously begotten offspring. The smaller nuclear family is symbolized in various cultures, as for example the Egyptian Osiris, Isis, and Horus; the holy family of Christianity consisting of Joseph (or God the Father), Mary, and Jesus; and the holy triad of the Yurak Samoyed: Nyebye-haha, the mother deity; Wesako-haha, her spouse; and Nyu-haha, their son. Curious variants appear in the enneads (triple triads) so characteristic of dynastic Egypt.

In its smallest possible configuration (apart from the single individual sometimes defined as family in the post-industrialized West)—as husband and wife—family appears in almost all mythologies. Universally, tales of the hieros gamos tell of the sacred marriage of Heaven and Earth from whom humanity springs, as illustrated by the Zuni Awitelintsita, the fourfold-containing Mother Earth, and Apoyan Ta ʾChu, the all-covering Father Sky. Often such etiological stories of how the world came to be tell how one or more of the children produced by the union separate the pair, often forcibly, to form the realms of earth and sky. Such is the case in the Vedic account of Dyaus and Prthivi.

Ancestors

Probably no members so fully embody both the ritual and the symbolic significance of the traditional extended family as do ancestors. From the Paleolithic period to the present, many cultures have venerated ancestors to varying degrees, although Herbert Spencer's theory that ancestor worship stands behind all religious practice has been generally discarded. For example, almost all Native American tribes believe that spirits of tribal ancestors return to earth to warn, protect, and instruct the living, although only specially trained shamans are capable of seeing them.

Babylonian mythology and artifacts incorporate important motifs of ancestor veneration. The failed attempt of the hero Gilgamesh to escape mortality by visiting his ancestor Utanapishtim, the Babylonian Noah who did escape it, indicates the salvific role hoped for from ancestors. Between the third millennium bce or earlier, when sacrifices were offered to the departed kings Shulgi and Gudea, and about 2500 bce, when Grimalsin of the second dynasty of Ur appears to have been deified while still living, two other important themes emerge: Ancestor worship by actual descendants tends to merge with homage paid by a whole people to departed rulers or "fathers." Thus, in many cultures ancestors function variously as objects of domestic and state devotion, a situation that became pronounced in the Roman Empire. Attribution of divine ancestry has been common for kings, as notably in post-Meiji Restoration Japan, where the emperor was officially proclaimed a direct descendant of the sun goddess Amaterasu Ōmikami. Such ancestry has even been assigned to whole peoples, as repeatedly shown in epic poetry.

So important are the honored ancestors in cultures such as those in China, Japan after Chinese contact (seventh century ce), and areas of Aryan influence, particularly India and Rome, that the traditional family often seems to exist more for their sake than for that of the living. This point indicates one theme present in the traditional family, its orientation toward death. Furthermore, because typically in these cultures ancestors collectively overpower and stifle the individuality now common in the Western world, ancestor veneration also highlights a second important theme: Family as corporate entity strongly opposes the individuality of its members.

Emphasis on ancestors indicates that family is not only the matrix within which an individual enters life but also the means by which he (less commonly she) achieves a kind of immortality. Paradoxically, this denial of death that leads to ancestor veneration makes the family a kind of perpetual cult of the dead enacted by the living.

Children

In contrast to dead family members, who are almost universally venerated, children are often treated ambivalently. Though desired in the abstract for perpetuating the family, children may be abused or even denied life, as in the ancient classical world. Hippocrates illustrates this point when he asks, "Which children should be raised?" The essentially universal theme of infanticide is clearly present in the biblical stories of Isaac and Moses, who were saved. To this day the practice continues sporadically for girls in parts of India and China, as historically had been the case almost worldwide.

The countertheme of life orientation surfaces most strongly in connection with those newborns elected to survive. Yet even here the tension of opposites is strained, for only some attain family membership. Commonly thought of as "natural," family construction is actually often highly artificial.

Birth, mating, and death, the three natural methods of creating, maintaining, and pruning families, are simultaneously both biological and socioreligious events. Successful delivery of a live baby does not guarantee the existence of a new family member. In many cultures, once a child is born (notably, in patriarchal Hellenic Greece or even contemporary China or India), the father must determine whether or not to keep it. Then it must be incorporated into the family. The contemporary Islamic Malays illustrate one variant of this once nearly universal practice: First, the father whispers into the infant's ear the Islamic call to prayer; next, a specially selected person touches certain objects to the baby's lips to guard against future lying and gossiping; then, forty-four days later, the father buries the placenta beneath a coconut palm seedling. These and other birth rituals help place the child in its familial and socioreligious context. Thus a new family member is "created" only in the most superficial way by its actual birth. Subsequent actions of family members, often other than the mother, bring the child fully into family life.

As a symbol of deity, the divine child appears in various traditions. Archaeological finds such as vase paintings and figurines depicting infancy themes and rituals place this concept at least as far back as the Neolithic and Chalcolithic periods (c. 7000–3500 bce) in Old Europe (roughly, southeastern Europe from Czechoslovakia to the Aegean). Motifs of birth and maturing of the infant later took shape mythically and cultically in many variations that recount the passion of the young god of vegetation. Representative is the cult of the infant Dionysos, originally Boeotian and Cretan but subsequently almost universal in Greece, in which the infant Dionysos-Zagreus is dismembered. According to myth, the Titans lured the child with rattles, knucklebones, a top, a ball, and a mirror, then cut him to pieces, cooked him, and devoured him. In some versions he is resurrected by the earth mother, Rhea. This death and resurrection theme, common to the complex of images central to agrarian religion, finds in the child (or alternatively in the seed) an appropriate image of renewal.

Worship of the divine child was originally shared by or even predominantly directed to the mother goddess, as in the case of Ishtar, Astarte, and Cybele, whose son-consorts were of secondary importance. With time, however, the child, originally of either sex as suggested by numerous Sumerian female Marduks, ceases to be merely the child or sacrificial consort and becomes more and more an object of veneration in its own right. Christianity epitomizes this process whereby the divine child eclipses its mother.

In a very different form, images of the divine child as divine hero are also common in Native American mythology. This pattern is typified by the Haida story of Shining Heavens. One day a Haida woman was digging on the beach. Hearing a cry from a cockle shell, she uncovered it and found the baby Shining Heavens. She took him home and soon discovered his supernatural power, manifested in his ability to grow up almost immediately. This common motif of the wonder child who grows almost instantly from baby to strong youth or man is also illustrated by the Siouan Young Rabbit and the Algonquian Blood-Clot Boy. Sometimes the child-hero even makes plans in the womb, as do the Iroquoian twins Good Mind and Evil Mind. Such "unnatural" capabilities illustrate the power of godhead to transcend nature.

Such capability is even more apparent in the Vallabha and Caitanya sects of Hinduism, in which worship of the cowherd Kṛṣṇa as the divine child has been popular from at least 900 ce. In the spontaneity of his laughter, pranks, dancing, disobedience, and play, the child Kṛṣṇa symbolizes the unconditional nature of divinity. In such activity, engaged in for no purpose beyond sheer joy, the play of the child metaphorically expresses an aspect of divinity less easily rendered by "adult" personifications.

Mothers

So important is a woman's role as mother in most societies that the biblical Hebrews, for example, insisted that a wife who failed to bear children was obligated to provide her husband with a concubine (Gn. 16:2). According to popular Islamic tradition, the main duty of a woman is to obey and serve her husband respectfully; her second duty is to give him male heirs. In traditional China with its strong Confucian ethic, life was meaningless without sons. Without sons, a wife could count on a second wife essentially replacing her.

Theorists assume that the discovery of stockbreeding and planting taught humans about male reproductive capability. That means that for only about twelve thousand of the million years of hominid existence have humans understood paternity and reckoned male as well as female lineage. Thus was the ancient mother-child kin tie challenged by the familial tie. The nineteenth-century belief in mother right, espoused by J. J. Bachofen, Robert Stephen Briffault, Henry Maine, and others, whereby women were thought to have held social and political power during a prepatriarchal era, has long been invalidated; but current scholarship makes indisputable the existence of a practice of prehistoric mother worship in Europe and Asia Minor.

Material evidence in the form of large numbers of "Venuses," often with exaggerated secondary sex characteristics and pregnant bellies, as exemplified by the well-known Venus of Willendorf, firmly roots the idea of divine motherhood in the Upper Paleolithic period in Eurasia (c. 22,000 bce). By the time of the Neolithic and Chalcolithic periods (c. 7000–3500 bce) in Old Europe and the Near East, the Great Mother, with her accumulated Paleolithic traits, is well established in the variant forms that generally characterize her in agricultural societies around the world. (In patrilineal totemic and patriarchal nomadic cultures she figures less prominently, as an adjunct to the dominant sky god.) Under various names she appears almost universally wherever agriculture develops—as Ishtar (Babylon and Assyria); Astarte (Canaan); Isis (Egypt); Cybele (Phrygia); Rhea, Gaia (pre-Hellenic Greece); Pṛthivī (Vedic India); Di (ancient China); Pachamama (Inca); and so on.

In cultures such as those of Old Europe, pre-Hellenistic Greece, and pre-Vedic India (Harappa, Mohenjo-Daro), which were not dominated by nomadic pastoral peoples, motherhood is typically aligned with a concept seemingly opposed to it—that of virginity. But this belief reflects the archaic notion that birth results from parthenogenesis, an understandable belief for those unaware of the male role. Far from being a moralistic concept, as it subsequently became in patrilineal and patriarchal cultures, it originally reflected an understanding of woman as creator and powerful figure in her own right.

Earth and related vegetal phenomena such as grains are not the only natural elements associated with motherhood. Water, the medium from which humans originally emerge onto land, also functions this way, as with the ancient Mexican goddess of the waters, Chalchiuhtlicue, and the water mother common to the ancient Karelians and other Finno-Ugric peoples. Sometimes, as with the Japanese sun goddess Amaterasu, or the pre-Islamic mother of the heavens Allat, or the Egyptian sky goddess Nut, the traditional association of earth with motherhood and of sky with fatherhood are reversed; consequently, the predominant associations do not always hold, as when the concept "down," normally affiliated with earth and motherhood, attaches to a male chthonic deity. The variant son-consorts of the mother goddess, such as Adonis and Tammuz, reflect this phenomenon.

Various interconnected processes particularly affect the ways motherhood is represented in divinity and vice versa. Specialization tends to separate qualities originally mixed together in a single great goddess figure into different embodying images, as exemplified by the goddesses of the Homeric pantheon. Artemis and Aphrodite, for example, both lose their original fullness of personality to become mainly associated with the hunt and with erotic love, respectively. In this way motherhood, especially in Western cultures dominated by monotheism, has typically been strictly separated from all other potential and actual attributes of womanhood. Thus hunting, wisdom, sex, and war, all attributes of the undifferentiated goddess, come to appear totally divorced from each other.

A related process polarizes "good" and "bad" qualities into beneficient and terrible goddesses. Such terrible mothers of death and destruction as the Hindu Kālī, the Aztec Tlamatecuhtli, and the Greek Medusa typify this process. Such splitting dichotomizes the originally unified cycle of birth and death, in which Mother Earth gives birth (often quite literally, as in the Greek story of Erichthonius and in the many Native American myths that portray humanity emerging from the womb of the Earth) and later takes back her dead for burial (as in the Pueblo belief that Shipapu, the underworld, is also the womb of the earth goddess, Natya Ha ʾAtse.) With polarization come goddesses of the underearth realm such as the Greek Persephone and the dread Sumero-Akkadian Ereshkigal, who are separate from beneficent counterparts such as Demeter and Ishtar.

In a variant process the single goddess multiplies, usually into a triad, as in the case of the Scandinavian Norns, the Greek Fates, or the strange matres and matrones figures from the Celtic and Germanic provinces of the Roman Empire. Such trinitarian representations often involve different stages of motherhood, as in various ubiquitous virgin-mother-crone triads (the Hindu Pārvatī, Durgā, and Umā and the Celtic Macha, Morríghan, and Badhbh, for example).

Fathers

Next to ancestors and frequently amalgamated with them conceptually, fathers hold the greatest power in traditional patriarchal families, whether or not their fatherhood is biological. This paradox is logical when fatherhood is divided into three categories: The genetic father fertilizes the ovum; the genitor contributes to the child's growth in the womb, as when the Holy Ghost causes Mary to conceive through her ear; and the social father, known as the pater, dominates family life. Whether as genetic father, adoptive father, or maternal uncle, the pater supplies the child's social position.

In a patrifocal extended family the pater, as oldest living father in a direct line of descent, firmly heads the family hierarchy. This pattern was thoroughly worked out in the Roman family, where patriarchal power was so complete that, until he died, the father retained limitless authority over unmarried daughters and grown sons and their children. A married daughter customarily joined the household of her husband and so came under the authority of his father. Such extreme paternal power distinguished the Roman father from fathers in other societies in degree, but not in kind.

In contrast to motherhood, which results from pregnancy and childbirth, fatherhood is not immediately self-evident. Nor can fatherhood be as readily represented in images. Development and evolution of the concept are consequently less certain and less easy to follow. Almost everywhere, the most archaic manifestation of divine fatherhood is the "high god" located in the sky. Typically this "father" is originally a creator whose traits include goodness, age (eternity), and remoteness from the world of human affairs. So transcendent is he that he often abdicates his role of creator, handing it over to a successor-demiurge. Consequently, he is seldom reverenced in cult and may even disappear entirely. Representative examples are the Australian All-Father deities Baiame, worshiped by the Kamilaroi, and Bunjil (of the Kulin tribes); the Andamanese Puluga; numerous African father gods such as Nzambi of the Bantu-speaking peoples and Nyan Kupon of the Tshis. Existence of a sky god of this sort is evident from Neolithic times on and may well go back to Paleolithic times, but hard material evidence to prove it is currently insufficient. Aside from images suggesting the bull-roarers universally associated with father gods, no images comparable to the Paleolithic Venuses have been discovered.

The fatherhood of such archaic deities is often less specifically biological than creative, as reflected by the terms Bawai and Apap, applied respectively by the African Chawai and Teso, which convey the fatherhood of God relative to creation. In this sense the supreme being is a "father" whether or not he creates in the well-known hieros gamos of Mother Earth and Father Sky or through powers entirely his, as Baiami does.

By contrast, in many less archaic mythological and ritualized conceptions, divine fatherhood is unmistakably biological. Here the archaic mating of Mother Earth and Father Sky, originally an abstract description of creation, becomes far more concrete. The sovereign father is typically eclipsed by his son, as is the Greek Ouranos by Kronos, the Australian Baiami by Grogoragally, the Tiv Awondo by the Sun. Thus, one theme typically connected to divine fatherhood in most mythologies is the generational conflict of fathers and sons. As the archaic father god recedes, the son who replaces him, even as he himself achieves fatherhood, seldom attains the stature of his own progenitor. This is indicated by his characteristic shift from sky god variously to solar or weather god (as when the weather god Zeus replaces Kronos) or agricultural deity (as when the Babylonian Marduk, both a solar and a vegetation deity, eventually supplants the Sumerian great triad of sky gods Anu, Enlil, and Ea), all of which lack the majestic connotations universally ascribed to the sky.

Particularly in the Chalcolithic cultures of the Near East (e.g., Sumer, Babylonia), where worship of the Paleolithic goddess developed strongly into the historic period, this shift is apparent. Here earth as mother, rather than sky as father, typically symbolizes the supreme being, rendering fatherhood a less exalted concept. The god is father solely as fecundator, being more often lover than spouse. Such vegetation gods as Adonis, Tammuz, and their myriad counterparts function this way.

In marked contrast to this biological, often chthonic, fatherhood is the refinement of sky-oriented fatherhood apparent in the monotheistic religions—Judaism, Christianity, and Islam—and in dualistic Zoroastrianism, all of which developed out of patriarchal nomadic herding societies that retained more of the archaic religion than did their matrilineal agricultural counterparts. The biblical Yahveh, for example, is thought to have emerged from the celestial West Semitic deity known as Ya, Yami, or Yahu.

One of the attributes frequently credited to father-gods in almost all patriarchal cultures is that of giving birth: The biblical God creates life without aid from a female deity; Zeus produces Athena from his brow and gestates Dionysos in his thigh; the Scandinavian giant Ymir and the Aboriginal Australian Great Father, Kakora, both give birth from their armpits; and the Egyptian Khepri variously spits and masturbates to produce Shu and Tefnut, respectively.

In ritual, too, fathers often mimic the maternal role. Particularly among various Australian Aboriginal groups, initiation rites for boys, frequently reveal the fathers of a tribe functioning as male mothers, as they ritually mimic menstruation and "giving birth" to the young male initiates. Such sexual crossing over introduces into the concept of fatherhood several conflicting themes. Variously, fatherhood is as self-contained as motherhood in its parthenogenetic form; or it projects a "maternal" nurturing quality far different from the remoteness of the archaic sky god; or sometimes it deemphasizes sexual differentiation by blurring, a theme explicit in fertility figures such as Marduk, whose sex changes. These are just some of the ways in which concepts of fatherhood and divine fathers have developed their complexity worldwide.

Siblings

Symbolically, relationships between siblings are almost as central to religion and mythology as those between parents and children. This is partly because brothers and sisters are frequently also spouses, like Zeus and Hera, especially in creation myths, making the theme of incest a common universal mythologem. Almost universally, cross twins (those of opposite sex) are believed to have been the first humans from whom all others descend.

The first couple of the ancient Egyptian ennead were the twins Shu and Tefnut; the second were the brother and sister Geb and Nut, the father and mother of the Osirian gods. The Vedic twins Yama and Yami and the Norse Askr and Embla functioned similarly.

Some idea of the possible meanings of sibling "marriage," whether of twins or not, is evident from the Japanese myth of Amaterasu Ōmikami, the Heaven-Illuminating Goddess, and her brother Susano-o no Mikoto, the Valiant-Swift-Impetuous Hero. The rule of the universe was divided between these two: The realm of light, including heaven and earth, was presided over wisely by the sun goddess, while the ocean and the domain of hidden things was ruled widely by her stormy brother. In consequence of her brother's evil behavior, Amaterasu hid in a cave, plunging the entire world into darkness. When she emerged, light triumphed over dark, and her brother was banished to a remote region.

Most variants of this rivalry show the siblings as two brothers, often twins, as in the ancient Persian Zoroastrian myth of the twins Ōhrmazd ("light") and Ahriman ("darkness") or in the Iroquoian myth of Ioskeha, the creator and preserver, and Tawiskara, the deadly winter god. Unlike the Iroquoian pair, however, the Persian dyad, representing the principles of good and evil, respectively, set the stage for a dualistic system of thought in which both principles are equal.

Often sibling rivalry incorporates the theme of fratricide, as in the case of the Egyptian Seth who kills Osiris or the Greek brothers Ismenos and Kaanthos through whom fratricide was first introduced into the world. This common theme dramatizes the invidious distinction most cultures make between elder brothers and all other children. An Australian creation myth about two brothers traveling together at the beginning of time vividly dramatizes this distinction. When the elder brother desires a wife, he operates on his sibling, making him into a woman. The younger brother-turned-wife simply continues in the subordinate position he had occupied all along, making clear the equivalent impotence of younger brothers and wives.

All these sibling tales ring changes on certain important familial themes. Battling twins represent identity altering into difference; fighting sisters and brothers depict familial opposition; cohabiting sisters and brothers embody familial unity; and battling brothers symbolize the struggle between equality and hierarchy, as brotherhood gives way to the rights of the elder brother, the patriarch-to-be.

Servants

Of all the traditional family members, none so emphasizes the way family functions as both example and symbol of a hierarchical order as does the servant, the hired or enslaved person contributing to family life and economics in both agrarian and commercial settings. Particularly in its most extreme form, as slavery, servitude emphasizes the hierarchical nature of the traditional family. In ancient Hawaiian culture, for instance, one outcast social group, the kauwa, were designated to serve the chiefs and touch them directly. They alone were exempt from the kapu ("taboo") that prohibited touching the chiefs on pain of death. Yet these kauwa were themselves untouchable: It was not proper to eat with them or sleep close to them. At the death of their masters, they were buried alive, often as sacrificial atonement for kapu violations committed by others. While extreme, this example, like others involving Indian untouchable servants, American black slaves, and Middle Eastern eunuchs, clearly embodies the themes of scapegoating, sacrifice, and hierarchy common to families in general.

Certain religious traditions overtly take up the themes implicit in servitude, stressing them as positive rather than negative attributes, as in the cases of Hanuman, the perfect Hindu servant, and Christ, understood as fulfilling the promise of the servant poems of "Second Isaiah." Others simply portray servitude as an institution as natural to divinity as to humanity: In Japanese mythology, for example, the fox functions as the messenger of the god of harvests, Inari, much as Hermes serves the Greek Olympians. Among the Haida of the Northwest Coast, Old-Woman-under-the-Fire serves as messenger of the supernaturals, going between this world and that of the spirits. Servitude, exemplifying a humility appropriate to worshipers, characterizes many traditions; thus the Vedic Hindus feel like slaves in the presence of Varuṇa (Ṛgveda 1.25.1).

See Also

Ancestors; Child; Deus Otiosus; Domestic Observances; Goddess Worship; Hieros Gamos; Home; Sky; Supreme Beings; Twins; Virgin Goddess.

Bibliography

Besides canonical scriptures, the most useful primary texts for students of the family are the ancient religio-legal codes developed by most literate cultures. Representative is the Institutes of Hindu Law, or The Ordinances of Menu, according to the Gloss of Culluca; Comprising the Indian System of Duties, Religious and Civil, translated from the original Sanskrit by William Jones (1794; 2d ed., London, 1876); the ordinances cover a wide range of family-related topics, including divorce, remarriage, status of wives, and the like. An excellent compendium of various issues of concern to contemporary students of family is the Spring 1977 issue of Daedalus, which ranges from articles on specific cultures to family policy issues in the United States to the study of the history of the family.

Of the hundreds of recent works on family studies, Household and Family in Past Time, edited by Peter Laslett with the assistance of Richard Wall (Cambridge, 1972), is most representative of the controversial demographic approach. In this work Laslett presents his provocative, ground-breaking argument that the nuclear family preceded the industrial revolution and hence was causative rather than resultant. Also representative of the new demographic scholarship on family is Michael Mitterauer and Reinhard Seider's The European Family: Patriarchy to Partnership from the Middle Ages to the Present (Chicago, 1977). For those who wish to pursue the historical aspects of family in depth, particularly by looking at small numbers of people in very precisely documented areas, the Journal of Family History (Worcester, Mass., 1796–) presents the most recent work.

Among numerous excellent sources of information on the mother goddess, two stand out for their lucidity: E. O. James's The Cult of the Mother-Goddess: An Archaeological and Documentary Study (New York, 1959) and Marija Gimbutas's The Goddesses and Gods of Old Europe, 6500–3500 b.c.: Myths and Cult Images (Berkeley, Calif., 1982). Somewhat more difficult to trace for lack of early material evidence and fewer books devoted exclusively to the subject is the concept of the father god. Helpful sources include E. O. James's The Worship of the Sky-God: A Comparative Study in Semitic and Indo-European Religion (London, 1963); Wilhelm Schmidt's The Origin and Growth of Religion: Facts and Theories (1931; New York, 1972), considered by many the locus classicus for its topic; and Mircea Eliade's Patterns in Comparative Religion (New York, 1958), in which see especially chapter 2, "The Sky and Sky Gods," and chapter 3, "The Sun and Sun-Worship." A helpful discussion of Kṛṣṇa as divine child appears in David Kinsley's The Sword and the Flute: Kali and Kṛṣṇa, Dark Visions of the Terrible and the Sublime in Hindu Mythology (Berkeley, 1975). Much useful information on siblings appears in Donald J. Ward's The Divine Twins: An Indo-European Myth in Germanic Tradition (Berkeley, Calif., 1968).

New Sources

The Histoire de la famille, edited by André Burguière, Christiane Klapisch-Zuber, Marin Segalen and Françoise Zonabend (Paris, 1986), is the basic work of reference. It contains historical chapters on Mediterranean antiquity, European Middle Ages and ancient Asia as well as anthropological surveys on family in today's Western and Eastern world. The three prefaces by an historian such as Georges Duby and two anthropologists such as Claude Lévi-Strauss and Jack Goody are very inspiring from the methodological point of view. The well-organized bibliographies and the ample indexes are very helpful. Among the classics, N. D. Fustel de Coulanges, La cité antique (Paris, 1872) and Edward Westermarck, A Short History of Marriage (London, 1926) are still worth reading as well as various essays on society and religion by Talcott Parsons. Philip Ariès, L'enfant et la vie familiale sous l'ancien régime (Paris, 1960); J. L. Flandrin, Familles. Parenté, maison, sexualité dans l'ancienne société (Paris, 1976); and Lawrence Stone, The Family, Sex and Marriage in England 1500–1800 (London, 1977) have been innovating historical studies. The Development of the Family and Marriage in Europe (Cambridge, U.K., 1983) and The Oriental, the Ancient and the Primitive. Systems of Marriage and the Family in Pre-industrial Societies of Eurasia (Cambridge, U.K., 1990) by Jack Goody are provoking comparative surveys of pre-modern family institutions throughout the world who challenge many traditional assumptions.

Günther Kehrer, "Familie" in Handbuch religionswissenschaftlicher Grundbegriffe, vol. 2, edited by H. Cancik, B. Gladigow and M. Laubscher (Stuttgart, Germany, 1990), pp. 404–414 is one of the very few accounts of the relationship between family and religion, written from the perspective of social scientific studies. Marzio Barbagli, "Famiglia. 1. Sociologia," in Enciclopedia delle scienze sociali, vol. 3 (Rome, 1993), is particularly useful for his ample bibliography.

For religions (especially religions of the Book) in the contemporary situations of American families, see the site of the "Religion, Culture, and Family Project," directed by Don Browning at the University of Chicago Divinity School (http://divinity.uchicago.edu/family/index.html), which addresses marriage, sex, and family issues from a range of theological, historical, legal, biblical, and cultural perspectives.

Kathryn Allen Rabuzzi (1987)

Revised Bibliography

Family

The family is perhaps the only societal institution that is regarded widely as both natural and essential. The biological basis of kin ties and the reproductive capacities of women historically have conferred that status on the family. This emphasis on biology has led to reductionist and functionalist accounts of the family that transcend cultural barriers. For example the sociologist Talcott Parsons (Parsons and Bates 1955), using a functionalist perspective, argued that the modern family has two main functions: to socialize children into a normative system of societal values and inculcate appropriate status expectations and to provide a stable emotional environment that will protect (male) workers from the psychological damage of the alienating occupational world. These functions are carried out by the wife and mother: She plays the affective, expressive role of nurturer and support, whereas the husband plays the instrumental role of earning the family's keep and maintaining discipline. In similar fashion, the Egyptian Islamist Seyid Qutb spoke of the family as "the nursery of the future," breeding "precious human products" (Choueiri 1990, pp. 127-128). According to Qutb, a man and a woman voluntarily enter into a relationship of marriage as two complementary partners, each discharging functions assigned by nature and biology. A woman fulfills her functions by being a wife and mother, and a man is the undisputed authority, the breadwinner, and the active member in public life (Haddad 1983).

THE ROLE OF SOCIALIZATION

Social scientists differ in their explanations of the biological, cultural, and social underpinnings of the family, but most agree that socialization plays a major role in the perpetuation of the sexual division of labor in the family and society. Socialization patterns ensure that girls and boys will be raised differently, with different expectations of their place in the family and the society. In this way attitudes and practices regarding gender and sex have shaped notions of the family. Similarly, the sexual division of labor in the family has reinforced attitudes toward and practices of gender and sex as well as state policies regarding women's roles.

THE MYTHIC GOLDEN AGE OF THE FAMILY

For some feminists the family is the site of women's oppression and gender inequality, whereas for some psychoanalysts it is the source of personality disorders or conditions such as the Oedipus and Electra complexes. Nevertheless, the family is regarded by many as a haven in a heartless world. Some have argued that in Europe and North America, this concept of the family emerged in the course of struggles against the market and the state.

Conservative commentators warn against the breakdown of the family and family values. In the former Soviet Union during the restructuring known as perestroika in the late 1980s, social problems were blamed on the overemployment of women and their forced detachment from the family under communism. The solution, in that view, was to reduce female labor-force attachment and increase female family attachment. In Eastern Europe and the former Soviet Union, a romanticization of the family, domesticity, and the private sphere, combined with an emphasis on women's maternal role, followed the end of communist rule.

According to the sociologist Rebecca Klatch (1994), for the conservative movement in the United States in the 1980s, the ideal society was one in which individuals were "integrated into a moral community, bound together by faith, by common moral values, and by obeying the dictates of the family and religion" (p. 369). In this ideal community, male and female roles are respected as essential and complementary components of God's plan but men are the spiritual leaders and decision makers in the family. Women's role is to support men through altruism and self-sacrifice. Similarly, a contemporary Muslim view sees the family as the fundamental unit of society and stresses the mother's role in the socialization of children, particularly in raising committed Muslims and transmitting cultural values. According to the Iranian Islamist thinker Murteza Mutahhari (1982), marriage and family life are central to social reproduction and are "a sublime manifestation of the Divine Will and Purpose." He argues that "mutual affection and sincerity, as well as humane compassion and tenderness," are highly desirable attributes in married couples and "are often in evidence in societies governed by Islamic moral and legal checks and balances. In the others, such as those in the Europe and North America, these qualities are seldom noticeable" (pp. 7, 31, 58).

The 1990s saw the formation of a coalition of conservative Muslim, Catholic, and Protestant governmental and nongovernmental organizations in defining and dealing with family values. That coalition first formed around what it saw as objectionable recommendations pertaining to women's sexual rights in connection with the United Nations (UN) International Conference on Population and Development (ICPD), which took place in Cairo in 1994, and the Fourth World Conference on Women (FWCW), which was held in Beijing in 1995. The alliance regrouped in June 2001 at the special session of the United Nations General Assembly on AIDS in New York to stop what it saw as the expansion of sexual and political protections and rights for gays being pushed by the European Union.

Lamentations about the current condition of the family imply that at an earlier period, the family was more stable and harmonious than it is in the early twenty-first century. However despite massive research historians have not located a golden age of the family, as Stephanie Coontz (1992) found in the case of the United States. John Caldwell (1982) notes that many writers have tended to romanticize the peasant family even though Russian peas-ant women and girls worked 1.21 times as many hours as men and boys. Teodor Shanin (1987) points out that despite their heavy burden of labor (both housework and fieldwork) and their functional importance in the Russian peasant household, women were considered second-class members of the family and nearly always were under the authority of men.

The family is a powerful metaphor that is projected onto communities and even nations. In some accounts the ethnic group or the nation is a kind of family writ large. Notions of ideal family forms, typically accompanied by notions of the ideal woman, are found in an array of religious and nationalist writings and discourses across cultures. Here the ideal is the traditional patriarchal family unit, sometimes extended and sometimes nuclear, in which women are responsible for the biological reproduction and hence the continuity of the community (whether an ethnic group, a religious community, or an aspiring nation), and men are in charge of political, military, and economic matters. Women also are cast as the means by which the group's values are transmitted to the next generation and as symbols of the group's identity; this explains why women's dress and comportment are emphasized and often controlled. During times of transition, conflict, or crisis, nationalist ideologies and the state will stress women's reproductive responsibilities, exhort women to have more children, and/or ban contraception and abortion.

The importance attached to biological reproduction and group identity and the strength of attitudes toward male and female roles are the principal reasons for the continued opposition to nonheterosexual affective relationships and family forms in many parts of the contemporary world. This also explains why there is ambivalence and sometimes hostility toward female-headed households, especially when the female heads are unmarried and have children. The family ideology also is behind the continued control of the sexual behavior of women and girls, especially in more patriarchal settings. Many social scientists, including feminist scholars, have attributed gender inequalities and hierarchies in societies to the persistence of traditional family upbringing and the sexual division of labor in the household. Scholars of second-wave feminism wrote about the oppression of women within the family, whether as daughters, sisters, or wives, and called for equality in the family as well as in other societal institutions.

PATRIARCHAL SOCIETY AND THE FAMILY: RISE AND DECLINE

In The Origins of the Family, Private Property and the State, Friedrich Engels (1972) wrote about the "world-historical defeat of the female sex" (Engels 1972, p. 68) in the wake of the agricultural revolution and the advent of civilization and class society. The historian Gerda Lerner (1986) opposed Engels's narrative by arguing that the subordination of women—the creation of patriarchy enforced by legal codes in the ancient Near East—enabled the development of private property and state power there and elsewhere.

The persistence of patriarchy is a matter of debate, and some feminist theorists argue that industrialized societies are also patriarchal. Sylvia Walby (1996) distinguishes between the private patriarchy of the premodern family and social order and the public patriarchy of the state and the labor market in industrial societies. In his work on South Korea, John Lie (1996) distinguished between agrarian patriarchy and patriarchal capitalism. Others have used the term patriarchy more strictly so that patriarchal society is cast as a precapitalist social formation that historically has existed in varying forms in Europe and Asia.

In the patrilocally extended household—commonly associated with the reproduction of the peasantry in agrarian societies—property, residence, and descent proceed through the male line. The senior man has authority over everyone else in the family, including younger men, and women are subject to control and subordination. Childbearing is the central female labor activity. Women's honor, along with, by extension, the honor of the family, depends in great measure on their virginity and good conduct. A study of "the values of Mediterranean society" described the importance of manliness, woman's sexual purity, and defense of family honor in Andalusia, Spain; in villages in Greece and Cyprus; and among the Kabyle in Algeria and the Bedouins of Egypt (Peristiany 1966).

As John Caldwell (2001) and Deniz Kandiyoti (1985) separately described it, the contemporary belt of classic patriarchy includes areas in North Africa, the Muslim Middle East (including Turkey and Iran), and southern and eastern Asia (Pakistan, Afghanistan, northern India, and rural China). In those places one finds practices such as adolescent marriage of girls, high fertility, a preference for sons, and abortion of female fetuses (largely restricted to India and China). In many areas the preoccupation with female virginity leads to honor killings when there has been real or perceived sexual misconduct by a female relative. Such killings of females who are considered to have transgressed the norms of proper sexual behavior are meant to cleanse the violation of the family's integrity and restore its respectability and honor.

STRATEGIES FOR REPRODUCTION AND EXPANSION OF THE FAMILY

Family strategies for reproduction and expansion are varied. One strategy is endogamy, the practice of marrying within the lineage. In many cultures cousin marriage is used to keep property within the lineage and is associated with adolescent marriage, especially for girls. Jack Goody (1990) argues that endogamy mitigates the view of women as property, rejecting the classic Claude Lévi-Strauss's view of women as pawns who embody transaction and exchange, a view adopted by Gayle Rubin (1975). Lévi-Strauss (1969) studied primitive groups, which were exogamous, whereas Arab-Islamic tribes are endogamous. Nonetheless, many scholars continue to view endogamy as being related to patrilineality, which privileges men in terms of property ownership and control over family resources. In such settings women are considered a form of property.

In a study of northern Africa, Germaine Tillion (1983) identified endogamy as setting the stage for the oppression of women in patrilineal society long before the rise of Islam. Endogamy, she argued, kept property (land and animals) within the lineage and protected the economic and political interests of men. Endogamy increases the tendency to maintain property within families by controlling women in tightly interrelated lineages.

Endogamous and exogenous marriages alike entail some form of bride price or bride wealth. Often this transaction is meant to compensate the father, extended family, or tribe for the loss of a girl's labor power. Sometimes it is the price of the girl's deflowering; at other times it signifies the respectability and value of the girl or that of her family. It can take the form of cattle, jewelry, or money. Studies on Afghanistan find that the bride price (walwar in Pashto) signifies an interfamilial exchange.

Nancy Tapper (1984) has described the mobility and migration patterns that revolved around the bride price in Afghanistan in the 1970s. Men from one region would travel to another area to find inexpensive wives, and fathers would travel in search of a higher price for their daughters. In the Muslim communities of the Middle East and North Africa (MENA) the dower (mahr) goes from groom to bride. In India it goes from the bride's family to the groom's family (dowry). In either case it can be an onerous financial burden. More important, dowry and dower are remnants of a patriarchal inegalitarian past in which parents sell their daughters. In all cases inheritance practices favor male kin, and those practices often are codified in law.

Another strategy of family (or kinship) reproduction and expansion is polygyny. The practice of multiple wives for a man is rooted in tribal imperatives and the agrarian economy, which requires the labor power of women and children. In the early twenty-first century polygynous households are most prevalent in sub-Saharan Africa, which remains largely rural and agrarian. Although Muslims claim that their religion permits men to marry up to four wives, urbanization and educational attainment have led to a decline in the practice and an expansion of nuclear families among Muslim communities in urban settings. Whereas family structure in the MENA region once was described as extended, patrilineal, patrilocal, patriarchal, endogamous, and occasionally polygynous, in urbanized countries in that region polygny has become a statistically insignificant family form. Monogamy is the norm, and the recent reform of family law in Morocco has made it extremely difficult for a man to obtain a second wife; however, only Turkey and Tunisia have banned polygyny outright.

CHANGES IN HOUSEHOLDS AND FAMILIES

Some of the most extensive studies on changes in household or family types and the impact of economic changes on women's status have been undertaken in Turkey. In the 1970s Kandiyoti delineated six socioeconomic categories of women: nomadic, traditional rural, changing rural, small town, newly urbanized squatter (gecekondu), and urban middle-class professionals and housewives (Kandiyoti 1985). Family form and household composition varied across those groups, as did the sexual division of labor. An interesting discovery was that the patrilocal extended household was being undermined by market incorporation, migration, and poverty, although patriarchal attitudes and practices remained strongest in the countryside. The patrilineal extended household is similarly characteristic of rural areas elsewhere in MENA but is less typical in large metropolitan areas, where neolocal residence is assumed upon marriage and the nuclear family form prevails.

Nuclear families, however, can continue to be patriarchal, partly as a result of the provisions of family law. The gender hierarchies and inequalities of the law may allow parents to beat their children and husbands to beat their wives. Brothers inherit more than sisters do, and a deceased man's wealth may be inherited by his male kin as well as by his widow or widows and children. Divorce may be obtained easily by a man but is very difficult for a woman, and after divorce, the children remain with their father or his kin. The husband is entitled to exercise his marital authority by restraining his wife's movements and preventing her from showing herself in public. This is the case because Arab-Islamic culture privileges patrilineal bonds and enjoins men to take responsibility for the support of their wives and children. The wife's main obligations are to maintain a home, care for her children, and obey her husband. Valentine M. Moghadam (2003) has referred to this as the "patriarchal gender contract." In the MENA region the patriarchal contract has been codified by the state in the form of Muslim family law or personal status codes, which usually are based on an interpretation by one of the Islamic schools of jurisprudence.

Studies of patriarchal family forms and kinship ties have been conducted by anthropologists, sociologists, and historians, including feminist scholars. The social anthropologist Jack Goody (1990) drew attention to two major kinship systems: African and Eurasian. The sociologist William Goode (1963) argued that the nuclear family form in the Europe and North America would be replicated in other parts of the world because it was best suited to the requirements of industrialization. More recently, Göran Therborn (2004) surveyed family systems across time and place and identified five geocultural patterns, each of which has been shaped largely by its predominant religion: European (including New World and Pacific settlements): East Asia, sub-Saharan African, West Asia/North Africa, and South Asia. Two additional interstitial family forms are religiously hybrid: the southeastern Asian and the Creole American. All traditional family systems, Therborn argues, include three regimes: patriarchy, marriage, and fertility. Echoing the work of other scholars, he argues that whereas at the turn of the twentieth-century patriarchy in the classic sense of male domination was a universal pattern, by the twenty-first century patriarchy had experienced a serious decline.

Many sociologists have attributed this decline to the effects of industrialization, urbanization, and educational attainment, along with the social changes caused by the youth rebellions of the 1960s and 1970s and second-wave feminism. Therborn also stresses the effects of wars and revolutions and especially the ideologies and political systems of communism and socialism, which transformed the family in Russia and China. He agrees with other scholars that the patriarchal holdouts are Muslim and Hindu communities in the Middle East, North Africa, and southern Asia. In Euro-American postpatriarchal societies social movements and equal rights have led to significant changes in family systems, although women remain unequal in terms of political and economic power.

The other two family regimes of marriage and fertility also have undergone change. Divorce, cohabitation, and extramarital birth are very common, and civil unions have been increasing in European and North American societies. In the United States 70 percent of women age twenty to twenty-four were married in 1960, but by 2000, that proportion had dropped to 23 percent. In Tunisia the average age at first marriage has risen dramatically, reaching twenty-eight years for women in 2005. However, Therborn points out that heterosexual marriage, along with monogamy, remains universal, the principal way of regulating sexual behavior and sexual bonding. In regard to fertility Therborn cites the major demographic literature to show how birth rates have been falling throughout the world and across the family systems, with some variations. Such a demographic transition, in which previously high-fertility countries such as Iran have seen major decreases in family size, is the result of the usual sociological and economic explanations of urbanization, educational attainment, and the increasing cost of caring for children, along with modern ideologies of socialism, secularism, and feminism that have promoted family planning or women's control over their own bodies.

THE DEMOGRAPHIC TRANSITION, THE FAMILY, AND SEXUALITY

The demographic transition involves a change from the high-mortality and high-fertility characteristic of preindustrial societies to a pattern of low mortality and low fertility. John Caldwell (1982) argued that in Western Europe, the economic and demographic transitions coevolved: The transition from the traditional peasant (family-based) economy to the capitalist economy entailed changes in decisions about and the need for reproduction. There was less rationality for having large families because the cost of each additional child increased. In England and France the rate of population growth increased up until 1780 and then slowed after 1820 and 1879 in France and England, respectively. Although lower fertility rates came about in European and North American societies over the course of industrialization and urbanization, another important source of instability in the family-based system of production and reproduction, according to Caldwell, was "the egalitarian strain in the modern European ideology, powerfully augmented by the spread of education" (Caldwell 1982, p. 176). Marriage patterns also changed in the course of the demographic transition in Western Europe.

In the 1960s modernization theorists predicted that fertility and household patterns in European and North American societies would be adopted in developing countries that wanted to enhance their social and economic development. The World Fertility Survey (WFS) conducted in forty-one countries between 1977 and 1982 (World Fertility Survey 1984) found that women with higher education living in urban settings tended to have fewer children. The WFS also found that high fertility persisted in a number of regions, notably the Middle East, North Africa, southern Asia, and sub-Saharan Africa. As late as 1988 Algeria reported 5.4 births per woman, and the Islamic Republic of Iran reported 5.6 births per woman. The 1990s, however, saw declines in the region as a whole and a dramatic decline in Iran.

Demographers who have studied global fertility decline since the 1960s offer varied explanations for the fertility transition: mortality reduction; reduced economic contributions from children; the opportunity cost of childbearing, especially for mothers; family transformation; vanishing cultural props for childbearing; improved access to fertility regulation; marriage delay; and the diffusion of certain ideas and practices. Caldwell (2001) cites the role of ideologies, attitudes, and the mechanisms of fertility control but points out that inadequate socioeconomic change may explain why some countries or some social groups within countries have been excluded from the global fertility decline. Karen Oppenheim Mason (2001) adds gender to the equation, arguing that the status of women and the family determines some of the explanations described above. The status of women is thus both an independent variable and a dependent variable in the demographic transition.

Socioeconomic development, state, gender, and class certainly play roles in fertility. The patriarchal family and the agrarian economy both favor high fertility. In protoindustrial societies the merchants (the traditional elite) organize their families much as farmers do and feel few, if any, ill effects from high fertility. By contrast, the fully developed labor market mode of production offers no rewards for high fertility. There is evidence that the labor status of the wife, especially if she works in the modern sector of the economy (the nonagricultural cash economy), is an important determinant of marital fertility. State policies may encourage or penalize large families. Women's lower status means restricted access to education and employment and hence higher fertility. Because women from elite families generally have the highest degree of access to education and employment, fertility also varies by class. Higher levels of education tend to result in more knowledge and use of contraceptives, although the availability of family planning programs is also an important variable. Salaried middle-class women have the fewest children.

These analyses help explain the demographic transition in MENA and its implications for the status of women, gender relations, and the family. As in other developing regions in the twentieth century, the demographic transition occurred more rapidly in MENA than it occurred in Europe, though it occurred later than in Latin America and southeastern Asia. The Demographic and Health Surveys statistics from the early 1990s in Egypt, Jordan, Morocco, Tunisia, Turkey, and Yemen, along with many other developing countries, found that rural versus urban residence, education, and socioeconomic status determined the number of children as well as the health of the mother and child. In general, the surveys found lowering fertility rates and a rising age of marriage in the MENA countries surveyed (survey data are available at the Demographic and Health Surveys web site).

On average, fertility in MENA countries declined from seven children per woman in around 1960 to 3.6 children in 2001. As Farzaneh Roudi (2001) has shown, only Yemen has been relatively unchanged, and the average number of births per woman there is close to eight. In Iran, Lebanon, Tunisia, and Turkey, the combined effects of socioeconomic development, women's educational attainment, and state-sponsored family planning programs have produced the lowest fertility rates in the region. Indeed, the average of about 2.5 children per woman in these countries is even lower than the fertility rate in many Latin-American countries.

Thus, urbanization, industrialization, proletarianization, and mass schooling, which were so important in the demographic transition and the decline of classic patriarchy in the Europe and North America are present in the MENA region and have altered the social structure and gender relations, including the family system. Developmentalist, welfarist, and revolutionary states also have helped bring about societal changes, including legal reforms that bolster women's position in the family. Those reforms are an important basis for the ability of women to act autonomously. However, perhaps most important has been the expansion of schooling for girls. As Fatima Mernissi (1987, p. xxv) has stated: "Access to education seems to have an immediate, tremendous impact on women's perception of themselves, their reproductive and sex roles, and their social mobility expectations." These social changes have led to differentiation among the female population and an expansion of the range of options available to women, including the right to make informed choices about marriage and childbearing. These trends are relevant to a growing proportion of the urban female population, and they have been visible enough to result in opposition by conservative forces. Those forces see the relative rise in the position of women as having the potential to undermine the traditional patriarchal family.

Caldwell (2001) has argued that mass schooling probably has had a greater impact on the family in developing countries than it had in Europe and North America. First, mass schooling came in many countries at an earlier stage of economic and occupational structure development than it did in Europe and North America. Second, schooling frequently means westernization, including European and North American concepts of family and gender. According to Caldwell, "Schools destroy the corporate identity of the family, especially for those members previously most submissive and most wholly contained by the family: children and women" (2001, p. 322). Mernissi similarly emphasized the role of state-sponsored education in creating two generations of independent women. These are the women, Moghadam (2003) notes, who are forming feminist organizations that seek further social changes, including the modernization of family law and the criminalization of honor crimes and domestic violence.

Algeria and Iran, two large MENA countries, are representative of the profound family changes that are occurring in that region. Whereas a few decades ago the majority of women married before age twenty, only 10 percent of that age group in Algeria and 18 percent in Iran were married by the early twenty-first century. This is despite the fact that the age of marriage was lowered to puberty after the Iranian revolution and is age fifteen. The surge in unmarried young people and the fear of illicit sex led some Islamist leaders, such as then-president Hashemi Rafsanjani, to encourage temporary marriage (muta'a in Arabic, sigheh in Persian) and an Islamic contractual arrangement for sexual relations. Temporary marriage is, however, highly unpopular in Iran's middle-class society, which associates it with legalized prostitution.

Elsewhere more young people are remaining unmarried. In Turkey 14 percent, in Morocco 13 percent, and in Tunisia 3 percent of young women age fifteen to nineteen were married in the 1990s. Mernissi has argued that the idea of a young unmarried woman is completely novel in the Muslim world, for the concept of patriarchal honor is built around the idea of virginity, which reduces a woman's role to its sexual dimension: reproduction within an early marriage. The concept of a menstruating and unmarried woman is so alien to the Muslim family system, Mernissi adds, that it is either unimaginable or necessarily is linked with fitna, or moral and social disorder. The unimaginable has become a reality. Young men faced with job insecurity or lacking a diploma to guarantee access to desired jobs postpone marriage. Women, faced with the pragmatic necessity to count on themselves instead of relying on a rich husband, further their formal education. Rates of higher education have increased, and in a number of MENA countries, women's enrollments exceed those of men.

REACTIONS TO RECENT CHANGES IN THE FAMILY

There is a consensus that the dramatic increase in education among North American women in the postwar era was a major cause of the women's movement. The baby boomers went to college in massive and unprecedented numbers. College education increased women's labor-force participation; at the same time there was an expansion of labor participation by married women. A similar pattern can be discerned in MENA countries. Activist women, married and unmarried, emerge from the ranks of the educated and employed. This rapid social change—the impact of industrialization, urbanization, and education on marriage, the family, and gender roles—has caused a conservative backlash and a rise in religious fundamentalist movements as well as lamentations about eroding family values.

BIBLIOGRAPHY

Caldwell, John C. 1982. Theory of Fertility Decline. London: Academic Press.

Caldwell, John C. 2001. "The Globalization of Fertility Behavior." In Global Fertility Transition, ed. Rodolfo A. Bulatao and John B. Casterline. New York: Population Council.

Choueiri, Youssef. 1990. Islamic Fundamentalism. London: Pinter.

Coontz, Stephanie. 1992. The Way We Never Were: American Families and the Nostalgia Trap. New York: Basic Books.

Demographic and Health Surveys. Available from http://www.measuredhs.com/.

Engels, Friedrich. 1972. The Origins of the Family, Private Property and the State. New York: Pathfinder Press. (Orig. pub. 1884.)

Goode, William Josiah. 1963. World Revolution and Family Patterns. New York: Free Press of Glencoe.

Goody, Jack. 1990. The Oriental, the Ancient and the Primitive: Systems of Marriage and the Family in the Pre-Industrial Societies of Eurasia. Cambridge, MA: Cambridge University Press.

Haddad, Yvonne Yazbeck. 1983. "Sayyid Qutb: Ideologue of Islamic Revival." In Voices of a Resurgent Islam, ed. John Esposito. New York: Oxford University Press.

Kandiyoti, Deniz. 1985. Women in Rural Production Systems: Problems and Policies. Paris: UNESCO.

Kandiyoti, Deniz. 1988. "Bargaining with Patriarchy." Gender & Society 2(3): 274-290.

Klatch, Rebecca. 1994. "Women of the New Right in the United States: Family, Feminism and Politics. In Identity Politics and Women: Cultural Reassertions and Feminisms in International Perspective, ed. Valentine M. Moghadam. Boulder, CO: Westview

Lerner, Gerda. 1986. The Creation of Patriarchy. New York: Oxford University Press.

Lévi-Strauss, Claude. 1969. Elementary Structures of Kinship, trans. James Harle Bell, John Richard von Sturmer, and Rodney Needham. Boston: Beacon Press.

Lie, John. 1996. "From Agrarian Patriarchy to Patriarchal Capitalism: Gendered Capitalist Industrialization in Korea." In Patriarchy and Economic Development: Women's Positions at the End of the Twentieth Century, ed. Valentine M. Moghadam. New York: Clarendon Press.

Mason, Karen Oppenheim. 2001. "Gender and Family Systems in the Fertility Transition." In Global Fertility Transition, ed. Rodolfo A. Bulatao and John B. Casterline. New York: Population Council.

Mernissi, Fatima. 1987. Beyond the Veil: Male-Female Dynamics in Modern Muslim Society. Bloomington: Indiana University Press.

Moghadam, Valentine, M., ed. 1994. Identity Politics: Cultural Reassertions and Feminisms in International Perspective. Boulder, CO: Westview Press.

Moghadam, Valentine M., ed. 1996. Patriarchy and Development: Women's Positions at the End of the Twentieth Century. Oxford: Clarendon Press.

Moghadam, Valentine M. 2003. Modernizing Women: Gender and Social Change in the Middle East. 2nd edition. Boulder, CO: L. Rienner.

Mutahhari, Murteza. 1982. Sexual Ethics in Islam and in the Western World, trans. Muhammad Khurshid Ali. Tehran, Israel: Bonyad Be'that Foreign Department.

Parsons, Talcott, and Robert F. Bales. 1955. Family, Socialization and Interaction Process. Glencoe, IL: Free Press.

Peristiany, J. G., ed. 1966. Honour and Shame: The Values of Mediterranean Society. Chicago: University of Chicago Press.

Pitt-Rivers, Julian. 1977. The Fate of Shechem: Or, The Politics of Sex: Essays in the Anthropology of the Mediterranean. Cambridge, UK: Cambridge University Press.

Roudi, Farzaneh. 2001. "Population Trends and Challenges in the Middle East and North Africa." Washington DC: Population Reference Bureau Policy Brief.

Rubin, Gayle. 1975. "The Traffic in Women: Notes on a Political Economy of Sex." In Toward an Anthropology of Women, ed. Rayna R. Reiter. New York: Monthly Review Press.

Shanin, Teodor. 1987. "A Peasant Household: Russia at the Turn of the Century." In Peasants and Peasant Societies: Selected Readings. 2nd edition, ed. Teodor Shanin. London: Blackwell.

Tapper, Nancy. 1984. "Causes and Consequences of the Abolition of Brideprice in Afghanistan." In Reveolutions and Rebellions in Afghanistan, ed. Nazif Shahrani, and Robert Canfield. Berkeley: University of California Berkeley International Studies Institute.

Therborn, Göran. 2004. Between Sex and Power: Family in the World 1900–2000. London: Routledge.

Tillion, Germaine. 1983. The Republic of Cousins: Women's Oppression in Mediterranean Society, trans. Quintin Hoare. London: Al Saqi Books.

Walby, Sylvia. 1996. "The 'Declining Significance' or the 'Changing Forms' of Patriarchy?" In Patriarchy and Development: Women's Positions at the End of the Twentieth Century, ed. Valentine M. Moghadam. Oxford: Clarendon Press.

World Fertility Survey. 1984. World Fertility Survey: Major Findings and Implications. Voorburg, Netherlands: International Statistical Institute.

                                     Valentine M. Moghadam

FAMILY

HISTORY    Andrejs Plakans

FUTURE    Frank F. Furstenberg

HISTORY

Historical research on individual families in the Western world was carried out by genealogists long before the field was systematized in the 1960s and 1970s. In those two decades, however, as a result of the confluence of new initiatives and organizational developments in several disciplines, family history established itself as a specialized endeavor with strong links to cognate disciplines such as demography, sociology, and anthropology. The "new social history," proceeding from the influence of the French Annales school, pursued the history of "structures of long duration," including micro-structures such as families, households, and kin groups.

The pioneering essays in Population in History (Glass and Eversley 1965) demonstrated the potential of the historical study of populations, and John Hajnal's essay on European marriage patterns in that volume put forth a seminal hypothesis about an important aspect of the long-term evolution of European family life. The formation of the Cambridge Group for the History of Population and Social Structure in 1967 resulted in an inclusive program for both historical demography and family/household structural analysis on the basis of historical sources such as parish registers and household listings. Independently of these initiatives, Philippe Ariès's Centuries of Childhood (1962) pointed to important historical shifts in the manner in which children were socialized.

By the early 1970s these initiatives had coalesced into a research agenda that dealt with the family, broadly defined, in both the distant and the recent past and used a wide variety of analytical concepts and approaches. Over the next 30 years the agenda widened to include questions not dealt with by the "founding generation."

Family Households and Historical Demography

In the decades between 1970 and 2000 historical demography and family history developed simultaneously, with findings on historical patterns of mortality, fertility, marriage, and migration creating in any historical period the context in which the family as a social structure had to be understood and with the historical sources used in family history (house-hold listings, parish registers) providing much of the raw data for historical-demographic generalizations. It was clear that decisions about marriage, childbearing, and geographical movement always involved the family in some fashion; death, by contrast, was not a result of personal decisions but did have wide ramifications for family structure and life.

Changes in marital, childbearing, and migration behaviors were shown to be interrelated with changes in family size and structure over time. In pre-modern, pre-contraceptive societies, in which births outside wedlock brought stigma for both the mother and the child, most births took place after marriage. Consequently, the age at first marriage was identified as an important variable influencing the ultimate size of the nuclear family group. Normally, women bore children every two years; hence, marriage in the late teens, in contrast to the late twenties, could result in a size difference in offspring groups of four or five, assuming the same levels of infant and child mortality.

Hajnal's (1965) hypothesis of "western" and "eastern" marriage patterns on the European continent proposed that in the western half of the continent marriages on average took place when both partners were in their late twenties, whereas in the eastern half women tended to marry in the late teens. The larger number of ever-born children per couple in the east, however, was offset by higher levels of infant and child mortality. In the east both long-and short-distance movement was restricted by institutions such as serfdom so that localities retained more of the human material out of which cultural choices often fashioned families that were more complicated than those in the west. The survival into later life of the initially somewhat larger sibling groups and the reduced pace of dispersion of those groups meant that there were more related adults in a given locality who could live together if they chose to do so. Mean family household size in Western Europe averaged four to five persons, whereas in the east it was closer to eight to nine. In the west the proportion of family households with more than two generations and with complex structures was low; in the east, by contrast that proportion tended to be high.

These pre-modern interrelationships between demographic patterns and family structures were dissolved during the "demographic transition," when mortality, and later fertility, started to decline. The timing of the onset of the transition and its duration were the subject of the Princeton Fertility Project that was initiated in the 1970s. Although the project failed to show convincingly the precise relationship between demographic, cultural, and socioeconomic change, it did demonstrate that the timing of the beginning of the decline varied substantially across the continent. The transition started first in Western Europe, notably in France, in the early nineteenth century. Central Europe experienced it in the middle decades of the century, and it did not take hold in the eastern part of the continent until the late nineteenth and early twentieth centuries. The project also noted the presence of "pioneering" localities where fertility decline began earlier than in the surrounding areas in all of Europe's regions.

At the family level, the transition meant that couples could count on more of their children surviving past the childhood years, increasing pressure on family resources. One response to this pressure was to begin to limit childbirth, initially by spacing births and then with the aid of artificial means of birth control. This had the effect of delinking age at first marriage from the start of childbearing: the end of a historical pattern. The increased probability of survival of each individual child, entailing educational expenses first for boys and later for girls as well, meant that the ideal number of children per couple fell steadily.

Coinciding with the demographic transition were political reforms that lifted restrictions on movement, especially in Eastern Europe. As the industrial and service sectors grew and as urban areas became more capable of absorbing in-migrants, the dispersion of sibling groups became a more common occurrence, diminishing the scope for complex families. Large multi-generational complex family groups remained important only in those areas, such as Serbia, where cultural imperatives for their creation were particularly strong. These shifts were not as meaningful in Western countries where even in earlier times family complexity had for the most part manifested itself in elderly parents coresiding with married children rather than as the coresidence of married siblings.

Relationships between demographic patterns and family structures still existed in the twentieth century but the connection was weakening. By the end of the century, in the domains of marriage, family formation, and childbearing, cultural imperatives and personal choice seemed to have moved into positions of dominance in all the regions of Europe. Marriage was no longer a prerequisite to sexual gratification, as the stigma of premarital sex had nearly vanished and cohabitation of unmarried persons had become widely accepted. The timing of marriage became increasingly disconnected from the question of offspring, which was determined more by the economic and professional readiness of couples to "begin having children." Divorce had lost it stigma as well. Net reproduction fell below one (the replacement level) first in Western and then in Eastern European countries. Low fertility, together with increased life expectancy, meant that most European populations were aging, with steadily increasing proportions of elderly persons and falling proportions of children.

Historical Changes in Family Structure

Although it was possible to trace from historical demographic trends the different consequences for different aspects of family life, changes in the structure of family groups as groups required different kinds of evidence. The systematic study of family household structure over time began in the late 1960s with the work of the Cambridge Group, particularly the investigations by Peter Laslett of the "listing of inhabitants." Thousands of those listings were uncovered throughout the Continent and were mined for their content. In some localities, nominal listings were randomly spaced over time; in others, they were made at systematic intervals.

In these listings it was possible to research family structure at moments in past time or in a series of moments. Occasionally communities recorded information about family groupings continuously, permitting the tracking of changes in family groups over long stretches of generations. Some listings carefully recorded the relationship between each member and the group head, whereas others left it unspecified. Some lists distinguished between groups, whereas in others such boundaries had to be interpreted. Successful record linkage involving household listings and parish registers sometimes could be used to enrich both kinds of evidence.

Research on these sources demonstrated the flaws in various earlier claims about structural changes, expanded geographical coverage of family research, and showed that the boundaries of the family in the past were much more porous than had been thought. Geographically, Europe and North America rapidly became the best-researched parts of the world, but historical sources revealed significant historical information about China and Japan as well. Within Europe the western countries and Scandinavia became the most thoroughly researched regions, with southeastern Europe, Italy, and Iberia in second place and Eastern Europe not yet explored fully. Several typologies of European patterns, an east-west division, and a four-part division, served as useful guides to research but were questioned with regard to oversimplification. Chronologically, research was most thorough for the centuries between the seventeenth and nineteenth, with the classical world remaining to be described fully. For the twentieth century, questions of family dynamics largely supplanted questions of structural change.

Broadly speaking, research on Europe since the early 1970s has shown that although nineteenth-century sociological theories of familial evolution describing a trajectory from simple to complex structures held in some regions, they misrepresented the history of other regions. The simply structured two-generational family (father, mother, children) everywhere and always accounted for a significant proportion of all family groupings, and in the European west this structure was predominant ever since the availability of historical records. Elsewhere in Europe the story was more complicated. In the European east (including Russia), under conditions of serfdom that limited movement, and in some regions of the Balkans where local traditions celebrated joint ownership of land and property, complex family structures (coresidence of married siblings and of parental couples with married children) historically represented a significant proportion of any community's total family households. Similar statistical importance of complex groups was found in localities of such widely dispersed countries as Finland, France, and Italy, where these patterns usually were associated with labor needs and inheritance patterns. Wealthier family households were generally more complex than poorer ones, though there are many exceptions to this generalization. The growing size of the "middle class" during the nineteenth century introduced value-based preferences that favored the small nuclear unit.

In most localities a family household almost always contained various nonrelated members: farm-hands, apprentices, lodgers, and paupers. Family structure varied with the age of the household head, exhibiting the "family developmental cycle" (from simple to complex to simple as the head aged). Over time until the twentieth century, complicated groups tended to change the nature of their complexity from horizontal to vertical: married siblings earlier in the developmental cycle and aging parents coresident with a married child later. A unilinear evolutionary pattern (from wholly complex to wholly simple) at the local level, however, could not be found anywhere, as family groups responded to crises by expanding their ranks, becoming simple again when times turned less threatening. The association of particular familial structures with particular ethnic or nationality groups was shown to be irrelevant, and so claims about the "typical French family" or the "typical German family" fell by the wayside. From the functional viewpoint, familial units everywhere reacted to plenty and to adversity generally in the same fashion.

During the twentieth century, however, there was a convergence of structural patterns throughout Europe, though this question has yet to receive a conclusive answer. Also, since the mid-twentieth century new forms of cohabitation have emerged that have required redefinition of what a "family" and a "household" are. The definitions that served to keep research comparable for the pre-twentieth century period have been revealed to be increasingly time-specific.

The early programs of the Cambridge Group to systematize historical family research and make it comparative eventually were criticized as researchers turned increasingly to the study of family dynamics in the past. It was suggested that the constant changes in family life made talk of "structures" irrelevant because any family could experience numerous structural changes in the course of its existence. Some researchers shifted the focus from the family group to individuals within the family group, underlining the interconnectedness between the group's evolution and the individual lives of the persons within the group. Indeed, in the course of time research on the history of the family tended to leave the question of "structure" behind, preferring to look instead at the intergenerational transmission of property (inheritance) within the family group, the distribution of power (patriarchal authority) within it, the effects on it of state policy (public welfare institutions), the experience of crises within the group (widowhood and widowerhood), and the play of emotions within it (parent–child relationships). Some researchers have begun to investigate other social groups of a quasi-familial nature, such as guilds and brotherhoods. The study the of history of the family presents a clear example of a field developing new research directions before early questions were fully answered.

Household, Family, and Kin Group

Successful studies of kinship within the family core-sidential group in the European past showed that when kin beyond the head's immediate family were present, they tended to be kin of a certain kind, for the most part patrilineal. Thus, in the European context there was a higher probability that the coresident parental couple would be the husband's parents, that the coresident siblings of the head would be the husband's brothers, and that the coresident married offspring would be the sons rather than the daughters. These configurations were all predictable from knowledge of the way patrilineal societies worked at any time and in any place. What was always of interest in the European past was who was excluded from the domestic group and on what basis. From the very beginning this question hovered just offstage: If kinship within the domestic group was important, what was the significance of kin ties that crossed household boundaries?

The strongest argument for looking beyond the family household for important family connections lay in already documented behaviors. Aristocracies of various kinds as well as wealthy urban patriciates had always had a keen interest in their lineages, and a similar preoccupation existed in some peasantries, in France for example. In some areas of Europe such as Albania, clan-like organizations were said to have continued to exist well into the twentieth century. Unfortunately, the functioning of larger kin groups is poorly documented in the European past. Whether the influence of the larger kin group on families was peripheral and weak or strong but subtle was an empirical question and could not be answered with any certainty unless such larger kin ties were mapped and the dynamics of domestic units within them were explored.

Several hypotheses about these matters emerged in the course of research. Laslett (1988) contended that people who lived in nuclear families encountered difficulties when faced with crises such as widowhood, unemployment, sickness, and senility. Accordingly, they sought support from their kin or, in the absence of kin, from friends, neighbors, or institutions in the community at large. Yet in the European historical record kin groups were not simply exemplars of the "ethic of amity" (Fortes 1969); they involved antagonisms and divisions as well, especially when disputes involved property, position, or other forms of wealth. Moreover, large kin-linked formations could experience internal shifts. In his explorations of kinship formations in the German village of Neckarhausen in the period from 1600 to 1900, David Sabean (1990, 1998) showed that large kin configurations could undergo transformations of emphasis even while retaining general characteristics such as "patrilineality." Marriage choices within large configurations changed, as did the persons whose job it was to cultivate and maintain kin relations. Sabean's larger point is that the characterization of a kinship system as bilateral or patrilineal was only the first step because historians confronted historical kinship not as a full-blown "system" but as a collection of concrete acts and transactions, each of which had to be understood and interpreted. The temporal changes that are worth knowing about could occur without changing the general "tilt" of the entire system. However, in order to understand the meaning of the change-producing decisions, those decisions have to be laid against a reconstructed network showing how the makers of everyday decisions, within the domestic group context and outside it, were related to each other. A full understanding of the growing autonomy of the domestic group and its growing tendency to make collective decisions without reference to any outside persons required knowledge of the larger group from which autonomy was sought. If the modern state gradually assumed many of the functions that supportive kin networks may have served in earlier times, this transfer of obligations did not cancel, even as it may have weakened, kin ties, and such ties continued to stand ready to be reactivated when the state failed in its duties. Important questions of this kind remain largely unanswered for most of the European continent.

See also: Family Reconstitution; Henry, Louis; Historical Demography; Household Composition; Laslett, Peter.

bibliography

Anderson, Michael. 1980. Approaches to the History of the Western Family 1500–1914. London: Macmillan.

Ariès, Philippe. 1962. Centuries of Childhood: A Social History of Family Life, trans. Robert Baldick. New York: Knopf.

Coale, Ansley J., and Susan Cotts Watkins, eds. 1986. The Decline of Fertility in Europe. Princeton, NJ: Princeton University Press.

Fortes, Mayer. 1969. Kinship and the Social Order: The Legacy of Lewis Henry Morgan. Chicago: Aldine.

Glass, David, and D. E. C. Eversley, eds. 1965. Population in History. Chicago: Aldine.

Hajnal, John. 1965. "European Marriage Patterns in Historical Perspective." In Population in History, ed. David Glass and D. E. C. Eversley. Chicago: Aldine.

Hareven, Tamara K. 1996. Aging and Generational Relations over the Life Course. Berlin: De Gruyter.

Kaser, Karl. 1995. Familie und Verwandschaft auf dem Balkan: Analyse einer Untergehenden Kultur. Vienna: Suhrkampf.

Kertzer, David. 1991. "Household History and Sociological Theory." Annual Review of Sociology 17: 155–179.

Laslett, Peter. 1966. "The Study of Social Structure from Listings of Inhabitants." In An Introduction to English Historical Demography. New York: Basic Books.

——. 1983. "Family and Household as Work Group and Kin Group: Areas of Traditional Europe Compared." In Family Forms in Historic Europe, ed. Richard Wall, J. Robin, and Peter Laslett. Cambridge, Eng.: Cambridge University Press.

——. 1988. "Family, Kinship, and Collectivity as Systems of Support in Pre-Industrial Europe: A Consideration of the 'Nuclear-Hardship' Hypothesis." Continuity and Change 3: 153–175.

Laslett, Peter, and Richard Wall, eds. 1972. Household and Family in Past Time. Cambridge, Eng.: Cambridge University Press.

Plakans, Andrejs. 1984. Kinship in the Past: An Anthropology of European Family Life 1500–1900. Oxford: Blackwell.

Ruggles, Steven. 1990. "Family Demography and Family History: Problems and Prospects." Historical Methods 23: 22–33.

Sabean, David. 1990. Property, Production, and Family in Neckarhausen, 1700–1870. Cambridge, Eng.: Cambridge University Press.

——. 1998. Kinship in Neckarhausen 1700–1870. Cambridge, Eng.: Cambridge University Press.

Segalen, Martine. 1985. Fifteen Generations of Bretons: Kinship and Society in Lower Brittany 1720–1980. Cambridge, Eng.: Cambridge University Press.

Smith, Daniel Scott. 1993. "The Curious History of Theorizing about the History of the Western Nuclear Family." Social Science History 17: 325–353.

Wall, Richard, J. Robin, and Peter Laslett, eds. 1983. Family Forms in Historic Europe. Cambridge, Eng.: Cambridge University Press.

Wetherell, Charles. 1998. "Historical Social Network Analysis." International Review of Social History 43, Supplement: 125–144.

Wrigley, E. A., ed. 1966. An Introduction to English Historical Demography. New York: Basic Books.

——. 1969. Population and History. New York: McGraw-Hill.

Andrejs Plakans

FUTURE

Changes in the institution of the family probably occur more rapidly in modern and modernizing societies than in the past, but historians and anthropologists have long been aware that shifts in kinship and marriage practices take place even in traditional societies, albeit at a slower pace. What is also true in the early twenty-first century, far more than in the past, is that through social science research, government reports, and stories in the mass media, people are acutely conscious of the changes that are taking place in family norms and behaviors. It is known that the family is changing, but it is nonetheless difficult to project the course of that change beyond a decade or two. Charting the future of the family, then, is an exercise in imagination or science fiction.

Having conceded an inability to read the future, there are certain straws in the wind that can provide some clues of what might be in store for the Western family. Also, it may be worth revisiting ideas about convergence in family forms that were popular in the middle decades of the twentieth century.

Future of Marriage in the West

The widespread practice of cohabitation, the rising age of marriage, and high levels of marital instability have lead some observers to question the viability of the institution of marriage. Certainly, the practice of lifelong monogamy, which became the cornerstone of the Western family with the spread of Christianity, has given way to more varied arrangements: consensual unions not sanctioned by state or church; single-parenthood; homosexual unions; and conjugal succession or "serial marriage." Such arrangements have always existed in many societies, but without the legitimacy that they are accorded today.

The conditions that have given rise to greater variation in family forms in which childbearing or, at least, childrearing occurs can be traced to many different factors. The decline of church and state authority to shape public morality is one important source of family change. The breakdown of strict gender roles that once created a high degree of interdependence between spouses is another powerful impetus for revising matrimonial arrangements. The spread of education and of the ideology of choice is a third reason for increasing variability in family forms and in the roles of family members.

It appears unlikely that any of these conditions that have undercut the hegemony of the nuclear family are going to recede. Yet it is entirely possible that customs and fashions, economic forces, or the growth of state authority may influence the distribution of family types within and between Western nations. In the United States, for example, politicians are mounting strenuous efforts to promote marriage. Whether public policies or official rhetoric are likely to have any effect on marriage practices is at best dubious.

According to David Ellwood and Christopher Jencks (2001), variation in family forms in the United States is much more conspicuous as one goes down the socioeconomic ladder. Over the past several decades, family behavior among the privileged has changed little, while in other social strata rates of marital instability and single-parenthood have increased, especially among the poor. This observation suggests that the flux in marriage may be partly produced by economic strains or, perhaps, by gender discord resulting from changing expectations of men and women. The exploration of class differences in family forms is an intriguing area for further investigation.

Future of Fertility in the West

If the fate of marriage is unpredictable, low fertility within marriage appears to have a more secure future. Technological developments in fertility control have increased the ability of couples to manage fertility effectively. Given the high cost of children and low levels of mortality during childhood, large family size is becoming a relic of the past. It is difficult to imagine conditions that will produce, once again, a demand for large families.

The challenge of raising fertility to replacement levels has become an urgent issue of public policy in many Western nations. More than any other, this policy problem is likely to have important effects on the family. The difficulty of combining work and family roles and the high costs involved in rearing children are leading many parents to severely restrict childbearing. It seems likely that societies will experiment with arrangements that alleviate the private costs of rearing children and with building institutions that enable parents to combine work and family roles more easily. Innovations in these areas are already evident, but there is likely to be a good deal more institutional invention as technology allows parents of children to work in the home, or as day-care arrangements permit parental monitoring of children's safety and comfort.

Techniques of Reproduction

Nowhere has reality come closer to science fiction, if not actually surpassing it, than in the area of reproductive technology. Fertility has become ever-more controllable through new medical and biological procedures. The capacity of parents to predetermine at least some of their children's physical characteristics is just around the corner; however, it is not at all clear how different societies will handle the potential benefits and abuses of new reproductive technologies. It seems likely that legal prescriptions will be developed to impose rules on the use of new reproductive technologies, and equally likely that such regulations will create a black market in the use of proscribed practices.

Future of Kinship in the West

High rates of divorce and remarriage reshaped kinship arrangements in last half of the twentieth century. Rising levels of cohabitation and nonmarital childbearing have added complexity to the family as broadly defined. The links across households produced by nonresidential parents and their partners, not to mention their siblings, parents, and children, have created wider but shallower family bonds. Moreover, gay couples and their families have established new kinship arrangements not formed by blood or marriage–the traditional ways of constructing a family. Western societies allow greater latitude in defining family but, in doing so, may be attaching lower levels of obligation to kinship.

Kinship has always been socially constructed, even if members of a society come to think of these ties as "natural." Parent and child relationships based on biological and genetic ties tended to be seen as fundamental, while in-law relations created by legal arrangements were accepted as socially binding. By the beginning of the twenty-first century, both of these axes of kinship had become more questionable in law and practice. The father who sires a child but never lives with him or her has less relevance than the sociological parent (man or woman) who is the mother's partner and helps in raising the child. Moreover, this partner's family becomes part of the child's family, with views of rights and obligations to the child that may be highly variable. Relatively little research exists on new family forms, especially on the way that these forms affect family members' relations with each other in every day practice.

The charting of kinship bonds over time within nontraditional families is an attractive way of understanding how such bonds are created and maintained, and illustrates the strength of relationships in families that are not established by blood or marriage. Students of the family should examine the transfer of property, the keeping of family albums, the frequency of family reunions, and many other everyday aspects of behavior as ways of establishing the meaning of kinship in alternative and traditional families. As yet, virtually no literature exists on this topic.

Convergence of Western and Non-Western Families

In the 1960s some social scientists argued that family systems were converging across the world, gradually moving toward a Western model. The sociologist William J. Goode argued that the fit between the nuclear family and the needs of a modern economy would ultimately force different kinship systems to take on the Western form. Although there is abundant evidence of change in kinship systems worldwide, the evidence suggesting convergence to a Western model of the family is equivocal at best.

It may be still too soon to detect the movement away from complex to simpler forms of the family. However, the thesis put forth by family sociologists in the mid-twentieth-century seems naive in light of what has occurred in the West. In the first place, the assumption of a uniquely appropriate fit seems doubtful in view of the vast changes that have taken place in the Western form of the family and the continuing stresses that are evident between work and family. Moreover, it is clear that traditional forms of the family persist even as economic change takes place.

Will plural marriage–polygamy–where it still exists survive economic development and the spread of Western corporate institutions? Can multi-generational households co-exist with modern economic markets that promote the interests of individuals over aggregates? It seems likely that some accommodations will occur as economic development advances and the market economy spreads to non-Western nations. Clearly, fertility has declined and may continue to drop, forcing changes in household structure and living arrangements. It remains to be seen, however, whether the kinship arrangements that result will have a Western look.

Variations in family forms worldwide have been more resilient than many observers predicted. Goode's thesis that the Western, nuclear family would be imported to nations of varied kinship arrangements has not yet come true, even though changes in marriage and divorce practices and fertility are evident in many developing nations. The hegemony of the nuclear family is less evident and variety is more apparent. In this respect, kinship is proving to be a more durable feature of culture than was thought by those who predicted the demise of the family or the convergence of family forms to a single model.

See also: Childlessness; Cohabitation; Divorce; Fertility, Below-Replacement; Fertility, Nonmarital; Marriage; Parenthood; Partner Choice.

bibliography

Ellwood, David T., and Christopher Jencks. 2001. "The Growing Differences in Family Structure: What Do We Know? Where Do We Look for Answers?" Paper prepared for the New Inequality Program. New York: Russell Sage Foundation.

Goode, William J. 1963. World Revolution and Family Patterns. New York: Free Press of Glencoe.

Johnson, Coleen. 1988. Ex Familia: Grandparents, Parents, and Children Adjust to Divorce. New Brunswick, NJ: Rutgers University Press.

Popenoe, David. 1996. Life without Father. New York: Free Press.

Waite, Linda, and Maggie Gallagher. 2000. The Case for Marriage: Why Married People Are Happier, Healthier, and Better off Financially. New York: Doubleday.

Weston, Kath. 1991. Families We Choose: Lesbians, Gays, Kinship. New York: Columbia University Press.

Wilson, James Q. 2002. The Marriage Problem. New York: Harper Collins.

Frank F. Furstenberg