Ann. Rev. Socio!. 1983. 9.125-49
Copyright © 1983 by Annual Reviews Inc. All rights reserved
GENERATION AS A SOCIOLOGICAL PROBLEM
David I. Kertzer
Department of Sociology & Anthropology, Bowdoin College, Brunswick, Maine 04011
INTRODUCTION
For many miliaria, in cultures around the world, the concept of generation has prospered. Its privileged place in Western societies is reflected in its codification in the Bible, while the most disparate societies of Africa, Asia and Australia have incorporated the generational concept in their notions of the social order. It is no surprise that the idea of generation should have come into prominent use in Western sociology, just as so many other terms have been preempted from popular to scientific vocabulary. It was perhaps also inevitable that this transformation from folk to analytical usage would occasion considerable conceptual confusion. The term's mulivocality, a virtue in popular discourse, became a liability in science.
In this review, I identify the sources of confusion in the sociological usage of "generation" as I examine the recent literature. Though others have identified many problems with the sociological usage of the generation concept, their strictures have thus far had only limited effect. The concept of generation is important to future sociological research, but progress can only be made if an acceptable definition of generation is employed and other usages are abandoned.
Here my focus is conceptual and methodological. I do not attempt a comprehensive review of substantive findings, though the bibliography should be useful to those interested in pursuing specific substantive interests. Along with the recent sociological literature, I examine closely related works by social psychologists, political scientists, and anthropologists. I first identify the diversity of meanings attributed to the generation concept and discuss the intellectual heritage bequeathed to contemporary sociology. I examine the literature of the past dozen years, identifying areas of concep-
125
126 KERTZER
tual confusion under a variety of general topics: the confusion of generation, cohort, and age; studies of intergenerational transmission of values; inter-generational mobility; and the use of the generation concept in studies of immigration. In the final section I identify some areas for future sociological research.
EARLIER USES OF GENERATION
Multiple Meanings
Social scientists have traditionally looked upon the diverse popular meanings of "generation" as an opportunity for extension of the term in social science, rather than as a source of imprecision to be avoided. Troll (1970), for example, lists five different concepts of generation, and finds them all useful. Altering her list slightly, I place these in four categories: generation as a principle of kinship descent; generation as cohort; generation as life stage; and generation as historical period. These meanings are all found in the sociological literature; indeed, many sociologists simultaneously use more than one.
In its sense of kinship descent, the concept of generation has a long tradition in social anthropology. Unlike sociologists, social anthropologists use it in referring not so much to parent-child relations as to the larger universe of kinship relations (Fox 1967; Baxter & Almagor 1978; Fortes 1974; Foner & Kertzer 1978; Jackson 1978; Kertzer 1978; Legesse 1973; Needham 1974; Stewart 1977). Demographers have utilized this sense of the term in attempting to develop measures for "length of generation." Here the interest is in population replacement, based on the reproduction of females (Preston 1978; Krishnamoorthy 1980).'
The use of "generation" to denote cohort is widespread. Demographers also had considerable influence in propagating this usage, with the term cohort only fully replacing this usage of generation among demographers in the past decade (Jacobson 1964). Here the "generation" refers to the succession of people moving through the age strata, the younger replacing the older as all age together. This usage is widespread beyond sociology as well and finds frequent expression in intellectual history, where, for example, "literary generations" may succeed one another each 10 or 15 years (Cowley 1978). This cohort notion of generation has been extended beyond that of birth cohorts to apply to any succession through time, so that we find reference to first, second, and third "generations" of health behavior studies (Weaver 1973; Farge 1977) or to marital "generations" (Hill 1977).
'For an extension of this notion, using mathematical techniques, to society-wide oscillations in value patterns, see Carlsson & Karlsson (1970). See also Levine (\9"11)
GENERATION 127
In its life-stage usage, we find such expressions as the "college generation." Sorokin's discussion of generation can best be understood in this sense, for he attributed the conflict between "younger and older generations" to the differential response of people of different ages to the same events (1947:192-93). Eisenstadt's (1956) classic study combined the descent and life-stage meanings of generation.
The use of "generation" to characterize the people living in a particular historical period is less common in sociology than in history, where books bearing such titles as The Generation of 1914 (Wohl 1979) and The Generation Before the Great War (Tannenbaum 1976) are numerous. In this sense, "generation" covers a wide range of cohorts. However, though it is the great historical event that defines such "generations," they are often linked in practice to the cohorts of youths and young adults thought to be particularly influenced by such events.
Various usages of the generation concept are commonly mixed together, sometimes intentionally. Laslett's (1977) Family Life and Illicit Love in Earlier Generations profits from the descent and period meanings of "generation." Moreover, the generation idea has great popular appeal (e.g. the "generation gap" concept of the 1960s); the term is thus used in many social scientific books written for the mass market (Cohen & Gans 1978; Franz-blau 1971; Jones 1980).
Mannheim and Ortega y Gasset2
While the roots of the current confusion in generational studies can be traced back millenia, the proximate antecedents can be identified in the works of Karl Mannheim and Jose Ortega y Gasset.
Mannhim's writing has heavily influenced sociological works on generation, and his own confounding of the genealogical meaning of "generation" with the cohort sense of the term continues to be reflected in current research.
Mannheim wrote that the "sociological phenomenon of generations is ultimately based on the biological rhythm of birth and death" (1952:290). Over time, a succession of waves of new individuals reach adulthood, coming at that time into contact with the prevailing culture and remodeling what they find. This is what Mannheim meant by "fresh contact." He identified these waves with generations but distinguished between those individuals within such generations who shared a common outlook on the basis of their common experience and those who did not. The former he labelled "generation units."
2A fuller discussion of Mannheim and Ortega y Gasset's influence on generational studies may be found in Kertzer (1982).
128 KERTZER
Ortega y Gasset formulated a similar concept of generation based on the notion that people born at about the same time grow up sharing an historical period that shapes their views. Arguing that generation "is the most important conception in history," Ortega y Gasset wrote that each generation has its "special mission," though this mission might be left "unachieved" (1933:15,19). Such followers of Ortega y Gasset as Marias (1968) have repudiated the kinship descent definition of generation, championing the historical cohort meaning alone. Once the concept was thus cut loose from its genealogical anchor, followers of Mannheim and Ortega y Gasset could claim that a new "generation" might appear as frequently as every year, depending on the rapidity of change new cohorts face as they come of age in their society (Rintala 1968; Berger 1959)
Conceptual Clarification
The polysemous usage of generation came under attack by Ryder (1965), who argued for restricting generation to its kinship descent meaning. There exists an unambiguous term—cohort—to refer to the succession of individuals who pass through a social system, and there exists a similarly clear term —life stage—to refer to a particular segment of the life course. Processes of family transmission should not be confused with processes of cohort succession and social change.
Ryder's argument found favor among demographers, who have for the most part embraced the cohort terminology, but it has not been heeded by many other social scientists (cf. Troll & Bengtson 1979). Ironically, many of the sociologists who employ "generation" in the sense of cohort cite Ryder's article as their authority. Only slowly, too, is the use of "generation effect" as a synonym for "cohort effect" dying out (Baltes 1968; Riley 1973, 1976). Riley et al (1972:5), in one of the most influential works on age in sociology, reiterate Ryder's plea that generation be restricted to its kinship reference. Generation, then, is a relational concept bound to the realm of kinship and descent; it is not an appropriate tool for dividing societies into segments or populations into aggregates.
GENERATION, COHORT, AND AGE
Generations as Cohorts
The continued use of "generation" to refer to cohort effects is apparent in much of the literature (Markides 1978). Faver (1981), for example, tests the hypothesis of "generational and life-cycle effects on women's career and family values," using a sample of women aged 22 to 64. By a "generational effect" Faver means a cohort effect, an effect exerted upon people by life experiences attributable to the historical slice of time in which they have
GENERATION 129
lived. Faver does not link this effect with "generation" in the genealogical sense of the term.
Many political studies have made similar use of "generation." Lipset & Ladd (1971) compared college cohorts of the 1930s with those of the late 1960s to determine if there were cohort differences in attraction to left-wing politics. They refer to these cohorts as "generations" of college students. Similarly, Claggett (1980) writes of the "generational model" of partisan allegiance, suggesting that the "cross-sectional variation of partisan strength is a function of the saliency or relevance of partisan politics at the time when individuals are young and malleable." As he describes it, "a generational effect explanation ascribes the cross-sectional relationship to birth cohorts entering the electorate with different values on the variable, which thereafter do not change as the cohorts age." Tsukashima & Montero (1976), writing of the difference in anti-Semitism between younger and older American blacks, employ "generation" in comparing these age groups. In seeking to distinguish between-cohort effects and aging (or life-course) effects, they write of a "generational" effect as opposed to a "maturational" effect. The three "generations" in question consist of those 20-29, those 30-49, and those 50 and above. A variation on this theme is provided by a recent exchange in the American Political Science Review involving the "generational replacement hypothesis" (Born 1979; Alford & Hibbing 1981). Here it is argued that the "generation" of American Congressmen who were elected for the first time in 1966 had characteristics different from the "generation" first elected before 1966.
Were the term "generation" used simply as a popular synonym for cohort the matter would not be of great importance. The problem is that when authors use the term in this sense they often retain the notion of genealogical relationships (Rosaldo 1980). In this way independent variables are confounded. For example, in a study of the cohort of Parisians who have experienced retirement in the past few years, Cribier (1981) begins by contrasting the experience of the present "generation" with that of the previous "generation," discussing differences between recent retirees and their parents. Yet he then shifts to a cohort meaning of "generation," writing of one "generation" born between 1918 and 1927 and "the older generation born between 1907 and 1911." Similarly, in a study of American families, Masnick & Bane (1980) distinguish three "very different generations" of adults, those born by 1920, those born between 1920 and 1940, and those born since 1940. In the sense of cohorts this makes sense, but it soon becomes clear that Masnick & Bane are employing "generation" in its genealogical sense as well, comparing "the family patterns of today's young adults with that of their parents' generation." This is not acceptable, as a large proportion of those born in the 1940s, for example, were born to
130 KERTZER
parents born before 1920. No justification for a length of genealogical generation of 20 years is given, and the two meanings of generation are confused.
Faced with the common double usage of "generation," the reader may not be certain what is denoted in generational discussions. Rosow's (1978) perceptive discussion of how best to define cohorts is partially marred by this problem. He writes that "Cohort effects are a central concern in the analysis of generations. By cohort effects, I mean the typical response patterns of members of various cohorts to the same thing. Those in one generation react the same way, but differently from members of another. So when responses to the same phenomenon are similar within, but different between generations, this is a cohort effect" (1978:72). It is not at all clear just what "generations" are in this usage (see below). A comparable problem of dual meaning is found among historians who write of "generations" in terms of particular cohorts and at the same time refer to particular historical periods as "generations" (Butterfield 1972).
The perils of confounding the genealogical and the cohort meanings of "generation" have already been persuasively articulated by Vinovskis and Elder, following from Ryder's earlier strictures. Vinovskis (1977) faults the well-known study by Greven (1970) on four generations of residents of colonial Andover, Massachusetts, and by Hill and associates (1970) on three generations of family life in Minnesota on the grounds that they did not properly distinguish between genealogical generation and birth cohorts. Elder has pursued the same point in critiquing these studies, as well as in examining the work of Bengtson & Lovejoy (Elder 1975, 1978b; Bengtson & Lovejoy 1973). The major problem Elder and Vinovskis cite, in brief, is that when a population is divided on genealogical principles into various generations, there is substantial overlapping in age among the various generations. To the extent that this is true, it is impossible to properly characterize the generations in terms of their common characteristics vis-a-vis other generations. It is, in short, inappropriate to refer to them in cohort terms; members of the same "generation" have lived through different historical periods.
Generation as Life Stage
In sociological studies involving age, the distinction between age, cohort, and period effects is now well known and guides the methodology of most research. We cannot attribute differences between people of different ages to their life-course position without determining whether these differences stem instead from cohort characteristics. Riley (1973) and others have pointed out the methodological fallacies commonly found in the social scientific literature in this regard, and the development of the sociological study of age in recent years has been anchored in these distinctions. The
GENERATION 131
generational concept is anachronistic in its polysemous usage by encouraging fuzzy thinking about these distinctions. By referring to a variety of diverse processes, from kinship descent, to cohort, to life stage, to period, it discourages the kind of analysis that is so necessary to research on age.
This problem is particularly evident when we review studies that employ generation to characterize age strata or particular stages in the life course. Based largely on cross-sectional data, and written in terms of "generational differences" and "relations among generations," these studies are generally not in a position to distinguish between age and cohort effects and, more disturbingly, often pay scant attention to this problem.
Adamski (1980), in a study of Polish workers, compared the values of the younger "generation" with those of the older "generation," distinguishing these simply by age. In finding that there were "significant differences between the generations," the author offers us no means of knowing whether to attribute these differences to life-course effects or to permanent cohort characteristics. Similar problems plague a number of studies involving political generations. Braungart (1974), who has hailed the birth of the "sociology of generations," tells us that "The fastest growing generation during the 1900-1975 period was the age group 65 and over." He clearly has in mind age groups, rather than either descent-defined units or cohorts. Wheeler (1974), in a study of German labor radicalism in the earlier part of this century, asks whether political differences among the union leaders can be attributed to "generational differences." He concludes that there was in fact a strong correlation between youth and radicalism, taking this as evidence for the importance of generational differences. Cohort and age effects are not distinguished.
In demographic studies similar identification of age groups with generations is sometimes found. Treas (1981) devotes a recent study to the "generational balance" in the United States, and its relationship to support for the elderly. By "generational balance" she means the relative sizes of the population over age 65 and the working-aged population of ages 18-64, though she also refers to intrafamilial generational relations. While Treas deals intelligently with this issue, the nongenealogical use of "generation" would best have been avoided. In social-psychological studies, too, this usage of "generation" continues to proliferate. In a study of "generational differences" in work orientations, Taveggia & Ross (1978) contrast factory workers whom they divide into four age groups (defined in relation to a 1974 survey): under 26 year old; 26-35 years; 36-45 years; and over 45 years. They find the notion of "generation gaps" inapplicable, for relatively little difference in work orientation is evident among the age groups. Of course, had they found such differences they would have been hard put to attribute them to cohort, as opposed to life-course, factors.
Studies of college students remain popular among social psychologists,
132 KERTZER
who have offered various "intergenerational" studies on the basis of such work. Fitzgerald (1978) compares college freshmen with a sample of unrelated old folks to identify differences in perceptions of old people by the young and old, and by males and females. He concludes that evidence that the college students "inaccurately perceived the older adults as dominant, competitive, and very aggressive" highlights "a basis for conflict between the generations...." A similar study was conducted by Crockett & Press (1981), limited, however, to college students who were shown photographs of elderly individuals.
The mixture of the life-stage meaning of generation and the descent-relationship meaning is apparent. We are left not only with the question of whether these differences have any cohort basis but also with the unjustified impression that aggregate differences between people at different stages of the life course can be directly translated into conclusions about relationships between parents and their children. Similar problems are raised in Collette-Pratt's (1976) study using a semantic differential measuring technique to investigate the devaluation of old age. Here the "multi-generational sample" consists of three non-kin-related age groups: the young adults (aged 18-29), the middle-aged adults (30-59), and the older adults (60 and over). Given this usage of "generation," it is difficult to know just what could be meant by such measures as number of "intergenerational contacts" (relations between a 29 and 30 year old?). Again, introducing generational terminology in describing age groupings adds nothing but confusion.
Mannheim's Legacy
The continued influence of Mannheim and to a lesser extent Ortega y Gasset in contemporary sociological usage of the generation concept is clear. Indeed, Jansen (1975) has suggested that a "Generation Theory" (meaning by this a fuller development of Ortega y Gasset's ideas) is fundamental to sociology. Mannheim's use of the generational unit has been embraced in a variety of studies, many making use of his notion of the "fresh contact" each such unit makes with the society at the time its members reach maturity (Back 1976; Balswick 1974; Lambert 1972; Wuthnow 1976; Krie-gel 1978). The Mannheimian perspective is particularly popular in studies of youth and youth movements, though the imprecision of the generation concept in this camp has not been without its critics (Smith 1981).
Perhaps the most prominent follower of Mannheim in contemporary sociology has been Vern Bengtson, who has published numerous papers in conjunction with a variety of colleagues in this sphere. The confusion generated by Mannheim's use of the generation concept is evident in these works. At times in writing of generations, Bengtson refers to attitudes held by grandparents, parents, and their children, as in his interesting discussion
GENERATION 133
of the perceptions one generation has of another (Bengtson & Kuypers 1971). At other times such genealogical generations are identified and then transformed into age groupings for analysis of "generations" that need have no connection to kinship descent at all (Bengtson & Lovejoy 1973; Bengtson 1975). In some papers "generation" is used in three distinct senses, ranging from discussions of eight-year birth cohorts, to age groups ("current youth and current adult generations"), to Mannheimian generation units ("the beat generation") (Cutler & Bengtson 1974). With "generation" used in so many different, analytically incompatible senses, what is being denoted is sometimes unclear, as when Bengtson & Troll (1978) cite "societal generational processes" in discussing the thesis that parent-child similarities may be the result of common environment rather than within-family socialization.
Through the 1970s Bengtson has attempted to marry the Mannheimian notion of generation units as agents of social change with the developing field of age stratification (Riley et al 1972; Elder 1975; Riley 1976). At times, the sociology of age has been identified with generational analysis (Bengtson et al 1974; Bengtson & Cutler 1976). However, the effort has involved insertion of an additional variable—that of generation—in addition to the familiar concepts of cohort, life course, age stratum, and aging. "Generational analysis" is distinguished from these concepts as being "concerned with age groups as agents of social change," with such "generation units" being composed of self-consciously active age-based groups (Laufer & Bengtson 1974; Bengtson & Starr 1975). Yet at the same time "generation" is used in the broader sense of cohort, contrasting within-family socialization processes that lead to family homogeneity with "between generation similarity at a broader societal level" (Bengtson 1975).
Another notable champion of generation as a key sociological principle, following in the Mannheimian tradition, has been Marshall, in collaboration with Tindale. In an effort heralded as an advance beyond the age-stratification approach of Riley and associates, they too have called for renewed focus on generation, in the form of a "generational conflict theory of aging" (Tindale & Marshall 1980). The claim is made that a cohort is "a statistical artifact" while "a generation is a sociological reality, consisting of a cohort, significant proportions of whose members have experienced profound historical events ..." (Marshall 1980a). The "reality" of generations has to be questioned, however, when it is recognized that history does not present itself as a simple series of chronologically distinct slices but rather as a variety of overlapping forces. Moreover, the core of Tindale & Marshall's thesis, that endemic conflict among age strata in society is due to differential access to wealth and power, is well accommodated in the age-stratification model (Riley et al 1972; Foner 1974, 1975; Foner &
134 KERTZER
Kertzer 1979). Introducing the concept of generation, in the absence of a specific argument linking age strata to parent-child relations, simply adds confusion. Indeed, when Marshall writes of the potential for "intergenera-tional tension" linked to the relative proportion of retired individuals and workers in society, and when he speaks of the costs of old age assistance as being "borne by the younger generation," he is not systematically using his Mannheimian definition of generation but is returning to the multiple signification that has bedevilled so many generational theorists. In writing of "intergenerational reallocation of resources ... as a feature of political economy," Marshall risks confusing two important, yet distinct, properties of economies—the question of transfer of resources from parents to children, and that of the continuous flow of resources among age strata in a society (Marshall 1980b). The analysis of each of these is important in sociological study, but they can and must be distinguished.
Of course since the 1960s much has been written about "generational conflict" (earlier often expressed as the "generation gap"), and most of this has been informed by Mannheimian principles as well. Many have profited by the dual meaning of "generation"—as parent-child relations and as life stage—to move freely from discussion of the one to discussion of the other. For example, Laufer (1971) writes of the "present generation of middle-class youth" and of the "parental generation," while dividing the population into "the younger and older generations." But most of the generational conflict literature, while making reference to parent-child relations, addresses the issue of the relations between people found in different age strata. Thus Shimbori (1971) analyzes student radicals in Japan in terms of intergenerational conflict, based on a Mannheimian definition of generation. Feuer (1969) employs a similar usage in his global study of youth movements. Generational conflict in this sense is also examined in relation to the pace of social change by Abrams (1970) and Brent (1978).
Generational conflict terminology has been employed in studies of successive cohorts entering certain fields of endeavor, relating subsequent changes in those fields to the particular characteristics of succeeding cohorts. Thus, LeVine (1968) wrote of "generational conflict and politics in Africa," distinguishing various "generations" of political leaders. Feuer (1974) extended his earlier Mannheimian approach to youth movements and generational conflict to the study of successive "generations" of scientists.
GENERATION AS DESCENT
So far I have focused on the several extensions of the genealogical generation concept that have flourished in the sociological literature, noting the ambiguities and imprecisions that have resulted. No review of the use of the
GENERATION 135
generation concept would be complete, however, without consideration of how sociologists have employed generation in its genealogical, or descent sense. Some of the same problems discussed in the previous section afflict these studies as well, for application of a kinship-relational principle to population-wide groupings is fraught with analytical dangers.
Applications of "generation" denoting descent include studies of value transmission, studies of social mobility, and studies of immigration.
Socialization and Values
PARENTS AND CHILDREN There is a long sociological tradition of studying parent-child value transmission, employing the concept of inter-generational continuity. The research design of such studies calls for selection of a sample of two-generation dyads, often varying by sex. Some of these studies include three generations, inquiring into continuities and discontinuities between grandparent and parent, as well as between parent and child.
Aldous & Hill (1965) attempted just such an inquiry by locating 88 three-generation triads of married couples living within 100 miles of Minneapolis. However, the representativeness of people who not only are members of three generations of living married couples but also remain geographically localized, remains a serious question in interpreting the results. Kalish & Johnson (1972) began with a sample of 53 women aged 14-29 and located their mothers and mothers' mothers in an attempt to identify generational differences in values. They found such differences to be modest and to characterize both sets of adjacent generations. However, they assume that these differences have a cohort-historical basis and do not test the possibility of a life-course, or aging basis. Simon & Gurevitch (1971) studied 30 father-son dyads of ultra-Orthodox Jews and 30 such dyads of urban Arabs, both in Israel, concluding that value congruity prevailed among the Jews while considerable tension existed between Arab fathers and sons. Skvoretz & Kheoruenromne (1979) inquired into the extent to which parents transmit the values of their social class to their children in a study of fifth grade students and their parents in Columbia, South Carolina. Their results were inconclusive. And in a study of women from Pennsylvania, Johnson & Stokes (1976) found that the number of siblings each woman had influenced the number of children she bore, with this relationship being strongest among women who were themselves eldest daughters.
In addition to studying values, parent-child studies have also inquired into social relations between the generations as the child proceeds through adulthood. Wilkening et al (1972) relate the geographical proximity of adult children to their parents in Wisconsin farm families to the extent of inter-
136 KERTZER
generational contact, and trace changes in distance and contact to aging of the two generations. However, they are unable to determine the nature of the cause-effect relationship between distance and contact. Leonard (1980), following a tradition of British family sociology best known from the work of Young & Willmott (1957), found continued close ties between children and their parents in Wales even after children marry. The importance of studying just how these intergenerational relations change as both parents and children age together has recently been stressed by Hess & Waring (1978).
Many of the intergenerational studies of value transmission in fact do not directly link parents and their children at all. Rather, taking data from the "parental generation" and the generation of their children, they compare the characteristics of the two groups. Such studies rest on shaky methodological premises, for they tell us nothing at all about value transmission from parent to child. Moreover, since generations thus denned tend to include wide and often overlapping age ranges, the data may not even be usable for cohort analysis.
Curiously, though many of these studies are based on samples composed of a particular age group of youths (most often college students) and their parents (and sometimes grandparents), their analysis is based on treating each "generation" as an aggregate (Penn 1977). Wake & Sporakowski (1972) selected students aged 14—23 and their parents, aged 30-62, comparing the attitudes of each "generation" toward support of aged parents. The result tells us nothing about intrafamilial value transmission, yet as a study of age differences in attitudes toward the elderly, the age groupings it uses are overly broad. Keeley (1976) administered questionnaires to college students and their parents, but analyzes the two "generations" as aggregates and, as is true with almost all such studies, provides no controls for differentiating between age and cohort effects. His ciaim that "the corpus of any culture must be relearned at least three times each century" confuses what is an intrafamilial socialization process with societal processes of cohort succession that are continuous, as Ryder clearly pointed out years ago.
Payne et al (1973) gathered data from college students, their parents, and their grandparents to determine value differences between generations. However, they too use "generational" groupings rather than linked triads in their analysis. The weakness of this approach is underlined by the fact that these generations are not from distinct age strata, with parents ranging up to 69 years of age, and those of the "grandparental generation" being as young as 58. In a similar three-generation study, Antonucci et al (1979) compare all-female triads (daughter, mother, grandmother) with all-male triads. They too analyze their data in terms of generational aggregates. They explain neither the narrow age range of the parental generation (39-50) nor
GENERATION 137
the seemingly impossibly high mean age (47.9) within this range. Thomas (1974) similarly sampled college students and their parents, treating the two in aggregate terms and attributing differences to cohort effects. He contends that "If one is to argue for the existence of a generation gap, it is necessary that there be more agreement within the youth cohort than between the younger and older generations." But this is not true if the genealogical notion of generation is to be considered, and the generation gap is used to refer to strained relations between parents and children. Every child in the sample could have values sharply divergent from those of his or her parents without there being any aggregate difference in values between the two "generations."
Some studies of intergenerational value relations in fact make no attempt to sample from descent-related dyads or triads at all. In a study of "congru-ency of belief across generations," Nelsen (1981) uses survey data from 1944, 1960, and 1973-8, dividing the population into fifteen-year birth cohorts. Pines & Kafry (1981) conclude that "tedium decreases with generation" among professional women, basing their study on three "generations" of women: preprofessional college students, professional women (mean age = 34), and retired professional women (mean age = 66). The confounding of cohort and life-course effects is complete. In an unusual approach, Thurnher et al (1974) selected 52 high school seniors and 54 parents of high school seniors from that school who were not parents of the students in their sample. They conclude that the generation gap is a myth, but again no data on parent-child "gaps" are presented. Finally, Markides et al (1981) have been studying older Mexican Americans in San Antonio who have adult children living in the area, to inquire into "perceptions of intergenerational relations." The study so far is limited to the perceptions of the older people.
POLITICAL VALUE TRANSMISSION Throughout the 1970s a number of political scientists and political sociologists heralded the generation concept as providing new insight into the classic problem of how political values are transmitted and how change in political attitudes takes place over time. Here again, some studies made use of actual parent-child attitudinal data, while others did not. The familiar confusion of the descent meaning and the cohort meaning of "generation" is commonly found.
In a discussion presaging in some ways Rosow's (1978) critique, Klecka (1971) complained that political sicentists had "concentrated too much on 'cohorts' and not enough on 'generations.'" By this he meant that the population should not be divided into equal-interval birth cohorts for analysis; rather, cohorts should be defined on the basis of their common exposure to historical events during their period of most intense political socialization. However, rather than define such intervals and test to see whether in
138 KERTZER
fact these historical events had led to meaningful cohort differences, Klecka took survey data from 1952-68 and applied an analysis of variance to divide the population into homogeneous cohort groupings based on their survey responses. The resulting 15 "generations" range from 3 to 8 years in length and show little differentiation, having all the appearance of being arbitrarily drawn.
In another often cited political generation study, Connell (1972) surveyed the evidence and distinguished between two kinds of generational comparisons: pair and group. Pair correspondence refers to homogeneity in political attitudes between parents and their children, while group correspondence refers to such homogeneity when the generation of parents and that of children are compared as aggregates. Connell concluded that in recent American history there has been low pair correspondence but high group correspondence between the generations, suggesting that the "older and younger generations have developed their opinions in parallel" rather than as a result of direct parental transmission. Friedman et al (1972) lend empirical support to Connell's view in a study of Columbia University students and their parents, finding "relatively small family effect upon political values." On the other hand, Abramson (1972), in a study of British voters, accounts for one third of the voters who were not supporters of the party of their social class by showing that they were socioeconomically mobile individuals who continued to support the party of the social class of their fathers (cf. Knoke 1973).
Even those scholars who are most conscious of the analytical distinction between the descent and cohort meanings of generation have been prone to confuse the two, or at least to be unclear about what meaning is intended at any particular time. Jennings (1976) has cogently pointed out the fallacy of equating these two meanings of generation, yet in their panel study of "two age generations"—originally high school seniors and their parents— Jennings & Niemi (1975) deal simultaneously with both meanings. They conclude that through the life course there is a "smoothing out of intergen-erational antagonisms." This is clearly a conclusion based on aggregated data on the parent and child groups. It could be interpreted in cohort terms, but then the method of data collection would be faulty, since the technique does not sample on the basis of cohort. It could also be interpreted in terms of parent-child relations over the life course, but the analysis is based only on a comparison of "the aggregate changes in each generation on some of the political orientations first tapped in 1965." A similar difficulty is encountered in the work of Cutler, another political scientist well versed in the sociological literature. In moving "toward a generational conception of political socialization," Cutler (1975) calls for cohort analysis but simultaneously uses such concepts as "parent and child generations "
GENERATION 139
Social Mobility
One of the most frequent uses of generation in contemporary sociology is in studies of social or occupational mobility. Indeed, generation is a key concept in social stratification research, referring to the process of status and occupational transmission from parent to child (Sorokin 1947). Researchers are interested in the nature and extent of such transmission, how this process differs among different segments of the same society and between societies, as well as how the transmission process itself changes over time.
While a substantive review of the mobility literature is well beyond the scope of this chapter (see Matras 1980; Featherman 1981), some of the implications of the use of the generation concept in mobility studies should be mentioned here. Of particular interest is the relationship between generational groupings, birth cohorts, and historical change.
Duncan (1966a, 1966b) has already cogently expressed some of the potential pitfalls in the use of the generation concept in this sphere, cautioning against attempts to base analyses of changes in social stratification processes over time on surveys that collect data on the occupation of men currently in the labor force and the occupation of their fathers. The fathers of these men are not representative of any population in the past, owing both to differential fertility (some produced more sons than others and are thus overrepresented, while others produced no sons and are thus not represented at all) and to variability in parental age. As Duncan (1966b) concludes, "The transformations (in the labor force) that occur via a succession of cohorts cannot, for basic demographic reasons, be equated to the product of a procession of 'generations'." Thus it is important to distinguish between (a) collecting data on the lifetime mobility of individuals in one's sample and (b) addressing the matter of how changes in the stratification system itself take place historically.
One of the problems mentioned by Duncan is that there is no neat correspondence between generational differences and age differences. Even where a limited birth cohort is selected, variability in parental age may be great and the span of grandparental ages would overlap with the parents' ages. Thus if our interest is in how processes of occupational transmission change over time, we could not take as units "generational" groupings selected in this way, particularly in characterizing grandparent-to-parent transmission. This is a problem that afflicts Goyder & Curtis' (1975) three-generational US study, as well as Gamier & Hazelngg's (1974) study of Frenchmen born in the period 1918-41 and their fathers. Such data have certain descriptive usages, but they cannot be used to study the historically changing stratification system.
140 KERTZER
To see how intergenerational transmission processes change over time it is necessary to specify limited historical periods in which the transmission took place. For example, lumping together men aged 20-64, thereby aggregating people who entered the labor force in widely differing historical periods (Erikson et al 1979), makes it impossible to inquire into the changing historical context of mobility processes. It is important, then, to pay careful attention to age groupings, relating specific historical periods to specific patterns of intergenerational transmission. In this way, historical changes in the mobility experience of successive cohorts can be determined and their causes assessed (Featherman & Hauser 1978).
Following this approach, many scholars have found it preferable to conceive of mobility in //tfragenerational terms, making use of limited birth cohorts (Sorensen 1975). This is in line with Duncan's view that father's occupation may more profitably be viewed as an origin occupation point in the life of the son rather than as a base for computing intergenerational changes (see also Duncan et al 1972; Sewell & Hauser 1972; Featherman & Carter 1976; Featherman 1980; Kessin 1971).
There is another problematical aspect to the relationship of generation to age in mobility studies, though, for even where we focus on a limited birth cohort—in the most extreme case limited to a single year (Britten 1981)— the parents of the sample members will be from different birth cohorts. If we ask the occupation of the father when the son was aged 15, we are asking the occupation of some fathers at age 35 and for other fathers at age 60. We are left with the question of life-course trajectories and how these may relate to occupational transmission from father to son, including occupational aspirations (Mortimer 1974, 1981; Spenner 1981).
In recent years there has been growing interest in women's occupational mobility, largely ignored in earlier studies. Mobility has been studied by comparing women's adult occupations with those of their own mothers, their fathers, and their husbands. DeJong et al (1971) found father-daughter occupational mobility to follow the same pattern as father-son mobility. They lumped together all females in their sample who were over 21 years old and had ever been in the labor force, so the historical processes in question are concealed rather than revealed by the study. Glenn et al (1974) compare women's fathers' occupations with their husbands' occupations, arguing against the common belief that women experience greater upward mobility through marriage then men do through "occupational attainment." Chase (1975), however, finds that women experience greater upward and downward status mobility, in terms of the difference between their fathers' and their husbands' occupations, than do men, who are more likely to have occupations similar to their fathers.' Chase's sample encompassed couples whose wife's age ranged from 22 to 61, so again we are left with
GENERATION 141
the issue of whether the historical processes relating women's origins to their marital destination were unchanged over a four-decade period, as this method implies. Finally, Rosenfeld (1978) finds that the work history of the woman's mother is more predictive of her occupational career than the occupation of her father. Here the sample includes women aged 30-44, further divided by the analysis into five-year age groups. Differences between mothers' and daughters' ages, however, are not considered.
Immigration
Certainly one of the workhorses of sociological studies of immigration is the concept of first, second, and third generations. If the primary theme of the American sociological literature has been immigrant assimilation, the primary way this has been analyzed has been in terms of progressive loss of cultural and social distinctiveness from first to second and from second to third generation (Nahirny & Fishman 1965). While these distinctions reflect obviously important aspects of migrant study, this usage of the generation concept is much more problematic than most scholars have acknowledged.
Five primary analytical problems can be distinguished in this usage: (a) Unless the migration was restricted to a brief period, people sharing the same generational location in fact belong to different historical periods, confronting different historical conditions at their arrival and coming from a society that itself was different from the one earlier migrants had left, {b) Parents often migrate with their children, and occasionally three-generation families migrate together. In such cases we are either left with the anomalous case of children and their parents belonging to the same "generation" or with some immigrants of a certain age being considered first-generation (having no parents accompanying them) and others of the same age being considered second-generation (arriving with their parents), (c) Related to this point, immigrants range in age from infancy to oc-tagenarians. Does it make sense to lump these together as members of the same generation? The cultural imprint of foreign birth on the 80-year-old is entirely different from the imprint on the infant, (d) What of the not infrequent cases in which the migrant, resident for some time in the new country, is joined by his or her parents? Does the younger migrant shift from first- to second-generation migrant, or are both parents and children members of the same generation? (e) Finally, as we move beyond the first generation, all these initial problems are magnified because marriages are not necessarily "generation" homogeneous or ethnically endogamous.
Some of these matters can be illustrated by a brief look at some of the recent generational immigration literature. Lieberson (1973) extends the concept of migrant generation to the study of northern blacks in the United States, differentiating between first-generation (those born in the South) and
142 KERTZER
second-generation (Northern-born). He urges that scholars consider generation in this sense as well as age in studying the characteristics of American blacks. However, the use of generational terminology for this birthplace distinction risks creating confusion when what is of interest is the transmission of occupation from father to son. In such cases it may be unclear what is meant by "generational differences in occupational status" (Hogan & Featherman 1977 103). It would be more prudent to simply use such labels as "Southern-born,' enabling the anlayst to relate this dichotomy to processes of generational transmission
Moore (1981) describes "second generation New York Jews" as a "cultural generation, not a chronological one." This is an admission of the fact that the mass of Jewish immigrants arrived over a 33-year period (1881-1914). While she contends that the second-generation members—presumably born over a period ranging from 1881 to the mid-1930s—occupied "a similar point in space," the failure to differentiate cohorts may leave the reader skeptical. In a study of Puerto Rican migrants, Rogler et al (1980) recognize the pitfall of lumping together as first generation individuals who arrived in the United States as infants and those who arrived as adults. They conclude that analysts must recognize the "fundamental importance of the age at arrival variable. " To this must be added the importance of distinguishing the period at arrival (Scourby 1980)
Special mention must be made of the numerous studies of Japanese Americans, for here we are dealing with a conception of generation that has great meaning for the immigrants themselves, the division into Issei (immigrants), Nisei (their children), and Sansei (children of the Nisei). The clarity of this distinction is facilitated by the fact that immigration, and particularly female immigration, initially occurred over a short period (for women, 1907-24) Various attempts have been made to characterize the nature of assimilation across these generations (Connor 1974; Kiefer 1974; Montero &Tsukashima 1977, Montero 1981; Woodrum 1981). However, there is age overlap among the generations thus defined, and the importance of generation identity versus cohort homogeneity has not yet been fully clarified.
THE FUTURE OF GENERATION IN SOCIOLOGICAL RESEARCH
While some progress has been made in introducing precision into the use of the generation concept in sociology over the past decade, the term continues to be employed in a polysemous manner guaranteed to sow confusion. The confounding of generation as a principle of descent relationship with concepts related to age and historical time has resulted in studies that are methodologically flawed and has impeded the development of the
GENERATION 143
study of generation within the sociology of age. The relationship between generation (as we have defined it), aging, and cohort processes is of great importance. But this relationship can only be addressed productively once the confusion of concepts is resolved.
As Elder (1978a, 1978b; Bennett & Elder 1979) has cogently argued, what studies we do have of intergenerational relations are "strikingly ahis-torical." If relations between generations are to be seen as historically changing, and these changes are linked to larger events taking place in the society, it is crucial that analysis isolate how these historical events actually affect people. Insofar as age, or life stage, is presumed to be an important variable, the absolute and relative ages (Hammel 1983) of the individuals affected by the historical context must also be considered. Elder has proposed the use of "generational cohorts" as a device for introducing greater precision into generational studies. In this method, individuals of a certain limited birth cohort are taken as the focus, with these individuals passing through their various life stages in a similar historical environment. Their parents, rather than being treated as a single older generation, and hence mixing together people from different cohorts, are grouped by year of birth into "generational cohorts." This allows us to look at relationships between parents and children that have both an identifiable historical locus and specifiable age characteristics. However, in using generation to refer to a superfamilial grouping, problems are created. Among other concerns is the fact that where mother and father are of different ages we are faced with the analytically uncomfortable choice of either placing them in different generational groupings or disregarding the age of one parent.
The scope of future generational studies may be somewhat restricted by limiting the concept of generation to relations of kinship descent. But such restrictions do not entail any limitation of substantive or theoretical inquiry; rather, they entail a more precise use of concepts.
Where does the future of generational research lie? Generational processes will remain of great importance to sociology, for they are at the heart of the social metabolism. Studies of social change, value transmission, social mobility, and the cultural and social integration of immigrants must all cope with generational relations. What is crucial to the future of such study, though, is that the generational processes be firmly placed in specific historical contexts—i.e. that they be analyzed in conjunction with the concepts of cohort, age, and historical period.
Examining generation in conjunction with age opens up a research agenda that may be obscured where age, cohort, and generation are used interchangeably. The issues likely to be of greatest interest depend on the theoretical orientation of the researcher. From a sociobiological viewpoint, generational relations are central to society, for they underlie the transmission of genes, a process thought to affect the structure of society itself. No
144 KERTZER
sociobiological calculus would be complete without the inclusion of variables depicting both generational and age relationships among individuals, for the reproductive "value" of any individual is a function of both kin relatedness and age. From a Marxian viewpoint, of course, the mechanism of transmission of class position is of great interest, as are changes in such processes in conjunction with changes at the societal level. Such transmission processes must be studied historically, in a way sensitive to how mac-rosocietal changes affect the transmission of status from parent to child.
In the traditional American sociological perspectives connected with social organization and social psychology, the liberation of generation from its age-bound connotations offers the opportunity to place parent-child relations in a more comprehensive perspective, one linked to the developing sociology of age. It may seem paradoxical that generation can profitably be studied in the sociology of age only after it has been divested of its age connotations. Yet while generation must be viewed as conceptually distinct from age, generational processes and age relations are intricately related. Relations between parents and their children, for example, change as both parents and children age. The study of these relations and their changes— both over the life course and through historical time—is just beginning to be undertaken systematically (Rossi 1980; Hagestad 1981). With the increase in longevity (Troll 1980; Plath 1980) and with the new complexities introduced in American society by increased divorce and remarriage (Fur-stenberg 1981; Fox & Inazu 1982), the need for clear analysis of generation in studies of American society is all the more pressing.
I advocate a role for the concept of generation more restricted than that championed by many other social scientists, but a role that is nonetheless important. Not all adult-youth conflict can be portrayed in terms of generational struggle, but this restriction permits us to distinguish between those aspects of such conflict that are related to parent-child relations and those that are linked to larger societal processes tied to the age stratification system. Similarly, we cannot broadly speak of intergenerational value continuity but must distinguish between the transmission of values from parent to child and the society-level processes of social reproduction. With the opening up of new research issues connected to the life course and age stratification perspective, none of these old research questions need be abandoned; rather, adoption of the analytical distinctions herein suggested should prove helpful in building on past work in these areas.
Acknowledgments
For their helpful comments on an earlier draft of this review I am most grateful to Glen Elder, David Featherman, Anne Foner, Craig McEwen,
GENERATION 145
and Matilda White Riley. I would also like to thank Aaron Weissman and the other members of the Bowdoin College library staff for their aid in tracking down needed materials.
Literature Cited
Abrams, P. 1970. Rites de passage: The conflict of generations in industrial society / Contemp. Hist. 5:1:175-90
Abramson, P. R. 1972. Intergenerational social mobility and partisan choice Am. Pol. Sei. Rev. 66:1291-94
Adamski, W. W. 1980 The life orientations of the younger and older generations of Poles. Int. J. Polit. Educ. 3.255-69
Aldous, J., Hill, R. 1965. Social cohesion, lineage type, and intergenerational transmission. Soc. Forces 43:471-82
Alford, J. R., Hibbing, J. R. 1981. Comment on Born. Am. Polit. Sei. Rev. 75 159-62
Antonucci, T, Gillett, N., Hoyer, F. W. 1979 Values and self-esteem in three generations of men and women /. Gerontol. 34:3:415-22
Back, K. W. 1976. Personal characteristics and social behavior: Theory and method. In Handbook of Aging and the Social Sciences, ed. H. Binstock, E. Shanas, pp 403-31. NY: Van Nos-trand. 684 pp.
Balswick, J. 1974. The Jesus people movement: A generational interpretation. / Soc. Iss. 30:23^2
Baltes, P. B. 1968. Longitudinal and cross-sectional sequences in the study of age and generation effects Hum Devel. 11.145-71
Baxter, P. T. W , Almagor, U. 1978 Introduction In Age, Generation and Time, ed. P. T. W. Baxter, U. Almagor, pp 1-35. NY: St. Martin's Press. 276 pp.
Bengtson, V. L. 1975. Generation and family effects in value socialization Am. Sociol. Rev. 40:358-71
Bengtson, V. L., Cutler, N. E 1976. Generations and intergenerational relations-Perspectives on age groups and social change. See Back 1976, pp. 130-59
Bengtson, V. L., Furlong, M. J., Laufer, R. S. 1974 Time, aging, and the continuity of social structure: Themes and issues in generational analysis. J. Soc. Iss 30 2.1-30
Bengtson, V L., Kuypers, J A 1971. Generational differences and the developmental stake. Aging Hum. Devel 2:249-60
Bengtson, V. L., Lovejoy, M C 1973 Values, personality, and social structure: An intergenerational analysis Am Behav Sei. 16:880-912
Bengtson, V. L., Starr, J M. 1975. Contrast and consensus: A generational analysis of youth in the 1970s. In Youth: The Seventy-Fourth Yearbook of the National Society for the Study of Education, Part 1, ed. R. J. Havighurst, P H. Dreyer, pp. 224-66. Chicago: Univ. Chicago Press
Bengtson, V L., Troll, L. E. 1978. Youth and their parents: Feedback and intergenerational influence in socialization. In Child Influences in Marital and Family Interaction, ed. R. M. Lerner, G B. Spanier pp 215-40 NY: Academic 360 pp.
Bennett, S. K., Elder, G. H. Jr. 1979. Women's work in the family economy: A study of depression hardship in women's lives. /. Fam. Hist. 4.153-76
Berger, B. M. 1959. How long is a generation? Brit. J. Sociol. 10:10-23
Born, R. 1979. Generational replacement and the growth of incumbent reelection margins in the U.S. House. Am. Pol Scl Rev. 73.811-17
Braungart, R G. 1974. The sociology of generations and student politics: A comparison of the functionalist and generational unit models J.SocIss. 30:31-54
Brent, S 1978. Individual specialization, collective adaptation and rate of environmental change Hum. Devel 21:21-33
Britten, N. 1981. Models of intergenerational class mobility: Findings from the National Survey of Health and Development. Brit. J. Sociol. 32-224-38
Butterfield, H. 1972. The Discontinuities between the Generations in History. Cambridge: Cambridge Univ. Press
Carlsson, G., Karlsson, K 1970 Age, cohorts and the generation of generations. Am. Sociol Rev. 35:710-18
Chase, I D. 1975. A comparison of men's and women's intergenerational mobility in the United States. Am. Sociol. Rev. 40-483-505
Claggett, W. 1980. The life cycle and generational models of the development of partisanship. Soc Set. Q. 60.643-50
Cohen, S. Z., Gans, B. M. 1978. The Other Generation Gap: The Middle-Aged and their Aging Parents. Chicago: Follett
Collette-Pratt, C. 1976. Attitudinal predictors of devaluation of old age in a multi-
146 KERTZER
generational sample. J Gerontol 31 2 193-97
Connell.R W 1972 Political socialization in the American family The evidence reexamined. Pub! Opin Q 36 323-33
Connor, J. W 1974 Value continuities and change in three generations of Japanese Americans Ethos 2 232-64
Cowley, M 1978 And I Worked at the Writer's Trade. NY Viking 276 pp
Cnbier, F. 1981 Changing retirement patterns The experience of a cohort of Parisian salaried workers Ageing and Society 151-71
Crockett, W H, Press, A N 1981 Relationships among generations In The Dynamics of Aging, ed F J Berghorn, D E Schafer, pp 267-88 Boulder West-view Press
Cutler, N E 1975 Toward a generational conception of political socialization In New Directions in Political Socialization, ed D C Schwartz, S. K Schwartz, pp 254-88 NY. Free Press
Cutler, N E, Bengtson, V l 1974 Age and political alienation Maturation, generation and period effects Ann Am Acad. Polit Soc Sci 415 160-75
DeJong, P Y., Brawer, M J , Robin, S S 1971 Patterns of female intergenerational mobilitv Am Sociol Rev 36 1033^2
Duncan, O D 1966a. Occupation trends and patterns of net mobility in the United States Demography 31-18
Duncan, O D 1966b Methodological issues in the analysis of social mobility In Social Structure and Mobility in Economic Development, ed N J Smelser, S M Lipset, pp 51-97 Chicago Aldme 399 pp
Duncan, O D , Featherman, D L , Duncan, B 1972 Socioeconomic Background and Achievement NY Seminar Press 284 pp
Eisenstadt, S N 1956 From Generation to Generation Glencoe, IL Free Press 357 pp
Elder, G H Jr 1975 Age differentiation and the life course. Ann Rev Sociol 1 165-90
Elder, G H Jr 1978a. Approaches to social change and the family Am J Sociol 84S1-S38
Elder, G H Jr 1978b Family history and the life course In Transitions, ed T K Hareven, pp 17-64 NY- Academic 304 pp.
Enkson, R , Goldthorpe, J H , Portocarero, L 1979 Intergenerational class mobility in three Western European societies
England, France and Sweden Brit. J. Sociol. 3 415-41
Farge, E J 1977 A review of findings from 'three generations' of Chicano health care behavior Soc. Sci. Q. 58.407-11
Faver, C A 1981 Women, careers, and family Generational and life-cycle effects on achievement orientation J Earn. Iss. 2.91-112
Featherman, D. L. 1980 Schooling and occupational careers Constancy and change in worldly success In Constancy and Change in Human Development, ed O G Brim, Jr, J Kagan, pp 675-738 Cambridge, MA Harvard Univ. Press
Featherman, D. L. 1981 Social stratification and mobility. Two decades of cumulative social science Am Behav Sci 24 364-85
Featherman, D L , Carter, T M. 1976 Discontinuities in schooling and the socioeconomic life cycle In Schooling and Achievement in American Society, ed W H Sewell, R M Hauser, D L Featherman. pp 133-60 NY Academic 535 pp
Featherman, D L , Hauser, R M 1978. Opportunity and Change NY Academic 572 pp
Feuer, L S 1969 The Conflict of Generations. NY Basic 543 pp.
Feuer, L S 1974 Einstein and the Generations of Science. NY- Basic
Fitzgerald, J M. 1978. Actual and perceived sex and generational differences in interpersonal style: Structural and quantitative issues J. Gerontol 33.394—401
Foner. A. 1974 Age stratification and age conflict in political life Am. Sociol. Rev 39.187-96
Foner, A 1975 Age in society Structure and change Am. Behav. Sci. 19 144-66
Foner, A., Kertzer, D. I 1978. Transitions over the life course lessons from age-set societies Am. J Sociol. 83 5 1081-1104
Foner, A , Kertzer, D I. 1979. Intrinsic and extrinsic sources of change in life c -.use transitions. In Aging from Birth to Death. Interdisciplinary Perspectives, ed M R. Riley, pp. 121-36. Boulder, CO Westview 196 pp
Fortes, M 1974. The first born. /. Child Psychol Psychiatr 15.81-104
Fox, G L , Inazu. J K 1982 The influence of mother's marital history on the mother-daughter relationship in black and white households J. Man. Fam. 44 143-53
Fox, R 1967 Kinship and Marriage. Baltimore. Penguin
Franzblau, R N 1971 The Middle Generation NY Holt. Rinehart and Winston
GENERATION 147
Friedman, L. N., Gold, A. R., Christie, R. 1972. Dissecting the generation gap: In-tergenerational and lntrafamilial similarities and differences. Publ. Opin. Q. 36:334-46
Furstenberg, F. F., Jr. 1981. Remarriage and rntergenerational relations. In Aging: Stability and Change in the Family, ed. R. W. Fogel et al, pp. 115-42. NY- Academic. 341 pp.
Garnier, M., Hazelrigg, L. E. 1974. Father-to-son occupational mobility in France: Evidence from the 1960s. Am. J. Sociol. 80:478-502
Glenn, N. D., Ross, A. A., Tully, J. C. 1974. Patterns of intergenerational mobility of females through marriage. Am. Sociol Rev. 39:683-99
Goyder, J. C, Curtis, J. E. 1975. A three-generational approach to trends in occupational mobility Am. J. Sociol. 81:129-38
Greven, P J. Jr. 1970. Four Generations: Population, Land, and Family in Colonial Andover. Ithaca: Cornell Univ. Press. 329 pp.
Hagestad, G. O. 1981. Problems and promises in the social psychology of intergenerational relations. See Furstenberg 1981, pp. 11-46
Hammel, E. A. 1983. Age in the Fortesian coordinates. In Age and Anthropological Theory, ed. D. I. Kertzer, J. Keith. Ithaca. Cornell Univ. Press. Forthcoming
Hess, B. B., Waring, J M. 1978. Parent and child m later life: Rethinking the relationship. See Bengtson & Starr 1975, pp. 241-73
Hill, R. 1970. Family Development in Three Generations. Cambridge. Schenkman 424 pp.
Hill, R 1977. Social theory and family development. In The Family Life Cycle in European Societies, ed. J. Cuisenier, pp. 9-38. The Hague: Mouton
Hogan, D. P., Featherman, D. L. 1977. Racial stratification and socioeconomic change in the American north and south. Am J. Sociol. 83:100-26
Jackson, M. 1978. Ambivalence and the last-' born: birth-order position in convention and myth. Man 13:341-61
Jacobson, P. H. 1964. Cohort survival for generations since 1840. Milbank Mem Fund Q. 42:36-53
Jansen, N. 1975. Generation Theory. Johannesburg: McGraw-Hill. 129 pp.
Jennings, M. K. 1976. The variable nature of generational conflict: Some examples from West Germany. Comp. Pol Stud 9:171-88
Jennings, M. K., Niemi, R. 1975. Continuity and change in political orientations: A longitudinal study of two generations. Am. Pol. ScL Rev. 69:1316-35
Johnson, N. E., Stokes, C. S. 1976. Family size in successive generations: The effects of birth order, intergenerational change in lifestyle, and familial satisfaction. Demography 13:175-87
Jones, L. Y. 1980. Great Expectations: America and the Baby Boom Generation. NY: Coward, McCann & Geoghegan
Kahsh, R. A., Johnson, A. I 1972. Value similarities and differences in three generations of women. / Man. Fam. 34: 49-54
Keeley, B. J. 1976. Generations in tension: Intergenerational differences and continuity in religion and religion-related behavior. Rev. Relig. Res. 17:221-31
Kertzer, D. I. 1978. Theoretical developments in the study of age-group systems. Am. Ethnol. 5:2:368-74
Kertzer, D. I. 1982. Generation and age m life course perspective. In Aging from Birth to Death: Sociotemporal Perspectives, ed. M. W Riley, R. P. Abeles, M. S. Teitelbaum, pp 27-50. Boulder: Westview
Kiefer, C. W. 1974. Changing Cultures, Changing Lives: The Ethnographic Study of Three Generations of Japanese Americans. San Francisco: Jossey-Bass
Klecka, W. R. 1971. Applying political generations to the study of political behavior: A cohort analysis Publ Opin. Q. 35:358-73
Knoke, D. 1973. Intergenerational occupational mobility and the political party preferences of American men. Am. J. Sociol. 78:1448-68
Knegel, A. 1978. Generational difference. The history of an idea Daedalus 107-23-38
Knshnamoorthy, S. 1980. A note on the length of generation. Genus 36:167-71
Lambert, T A. 1972 Generations and change: Toward a theory of generations as a force in historical process Youth and Society 4:1:21-45
Laslett, P. 1977. Family Life and Illict Love in Earlier Generations. Cambridge: Cambridge Univ. Press. 270 pp
Laufer, R. W 1971. Sources of generational consciousness and conflict. Ann. Am Acad. Pol. Soc. So. 395:80-94
Laufer, R. S., Bengtson, V. L. 1974. Generations, aging, and social stratification- on the development of generational units / Soc. Iss. 30(3): 181-205
Legesse, A. 1973. Gada. NY: Free Press 340 pp.
148 KERTZER
Leonard, D 1980. Sex and Generation: A Study of Courtship and Weddings. London Tavrstock
Levine, D 1977 Family Formation in an Age of Nascent Capitalism. NY. Academic
LeVine, V T. 1968. Generational conflict and politics in Africa. A paradigm Civilisations 18.399-418
Lieberson, S 1973 Generational differences among blacks in the north Am. J Sociol 79.550-65
Lipset, S M , Ladd, E C 1971 College generations—from the 1930's to the 1960's Publ Int. 25 99-113
Mannheim, K 1952 The problem of generations In Essays on the Sociology of Knowledge, pp 276-320 NY Oxford Umv Press 319 pp
Manas, J 1968. Generations: the concept Int Encycl. Soc Sa. 6 88-92
Markides, K S 1978 Disentangling generational and life-cycle effects on value differences Soc. Sci. Q. 59-2.390-93
Markides, K S, Costley, D. S., Rodriguez, L 1981 Perceptions of intergenerational relations and psychological well-being among elderly Mexican Americans A causal model Int. J. Aging Hum. Devel. 13 43-52
Marshall, V W. 1980a State of the art lecture The sociology of aging. Presented at Ann Meet Can Assoc Gerontol., Saskatoon
Marshall, V W 1980b Social toleration of aging, sociological theory and social response to population aging Presented at IX Int Conf Soc Gerontol, Quebec City, Canada
Masnick, G S , Bane, M J 1980 The Nation's Families: 1960-1990. Boston Auburn House
Matras, J 1980 Comparative social mobility. Ann. Rev Soc. 6 401-31
Montero, D 1981. The Japanese Americans. Changing patterns of assimilation over three generations Am. Socio!. Rev. 46:829-39
Montero, D , Tsukashima, R 1977. Assimilation and educational achievement The case of the second generation Japanese Americans. Sociol. Q. 18:490-503
Moore, D D. 1981. At Home in America. Second Generation New York Jews. NY Columbia Umv Press
Mortimer, J. T. 1974 Intergenerational occupational movement. Am. J. Sociol. 79 1278-99
Mortimer, J. T 1981. Comment on Kenneth Spenner's Occupations, role characteristics, and intergenerational transmission Sociol. Work Occup. 8.113-17
Nahirny, V. G, Fishman, J A 1965. American immigrant groups, ethnic identification and the problem of generations Sociol. Rev. 13:311-26
Needham, R. 1974 (1966). Age, category, and descent In Remarks and Inventions, pp. 72-108. London. Tavistock
Nelsen, H. M. 1981 Life without afterlife: toward congruency of belief across generations. / Sci. Stud. Relig. 20:109-18
Ortega y Gasset, J. 1933 The Modern Theme. NY Norton
Payne, S., Summers, D A., Stewart, T R 1973. Value differences across three generations. Sociometry 36.20-30
Penn, J R. 1977. Measuring intergenerational value differences Soc. Sci. Q. 58 2 293-301
Pines, A., Kafry, D. 1981. The Experience of tedium in three generations of professional women Sex Roles 7:117-34
Plath, D W 1980 Long Engagements. Maturity in Modern Japan. Stanford Stanford Umv Press
Preston, S. H. 1978. Introduction. In The Effects of Infant and Child Mortality on Fertility, ed. S. H. Preston, pp 1-18. NY. Academic
Riley, M W 1973 Aging and cohort succession, interpretations and misinterpretations. Publ. Opin. Q. 37-35-49
Riley, M. W. 1976 Age strata in social systems See Back 1976, pp. 189-217
Riley, M. W, Johnson, M , Foner, A 1972. Aging and Society. Volume Three: A Sociology of Age Stratification. NY Russell Sage Foundation. 652 pp
Rintala, M. 1968. Political generations. Int. Encycl. Soc. Sci. 6:92-96
Rogler, L H„ Cooney, R. S., Ortiz, V 1980 Intergenerational change in ethnic identity in the Puerto Rican family Int. Migr. Rev 14193-214
Rosaldo, R 1980. Ilongot Headhunting 1883-1974 Stanford- Stanford Univ. Press
Rosenfeld, R. A. 1978. Women's intergenerational mobihty Am. Soc. Rev. 43:36-46
Rosow, I. 1978. What is a cohort and why? Hum. Devel. 21 -65-75
Rossi, A. S. 1980. Aging and parenthood in the middle years. In Life-Span Development and Behavior, ed. P B. Baltes, O G. Bnm, Jr., 3:137-205. NY: Academic
Ryder, N B. 1965 The cohort as a concept m the study of social change. Am. Sociol. Rev. 30:843-61
Scourby, A. 1980. Three generations of Greek Americans: A study in ethnicity Int. Migr. Rev. 1443-52
Sewell, W. H., Hauser, R. M. 1972. Causes and consequences of higher education-
GENERATION 149
Models of the status attainment process. Am. J. Agric. Econ. 54:851-61
Shimbori, M. 1971. Student radicals in Japan. Ann. Am. Acad. Poht. Soc. Sci. 395:150-58
Simon, R. J., Gurevitch, M. 1971. Some intergenerational comparisons in two ethnic communities in Israel. Hum. Org. 30:79-87
Skvoretz, J. V, Kheoruenromne, U. 1979 Some evidence concerning the value hypothesis of intergenerational status transmission: A research note. Soc. Sci. Res. 8:172-83
Smith, D M. 1981. New Movements in the Sociology of Youth Brit. J. Social. 32: 239-51
Sorensen, A. B. 1975. The structure of uitra-generational mobility. Am. Sociol. Rev. 40:456-71
Sorokin, P. A. 1947. Society, Culture, and Personality. NY: Harper
Spenner, K. I. 1981. Occupations, role characteristics, and intergenerational transmission. Sociol. Work Occup. 8.89-112
Stewart, F. H 1977. Fundamentals of Age-Group Systems. NY: Academic
Tannenbaum, E. R 1976. 1900: The Generation Before the Great War. Garden City Anchor Press
Taveggia, T. C, Ross, B. 1978. Generational differences in work orientations. Fact or fiction? Pac. Sociol. Rev. 21-331^9
Thomas, L. E. 1974. Generational discontinuity in beliefs, an exploration of the generation gap. /. Soc. Iss. 30:1-22
Thurnher, M., Spence, D., Lowenthal, M. F. 1974. Value confluence and behavioral conflict in intergenerational relations. /. Marr. Fam. 36.308-19
Tindale, J A., Marshall, V W. 1980 A generational-conflict perspective for gerontology. In Aging in Canada, ed. V. W Marshall, pp 43-50. Ontario- Fitz-henry & Whiteside
Treas, J. 1981. The great American fertility debate Generational balance and sup-
port of the aged. Gerontologist 21.98-103
Troll, L E. 1970 Issues in the study of generations. Aging Hum. Devel. 1.3:199-218
Troll, L. E. 1980. Grandparenting. In Aging in the 1980s: Psychological Issues, ed. L. W. Poon, pp. 475-81 Washington DC. Am. Psychol Assoc.
Troll, L E., Bengtson, V 1979. Generations in the family. In Contemporary Theories about the Family, ed. W R Burr et al, 1127-61. NY: Free Press
Tsukashima, R., Montero, D. 1976. The contact hypothesis: Social and economic contact and generational changes in the study of black anti-semitism. Soc. Forces 55:149-65
Vinovskis, M. A. 1977 From household size to the life course: some observations on recent trends in family history Am. Behav. Sci 21(2):263-87
Wake, S. B., Sporakowski, M. J. 1972 An intergenerational comparison of attitudes toward supporting aged parents. /. Marr. Fam. 34:42-48
Weaver, J. L. 1973. Mexican American health care behavior- A critical review of the literature. Soc. Sa. Q. 54:85-102
Wheeler, R. F. 1974. German labor and the Comintern: A problem of generations? / Soc. Hist. 1 304-21
Wilkemng, E. A., Guerrero, S., Ginsberg, S. 1972 Distance and intergenerational ties of farm fannies. Sociol. Q. 13: 383-96
Wohl, R. 1979. The Generation of 1914. Cambridge: Harvard Univ. Press. 307 pp.
Woodrum, E. 1981. An assessment of Japanese American assimilation, pluralism, and subordination. Am. J. Sociol. 87:157-69
Wuthnow, R. 1976. Recent patterns of secularization: A problem of generations? Am Sociol Rev. 41:850-67
Young, M., Willmott, P 1957 Family and Kinship in East London. Glencoe, IL. Free Press. 232 pp.
Copyright © 2003 EBSCO Publishing
Copyright of annual Review of Sociology is the property of annual Reviews Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.