KINSHIP SYSTEMS AND FAMILY TYPES

Kinship systems are mechanisms that link conjugal families (and individuals not living in families) in ways that affect the integration of the general social structure and enhance the ability of the society to reproduce itself in an orderly fashion. Kinship performs these social functions in two ways. First, through relationships defined by blood ties and marriage, kinship systems make possible ready-made contemporaneous networks of social ties sustained during the lifetimes of related persons and, second, they enable the temporal continuity of identifiable family connections over generations, despite the limited lifespan of a family's members. Variations in norms governing the structure of contemporaneous networks and the modes of temporal continuity compose the basis for the typologies of kinship systems described in this article.

In conceptualizing connections between kinship systems and family types, social scientists have applied either of two approaches. Some have developed typologies from historical analyses (and evolutionary schemes) that depict the transition of Western societies from ancient or medieval origins to modern civilizations. Other social scientists construct typologies that cut across diverse historical periods. Each historical era then constitutes a unique medium in which the structural typologies are expressed.

MODERNITY, FAMILY PATTERNS, AND KINSHIP SYSTEMS

There are at least three ways to develop historical typologies related to kinship and family. One way is to hypothesize a linear historical progression, which includes a family type existing at the beginning point in time, a particular historical process that will act upon the family and kinship structures (e.g., urbanization or industrialization), and a logical outcome at the end of the process. A second approach builds upon the above approach by positing a transitional family type that emerges during the historical process and gives way in the final stages of the process to another family type. A third approach, which includes devising a family type based upon a configuration of attributes peculiar to a particular historical era (e.g., the Victorian family, the American colonial family), implies that any historical era represents a unique convergence of diverse factors.

Bipolar Typologies. By and large, sociologists have drawn a connection between kinship and family on the basis of a distinction between traditionalism and modernity. Generally, this distinction draws upon Henry Maine's ([1861] 1963) depiction of the transformation of social relations in early societies. Maine argued that social relations changed from those based on ascriptive status (deriving from birth) to relations created and sustained through voluntary contractual arrangements. Maine's theory has evoked a series of typologies that, in large measure, refine the status–contract distinction. For instance, an ideal type developed by Ferdinand Toennies ([1887] 1957) has provided a backdrop for later typologies. The Toennies typology itself refers to a shift from Gemeinschaft (community) as a form of social organization based upon an existential will (Wessenwille), which is suited to feudalism and peasant society, to Gesellschaft (society) as a social form based upon rational will (Kurwille), which fits an urban environment under modem capitalism. Contemporary family typologies, in building upon Toennies's conceptual scheme, portray a weakening of kinship obligations and constraints.

One position, rooted in George P. Murdock's (1949) analysis of cross-cultural archives, has resulted in the main sequence theory of social change in kinship structure (Naroll 1970). Main sequence theory pertains to the way differential gender contributions to production of material resources affects the use of kindred as human resources/property. This theory holds that basic changes in kinship are initiated by a shift in the relative importance of men and women to the economic life of the society. First, there is a modification in the economic division of labor by gender. (For example, in hoe cultures, women tend to do the farming; when plows are introduced, men become the farmers.) Second, the shift in sexual division of labor generates a change in married couples' choices of residence, the major alternatives being near the husband's relatives (patrilocal), the wife's (matrilocal), or anywhere the couple desires (neolocal). (Plow cultures tend toward patrilocal residence.) Third, the change in choice of residential site affects the line of descent and inheritance favored in the kinship system: the husband's side (patrilineal), the wife's (matrilineal), or both sides (bilateral). (In line with the shift in residence, plow cultures show a greater inclination toward patrilinearity than do hoe cultures.) Fourth, the transfer to lineage affiliation generates a change in kinship terminology, particularly in ways that show tribal or clan membership, or, in modern societies, the dissolution of larger kinship structures. As applied to the emergence of modernity, main sequence theory predicts a continual emancipation from kinship constraints. An increase in the proportion of women in the labor force will produce a trend toward neolocal residence, which in turn will lead to increased emphasis upon bilaterality, weakening sibling ties and obligations to both sides of the extended family, and in the long run to changes in kin terminology and identity [e.g., voluntarism in choice of surnames as an indicator of preference as to line(s) of descent].

In a variation of main sequence theory, urban sociologists such as Wirth (1956) and Burgess and associates (1963) wrote on the effects of transferring the economic base of societies from the land to urban centers. The theme of their work is to be found in the German proverb "Stadt Luft macht frei" ("city air makes one free"). For example, Burgess and associates described a progression from what they named the institutional family to the companionship family. In this conceptualization, the institutional family, embedded in a larger kinship group, is characterized by patriarchy, clearly defined division of household labor by sex, and high fertility. Its unity is derived mainly from external constraints—social mores, religious authority, fixity in location, position in the social structure, and the value of familism (i.e., values giving priority to the collective welfare of the family over that of individual members). Burgess and associates regarded the institutional family as an adaptation to relatively immobile, rural, agricultural societies and believed its way of life was fixed over time. By way of contrast, urban society, which is characterized by mobility, anonymity, and change, makes inoperative the social control mechanisms developed to maintain stable, rural societies. With the withering of these external controls on rural family life, Burgess, Locke, and Thomes proposed that the companionship family is bound together by internal forces—mutual affection, egalitarianism, a sense of belonging, common interests—and affords freedom from the demands of traditional family and kinship ties.

Unlike the urban sociologists, structural functionalists such as Talcott Parsons (1954) place considerable emphasis on the interaction of subsystems in the larger social system. In part, structural functionalists are concerned with economic and kinship factors in structuring nuclear family relationships. Parsons described American kinship as "a 'conjugal' system in that it is made up exclusively of interlocking conjugal families" (1954, p. 180) and is multilineal (i.e., bilateral) in descent. Parsons associates kinship solidarity with unilineal descent, that is, with a "structural bias in favor of solidarity with the ascendant and descendant families in any one line of descent" (1954, p. 184). The absence of such bias in the American descent system, Parsons suggests, is in large measure responsible for "the structural isolation of the individual conjugal family" (i.e., its autonomy).

The importance Parsons attributes to unilinearity as a factor in facilitating strong dependence upon kin ties is exemplified by his highlighting two exceptions to the structural isolation of the conjugal family in America—the upper-class elements, whose status depends on the continuity of their patrilineages' solidarity, and the lower-class elements, in which there is "a strong tendency to instability of marriage and a 'mother-centered' type of family structure" (Parsons 1954, p. 185). However, Parsons regards the urban middle class as characterizing "the focal American type of kinship." Since in the middle class the residence of the conjugal family typically is neolocal, and the conjugal family is economically independent of "the family of orientation of either spouse," the role of the conjugal family in U.S. society can be, for theoretical purposes, understood as master of its own destiny, rid of the impediments of extended-family ties.

In reaction to those sociologists who see modernity as inimical to bonds of kinship, other social scientists (e.g., Adams 1968; Firth et al. 1969; Litwak 1985; Mogey 1976; Shanas et al. 1968; Sussman 1959) turn their attention to the attenuated functions of kinship in contemporary society. Just as Goode (1963) notes a "fit" between the needs of modern capitalist society for a socially and geographically highly mobile population and the flexibility of the isolated conjugal family system, the revisionists indicate a similar fit between the existence of a highly mobile population and the presence of kin who give emergency aid and social support to relatives. The revisionists shift our attention away from constraints imposed by kinship loyalties and obligations and direct it instead to sources of services, goods, and emotional support that cannot readily be supplied by bureaucracies, markets, or other agencies. In his typology, Litwak (1960a, 1960b) distinguishes the isolated nuclear family (without kiship resources) from the traditional extended family (implying a hierarchy of authority), on the one hand, and from the modified extended family (which consists of a network of related but autonomous nuclear families), on the other. Although the revisionists have not destroyed the foundation of the bipolar family typologies, they do focus on a previously neglected area of analysis.

Three-Stage Typologies. Some modernization typologies introduce a third, transitional stage between traditional and modern kinship and family structures. These typologies accept the position that initially there is an emancipation from traditional kinship constraints and obligations, but they also propose that at some point new values of modernity emerge to fill the vacuum left by the dissipation of the old kinship constraints. For example, building on the work of LePlay, Zimmerman and Frampton (1966) offer a scheme of transformation in which families change from a patriarchal form to a stem-family structure and thence to an unstable family type. Zimmerman and Frampton begin with the premise that each social organization derives its "essential character" from a triad of "imperishable institutions"—family, religion, and property. However, in their view, "familism is necessary in all complete social organization to a degree more imperative than the need for property" (1966, p. 14; 1947). Zimmerman and Frampton regard the patriarchal family as the most familistic form. The patriarchal type is rooted in idealistic religious values and is characterized by a common household of a patriarch and his married sons and their families, wherein the property is held in the name of the "house," with the father as trustee. They identify the patriarchal form as having been prevalent among agriculturists in the Orient, in rural Russia, and among Slavonic peasants.

With urbanization and industrialization, however, the unstable family becomes predominant. Zimmerman and Frampton associate the unstable family with materialism and individualism and the resulting atomization of social life. Individuals are "freed from all obligations toward their parents and relatives" (1966, p. 15), and the identity of each conjugal family as a social unit ends with the death of the parents and the dispersal of the children.

The stem family represents a transitional state between the patriarchal and unstable forms. The stem family extends branches into urban centers while retaining its roots in the ancestral lands. As a result, the stem family provides a balance between the security of the traditional influences and resources of the "house" and the freedom and resources of the cities. (However, historical researchers yield less idyllic descriptions of the stem family than the Zimmerman and Frampton portrait. See Berkner 1972.)

A less romantic depiction of a transitional family type is drawn by Lawrence Stone (1975) in his typology of the English family's movement from feudalism to modernity. Stone posits the existence of a dual historical process. He places the decline of the importance of kin ties in the context of the emergence of a powerful, centralized state, and he then regards the rise of the modern family as an ideological emergence accompanying the development of capitalism.

According to Stone's typology, feudal England emphasized (1) kin-group responsibility for crimes and treasonable acts of members and (2) the institution of cousinship with its broad obligations. As political and economic power moved away from the traditional, landed elite to the state and the entrepreneurial class, the common law of the courts no longer recognized criminal and civil deviance as a kin-group responsibility, and cousinship lost its effectiveness. To fill the vacuum left by the decline of kinship as a factor in one's destiny, the relatively denuded conjugal family had to take over the task of guiding the destiny for its members. Consequently, by the sixteenth century, as an intermediate step toward the modern family, there was a trend toward authoritarianism in husband–wife interaction, and governance in the conjugal family took the form of patriarchy.

Stone (1975, p. 15) suggests that it was not until the eighteenth century that the spread of individualism and utilitarianism gave rise to a more companionate and egalitarian family structure. This last family form has been designated by Alan Macfarlane (1986) as the Malthusian marriage system, in which (1) marriage is seen as ultimately the bride's and groom's concern rather than that of the kin group; (2) marital interaction is supposed to be primarily companionate; and (3) love is supposed to be a precursor of marriage. Functionally, the Malthusian system yields relatively fewer children—by choice—than earlier family forms.

The Problem of Structure in Modernity Typologies. Typologies depicting historical transformations in family and kinship place much emphasis on the "fit" between the needs of modern industrial society and the presence of the conjugal family type (Litwak 1960a, 1960b; Parsons 1954). Despite this conjecture, Parsons (1954, p. 184) suggests that in Western society an "essentially open system" of kinship, with its "primary stress upon the conjugal family" and its lack of larger kin structures, has existed for centuries, long before the modern period. Like Macfarlane (1986), Parsons dates its establishment in late medieval times "when the kinship terminology of the European languages took shape." Moreover, Goode's (1963) analysis of family trends in eleven societies indicates that acceptance of modern, conjugal family ideology may precede economic and industrial development rather than come as a subsequent adaptation. Such findings cast doubt on the validity of the dichotomy between traditional societies and modernity as providing a theoretical basis for the typologies discussed above.

Parsons argues that (1) there is an incompatibility between corporate kinship and multilineal systems, and (2) in large measure, this incompatibility accounts for the prevalence of highly adaptable, structurally independent conjugal households in modern societies. However, findings by Davenport (1959), Mitchell (1963), Pehrson (1957), Peranio (1961), and others that corporate structures of kinship (such as clans) do exist in some multilineal kinship systems undercut Parsons's argument that such structures are to be found only in unilineal systems. Nevertheless, if multilateral kinship systems can accommodate corporate structures, then they can also include other kinship elements that sustain loyalties to descent groups and facilitate segmentation of the society.

The Problem of Connecting Kinship and Family in Modernity Typologies. Revisionists of the isolated conjugal family position have presented considerable evidence of residual elements of kinship ties in contemporary society. However, they do not adequately explain the connections between types of kinship systems and variation in performance of family functions in different parts of the social structure. Their main concern is with changes in kinship and family, changes that are consistent with the general loosening of tradition in modern society. But their focus on emancipation from tradition diverts their attention from (1) the influence of emerging ethnic, religious, or class interests upon patterns of integration of family networks in the larger social structure and (2) the temporal dimensions of kinship, which go beyond living kin to departed ancestors and generations yet to come.

Additionally, given the fact that the family–kinship typologies described above have their roots in the distinction between tradition and modernity, they overlook those nonindustrial, primarily nonurban societies in which families approach the companionship model as well as those ethnic and religious segments of industrial, primarily urban societies where strong familistic tendencies persist. Except for Stone (1975) and Zimmerman and Frampton (1966), these typologies are based on the concept of emancipation from tradition, and they do not deal explicitly with the emergence of new family values (other than flexibility and freedom). Most of all, their emphasis on emancipation from the constraints of tradition precludes their explaining why cohesive forces of family and kinship may remain strong (or increase in strength) in the face of an economic and social environment that is hostile to stable family life. (Exceptions are Sennett 1970 and Harris and Rosser 1983.)

Critical Commentary on Historical Typologies. Family typologies describing historical trends from one period of history to another are vulnerable to criticism of their teleological assumptions. Criticisms often involve (1) the definition of polar concepts and (2) the problem of inevitability.

1. Definition of polar concepts. The definition of polar concepts depends upon the value commitments of the analyst. For example, those analysts who view trends in kinship and family as movement toward liberation from traditional constraints and from obstacles to personal independence define the original state as confining and generally unjust and the future state as enabling emancipation from these obsolete social structures.
   
   Family-theorist Ernest Burgess and associates (1963) view the evolution of family structure as going "from institution to companionship"—from external community constraints upon family relations to voluntaristic unity that derives from affection, domestic peace, and common goals. Similarly, Marxists define the transition as being away from family structures required to sustain an economic system based on unearned rewards of the dominant class and suppression of the laboring class. Their claim is that following the rise of future true communism, the dissolution of economic classes would liberate family life from the constraints and suffering imposed by economic position; for Frederick Engels ([1885] 1942), under true communism, family life would be liberated from economic demands, and, founded on personal bonds, families would endure only as long as these bonds lasted.
   
   By way of contrast, analysts favoring traditional values define the trend in family life as a steady decay of family structure. Pitirim Sorokin (1937, vol. 4, p. 776), upon whose work Zimmerman and Frampton base their typology, notes that "the family as a sacred union of husband and wife, of parents and children, will continue to disintegrate . . . . [T]he home will become a mere overnight parking place mainly for sex-relationship." From another perspective, the behaviorist John Watson (1927) predicted that "in fifty years [1977], unless there is some change, the tribal custom of marriage will no longer exist. Family standards have broken down . . . The mystery and beauty of marriage and the rearing of children has pretty well broken down." In the Aldous Huxley's science fiction novel Brave New World (1955), all functions now performed by families would be community undertakings, and the word "mother" would be regarded as obscene.
   
   The distinction between typologies focusing on personal liberation and those portraying decay highlights the fact that each approach deals with partial realities. Liberation typologies tend to slight disruptive activities of emerging family structure (e.g., spouse abuse, child abuse, splitting into factions, isolation from resources of kin and family). Instead, they tend to associate these activities with traditional family structures. In the Soviet Union, family problems were generally attributed to survivals of the traditional pre-Revolution family forms. Decay typologies do the opposite. They tend to underestimate unifying elements and personal satisfactions associated with the emerging family types and to overestimate the chaos associated with these types.
2. Problem of inevitability. Some typologies posit a straight-line progression from a beginning state to an end state, while others at least imply a degree of indeterminacy in movement. For any particular typology, the degree of indeterminacy depends on the analyst's conception of history. For example, the Burgess institution-to-companionship family typology assumes there to be a major historical evolution from fairly isolated rural communities to societies with a high degree of industrialization; the family and kinship institutions evolve through a long series of adaptations to keep up with the industrialization of societies and a movement from a rural to an urban way of life. Similar typologies—like Toennies's Gemeinschaft to Gesellschaft ([1887] 1957) and Redfield's folk society to civilization (1947)—are based on comparable assumptions deriving from the dictum of Henry Maine, namely, that, as an overarching historical trend, the basis for social relations has been evolving from status to contract.

Despite the inevitability of trends implicit in the definition of polarities of family and kin structure in typologies of liberation (or decay), with the passage of time, definitions of polarities change. For example, analysts have redefined the concept of companionship as an end-state. In the 1940s, Burgess (1948; Burgess et al. 1963) regarded the future end-state as one in which the husband and wife (1) would be married without interference from family and community constraints, (2) would remain united through affection and common interests, (3) would maintain an equality in decision making and other aspects of family status, and (4) would orient their parenthood toward producing children with healthy personalities. But variations in family life included under the "companionship family" definition have been broadly expanded over time. By the end of the twentieth century, the end-state of the companionship family (as well as the unstable-family concept) has been redefined to include a diversity of household arrangements, such as (1) couples living together without formal marriage, (2) same-sex couples and their children (by adoption or by birth from previous or supplementary liaisons), and (3) voluntary single-parent households. Given these modifications in the concept of the companionship family, the very nature of the typology has been transformed. As a result, it is difficult to determine what family and kinship theorists will consider to be the evolutionary outcome twenty-five years from the present.

According to Murdoch's (1949) depiction of main sequence theory (described earlier), the changing pattern of employment has facilitated the widespread movement of women into broad sectors of occupations. This change has affected the composition of residences and, subsequently, will affect the descent structure and eventually kinship terminology. Evidence of this development can readily be seen. Around 1960, in an offhand comment during a lecture, Murdoch predicted that the control over wealth in America (1) would flow increasingly into the hands of women, (2) would at some point create shifts in household patterns, and (3) in the long run would produce a kinship structure dominated by women. Regardless of the accuracy of Murdoch's prediction, changes in practices pertaining to kinship are appearing in various ways: (1) Newspapers obituaries have routinely begun to include "life companions" (of either gender) in the list of related survivors; (2) public policy pertaining to health insurance coverage has been modified in some communities to include unmarried domestic partners; (3) in some countries (e.g., Russia, Israel), intestacy laws have been amended to include unrelated household residents; (4) the issue of legally recognizing same-sex marriages (or domestic partners) as a valid arrangement has emerged in a wide range of communities.

Unlike the theoretical inevitability of collectively rational adaptations assumed by evolutionary theorists, the typologies formulated by cyclical theorists lead away from regarding their end-states as inevitable. In their portrayal of historical processes, the cyclical theorists have the burden of explaining conditions for triggering reversals in historical cycles. These reversals imply that critical periods arise through cultural innovations and conflicts. The effects of novelty and conflict in these critical periods introduce an indeterminancy into the historical process. This indeterminacy brings to the foreground the problem of the inhibition of change: What introduces a new cycle, and what brings the cycle to a halt? For example, as discussed earlier, Zimmerman and Frampton (1947) see the history of the family as a series of repetitive cycles: a decay from corporate family forms (based on idealistic values) to unstable, chaotic families (based on materialistic values and individualism), followed by a regeneration of familism. Their scheme of analysis explains the oscillations between various degrees of familism and individualism in terms of a conflict between maintaining an enduring, traditional social structure and attending to persistent personal yearnings. The idealism of religious or ascetic values facilitates social stability in corporate family settings. However, the stifling of personal aims and desires, without idealism, encourages the adoption of materialistic values and sensuality associated with the unstable family. Hence, there is no guarantee that an old cycle will end or that new ideals supporting familism will again emerge.

A TRANSHISTORICAL TYPOLOGY OF KINSHIP AND FAMILY SYSTEMS

For well over a millennium, church intellectuals have been aware of variations in marital selection and their implications for family structure and kinship ties as well as for social structure. Early in the fifth century, in his De Civitate Dei (City of God), Saint Augustine of Hippo (1984, pp. 623–625) noted that in early biblical times demographic insufficiencies made it necessary for Jews to practice kinship endogamy. However, he proposed that marrying close relatives, and thereby creating multiple family ties with the same people, restricted the potential expanse of social circles that could be tied into a coherent community. Kinship endogamy tends to divide societies into segments. On the other hand, marrying persons from previously unrelated families would "serve to weld social life securely" by binding diverse peoples into an extensive web of relationships. Later, in the twelfth century, Gratian suggested that God commanded the Hebrews to select relatives as mates "because the salvation of man was realized in the pure Jewish race" but that the Christian faith, which could be readily spread through teaching, made kinship endogamy obsolete (Chodorow 1972, p. 74).

Gratian's argument suggests that the differences between Judaic and Christian marriage systems have broad implications for contemporaneous functions of kinship as well as for temporal functions, connecting past and future generations. The discussion that follows presents a kinship and family typology derived ultimately from Augustine's and Gratian's depictions of marriage systems as well as from issues pertaining to descent. This typology involves theoretical concerns drawn from sociology and anthropology.

Contemporaneous and Temporal Functions of Kinship Systems. Both marriage systems and descent rules affect the character of links between contemporaneous networks of families. A major controversy that at one time occupied many social anthropologists was whether marriage systems (i.e., marital alliances between groups) are more fundamental in generating forms of social organization than are descent rules or vice versa. At stake in the controversy was the issue of whether the social solidarity undergirding descent rules is more fundamental than the ideas of reciprocity and exchange involved in marriage systems. In the end, Africanists favored descent rules, while Asianists leaned toward marital alliances. In their assessment of the controversy, Buchler and Selby (1968) found evidence for the validity of both views.

However, despite the chicken-and-egg character of the controversy, the alliance–descent issue highlights the contradictory nature of kinship structure. This contradiction is depicted in the opposing views of structuralists such as Claude Levi-Strauss (1963), who supports the alliance position, and functionalists such as Meyer Fortes (1969), who argues for the descent position.

Alliance theories of kinship systems identify the primary function of kinship as the integration of networks of related families into the contemporaneous social fabric. Alliance adherents begin with marriage as the central element in structuring the way kinship operates. To alliance theorists, the significance of marriage lies in the idea that marriage is essentially a mode of exchange whose primary reason for existence is to inhibit conflict in society. In their view, kin groups exist as organized entities to effect marital exchanges. According to Levi-Strauss, the leading figure in alliance theory, "exchange in human society is a universal means of ensuring the interlocking of its constituent parts" (1963, p. 2). In unilineal systems, women are exchanged for equivalent valuable property, services, or both; in bilateral systems (which by their nature become multilateral in the long run), commitments to each other's relatives are exchanged. In bilateral kinship, bride and groom are of presumably equivalent value. Thus, in general, alliance theorists regard descent groupings primarily as a necessary ingredient for sustaining the marriage exchange system over the generations.

The descent theory of kinship systems rests on the assumption that the continued welfare of kindred over the generations is the primary function of kinship. In particular, Fortes regards "filiation"—being ascribed the status of a child of one's parents, with all the lifetime rights and obligations attached to that status (1969, p. 108)—as the "crucial relationships of intergenerational continuity and social reproduction" (pp. 255–256). He proposes that, as a concomitant of filiation, "the model relationship of kinship amity is fraternity, that is sibling unity, equality, and solidarity" (p. 241), and he provides a biblical example of the tie between David and Jonathan. But he also notes that "the Euro-American kinship institutions and values of Anglo-Saxon origin are imbued with the same notion of binding force of kinship amity" (p. 242), and he cites the mother–daughter relationship in England (in research findings by Young and Willmott 1957) as exemplifying that same moral code of diffuse but demanding reciprocal obligations.

On the one hand, alliance theory postulates that the basic drive in kinship organization is derived externally, from the kind of alliances appropriate to the structure of power in the community. Collectively, marital alliances create between families a network of links that integrate them in reference to overarching religious, economic, and political institutions. On the other hand, descent theory ascribes the bases of organization to internal demands, structural factors in the persistence of the kindred: rules governing residential location, division of labor and authority among members, and the various economic and political functions to be performed by the kinship system (Buchler and Selby 1968, p. 129).

Given the contradiction in the impulse for kinship organization, there is an apparent "impasse between the alliance and filiation point of view" (Buchler and Selby 1968, p. 141). What appears to be at issue is the depiction of the kinds of reciprocity norms that define the character of kinship. Descent theory presumes that an axiom of amity (i.e., prescriptive altruism or general reciprocity) is basic to the coherence of kin groups; alliance theory holds that balanced reciprocity (i.e., the rightness of exchanges for overt self-interest, opportunistic individualism, or noumenal norms) is in the final analysis the glue that integrates families and kin groups into a coherent whole.

The contradiction is apparent in many ways. For example, in biblical references and religious writings, the Ten Commandments enjoin one to honor parents and, conversely, to "cleave" to one's spouse and maintain peace in the household. In terms of kinds of reciprocity, one commandment involves unconditional giving or honoring, while the other concerns maintaining domestic peace (implying fair give-and-take).

Similarly, contemporary writers on marriage generally find the concept of balanced reciprocity appropriate in describing the quality of husband–wife ties. For example, Walster and Walster (1978) report that marriages work best when both husband and wife (as well as lovers) believe that each is receiving a fair exchange for what he or she offers in the relationship. Moreover, in their review of research on the quality of marriage, Lewis and Spanier (1982) note the importance of the symmetry of exchange in establishing and maintaining strong marital ties. However, in the socialization of children and in the allocation of resources, the rule of amity (or prescriptive altruism) is supposed to prevail. For example, parents are ordinarily expected to make "sacrifices" for their children when necessary; to do otherwise is to be a "bad" parent. In the American court system, the general rule for the disposition of children in cases of divorce, child neglect or abuse, or adoption is that the court should base its decision on the welfare of the child rather than on the interests of the parents or other parties.

To some extent, the descent–marriage contradiction can be obscured by compartmentalizing marital, parental, and filial conduct and by dividing responsibilities of husband and wife. However, conflicts in norms for dealing with family members and kindred may occur for several reasons, but they occur principally because of scarcities of time and resources required to carry out duties and obligations in the face of a wide range of simultaneous and conflicting demands. Since the resulting dilemmas are widespread in the society, there is a need for a general rule. Because contradictory alliance and descent impulses are operative, each group is pushed to establish a coherent kinship scheme that gives priority to one impulse over the other or at least establishes some form of compromise between them.

There is evidence that rules governing marital functions conflict with those pertaining to descent functions, paralleling the alliance–descent controversy in kinship systems. Where descent functions are given precedence in family organization, marital functions are subordinated (and vice versa). Examples of this inverse relationship are (1) if husband–wife unity is central, then the unity between siblings is peripheral (and the reverse), and (2) if marriage between close affines is forbidden, first-cousin marriage is permitted (and vice versa). These examples are discussed in the sections that follow.

Marital Unity Versus Unity of the Sibling Group. Comparisons between societies indicate that ties between siblings have an inverse relationship to husband–wife ties. Where descent is valued over alliance or marriage in kinship relations, brother–sister bonds are particularly close (Parsons 1954), while the husband–wife relationship is relatively distant. In such family systems (whether or not its therapeutic implications are true), parents are expected to remain together for the sake of the children, and this expectation expresses the priority of descent over marital ties. Conversely, in family systems where the marriage function is more valued, the husband–wife relationship is intense (e.g., the importance of the give-and-take of love and of companionship for marriage) and the brother–sister relationship is competitive, distant, or both and the incest taboo justifies their apartness (see Lopata 1973 on widows and their brothers). In societies where priority is given to marital bonds over descent ties, the presence of children is of less importance in dissolving an unhappy marriage, and there is greater ambiguity about what is best for the children. The mere fact that the strength of brother–sister ties and that of marital ties vary inversely in different societies lends support to the proposition that there is a contradiction in the family system between its marital functions and its descent functions.

Affines and Cousins in American Marriage Law. The opposition between marital and descent functions in the family is also illustrated by the inverse relationship in American law of marriages considered to be incestuous: As a general tendency, states that forbid second marriages between a person and certain affines (such as that person's parents-in-law and sons- or daughters-in-law) allow first cousins to marry, while those that permit marriage between close affines forbid first-cousin marriage (Farber 1968). If the preferred function of marriage is to reinforce close consanguineous kinship ties, then this pattern of marital prohibitions signals a subordination of affinal bonds to those of consanguinity. Marrying into the family of the former spouse will not reinforce any of the other existing bonds of consanguinity. Consequently, although first-cousin marriage is to be permitted in order to reinforce intimate kinship ties, marriage with close affines should be avoided. However, if marriage is considered to be primarily a mechanism for creating new bonds between previously unrelated families, then a second marriage into the same family merely serves to maintain the affinal bonds initiated in the first marriage.

Social Structure and Kinship Systems. The presence of contradictory impulses in organizing kinship ties produces a predicament in establishing priorities between them. This contradiction evokes a question: Which circumstances lead some societies (and ethnic and religious subgroups) to give priority to descent and others to favor alliance assumptions in their kinship and family organization (Farber 1975)? In their analyses of the relationship between kinship organization and social structure, both Paige (1974) and Swanson (1969) distinguish between societies that feature the legitimacy of special interests—factionalism—in organizing social life and those that feature the importance of common interests—communalism—as an organizing theme.

Factions are a means for gathering forces and mobilizing members for conflict or competition with other factions. They emerge as a reaction to perceived danger to their well-being from other groups (cf. Douglas 1966). Factions emerge where either (1) special interest groups vie for superiority over other groups for access to power, wealth, or some other property, or (2) groups sense a danger to their continued autonomous existence as an ethnic or religious entity.

In kinship organization, the continual mobilization of family and kin results in the generation of norms that are centripetal in nature, that is, they facilitate the pulling inward of human, symbolic, and material resources. This centripetal tendency permits each kin group to separate itself from competing groups in order to endure. As a result, centripetal kin groups favor norms strengthening descent relationships over norms facilitating new alliances with other groups through marriage. Insofar as descent-group norms are rooted in the axiom of amity, one would expect centripetal kinship organization to feature the norm of prescriptive altruism over balanced reciprocities in kinship and family relations (see Farber 1975).

Jewish family norms provide some insight into the relationship between centripetal kinship systems and the application of the axiom of amity. In its basic ideology and in the code of laws supporting that ideology, Judaism assigns a major significance to the concept of nurturance (Farber 1984). Since nurturance is a central feature of maternal giving, it can be regarded as a metaphor for the axiom of amity. The Code of Jewish Law (Shulkhan Arukh) offers numerous instances that signify the place of nurturance in Judaism (Ganzfried 1963). For example, the code sublimates feeding and eating into sacred, ritualistic acts. The act of eating is invested with holiness, to be enjoyed in abundance, particularly on feast days and the Sabbath. A connection is made in the code between providing food and giving gifts and charity. It proposes that festive occasions are also times for charity to the needy and for sending gifts. In addition to drawing a connection between food and charity, the code applies the metaphor of the parent–child relationship to charity giving and assigns a priority to family in its general concept of nurturance: First parents, then offspring, and "other kinsmen take precedence over strangers" (Ganzfried 1963, chap. 34). The injunction to nurture children involves an emphasis not only on food but on other aspects as well (for example, an exaggerated emphasis on elaborated linguistic codes for use in child rearing). Zena Smith Blau (1974) writes that "whatever Jewish mothers did for their children—and they did a great deal—was accompanied by a flow of language, consisting of rich, colorful expressive words and phrases" (p. 175). The aim of socialization is presumably to turn the child into a Mensch—to transform the child from a receiver of nurture to a giver of nurture (Zborowski and Herzog 1952). Hence, in traditional Judaism, the concept of nurturance seems to tie together the kinship emphasis on descent and the axiom of amity in organizing family relationships.

If nurturing the next generation is a form of prescriptive altruism, this nurturing can also occur in symbolic form. Like the transmission of physical wealth and nurturing, the parents can also transmit a "symbolic estate" to the next generation. In a real sense, along with material resources, people inherit a collection of living and dead relatives connected to them by birth and/or marriage. These relatives constitute a trove of heroes and villains whose personal qualities, exploits, and ideas are remembered in socializing succeeding generations. This "symbolic estate" defines for individuals (1) a sense of belonging to an identifiable "family," (2) role models to emulate (or disown), (3) a legitimation of one's place in community and society (Farber 1971). Craig (1979) sees the symbolic estate as a vehicle for achieving personal and familial immortality. Implicitly, it is one's duty in centripetally-oriented kinship systems to contribute to the symbolic estate by living an exemplary life (however this way of life is defined in particular historical circumstances).

To be operative as memorials (or reminders), the content of symbolic estates must have some bearing upon the personal identities (or destinies) of family members. In Judaism, historically this meant assessing the "quality" of one's ancestry (yachas), however defined; this assessment was particularly important in eras of arranged marriages. However, Yerushalmi (1982) notes the general importance of collective memory for the endurance of Judaism. Its centrality is suggested by the appearance of the verb zakhar (to remember) "in the [Hebrew] Bible no less than one hundred sixty-nine times" (Yerushalmi 1982, p. 5). Moreover, numerous memorials have been incorporated into holy day observances (e.g., the retelling of the story of the Exodus annually at the family seder at Passover). Especially significant for sustaining symbolic estates among Jews is the ritualizing of the remembrance of dead relatives through (1) memorial prayer services (yizkor) on four major holy days, and (2) partly as a means to continue to honor one's parents after their death, the recitation of the prayer for the dead (kaddish) on anniversaries of the death of each family members. The concept of symbolic estates connects collective family memories—such as legends, myths, and moral ideas—to the continuity of "family" from one generation to the next. In her study of Genesis, Steinmetz (1991) applies the concept of "symbolic estates" to the succession from father to son of the obligation to ensure the realization of God's command to found and then maintain a Jewish nation. She regards the entire structure of Genesis as resting upon the transfer of this ideal to worthy heirs in the family line. Her emphasis upon the transmission of "symbolic estates" is echoed in an investigation by Bendor (1996) of the social structure of ancient Israel. Bendor concludes that Israeli social stratification is derived to a large extent from the kinship ideology of familial perpetuity—rather than from the influence of economic factors upon kinship and family life. The findings on ancient Israel by Steinmetz and Bendor bear upon historical and contemporary studies of kinship and family. Examples are the research reports by Pina-Cabal (1997) on family legends in urban Portugal, Attias-Donfut (1997) on home-sharing in France, Hastrup (1982) on establishing Icelander ethnicity, and Weigert and Hastings (1977) on maintaining family archives of photographs, old records, letters, and other memorials. The focus in these studies is upon symbolic mechanisms for sustaining family continuity. (The discussion of centrifugal kinship systems in the next section will describe obstacles to the perpetuation of "symbolic estates.")

As opposed to factionalism, communalism implies a situation in which special interests are subordinated to common concerns of diverse groups. In stateless societies, these common concerns may well emerge from economic interdependence or the presence of a common enemy. In societies with a centralized government, the state presumably symbolizes a concern for the common welfare of the populace. Other unifying concerns may exist as well, for example, the presence of a universal church (as opposed to competing sects and denominations), nationalism (as opposed to ethnic self-determination), a centralized bureaucracy or market (as opposed to regional competition for dominance), and so on. The common concerns would best be served if members of kin groups were to be dispersed by marriage to previously unrelated people living throughout the society. This dispersal would maximize the number of diverse kin groups with which any family is connected, and it would thereby scatter kinship loyalties, obligations, and property as widely as possible. Consequently, this kind of kinship system, associated with communalism, can be identified as applying an outward pressure upon its constituents; it is centrifugal in nature.

In contrast to the centripetal system, the centrifugal system subordinates kinship ties to conjugal family ties and extends marital prohibitions widely in order to inhibit marriages that would merely reinforce existing consanguineous ties. According to the theory outlined above, in centrifugal kinship systems, in which marriage functions are given priority over descent functions, the appropriate norm for defining family interaction is balanced reciprocity—exchange rather than the axiom of amity.

In the United States, although the centrifugal kinship system appears in a wide range of socioeconomic, religious, and ethnic groups, it is found disproportionately at lower socioeconomic levels, where families seek improved integration into the larger society (Farber 1981).

The application of balanced exchange as a norm in family and kinship is exemplified in a study of poor families by Stack (1974). She describes the prevalence of "swapping" as a named, bartering norm governing both ties between kin and between family members in their struggle for survival. Stack notes that "reciprocal obligations last as long as both participants are mutually satisfied" and that they continue such exchange relationships as long as they can "draw upon the credit they accumulate with others through swapping" (p. 41). Indeed, according to Stack, "those actively involved in domestic networks swap goods and services on a daily, practically an hourly, basis" (p. 35). But this exchange does not constitute a playing out of the axiom of amity since "the obligation to repay carries kin and community sanctions" (p. 34) and it extends beyond family and kin to friends. Although swapping may involve some element of trust, it exists to ensure exchanges in the lean times that predictably recur in domestic networks that are too marginal in resources to be magnanimous. It pays to create numerous bartering arrangements rather than to accumulate obligations within a very small network of intimate kin. Thus, in its own way, swapping mimics the proliferation of networks of previously unrelated families characteristic of centrifugal kinship systems.

In contrast to the importance of "symbolic estates" for facilitating the "immortality" of families in centripetal kinship systems, families in centrifugal systems are often characterized by a "legacy of silence." This silence may signify the existence of shameful or immoral acts of relatives, or it may simply reflect an emphasis upon individualism in these families. In Germany after World War II, this "legacy of silence" functioned to erase the collective memory of parental activities and ideas they held during the Nazi era (Larney 1994, pp. 146–162). For victims of torture and displacement under the Nazi regime, the legacy of silence enabled them to wipe their degradation from memory (Bar-On 1989). In either case, whereas symbolic estates provide a vehicle for family continuity, the legacy of silence established a discontinuity.

The German experience may result in a single break in family continuity—to permit starting afresh. However, the institutionalization of the legacy of silence in centrifugal kinship systems perpetuates this discontinuity between generations of nuclear families. This legacy has been found to be prevalent in low socioeconomic-level families populating urban slums (Farber 1971). Families tend to exchange little information about one another; in fact what is hidden may permit closer ties between kin than the revelation of illicit or immoral acts. Then too, in families where welfare agencies and police intrude, silence serves to maintain the privacy of the household.

The tacit norm of collective forgetting in these centrifugal kinship systems places the onus for kinship unity upon mutual assistance, friendship, and availability of kin. Yet, in her study of kinship among poor racial and ethnic minorities, Roschelle (1997) found that degree of mutual assistance between families and extent of interaction among relatives depend largely upon availability of kin. Migrant families frequently are isolated in time of need and the legacy of silence may thereby be enhanced.

In a society marked by much internal migration and social mobility, there are many opportunities for a proliferation of centrifugal tendencies in kinship. Whether centrifugal systems actually emerge through mobility may depend upon a variety of factors. Gullestad (1997) notes a shift in the meaning of kinship in urban Norway. She describes a social transformation from norms regarding "being of use" and social solidarity to self-realization and "finding oneself" (or "being oneself"), that is, from norms sustaining family continuity to norms fostering separation and discontinuity. She attributes this shift to "transformed modernity" involving "fundamental restructurings of home and neighborhood because women and children are not present in the same way or to the same extent as before" (Gullestad 1997, p. 210). Transformed modernity, as well as advances in reproductive technology, is identified also as a factor in the proliferation of diverse forms of kinship structure in contemporary society (Strathern 1992).

One can interpret the emergence of feminist movements as both stimulating and stimulated by the "transformed modernity" cited by Gullestad. For some forms of feminism, post-modern thought provides a rationale for denigrating traditional symbolic estates. Post-modern writings propose that the framing of "factual" and theoretical statements have an exclusionary element—that is, they mark a population segment for exclusion from free participation. These "factual" statements justify this exclusion. As "factual" statements, posing as objective discourses, these statements have a hidden core. This core reflects the special interests of those with the power to define "truth" for the society. (See Foucault [1971] 1996.) Certain feminists claim that the hidden core of meaning in statements justifying exclusion of women from full participation in society is to promote male dominance in social structure (Barnard 1993). The symbolic estates that facilitate the endurance of existing lines of descent are thus seen as supporting patriarchy. Consequently, they are regarded as an obstacle to the full participation of women in society. Yet, as women's participation in economic and political spheres continues to expand, it is likely that symbolic estates will eventually be infused with a marked increase in content pertaining to exploits and interests of women.

Kinship-Map Typology. Although mapping of kinship ties cannot express all aspects of kinship relations, it can generate models expressing general orientations implicit in various patterns of kinship structure. Basically, genealogical maps of consanguineal ("blood") relationships merely locate positions in an ideal web of biological connections. Variations in mapping come into play when these maps are used to describe how one's obligations and proscriptions vary in different kinship structures. Such obligations and proscriptions pertain to marriage, remarriage, birth and adoption, inheritance, relations between generations and genders, and so on.

One advantage of models of genealogical mapping is that these models express the logical connections between functions of kinship in a particular society and priorities assigned to different kin statuses. For instance, a kinship type with a prohibition to marry a first cousin generally has a different function in society as compared to one permitting such marriage. This pattern of marital prohibitions will likely be related to priorities in inheritance. For instance, in American state laws, permitting first-cousin marriage would be associated with giving a niece or nephew precedence over a grandparent in intestate inheritance (i.e., when there is no written will). The opposite will likely be true where first-cousin marriage is forbidden.

The relationship between genealogical mapping and functions of kinship has a long history in Western civilization. Atkins (1974) has explored a wide range of formulae for generating different patterns of priorities in mapping genealogical relationships. Implied in genealogical mapping is the principle that the smaller the number of links (by birth or marriage) between relatives, other things being equal, the greater is the degree of obligation between them. Thus, in such matters as succession to estates, when a choice is to be made among kin, genealogically close relatives are presumed to be given priority over more distantly related kin. However, since the various formulae differ in the patterns of priority among kin generated, choice of an appropriate pattern of mapping depends on the role of kinship in the particular society.

In general, three patterns of priority for mapping kin have been applied in the Western world (mainly in laws of intestacy and marriage). However, in practice, each society makes modifications in these patterns to fit its needs. Examples of these patterns occur in (1) Catholic canon law and the state of Georgia, (2) the civil code of the Twelve Tables of the Roman Republic and more recently in Napoleonic Code and Louisiana law, and (3) the parentela orders in the Hebrew Bible and in abbreviated form in Israel, Germany, and various states (e.g., Arizona) (Farber 1981).

At one pole, the canon law of the Catholic Church stipulates that a function of the church is to create a unity that ties together diverse segments of its constituency in a web of extensive relationships (including family bonds). This aim implies that collateral ties between families are equal in importance to ties between ascendants and descendants (i.e., between generations). Under such conditions, ties between are extended outward in a centrifugal fashion. In laws governing marital prohibitions, marriage is discouraged within the second degree of distance of collateral kin (i.e., first cousins). In earlier generations, marital prohibitions in Canon Law were even more inclusive; for example, in thirteenth century, consanguineous marriages were prohibited within the fourth degree of relatedness.

In assigning distances from Ego in the canon law genealogical model (e.g., for priorities in inheritance), (1) all consanguineal members of Ego's nuclear family (parents, siblings, and children) are one degree of distance from Ego, (2) relatives just outside the nuclear family are two degrees of distance (grandparents, aunts and uncles, first cousins, nieces and nephews, and grandchildren), and so on. The canon law model thereby expresses the general principle that neither line of descent nor collateral distance is given special emphasis—only degree of distance from one's nuclear family is significant.

At the opposite pole, the parentela orders genealogical model places much emphasis upon line of descent (and among collateral relatives, the closeness of line of descent). Like the sociobiological ideal, the parentela orders model is oriented toward the survival of any given line of descent (or failing that, the next closest line of descent). This model expresses centripetal tendencies in kinship structure. In marriage law, collateral prohibitions are minimal, and marrying someone in the closest line of descent (first cousin) is preferred.

As the parentela orders model is applied to intestacy law, the centripetal principle is expressed in the Hebrew Bible in Numbers 27:8–11 and 36:7–9. In this model, priorities among relatives are allocated by line of descent: (1) Direct descendants of Ego are given first priority (children, grandchildren, etc.); (2) if there are no direct descendants, those of Ego's parents are given next priority (siblings, nieces and nephews, etc.); (3) next in priority are descendants of Ego's grandparents (aunts and uncles, first cousins, etc.)—and so on. In theory, Ego's estate will be passed on to the closest survivor in the closest line of descent to Ego's.

Between the extremes of centrifugality of the canon law model and the centripetality of the parentela orders model stands the civil law model. This model gives somewhat more weight in assigning closeness in kinship distance to direct-line ascendants and descendants than to collateral relatives (i.e., those related to Ego through a common ancestor). In computing kinship distance from Ego, the civil law model counts generations between Ego and the common ancestor as well as generations between the other relative and the common ancestor; for direct-line relatives, only those generations between Ego and the other relative need be counted. Obviously, the nearer the common ancestor is to Ego, the closer is the collateral relative in genealogical distance (and vice versa). Thus, Ego's grandparent is closer genealogically than a niece or nephew; the reverse is true for parentela orders priorities, and both are equidistant from Ego in the canon law model.

Several social surveys have been undertaken to test empirically the above propositions about ways in which people's conceptions about priorities assigned to different relatives in kinship mapping are actually reflected in their lives—religious affiliation, socioeconomic status, minority status, and so on. In these surveys, the respondents were asked to choose priorities among kin (for which the kinship-map models differ) if they were to write a law to govern intestacy (i.e., where there is no written will). For ten pairs of relatives for whom the kinship models differed in assigning a priority, within each pair, the respondents were to select the relative they thought should have precedence (as a general rule). (Equal priority was one alternative.) Respondents were then classified according to the kinship model to which a majority of their choices conformed.

The first surveys were undertaken in the United States (Farber 1977, 1979). The results indicate that Jewish respondents do indeed tend to view priorities from the perspective of the parentela orders model, while Catholics tend to be overrepresented in the canon law category. Of course, these are tendencies and not blanket findings covering all Jews or Catholics. For example, the degree to which a religious grouping adheres to scripture and/or ritual practices seems important in influencing kinship mapping. When religious branch was taken into account, responses of Jews who identified themselves as Conservative (a fairly traditional branch) tended to conform to the parentela orders model and none conformed to the canon law model, while those in the Reform category more often conformed to the canon law model than to than to the parentela orders model. Similarly, among Mormons whose marriage was sealed in the Temple, their responses were like those of the Conservative Jews, whereas those whose marriage was not sealed for time and eternity responded like Reform Jews. Moreover, neofundamentalist Protestants were the only other religious grouping overrepresented in the parentela orders category (Farber 1981, pp. 73–75).

Taken together, the above findings suggest that the parentela orders model tends to be prevalent in groupings where endurance of the particular religious community into the distant future may be problematic. The community would then be motivated to intensify its inward pull—its centripetal incentive—to keep succeeding generations within the fold.

In the course of one investigation (Farber 1981), a reanalysis of findings yielded a fourth kinship model. This model, whose computation is the reverse of the parentela orders model, emphasizes obligations to ancestors who have been responsible for preparing the groundwork for Ego's place in society. In the serendipitous model, Ego's direct ancestors are given priority over any descendants—first priority is given to parents, grandparents, and so on; the next set of priorities consists of Ego's children, then Ego's brothers and sisters, aunts and uncles, great-uncles and great-aunts, and so on; following these, Ego's grandchildren, nieces and nephews, first cousins, and on and on (Farber 1981, p. 50).

The serendipitous model was disproportionately prevalent in several sectors of respondents—nonminority Protestants, those in professional and managerial occupations and at higher income levels, and those persons with U.S.-born fathers. In addition, persons who conformed to this model tended to come from smaller families (Farber 1981, p. 217) and expected to have fewer offspring than did other respondents (Farber 1981, p. 147). Since almost half the sample studied conformed to this model, it seemed appropriate to name it the Standard American model. The U.S. findings on the standard American model are consistent with Alexis de Tocqueville's observation made almost two centuries ago in Democracy in America ([1850] 1945), namely, that compared with Continental Europeans, Americans live in the present and show little interest in the perpetuation of family lines.

European data on the genealogical models throw further light on differences in the conception of kinship priorities between U.S. and Continental populations. An investigation in central Europe (Vienna, Bremen, and Cologne) shows parentela orders to be by far the most prevalent kinship model, especially among those families at upper socioeconomic levels (Baker 1991). By way of contrast, Baker's (1991) data from Dublin, Ireland, tend to be similar to the American findings: Jews display a strong tendency to conform to the parentela orders model, while Protestants and Catholics favor the standard American model (called by Baker the intercultural bourgeois model). Despite all the changes that have occurred over the generations, traditional perceptions of priorities in kinship claims still persist.

The typology of kinship maps (or collaterality models) is a heuristic for understanding an implicit theory of the workings of kinship structure. In itself, the typology is too simplistic to denote the complexity of norms and values and the operation of mechanisms involved. But these criticisms about the heuristic character of the types of collaterality models can be applied to all typologies used in kinship analysis. They are merely methodological tools for gaining insight into what is going on. To gain this insight, one forgoes the many nuances that give color to understanding the functioning of kinship.

Related Transhistorical Typologies. Variations on issues pertinent to the structural contradiction typology have been developed in other transhistorical schemes associated with the role of marriage and descent systems in organizing family and kinship systems. For instance, Guichard (1977) distinguishes between Eastern/Islamic and Western/Christian kinship systems. According to his typology, in the Eastern system, (1) descent is patrilineal; (2) marital ties are weak, and polygyny and easy divorce are permitted; (3) close ties exist between kin related through male lineage groups; (4) strong preference is given to endogamy within patrilineages; and (5) the sexes are segregated and women are relatively secluded within the home. In contrast, in the Western system, (1) kinship is bilineal or bilateral/multilateral, with ties to the maternal family considered important and with an emphasis on affinal connections as well; (2) marital bonds are the dominant unifying feature in family and kinship, with monogamy as prescribed and with extended kin ties as weak; (3) kin ties are defined according to individual connections rather than by lineage groups, with an emphasis on the ascending line rather than the descending line and with little importance attached to lineal continuity or solidarity; (4) kinship exogamy is prescribed, with endogamy permitted primarily for economic reasons; and (5) interaction between the sexes occurs in a wide range of circumstances.

In his reaction to Guichard, Goody (1983) revives the anthropological controversy between alliance theory and descent theory. Goody criticizes Guichard for basing his typology on marital norms (i.e., the endogamy–exogamy distinction) and suggests that by not starting with descent factors (i.e., inheritance practices), Guichard has overlooked a more fundamental distinction—that between kinship systems in which property is passed from one generation to the next through both sexes (by means of inheritance and dowry) and those systems in which property is transmitted unisexually (usually through males). Goody contends that passing property down unisexually encourages the development of corporate kinship groups (e.g., African systems). However, the use of bilateral devolution discourages such corporate structures, and Goody places both Eastern and Western systems in Guichard's dichotomy in the bilateral category. He faults Guichard for overstating the existence of corporate structures in Eastern kinship and proposes that Guichard's Western type represents merely a later historical development away from its roots in the Eastern system. Goody sees the primary problem of explaining the character of family and kinship in Western society as one of discerning how European societies shifted from preferred kinship endogamy (e.g., first-cousin marriage) to prescribed exogamy.

In his analysis of European kinship, Goody considers the changes introduced by the Christian (i.e., Roman Catholic) church from its beginnings to the late medieval period. He interprets the shift from kinship endogamy to exogamy mainly as a strategic move by the church to gain control over the lives of its members. As part of this effort, it had to wrest access to resources (especially productive land) from enduring control by family and kin. As a result, church laws evolved favoring those norms that might enhance allegiance to the church and weaken competition from the family and the state. In consequence, the church favored (1) the use of testation permitting bequests to the church; (2) the prescription of kinship exogamy as a means for inhibiting both the reinforcement of close kin ties and the passing down of resources exclusively within lineages; (3) the requirement of the consent of both bride and groom in marriage; (4) late marriage as a means for weakening family control over mate selection; (5) prohibition of divorce even for childless couples; and so on.

Goody seems to overstate his case in trying to interpret the shifts in kinship in ways that are consistent with his basic typology. For example, in giving primacy to inheritance patterns, Goody asserts that the ban on divorce in Roman Catholicism was devised primarily to encourage bequeathing estates to the church in case of childlessness. But, in fact, when there were no children, bequests usually were made "to brothers and sisters and to nieces and nephews" (Sheehan 1963, p. 75). Moreover, Goody's explanation of the ban ignores the widespread practice of bequeathing a portion of one's estate to the church even when one left a widow, children, or both. Sheehan (1963) reports that these bequests were made for the good of the soul: "Among the Anglo-Saxons, bequests to the palish church became so general that they were eventually required by law" (p. 292). This practice was not restricted to England. According to Sheehan, "Christians in the Mediterranean basin had developed the practice of bequeathing part of their estate in alms" (p. 303). Thus, church heirship in medieval Christian Europe was tied to repentance regardless of the existence of familial beneficiaries. Since church acquisition did not have to depend on bequests from childless couples, it is unlikely that the ban on divorce derives primarily from the desire of the church for additional benefices.

In addition, Goody dismisses the intermittent presence of kinship endogamy in medieval Europe as opportunistic deviations from the moral injunctions of the church. Yet, as Duby (1977) indicates, in medieval Europe the ebb and flow in kinship endogamy was tied to the amount of emphasis given to strengthening lines of descent. For example, Duby notes that in northern France, from before the tenth century to about the middle of the eleventh century, there was little utilization of the concept of lineage and only vague awareness of genealogy and knowledge about ancestors. Prior to that time, even members of the aristocracy considered their family to consist of "a horizontal grouping" of neighbors and kin "whose bonds were as much the result of marriage alliances as of blood" (Duby 1977, p. 147). Then, beginning in the tenth century, there was a change in ideas and norms regarding kinship—a conscious strengthening of lineage by controlling marriage, which frequently took place between close relatives despite impediments in canon law (Canon Law Society 1983). To summarize, Goody's argument is that medieval deviation from canon law consisted of opportunistic economic decisions and did not derive from a different set of norms. But Duby describes the coordination of kinship endogamy with the emerging notion of the legitimacy of lineage—a complex of ideas that requires a consensus among the kin in order to be effective. Hence, it appears that the change in marriage rules and the significance of lineage signaled more that ad hoc departures from church law.

There is still another reason for questioning Goody's conclusions: Goody makes the point that through bequests the Catholic church became the largest landowner in Europe. In his focus on the growth of exogamy as a consequence of the devolution of estates to both sexes, he has over-looked the church's own involvement as a major heir in the inheritance system. Particularly in the light of the church's view that ties through faith are equivalent to blood ties, the church is identified with spiritual kinship (Goody 1983, pp. 194ff). However, if it is legitimate to consider the church as an heir on a par with familial heirs, the system becomes one of trilateral devolution—sons, daughters, and the church. In that case, the European system differs markedly from the Eastern kinship system described by Guichard. Indeed, in contrast to Judaism and Islam, Christianity, at least until the end of the medieval period, saw family and kinship ties as competitive with church interests, and the strategies the church applied to weaken these ties altered both the marriage and the inheritance systems. The data imply that, despite their contradictory implications, the marriage, the alliance component, and the descent component should be addressed as equal factors in organizing family life. A task that remains is to integrate typologies of the emergence of modern kinship systems with transhistorical, structural typologies.

(see also: Alternative Life Styles; American Families; Family and Household Structure; Family Roles)

references

Adams, Bert N. 1968 Kinship in an Urban Setting. Chicago: Markham.

Atkins, John R. 1974 "On the Fundamental Consanguineal Numbers and Their Structural Basis" American Ethnologist 1:1–31.

Attias-Donfut, Claudine 1997 "Home-Sharing and the Transmission of Inheritance in France." In M. Gullestad and M. Segalen, eds., Family and Kinship in Europe. London: Pinter.

Augustine, Saint 1966 The City of God Against the Pagans. New York: Penguin Books.

Baker, David J. 1991 Conceptions of Collaterality in Modern Europe: Kinship Ideologies from Companionship to Trusteeship. Unpublished doctoral dissertation, University of Illinois, Urbana–Champaign.

Barnard, Malcolm 1993 "Economy and Strategy: The Possibility of Feminism." In Chris Jenks, ed., Cultural Reproduction. New York: Routledge.

Bar-On, Dan 1989 The Legacy of Silence: Encounters with Children of the Third Reich. Cambridge, Mass.: Harvard University Press.

Bendor, S. 1996 The Social Structure of Ancient Israel. Jerusalem: Simor.

Berkner, Lutz 1972 "The Stem Family and the Developmental Cycle of a Peasant Household: An Eighteenth-Century Example." American Historical Review 77:398–418.

Blau, Zena Smith 1974 "The Strategy of the Jewish Mother." In Marshall Sklare, ed., The Jew in American Society. New York: Behrman House.

Buchler, Ira R., and Henry A. Selby 1968 Kinship and Social Organization. New York: Macmillan.

Burgess, Ernest W. 1948 "The Family in a Changing Society." American Journal of Sociology 53:417–422.

——, Harvey J. Locke, and Mary Margaret Thomes 1963 The Family: From Institution to Companionship. New York: American Book Company.

Canon Law Society of Great Britain and Ireland 1983 The Code of Canon Law. London: Collins Liturgical Publications.

Chodorow, Stanley 1972 Christian Political Theory and Church Politics in the Mid-Twelfth Century. Berkeley: University of California Press.

Craig, Daniel 1979 "Immortality through Kinship: The Vertical Transmission of Substance and Symbolic Estate." American Anthropologist 81:94–96.

Davenport, W. 1959 "Nonunilinear Descent and Descent Groups." American Anthropologist 61:557–572.

Douglas, Mary 1966 Purity and Danger. London: Routledge and Kegan Paul.

Duby, Georges 1977 The Chivalrous Society. London: Edward Arnold.

Engels, Frederick (1885) 1942 The Origin of Family, Private Property, and the State. New York: International Publishers.

Farber, Bernard 1968 Comparative Kinship Systems. New York: Wiley.

——1971 Kinship and Class: A Midwestern Study. New York: Basic Books.

——1975 "Bilateral Kinship: Centripetal and Centrifugal Types of Organization." Journal of Marriage and the Family 37:871–888.

——1977 "Social Context, Kinship Mapping, and Family Norms." Journal of Marriage and the Family 39:227–240.

——1979 "Kinship Mapping Among Jews in a Mid-western City." Social Forces 57:1107–1123.

——1981 Conceptions of Kinship. New York: Elsevier.

——1984 "Anatomy of Nurturance: A Structural Analysis of the Contemporary Jewish Family." Paper presented at Workshop on Theory Construction and Research Methodology, National Council on Family Relations, San Francisco, October.

Firth, Raymond, Jane Hubert, and Anthony Forge 1969 Families and Their Relatives. New York: Humanities Press.

Fortes, Meyer 1969 Kinship and Social Order. Chicago: Aldine.

Foucault, Michel (1971) 1996 "The Discourse on Language." In R. Kearney and M. Rainwater, eds., The Continental Philosophy Reader. New York: Routledge.

Ganzfried, Solomon 1963 Code of Jewish Law (Kitzur Shulkhan Aruhh), rev., annot. ed. New York: Hebrew Publishing Company.

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

Goody, Jack 1983 The Development of the Family and Marriage in Europe. New York: Cambridge University Press.

Guichard, P. 1977 Structures sociales 'Orientales' et 'occidentales' dans l'Espagne musulmane. Paris: Mouton.

Gullestad, Marianne 1997 "From 'Being of Use' to 'Finding Oneself:' Dilemmas of Value Transmission between Generations in Norway." In M. Gullestad and M. Segalen, eds., Family and Kinship in Europe. London: Pinter.

Harris, C. C., and Colin Rosser 1983 The Family and Social Change. Boston: Routledge and Kegan Paul.

Hastrup, Kirsten 1982 "Establishing an Ethnicity: The Emergence of the 'Icelanders' in the Early Middle Ages." In David Parkin, ed., Semantic Anthropology. New York: Academic Press.

Huxley, Aldous 1955 Brave New World. New York: Bantam.

Larney, Barbara Elden 1994 Children of World War II in Germany: A life course analysis. Unpublished doctoral diss.

Levi-Strauss, Claude 1963 Structural Anthropology. New York: Basic Books.

——1969 The Elementary Structures of Kinship. Boston: Beacon Press.

Lewis, Robert A., and Graham B. Spanier 1982 "Marital Quality, Marital Stability and Social Exchange." In F. Ivan Nye, ed., Family Relationships: Rewards and Costs. Beverly Hills, Calif.: Sage.

Litwak, Eugene 1960a "Occupational Mobility and Extended Family Cohesion." American Sociological Review 25:9–21.

——1960b "Geographical Mobility and Extended Family Cohesion." American Sociological Review 25:385–394.

——1985 Helping the Elderly: The Complementary Roles of Informal Networks and Formal Systems. New York: Guilford Press.

Lopata, Helena Znaniecki 1973 Widowhood in an American City. Cambridge, Mass.: General Learning Press.

Macfarlane, Alan 1986 Marriage and Love in England: Modes of Reproduction 1300–1840 New York: Basil Blackwell.

Maine, Henry S. (1861) 1963 Ancient Law. Boston: Beacon Press.

Mitchell, William E. 1963 "Theoretical Problems in the Concept of the Kindred." American Anthropologist 65:343–354.

Mogey, John 1976 "Content of Relations with Relatives." In J. Caisenier, ed., The Family Life Cycle in European Societies. Paris: Mouton.

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

Naroll, Rauol 1970 "What Have We Learned from Cross-Cultural Surveys?" American Anthropologist 75:1227–1288.

Paige, Jeffery M. 1974 "Kinship and Polity in Stateless Societies." American Journal of Sociology 80:301–320.

Parsons, Talcott 1954 "The Kinship System of the Contemporary United States." In Talcott Parsons, ed., Essays in Sociological Theory. New York: Free Press.

Pehrson, R. N. 1957 The Bilateral Network of Social Relations in Konkama Lapp District. Bloomington: Indiana University Research Center in Anthropology, Folk-lore, and Linguistics.

Peranio, R. 1961 "Descent, Descent Line, and Descent Group in Cognatic Social Systems." In V. E. Garfield, ed., Proceedings of the Annual Meeting of the American Ethnological Association. Seattle: University of Washington Press.

Pina-Cabral, Joao de 1997 "Houses and Legends: Family as a Community of Practice in Urban Portugal." In Marianne Gullestad and Martine Segalen, eds., Family and Kinship in Europe. London: Pinter.

Redfield, Robert 1947 "The Folk Society." American Journal of Sociology 52:293–308.

Roschelle, Anne R. 1997 No More Kin: Exploring Race, Class, and Gender in Family Networks. Thousand Oaks, Calif.: Sage Publications.

Sennett, Richard 1970 Families Against the City: Middle Class Homes of Industrial Chicago, 1872–1890. Cambridge, Mass.: Harvard University Press.

Shanas, Ethel, Peter Townsend, Dorothy Wedderburn, Henning Friis, Paul Milhoj, and Jan Stehouwer 1968 Old People in Three Industrial Countries. New York: Atherton Press.

Sheehan, Michael M. 1963 The Will in Medieval England: From the Conversion of the Anglo-Saxons to the End of the Thirteenth Century. Toronto: Pontifical Institute of Medieval Studies.

Sorokin, Pitirim 1937 Social and Cultural Dynamics. Four Volumes. New York: Harper.

Stack, Carol B. 1974 All Our Kin. New York: Harper Colophon Books.

Steinmetz, Devora 1991 From Father to Son: Kinship, Conflict, and Continuity in Genesis. Louisville, KY: Westminster/John Knox Press.

Stone, Lawrence 1975 "Rise of the Nuclear Family in Early Modern England: The Patriarchal Stage." In Charles E. Rosenberg, ed., The Family in History. Philadelphia: University of Pennsylvania Press.

Strathern, Marilyn 1992 After Nature: English Kinship in the Late Twentieth Century. New York: Cambridge University Press.

Sussman, Marvin 1959 "The Isolated Nuclear Family: Fact or Fiction?" Social Problems 6:333–340.

Swanson, Guy E. 1969 Rules of Descents: Studies in the Sociology of Parentage. Anthropological Papers, no. 39. Ann Arbor: Museum of Anthropology, University of Michigan.

Tocqueville, Alexis de (1850) 1945 Democracy in America. New York: Knopf.

Toennies, Ferdinand (1887) 1957 Community and Society. East Lansing: Michigan State University Press.

Walster, Elaine, and G. William Walster 1978 A New Look at Love. Reading, Mass.: Addison-Wesley.

Watson, John 1927 Chicago Tribune. March 6, p. 1.

Weigert, Andrew J., and Ross Hastings 1977 "Identity Loss, Family, and Social Change." American Journal of Sociology 82:1171–1185.

Wirth, Louis 1956 Community Life and Social Policy. Chicago: University of Chicago Press.

Yerushalmi, Yosef Hayim 1982 Zakhor: Jewish History and Jewish Memory. Seattle: University of Washington Press.

Young, Michael, and Peter Willmott 1957 Family and Kinship in East London. London: Routledge and Kegan Paul.

Zborowski, Mark, and Elizabeth Herzog 1952 Life Is with People: The Culture of the Stetl. New York: Shocken Books.

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

——1966 "Theories of Frederic LePlay." In Bernard Farber, ed., Kinship and Family Organization. New York: Wiley.

Bernard Farber