Development During Middle Childhood: The Years From Six to Twelve (1984)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

CHAPTER 8 Ecocultural Niches of MicicIle Childhood: A Cross 'Cultural Perspective Thomas S. Weisner Imagine a satellite that could randomly sample the culture areas of the world. This imaginary satellite can focus on households with children ages 6-12 and can take audio and video recordings of their daily routines. The satellite can record with whom children associate; how far they venture from home; what work they do at what ages; the nature and difficulty of the tasks; with whom they work and how that work is shared; and the characteristics of the play group, household, and domestic group surrounding them. From the recordings we can assess the sources of child stimulation; how children explore the settings within their community; and with whom they talk and their topics of discourse and interaction. Children and adults in the sample communities could interpret the recordings and add to our understanding of settings and environments by bringing their subjective meanings to our interpretations. Together, the objective and subjective data would provide a systematic assessment of the social ecologies of childhood and development around the world. For any group of children we would be able to define their ecocultural niche (Super and Harkness, 1982~. The term ecocultural niche defines what Bronfenbrenner (1979) called the ecology of child development, going back to the tradition of Barker and Wright ~ 1954~. In comparative and cross-cultural studies (e.g., Berry, 1979:121-125; Konner, 1977; LeVine, 1977; OgLu, 1981; Super and Hark- ness, 1980; B. Whiting and ]. Whiting, 1975; ]. Whiting and B. Whiting, 1978; and others), the ecocultural niche describes the sociocultural envi 335

336 DEVELOPMENT DURING MIDDLE CHILDHOOD ronment surrounding the child and family. The term niche implies that this context has evolved through time and has adapted to the constraints imposed by the subsistence base, the climate, and the political economy of the region. The term niche connotes a somewhat different view of the environment than is implied by the proximal home reaming environment or the social structure, although it includes these. Parents, children, and families adapt to a niche and shape it to some extent as well. The niche includes the features of the environment, as conventionally defined, and also the scripts, plans, and intentions of the actors. Thus, the ecocultural niche includes variables inside as well as outside the person. Its most important elements are the relationships between participants in organized behavior settings or activity units actors with goals and intentions in a context. The study of the niches of childhood, then, includes the study of the actions, motivations, and goals or purposes shaped by those niches. As Super and Harkness ~1982) emphasized, these contexts, or scaffolds, for children's development change over time, just as individuals change and develop. Thus, the goals of a ctevetopmentat analysis include not only the study of individual and group . ~. ~ _ 1 . 1 ~ . . ~ 1 differences but also the study ot changes in the scaffolding surrounding children over time. Many features of the niche have been shown to affect children directly, or indirectly through the child's participation in the family or community. Whiting et al. (in press) developed an inventory of cultural features that influence child development. Their list, which appears below, is derived from cross-cultural as well as American studies and so includes some domains and activities that are not relevant to American children. The domains themselves, however the work cycle, health status, children's work and chores-probably represent pancultural features that affect all children: . ~ . . . . _ 1. The characteristics of the subsistence work cycle and the economic and technological system that produces it, including wage work, tending crops or animals, distance from the home, migration, etc. 2. The health status and demographic characteristics of the community, including mortality risks, availability of health care, birth control, family size, etc. 3. Overall community safety other than health and mortality, such as dangers from motor vehicles, intra- and intercommunity violence and war- fare, etc. 4. The division of labor by age and sex and perhaps other criteria like caste or race in childhood, adolescence, and adulthood, including the rel- ative importance of various activities for subsistence and prestige. 5. The work that children are expected to do beginning as a toddler through adolescence.

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 337 6. Child-rearing and chip care tasks in particular, including the personnel available and used for caretaking. 7. The role of the father arm older siblings in child care as a special issue of nonmatemal child care. 8. The composition of children's play groups by age, sex, and kinship category (siblings, cousins, relatives, and nonrelatives). 9. The autonomy, independence, and role of women in the community. 10. Institutionalized women's support groups, both formal and informal, such as work groups, church clubs, mutual aid societies, etc. 11. Various sources of child stimulation; more generally, the available sources of cultural influence on children from both literate and oral sources, including the child's contact with the media, the outside world, and toys. 12. Parental sources of information concerning child health, nutrition, new methods of subsistence activities, and new methods of child care; the avail' ability of novel or contrastive beliefs about childhood in the community. 13. Measures of community heterogeneity and change, including the presence of subethnic communities, bilingualism, subcastes, social~ciass differences and social solidarity; the role of minorities; group oppression and lack of community commitment among some subgroups; information on migration; and the number of generations that families have lived in the community. Ecocultural variables like these have been developed from some basic ideas about how the econiche has been formed; they are influenced by their functions for community adaptation and by the overall level of cultural complexity. The domains in this list, for example, can be grouped into five clusters on the basis of how they help children and families to adapt and survive. One cluster influences health and mortality (health and community de- mography, safety, defense and protection). Another affects provision of food and shelter (the subsistence work cycle, chores). Another influences the personnel likely to be around children and what those people are likely to be doing (daily routines, division of labor, child care system, play groups). A fourth focuses more specifically on the role of women and mothers in the community as the primary responsible caretakers (support, women's status, fathers' and siblings' roles). A fifth assesses cultural altematives available in the community (heterogeneity, outside influence and information). Cultural complexity is another widely used summary dimension that in- fluences econiche constraints and opportunities. Cultural complexity in- cludes an extensive cash economy, technological specialization, permanent urban settlement pattems, a centralized political and legal system, a priest- hood and other specialized religious roles, literacy, hierarchical status dis- tinctions (such as a caste organization or social classes), and a diversity of

338 DEVELOPMENT DURING MIDDLE CHILDHOOD altemative cultural models in the community (such as in bicultural or mul- tilingual settings) (Murdock and Provost, 1981~. Complexity does not nec- essarily imply a more elaborate, rituaVsymbolic world, nor an easier, more effort-free or stress-free life; but the size of the population and the scale of . . . . aCtlVltleS IS greater. American children, of course, grow up in one of the most complex cultures in the world, in this sense of hierarchy, stratification, technology, and altemative choices available. The score for an American family on nearly every niche feature is affected by that fact. Complex environments appear to promote increased nonaffiliative, individual achievement striving in chil- dren's social behavior and in parental goals (Gallimore, 1981~; more per- sonalized competition between children (Seymour, 1981~; more egoistic and dominant social behaviors in children (B. Whiting and J. Whiting, 1975~; lower rates of nurturant and prosocial behaviors, which are emphasized later in childhood (B. Whiting and J. Whiting, 19751; less sex-role segregation in family roles; a less shared-function, more specialized family role system; a more democratic family id. Whiting and B. Whiting, 1975~; and a general decline in the use of nonparental care by kin, especially sibling caretaking (Weisner and Gallimore, 1977~. Adaptation and complexity are also related to the ability to accumulate and store food and other resources for family use. In modem, complex societies, year-Ion", stable availability of food and many other resources is taken for granted, although the ability of families to purchase these resources is problematic. In much of the world, however, families face regular un- certainty in this matter. For example, early and strict responsibility and compliance training and high peer affiliation orientation appear more fre- quently as socialization goals and as child-rearing practices in societies that emphasize the accumulation of resources (Barry et al., 1959~. Berry (1976) contrasted "loose" societies (Iow accumulation, often based on a hunting economy, high mobility, dispersed settlement pattems, unstratified, and egalitarian) with "tight" ones (high accumulation, dense settlement, strat- ified, etc. ). He suggested that psychological differentiation and field inde- pendence characterize low food accumulating, low-density, migratory peoples. Regardless of the different ways to generate and cluster the variables that make up the niche description, certain dimensions recur as powerful eco- cultural determinants of child development: ( 1 ) the personnel available in the family which individuals, ages, sexes, kin; (~) the goal requirements, or tasks, to be done, which provide the reasons for children and others to be there; and (3) the cultural scripts, plans, and schemata that give meaning, create people's motivation, and give cues to intentions and purposes. Many studies in developmental psychology use home reaming environ- ment or microsystem as the unit of analysis to study the effects of niche

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 339 influences. The ecocultural niche includes the proximal home reaming en' vironment but is broader. The ecocultural niche helps to account for the existence of a particular home reaming environment in the first place. It accounts for the limited child caretaking personnel available to assist Amer- ican parents, for instance. It explains why the timing of bedtime and meals is so problematic for many American families. It identifies the source of the varying cultural ideas that appear in popular books on child stimulation or on television (Beckman, 1977; Wertz and Wertz, 1977~. The immediate home environment, then, is a result of the interaction between family goals and ideals, child characteristics, and constraints and opportunities within the ecocultural niche. Developmental research already uses econiche measures, such as social class or socioeconomic level, level of formal education, race, ethnicity, religion. These kinds of measures lump together many disparate features of the niche, drawn from different domains and functions, into a single pack' aged variable (B. Whiting, 1976~. Econiche variables decompose global descriptors like socioeconomic level into a much more complex set of mea' surest In addition, measures derived from ecocultural niche domains are more likely to reveal the mechanisms by which class or education produce their effects on children. One reason for this is that the niche features outlined here each have specific links to the daily routine of the child and the family. The daily routine of a child includes all the varied activity settings, with their personnel, cultural scripts, and plans and tasks, that the child experiences (Cole, 1981; B. Whiting, 19801. The use of the ecocultural niche mode! depends on an analysis of these activity settings, for they are the immediate situational circumstances that provide the social scaffold for assisting children to think, speak, and act (see Fischer and Bullock, in this volume). These scaffolds have their own developmental course in every culture. The developmental course of the individual is paralleled by the development of familial scripts and activity settings. These settings change with mature ation, just as the child is changing. Children's behavior between 6 and 12 results from the interplay between the child's development, on one hand, and the development of a culture's activity settings or scaffolds, on the other (see Super and Harkness, 1982~. Thus, the ecocultural mode} has a theo- retical and a comparative implication. For theory, research is needed on the development of activity settings that will parallel studies of individual dif' ferences in children's development. For comparative research, the range of activity settings available for American children must be viewed in the context of the range of such settings for children around the world. Child development studies done in Westem cultures rarely compare the data collected to data from other cultures around the world. Weisner et al.

340 DEVELOPMENT DURING MII:)DLE CHILDHOOD ( 1983) studied American parents who were attempting to be more "natural" and emotionally expressive in child-rearing practices with their infants and young children or who were voluntarily poor and emphasized loose, flexible discipline and compliance patterns (Weisner, 1982a). Although pronatural and voluntarily poor parents did differ in a number of child care practices from a comparison group, interactional styles often did not. More impor- tantly, the practices on which innovative families differed (such as more frequent breastfeeding, later weaning, or sleeping with the child) did not cliffer very much when compared to the range of such practices around the world. Thus American parents who weaned "late" did so by the time the child was 18 months; however, most cultures and mothers around the world do not even begin the weaning process until after age 18 months. Although strictness of discipline and the extent of immediate compliance to parental requests varied in the American sample, the cross-cultural evi- dence indicates that our culture is unusually flexible and permits children more autonomy and latitude in negotiations with parents over compliance than do most cultures around the world (Minturn and Lambert, 1964; Lam- bert et al., 1979~. The absolute amount of delayed compliance or negotiated requests is high in American samples, compared with comparable samples from Africa, for instance (Weisner, 1979~. Many statistically significant differences between Western samples may be of a similar character: they may produce only very small substantive differences in behavior, which are of small magnitude, with outcomes that are not sustained for very long. One powerful reason may be the fact that on a pancultural scale the magnitudes of the intracultural differences are not! very large. The only way to test this would be to systematically and routinely compare developmental data collected within our own niche, with data collected from a wide range of econiches and cultures around the world. Such a practice would, ~ expect, have the same importance in interpreting developmental data as the currently routine expectation in scientific studies of reporting test and instrument norms or statistical variance within a sample. The influence of the niche is subject to empirical test, as are features of the child or parents, such as gender, age, and temperament. This point is important to emphasize because culture is so often treated in just the opposite way as an untested, packaged variable. The ecocultural niche approach must not assume what is often exactly what needs to be proven: that cultural factors indeed have important effects. Culture must be used as a set of variables like others whose specific character and effects can be measured and tested. The same point is true for determining which aspects of the econiche have the strongest effect ethnic or cultural membership itself or subsistence

ECOCULTUR'4L NICHES OF MIDDLE CHILDHOOD 341 and environmental constraints. For instance, Ecigerton ~197 ~ ~ systematically tested values and personality characteristics in adults from four tribal cultures in East Africa. Within each culture, some individuals lived primarily as pastoralists and others as horticulturalists. Edgerton was thus able to compare the effects of cultural membership versus subsistence adaptation pastoral or horticultural for each of eight sample groups (four cultures x two subsistence modes). Results showed that both cultural group membership and subsistence modes differentiated between clependent variables in his study. Pastoralists were more concerned with displays of affection, direct aggression, divination, and independence than were farmers. Farmers em- phasized disrespect for authority and favored conflict avoidance, indirect aggression, emotional constraint, and other values, compared with pastor- alists. Tribal membership, however, was the best overall predictor of these sample differences, and subsistence mode was next best. In brief, the ecocultural niche defines the contexts for development; these contexts represent evolved, adapted family responses to opportunities and constraints of the environment; the activity settings that result are the measurable, visible features that can influence children and families. The study of child development, then, should include the study of the relation- ships between the activity settings provided for children within the niche, on one hand, and the maturational uniformities and individual differences children and parents bring to these activity settings, on the other. Basic research on the 6-12 age period should pursue new knowledge regarding the development and influence of activity settings of children at these ages and study a far broader range of such settings in American society and around the world. Finally, ~ believe that high-quality description of the lives of children in other cultural settings is in and of itself of basic scientific value in providing a mirror for ourselves. Lambert (1971:61) commented on the intrinsically valuable character of cross-cultural data its ability to awaken us to new altematives: Since no one culture has managed to achieve a monopoly of all the "good" or "bad" conditions for parent~child relations, then we are going to be delighted as we travel about the world. We are always going to find some facet of human personality or personality organization which glows with a serene excellence that we have never met before. And lying below the fact of that fresh, though partial and perhaps even fleeting, excellence, is new knowledge about how to make some future generation (and its parents) better, more happy, or more free. Although comparative work is widely accepted in principle or as a pro- grammatic need, it is not being done. LeVine (1980) reviewed every article and research note published over a 5-year period in four developmental

342 DEVELOPMENT DURING MIDDLE CHILDHOOD joumals and six anthropology journals. These represented the ten major joumals in the two disciplines. Only 9.3 percent of the articles in the developmental joumals included any data on subcultural variations in the United States, Europe, Israel, or anywhere else (171 of the 1,843 articles). Of these 171, 75 percent reported data from other Westem industrial so- cieties, leaving only 42 articles (2.3 percent of the total) with any data from Latin America, Asia, or Oceania. Similarly, the six anthropology journals published 911 articles during this 5-year period, of which only 70 (7.7 percent) gave any consideration to child care or development. The situation is even worse than this implies, since one of the anthropology joumals, Ethos, by itself published almost one-third of all the articles on children in the anthropology articles reviewed. The regions of the world are also very unevenly represented; Latin America, for example, is far more frequently mentioned than other parts of the world. WESTERN AND NON-WESTERN CULTURES In the discussion of ecocultural niches of middle childhood in this chapter, American children ages 6-12 are often lumped with Westem children or those living in complex societies. This is a gross oversimplification and of course does not mean that there are not large differences in the experience of children across Westem societies. Similarly, children in non-Westem cultures are often lumped together to contrast with American children. This is an even grosser oversimplification, since the range of cultures is even greater within this category. References to non-Westem societies should be understood as referring primarily to middIe-range horticultural and simple agricultural societies, unless otherwise noted. Most examples are drawn from Polynesia (Tahiti, Hawaii) and sub-Saharan Africa (East Africa, Ghana, Botswana). It is not possible or appropriate to present an ethnographic overview of the patterns of child care during the 6-12 age period around the world. The emphasis on broad, cross-cultural contrasts in this chapter is not intended to homogenize the rest of the world, nor to imply that there are not enormous social-cIass, racial, and ethnic differences in Westem societies, nor to suggest that non-Westem societies are uniform. To the contrary, the point is to search for the niche and activity settings that influence child development and that are the result of just such class and ethnic differences. The chapters in this volume reflect many of the central themes of middle childhood: caretaking pattems, schooling, health, cognition, and self-un- derstanding. Cross-cultural perspectives are especially important in the study of cultural conceptions of the person and the self; children's own theories

- - ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 343 of their development and roles; the differing structures for child caretaking during this period, particularly nonparental care; the socialization of appro- priate emotional expression; the influence of deviance and psychopathology in middle childhood; the influence of schooling and literacy; the effects of urbanization and modemization; sex-role and gender-identity development; the transition to adolescence; and others. ~ have selected four of these themes to illustrate a comparative niche approach: the structure of caretaking; de- velopment of the self; troublesomeness in children; and schooling effects. Each of these domains is covered in the next four sections, followed by a discussion of methods. A goal of this volume is to suggest areas for new basic research. The topics covered in this chapter are those that are not already covered in other cross- cultural reviews, such as the Handbook of Cross-Cultural Human Development (Munroe et al., 19811; that do not as yet have extensive comparative research but look promising; that appear to be important during the 6-12 age period; and that are emphasized in the other chapters in this volume. PATTERNS OF CARETAKING OF CHILDREN A central issue in American families with children ages 6-12 is the gradual shift in direct control of the child's behavior and activity settings from parents to the world of the school and peers. Medrich et al. ( 1982:102-103) reported that parents in Oakland, California, fee! they need to devote less of their time to either direct physical care or nonphysical care of their children during this period. The papers by Maccoby and Markus and Nurius in this volume emphasize that an important general developmental''task for Amer- ican children in this period of life is to accomplish a gradual change in processes of control and regulation. Children appear to gradually move from coregulated activities to self-regulated ones. Parents retain overall managerial influence, but children are increasingly capable of self-regulation of their activities for long periods of the day. American parents encourage individ- uation and self-control during this period but also attempt to negotiate with children and withhold resources in order to retain overall managerial control within the family. The developmental task or agenda that faces parents and children in many non-Western cultures is related to but different in many ways from the American one of individuation and separation from the parent as an exclusive controller. Rather, the task involves children gradually moving from under the responsibility of older children and other nonparental members of the household (e.g., grandparents or aunts) to becoming a responsible caretaker, in charge of younger siblings and cousins. Children ages 6-12 take on

344 DEVELOPMENT DURING MIDDLE CHILDHOOD increasingly responsible family roles, including those relating to child care. Parents' roles include the managerial and disciplinary ones familiar in Amer- ican families, but child care is more diffused and shared. This caretaking pattern has as its goal to produce an interdependent, responsible child, rather than an independent, self-directed, highly individuated child. Many cultures also share the belief that between age 5 and age 7 children begin to acquire reason or sense, the ability to understand cultural rules and to carry out directions. Rogoffet al. (1975), Super (1981), and I. Whiting and B. Whiting (1960) identified this age period from cross-cultural samples, and NerIove et al. (1974) did so with data from Guatemala. NerIove et al. identified two natural indicators of cognitive skill that develop before or during this period, which are both important in shared child management activities: self-managed sequencing of activity and voluntary social activities. Self-managed sequencing refers to the child's ability to follow a precise sequence or series of acts autonomously e.g., washing clothes, which en- tails gathering up a basket, clothes, and soap, putting the clothes in a basket, going to water or the river, etc. These tasks require, in correct order, "a scanning of the mode! and mapping of that model onto altematives, ret membering what one had already tried and how well it fit" (NerIove et al., 1974:287~. Voluntary social activities involve self-directed, shared activity with others, which assumes a shared goal and rule understandings. For [anguage-related voluntary social activities, reaming "to name, recognize, and verbally relate functions or attributes of objects" to others (p. 287) is crucial, including reaming kinship rules and cultural and family standards. Voluntary social activities thus include understanding and storing multiple roles and social scripts as well as the ability to lead and direct them. Sibling caretaking exemplifies the application of both of these skills to an important family Unction. Rogoff et al. (1980) extended their cross-cultural work to the 8-10 age period, suggesting that children appear to be developing skills at performing more complex tasks, which require more elaborate understanding of context- appropriate behaviors and more complex understanding of causality and intent which increase the child's ability to consolidate and integrate the separately acquired skills reamed in the 5-7 transition period. Effective performance of child care, as a part of the competencies needed to perform domestic chores and even manage the domestic routine, requires a minimum level of both these kinds of skills in childhood, and in turn domestic duties help train children in more general skills. Thus the age of greatest involvement in and responsibility for shared child and domestic task management corresponds to the 6-12 developmental period, when these social and cognitive skills become available to children.

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 345 The contrasts between coregulation and self-regulation and between in- terdependence and independence are certainly expressed as cultural goals and emphasized in parental talk and metaphor. The degree to which chil- ciren's and parents' behavior in fact reflects these metaphors seems to vary widely. The role of socialization and intemalized behavioral tendencies, in addition to differences in activity settings as influences on metaphor and behavior, needs new research. There is a sense in which all children are interdependent within their family and community and a sense in which autonomous self-regulation is more of a Westem cultural myth than a be- havioral reality. The differences, however, between Westem and non-West- em children in nurturance, prosocial responsibility, and affiliative orientations have been well established (e.g., B. Whiting and J. Whiting, 19751. Sibling Caretaking Barry and Paxson (1971) surveyed 186 societies in the cross-cultural sam- ple of the Human Relations Area Files and concluded that mothers were considerably less frequently the primary caretaker of children than either siblings, older children, or female adults other than the mother. Thus, children ages 6- 12 in most of the non-Westem world continue what already has been a common experience for them earlier in life: They participate in peer and sibling caretaking systems and are not usually under the direct, personal care and supervision of their mothers. Gallimore et al. (1974), Leiderman and Leiderman (1973; 1977), Levy (1973), Mead (1961), Mintum and Lambert (1964), Weisner and Gallimore (1977), B. Whiting andJ. Whiting (1975), anal. Whiting and B. Whiting (1973) have all recorded comparative data and developed theories about sibling care. Sibling care is associated with the following ecocultural con- ditions: horticultural, pastoral, and simpler agricultural societies in which the family workload is high; mothers are responsible for work outside the home; residence patterns establish sets of neighboring, extended family groups with children available for shared care; and shared work roles and task allocations within families promote joint care of younger children. Shared functioning is a useful term for describing such flexible, nonexclusive family work roles and child care responsibilities (Gallimore et al., 19741. Children ages 6-12 are cared for by older children; then, through participation in pivot roles (Levy, 1973), they move to caretaking supervision of still younger children. Mothers' roles are as indirect managers of the sibling and family group-assigning duties, overseeing the senior sibling caretaker, jointly doing chores and activities with children, providing discipline and occasional in- struction, or simply modeling correct behavior.

346 DEVELOPMENT DURING MIDDLE CHILDH001) Children ages 6-12 are also sent away from their own natal home, for years at a time or permanently, to live with others. In many West African societies, for instance, fostering is common (Goody, 1982~. In such systems, children are sent away to live with other kin; to live with kin or nonkin to whom parents owe a debt (debt fosterage); to live as apprentices and work for a craftsman; to work for an important personage or chief as a means of developing a political alliance; or to be fostered for educational advantage or service obligations. In many societies, including East African, children are sent to live with their kin to assist elderly grandparents or the child's mother's sister during difficult times (Weisner, 1982c). In Polynesia, adop- tion is frequent (some 23 percent of children in one Hawaiian community were adopted) and can extend throughout childhood (Levy, 1973; Ritchie and Ritchie, 1979~. Such practices are seen by parents as a necessary and expected reallocation of an important family resource their children. The institutions of fostering and adoption when children are ages 6-12 encourage socialization practices similar in many ways to those involved in sibling caretaking: diffusion of affect, attachment to community, early expectations of prosocial, mature behaviors, strong compliance and deference expecta- tions, work and responsibility expectations imposed early in life, and others. Weisner and Gallimore ( 1977) suggested a number of child outcomes that might be predicted to differ in settings with extensive peer or sibling care socialization systems of this kind. For example, potymatric systems (Leiderman and Leiderman, 1973, 1977), or multiple caretaking of infants and young children, may diffuse early attachment. It can also increase the child's secure sense that others in the community will care for him or her. There appears to beta balance between the insecurity of diffused and variable attachment to several caretakers and security in having several reliable alternative care- takers. Children in polymatric caretaking systems become responsible, active participants in their wider community during the 6-12 age period. This early maturity may be in part the result of more exposure to the security of wider nonparental community care and obligations. Children's play groups also depend on caretaking obligations: Child care- takers pool their younger charges, and they do chores along with them. Children ages 6-12 frequently divicle into groups of mixed-age girls and younger boys; younger boys are usually separated from older boys, who are off in small packs (Weisner, 1979~. This diffusion of child care responsibilities through the sibling hierarchy and among other kin relieves the parent of the constant monitoring char- acteristic of many Western domestic care situations. Children can be out of sight and out of hearing among peers and still be considered safe and acting responsibly. Of course mischief, teasing, and worse certainly occur.

- - ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 347 But the generalized, expectable cultural climate for compliance in tasks and child care seems to constrain most children during middle childhood (see section on troublesomeness below). Caretakers also leam responsibility, nurturance, and prosocial behavioral styles girls more so than boys (Edwards and Whiting, 1980; I. Whiting anc! B. Whiting, 1973~. Sibling care may reduce the strength of the inter- nalized parental role mode! and increase the influence of community con- straints based on shared function. It has also been suggested that individual differences across children are reduced as a result (Levy, 1973; Mead, 19611. Shared caretaking systems have also been linked to differences in cognitive style in children, such as field dependence and increased social empathy. Such children emphasize affiliation and cooperation rather than individual achievement in their play styles and in responses to standardized tasks (Gal- limore et al., 1 974; Madsen, 1 97 1; Madsen and Shapira, 1 970~. Children in these kinds of caretaking settings may also differ in their classroom learning styles and in the ways they interact with teachers and peers having fewer verbal interactions with teachers or directing help seeking away from the teacher and towards peer reaming in the classroom (Jordan and Tharp, 1979; Weisner, 1976~. Dunn (1983), Gottman (1983), and Hartup (in this volume) review, the evidence on the development of peer friendships and associations, from acquaintanceship through the various stages of friendship. North American studies emphasize dyadic themes of reciprocity, equity, faimess, mutuality, and intimacy in chil(lren's choosing and developing friendships. There is no question that making and retaining friends is crucial in the school-age period for American children. In the North American niche, this~process occurs among strangers for the most part. The task for the American child involves entering new and strange settings (primary schools, sports teams, churches, new neighborhoods) and selecting, with parental monitoring and influence to some degree, others to play with from among these groups of children. The American friendship relationship usually involves the creation of new, personal, independent alliances. Contrast this with the task of the 6- to 12-year-old in more kin-based, shared-function, sibling care cultures. There are very few activity settings in which other children who are strangers would ever be present. If so, older children or adults would almost certainly be there to carefully monitor what could be a potentially dangerous, uncertain situation. For most children in such societies, the task and activity setting are quite different than for the American child: Children already are participating in established, probably permanent relationships with cousins, clan-mates, or vilIage-mates. The task is not to construct a new acquaintance, but rather, first, to team precisely

348 DEVELOPMENT DURING MIDDLE CHILDHOOD who the other child is and how he or she is related (e.g., kinship ties, parents' status vis-a~vis one's own parents, the family's reputation, etc. ~ and, second, to observe carefully how others in one's own sibling and courtyard cousin group relate to that child. Thus the first steps do not require the child to personally initiate interaction, then display such personal skills as mutuality and verbal appropriateness. Rather, the child first must understand how the culture has already classified him or her into a preexisting set of alliances and feuds; second, the child must incorporate the acquaintance into his or her immediate sibling group. In addition, children ages 6-12 will very early have to consider what work, tasks, and chores must be done along with this child as a friend. Friendship will require much more than play, games, and childhood intimacy. The childhood relationships formed in the sibling care system will persist throughout the fire span. The same groups of children who are involved in shared care and not-always-benevolent authoritarianism in their families will in turn become adults who must cooperate in managing land, cattle, busi- nesses, and other family resources. For example, these children will soon engage in extensive marriage negotiations. Thus the shared responsibilities of children ages 6-12 anticipate the subsequent adaptations of siblings to their adult-life roles. Nonmaternal sibling care systems should not be seen simply as a big, cooperative, shared-play group. Fierce feuds, bitterness, and competition can characterize sibling relationships in childhood and adulthood. Teasing, benign neglect, and the domination by older children of younger ones are frequent. These experiences create rivalries and competition as well as co- operation in shared defense and survival needs of the community. New basic research on the development of friendship and childhood in- timacy in the Western econiche, compared with those of non-Westem societies, probably would be especially fruitful. All children probably share the developmental task during this period of learning how to sustain shared intimacy with peers. In much of the world, however, this task is undertaken by relatives and child acquaintances who are already known and by groups of children who are going to be lifelong sources of support for each other or lifelong opponents or both. Sibling caretaking may assume increasing importance in this culture, and American children ages 6-12 will probably participate in shared responsi- bility even more than they have in the past. American mothers of children ages 6-12 are steadily increasing the number of hours they work outside the home in wage employment. And Bane et al. (1978) have pointed out that American parents utilize kin and in-home shared care more frequently and prefer it to nonhome, nonparental care. An important issue for new basic

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 349 research on school-age children in American society will be to assess the effects of shared and sibling caretaking in the American context. Father Roles There is an apparent secular trend in the United States for fathers to take on a greater role in domestic work, including more involvement in direct child care (Lamb, 1981~. Although the degree of such change may be less than what some parents and others hope for (fleck, 1979; Weisner, 1982a), there is nonetheless a persistent and growing expectation for increased par' ticipation by fathers. American white middle-cIass fathers, however, are already actively involved in the domestic routine and in supplemental child care, compares] with fathers in most other cultures. Many features of our ecology and culture already encourage this relatively high involvement. The features of the ecocultural niche that have been identified in other cross- cultural studies (Katz and Konner, 1981; Weisner, 1982b) as associated with higher father involvement include: · nuclear or conjugal family model, not stem, extended, or joint families; · neolocal residence customs (couples live apart from their own kiln); · relatively high geographical mobility of parents; · a high expectation of intimacy between couples (emotional, interper- sonal, sharing of information, sharing resources); · relatively low availability of support from kin for assistance and support; · relatively low neighborhood solidarity and support for parents; · relatively low threat of violence to the home or community; and · infrequent use of sibling caretaking, adoption, fosterage' child lending, or other institutions for sharing children. Some of these features affect the availability of different kinds of caretakers (e.g., fathers, siblings, nonparental caretakers). Other cultural features in- fluence the quality of parent's involvement with their spouse and with their children; for example, greater sharing and intimacy between American cou- ples may translate into more shared activities and emotional involvement between fathers and children. There are surprisingly few cross-cultural studies of fathers using direct, naturalistic observation with quantitative methods for description and anal- ysis of patemal behavior. (There are relatively few such studies of families in the cross-cultural literature to begin with.) The reason usually given for this is that the base rate for occurrence of father-child interaction is typically very low compared with interaction rates for peers, siblings, and mothers. For example, Weisner (1979) and B. Whiting et al. (in press) dropped

350 DEVELOPMENT DURING MIDDLE CHILDHOOD father-child dyads (children ages 2-11) from statistical analyses for this reason. Fathers were present only between 3 and 14 percent of the time (in New England the figure was 9 percent) in the Six Cultures study (B. Whiting and ]. Whiting, 1975:45~. Only 1 percent of the interaction of boys and girls ages 3-11 was with fathers, compared with 10 percent with mothers, 54 percent with peers, and 15 percent with infants and toddlers (p. 153~; limited data on father interaction prevented direct comparisons between mothers and fathers. Konner (1975) observed !Kung infants and children ages 2-6 throughout the year during daylight hours. Fathers interacted with infants in 13.7 percent of the observations compared with an equivalent 99 percent for mothers. !Kung fathers were present in 30 percent of the observations of children ages 2-6. In contrast, Blurton-Jones and Konner (1973) reported fathers present 19 percent of the time during similar observations made in London. Fathers' participation in infant care expressed as a percentage of mothers' care is only between 2.3 percent and 6.3 percent for children ages 2-6. Although these studies find fathers home much less than mothers and find fathers interacting relatively infrequently with infants and young chil- dren when they are home, these studies do not demonstrate the true pro- portions of father involvement with older children. Relatively few studies have included older children, and observations are not made at times when fathers are most likely to be home: early momings, evenings, weekends, and holidays. Most settings for observational work focus on the home and its surroundings and perhaps a school-but these are not the places where fathers are likely to be found during much of the day. Field observational work.should be designed to include the times and places where fathers and children are most likely to be together. Basic research on new fathering styles among American families (Lamb, 1982; Radin, 1982; Weisner, 1982b) has also shown that children are ex- periencing a wide variety of father roles. Some fathers appear to truly share active management of the home and domestic routine with the mother. Others have increased their shared involvement and support of what remains primarily a maternal responsibility. Other &there are largely absent, involved in careers that take them away from either shared support or true equal management. Still other children divide their time between parents who share their custody but are no longer living together. Similar data need to be collected on the diversity of actual patterns of involvement of fathers with children ages 6-12 elsewhere in the world. The simple assessment of task or chore responsibilities of fathers or their presence in the home needs to be replaced with a more complex picture of the types of emotional in- volvements and coregulation of activities that fathers share with their sons and daughters during middle childhood.

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD Conclusion 351 The ecocultural niche of middle-range societies (those based on horti- culture or simple agriculture with family groups as important primary sub- sistence producers and consumers) frees mature, adult working-age women from exclusive child care responsibility. The niche also provides many of the essential conditions that encourage the sharing of caretaking and other work roles with children ages 6-12 (such as larger family size, joint and stem family pattems). Children ages 6-12 experience an active participatory role in family survival tasks in such settings, and they are important partic- ipants (although still lowly in status and power) in the tasks of the wider community outside the family. The activity settings for such children in these kinds of middle-range societies expose them to a wider circle of kin and community members and to greater responsibility and expectations for prosocial behavior than is the case for American children. Coregulation is also encouraged, while autonomy and individualism are discouraged. Obe- dience and compliance are expected, rather than being the subject of con- tinual negotiation between parents and children regarding work or free time. Basic research on nonmatemal care of children more generally should begin by mapping the true range of kinds of such care provided for children ages 6-12 around the world and the niches in which these occur. A number of plausible outcomes of nonmatemal care (such as emotional or attachment differences and affiliation motives) have been suggested, but these have rarely been measured directly or concurrently with direct measures of care- taking patterns and styles. In addition, the importance of shared, conjoint caretaking, with both adults and children participating, needs closer study. Descriptions of mutually exclusive types of caretaking (matemal; polymatric) often do not capture the variability in the strategic use of conjoint caretaking so characteristic of many societies, including the United States. SELF-CONCEPTIONS AND ECOCULTURAL VARIATIONS Markus and Nurius (in this volume) emphasize that the study of the self in children ages 6-12 includes the study of self-regulation in children that is, how children monitor and manage their behavior in ongoing interaction. Understanding how children acquire their skills at self-regulation in rum will require new basic research on the activity settings that surround them. American cultural goals emphasize egalitarian ideals and universalistic moral convictions regarding sharing and faimess. But the day-to-day activity settings in which American children typically find themselves (e.g., ciass- rooms, sports, individual homework) in fact encourage individualism, au- tonomy, competitiveness, self-direction, and self-regulation. In contrast,

352 DEVELOPMENT DURING MIDDLE CHILDHOOD while many non-Western cultures have public overt beliefs emphasizing differences between clans, castes, religious groups, or regions, in their daily routines, these children participate in activity settings that emphasize shared functioning, coregulation of behavior, compliance to adults and older chil- dren and that discourage exploration or private self-aggrandizement. It is arguec! in this section that different kinds of activity settings produce differing patterns of self-regulation, self-unclerstanding, and conceptions ot person- hood. The same contrasts between Western and non-Western activity set- tings clescribed above for the consequences of caretaking are also relevant in shaping the development or the sets. {n`]ivi~uation an] ~n~ivi~ua~ism Franz Boas (cited in Miller, 1963:280) reported in 1911 that"the three personal pronouns-I, thou, and he- occur in all human lan- guages.... The underlying idea of these pronouns is the clear distinction between the self as speaker, the person or object spoken to, and that spoken of." Every child, then, reams to distinguish between him- or herself, the dyed, and others separate from the self or the immediate group. In this sense, individuation is a universal process. Shweder and Boume (1981) distinguished between individuation and individualism; between the development of the ~ or me and egoism; between a sense of uniqueness as a person and a claim of personal autonomy from others. Individuation seems to be a universal process, but individualism, egoism, and autonomy from others seem far more characteristic of Westem self-conceptions during the 6-12 age period and throughout life. The idea of the individual self as contained inside a private mind. within "a bag of skin," looking out on the world, seems not to be universal t~hweder and Boume, 1981~. Children in many non-Westem cultures certainly do leam about themselves as unique individuals, but without the accompanying ego- ism, self-aggrandizement, autonomy, or concerns over invidious comparisons and esteem that seem to characterize the relatively few available Western reports of children's self-concepts. Levy, for example, described a typical man, called Poria, in a Tahitian village, Piri (Levy, 1973:2171; Poria is a unique individual, but neither autonomous nor part of an autonomous group: In Piri he is simply Poria, a unique individual with his own ways of getting on with others and of organizing his life. He is one of a large number of different kinds of individuals in the world of Piri, and he feels himself to be not "of the same measurements" as the others. People in Roto [an urban area] have, perhaps, more occasion to identify

- ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 353 themselves as "types," Tahitians, in contrast with the many non-Tahitians surrounding them, or to solidify themselves with others. Tahitians emphasize the role of shared life stages in accounting for their behaviors much more than Americans would. They are likely to describe whole periods of their life, especially adolescence and the early years of marriage or later childhood by saying "I was in the period at that time." Shared life stage categories or kinship groupings, rather than nonshared categories, are used to compare people. The intimate, face-to~face, small-scale character of such societies seems to promote shared, nonprivate self-concepts in children. Typological descriptions or labeling of others in the village context is infrequent except in terms of shared categories like life stages. Hallowell (1955) describer! the Ojibwa Indian self as part of a world of inclividuals seeking power over each other, beset by fears and anxieties con- cerning magic or sorcery. OjiDwa worry about giving offense to others and seek power from supematural spirits who operate in the worIcI only through men. The spirits are largely beneficent, but the men and women sometimes are not. For Hallowell, each cultural environment must be studied anew to search for the particulars that define self-conceptions. The self in this view is formed as a part of the larger construction of cultural meaning, which is a universal process in Hallowell's view. Hallowell's approach is at once pancultural and universal with regard to processes of self-understancling, but highly particularistic with regarc] to the content and substance of self-identity. This view reflects a relativistic position on development of the self and emphasizes the importance of the content of ideas within a specific cultural-historical period, or behavioral environment. Shweder (1981) also saw self-concepts acquired in childhood to be part of the larger cultural solutions to the problem of meaning, but he proposed in addition a set of specific topics for self-understanding in a local niche. He presented a list of 10 themes about social existence, which (a) resolve moral clilemmas; (b) are possibly universal problems of meaning; and (c) appear to be taught to all children as they acquire their self/cultural/moral understandings of their world (1961:~: the problem of personal boundaries what's me versus what's not me; 2. the problem of sex identity what's male versus what's female; 3. the problem of maturity- what's grownup versus what's childlike; 4. the problem of cosubstantiality who is of my kind and thus shares food or blood or both with me versus who is not of my kind; 5. the problem of ethnicity what's our way versus what's not our way;

354 DEVELOPMENT DURING MIDDLE CHILDHOOD 6. the problem of hierarchy why do people share unequally in the burdens and benefits of life?; 7. the problem of nature versus culture- what's human versus what's animal-like; 8. the problem of autonomy am ~ independent, dependent, or inter- dependent?; 9. the problem of the state what ~ want to do versus what the group wants me to do; and 10. the problem of personal protection how can ~ avoid the war of all against all? There are as yet far too few cross-cultural data on children's self-concepts and identity formation to support even descriptive generalizations about each of these issues for children ages 6-12. These themes, however, are surely widespread and take comparative research beyond the occasional collection of interesting but nongeneralizable case studies. Shweder's (1981) topical domains provide a framework for comparative data collection on the content of children's ideas about the self and social personhood in Westem and non- Westem settings. What are some of the features of activity settings for children ages 6-12 that might be linked to differences in self-conception? These characteristics are probably related to Damon and Hart's ( 1982) listing of proposed universal mechanisms in the acquisition of self-knowledge (internalization of cultural standards and scripts; development of schemata for the self; modeling; and identity formation). And they should also be linked to differences in chil- dren's' processes of self-regulation, following Markus and Nurius (in this volume). One niche difference with possible consequences for self-concep- tions concerns the degree of privacy, or separation of role performances, available to children ages 6-12 in American versus non-Western activity settings. Non-Western children do not seem to experience the degree of privacy that American children do during this period. A second major difference has already been considered: the degree to which coregulated, shared functioning is expected of children ages 6-12 in non-Westem set- tings. The next section briefly considers some basic research issues regarding the interrelationships between niche features, such as privacy and shared functioning, and self-conceptions. Private and Public Selves Westem children live in a remarkably private culture. Most cultures are vastly more public than is ours in their geography, architecture, and daily

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 355 routine. The opportunity to be private is so much greater in the West, and we encourage such privacy early in children's lives by giving them their own rooms and spaces, their own toys and other possessions. We allow children some degree of choice over their food, playmates, TV shows, clothes. We negotiate with children over space and possessions and make children at least partly coequal interlocutors concerning these and other matters. Amer' ican children learn how to make behavior and possessions private or at least capable of being kept private as a matter of their choice. Self~presentation in activity settings without privacy of these kinds is like being on a social stage with no private dressing rooms, where the stage wings are visible to nearly all of the audience, and where the same audience and cast comes to every show. The formation of social character in children ages 6-12 in such public cultures occurs in contexts in which nearly all the behaviors a child displays are potentially known to everyone else. The public character of life extends to the sharing of resources. Food, knowledge, child care responsibilities, and material possessions are shared in order to survive. Stinginess is one of the profoundly negatively sanctioned traits. "Visualize the kind of sharing that occurs around the dinner table in a Western household but expanded in scale to include a group of 15-30 people, and you have some idea of the nature of sharing in a !Kung camp" (Lee, 1981:98~. Foraging peoples like the !Kung exemplify, albeit in extreme form, other fundamentals of the development of the self in non-Westem settings: far more aspects of life are public; roles and settings are perpetually undermanned due to the small scale of village life; relationships are multiplex and face-to-face; and children participate in or observe most of adult life from age 6 on. The Mung also illustrate how public are the most intimate details of life. The private, separated self undramatized, carrying secrets one intends to reveal to no one or to only a few selected intimate others is relatively undeveloped in children given the way !Kung foragers deal with personal troubles (Lee, 1981:99~: Daily life goes on in full view of the camp. People rarely spend time alone, and to seek solitude is regarded as a bizarre form of behavior. Even marital sex is carried on discreetly under a light blanket shared with the younger children around the family fire. It is considered bad manners for others to look. Sullen, withdrawn behavior is regarded with concern and not allowed to continue. The person showing it is pestered and goaded until he or she loses his temper and the anger that follows helps to clear the air and reintegrate the outsider. When people are depressed or their feelings are hurt, they express it by awaking at night to compose sad songs, which they play for themselves on the thumb piano. These poignant refrains form a counterpoint to the night sounds of the crackling sleeping fires and the calls of the night-jars, and no one tells the players to pipe down or shut up.

356 DEVELOPMENT DURING MIDDLE CHILDHOOD This is not to say that private self-reflection does not occur in shared- function, non-Westem settings (Shostak, 19811. But public debate, shared conversations, and family negotiations about children's personal life course, viewed apart from the child's role in the family and community, are not common in the non-Westem world for children ages 6-12. In contrast, first graders in Los Angeles are asked to describe themselves, to tell why they are different from everyone else, and to think of good things about them- selves including drawing pictures, writing stories, and speaking in public to their classmates. They come home and ask their parents for pictures of themselves to bring to school and put up on the wall; they make an outline of their hanc! and print their name to go up on the board with the picture. Children 6-12 in much of the non-Westem world do not engage in these kinds of self-contemplation and public displays. If these activities occur at all, they come much later in life than in the West. Langness and Frank (1981:101-105) confirmed this view of cultural pre- sumptions about private selves in their review of self-conceptions as revealed in non-Westem life histories, biographies, and autobiographies. The West- em conception of biography sees the self and the life course as a unified, continuous progress, as the stories of lives viewed or reviewed. This seems inevitable to us: Clues to later life are to be found in childhood experiences; life is a chronology through time; a person's own feelings, perceptions, and life events are the central things; there is a search for causes and effects of one's actions that depend on the contrast between an inner and an outer world. Yet most cross-cultural biographical accounts seem largely concemed with the public self: one's role in public affairs; a sense of community in- tegrity' one's role in family activities; the esteem with which one is viewed in the community. For example, Geertz (1973) clescribed the Balinese as presenting the social and public self as the real self; that self is what is of personal importance, providing the focus of efforts toward self-expression, and it is reported to ethers as one's true self. It is worth emphasizing the point made at the beginning of this chapter: neither the Westem nor the non-Westem generalized patterns ~ have out- lined are meant to exclude the obvious diversity in the acquisition of an individualistic sense of self. There are boastful, self-aggrandizing childhood socialization patterns in the non-Westem world (e.g., the Northwest Coast), and there are many cultures and subcultures in the Western world without the pressures for individualism that ~ have described. Religion, ethnicity, local patterns of class or cultural dominance, and other features clearly modify the development of the individualistic self during the 6-12 age period. And individual differences within every community produce variations in self- esteem and egoism, apart from any general effect of the local ecocultural

ECOCUL~URAL NICHES OF MIDDLE CHILDHOOD 357 setting. The goal of new basic research on self-acquisition should be to disentangle these features wherever possible and to put them to systematic, comparative test. Ecocultural Origins of Differences in Self-Regulation Differing patterns of self-concepts and self-regulation have their origins, in part, in the demographic facts that so profoundly separate the Western and non-Western worlds. In non-Western cultures, family size is large; child mortality is high; a child's own parents, and some nonparental caretakers as well, may die before the child reaches marriage age; dangers from raiding, feuds, or outright warfare are all vivid realities for children in many parts of the world. In such circumstances, the unstated presumptions about the life course that buttress the Westem focus on personal self-investment may not hold namely, the assumption that one's family and neighbors and the child himself or herself will survive to adulthood. A child's concern for self- expression, and parental searches for behavioral evidence of it, in part may be a modern luxury of safe, healthy, low-mortality environments. Family survival also involves putting children ages 6-12 into the public world of work. These children are subject to strong, early compliance and discipline training (Mintum and Lambert, 1964~; the use of stem discipline and physical punishment for transgressions; low warmth and positive affect in chil`d care (Rohner, 19751; relatively little direct parental involvement with children during this period; and infrequent direct praise. In contrast, research on American child care environments, which encourage positive self-esteem; self-confidence; egoism; individualism; and an open, exploratory demeanor in new social situations, shows the opposite pattem. American family settings promoting such a pattern in self-concept include parental warmth, personal attention to children, family democracy and negotiation, and an absence of overcontro! in family discipline styles (Maccoby, in this volume). S. LeVine's ( 1979) description of the socialization of Gusii girls illustrates the difference. The Bantu Gusii are a patrilineal, patrilocal, horticultural society in Westem Kenya. Children are strictly controlled and punished and expected to assist in family subsistence and in child care (p. 384~: If a girl is taught to fear the unknown from infancy, required to be an obedient worker in the family labor force from early childhood, given no praise for conformity or accom' plishment, provided no parental approval for play or noninstrumental activities, and punished for deviation from parental command, she will become wary from calling attention to herself. Having learned that displaying good behavior does not result in positive attention, whereas misdemeanors particularly seeking fun with friends instead

358 DEVELOPMENT DURING MIDDLE CHILDHOOD of performing chores inevitably result in punishment if discovered, she will adaptively develop the "low-profile strategy" of social interaction. These child-rearing practices do appear to produce a child who is non- egoistic, not given to self-aggrandizement, and who functions by accom- modating to shared family obligations. Conclusion Children's self-conceptions are infrequently studied in cross-cultural re- search. Similarly, how the self is shaped by constraints of the ecocultural niche is not a common way to study the self in Westem research. But it Is precisely at the intersection of these two domains that new research on the self should be done with children ages 6-12. Clearly, the development of self-understanding is central during this period; for this reason, despite the dearth of comparative research on the subject, ~ have included some material on the self in this chapter. ~ have focused on the public nature of self- presentation and the effects of shared functioning, coregulation, an`] strict compliance training on the development of the self in non-Westem settings. New research is needed on what other aspects of community ecology may be related to self-concepts. Children's own views and ideas about themselves and their community are especially needed. Children's own voices are infrequent in the cross- cultural record. This is true in many respects of Westem research as well, to the extent that this work has relied exclusively on paper-and-pencil tests or formalized procedures. In addition, the dimensions of high and low self- esteem or coregulation versus self-regulation will perhaps need to be revised. Children in non-Westem settings are not always well described by "low self- esteem" or "shared" functioning. More elaborated, empirically based de- scriptions of the development of self in children of these ages in a variety of econiches are needed. TROUBLESOMENESS IN CHILDREN As American children enter schools, they come to the attention of teach- ers and others in public institutions. Shonkoff (in this volume) and Ach- enbach (in this volume) review the process of referral, diagnosis, and labeling for physical, mental, and reaming-related problems associated with this period. When children are in need of treatment or consultation of some kind, American parents intervene directly themselves or seek outside profes- sional services or both. The American goal is to change the child and the situation: if the child is having troubles in school, then improve his or her

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 359 school skills; if the child has behavior problems, provide counseling or special medical help. There seem to be many aspects of behaviors inside and outside school that do worry parents of children ages 6-12. Achenbach and Edel- brock (1981) identified ll8 such behavior problems and 20 social compe- tence items in their Child Behavior Checklist. Items ranged from truancy to cannot concentrate; refuses to talk; nervous; disobedient at home; feels unloved; and many others. Thus, many American children ages 6-12 have difficulties adjusting to outside institutions; they are the target of individually focused treatments; and there seem to be many areas of children's behavior during this period that are potentially troublesome. The non-Westem contrasts to the American child's experience provide a final example of the potential usefulness of ecocultural and comparative research for the study of children ages 6-12. It seems to be the case that: ( 1 ) children ages 6-12 in many non-Western settings are integrated smoothly (perhaps a better description is, without question) into the world of work, schooling, and community life outside the home; (2) a widely used mode for dealing with troublesomeness in children when it does occur is to change the child's family situation or activity setting rather than to focus on trying to change the individual child; and (3) compared with the large number of reported American parental concerns about their children, there are far fewer such troublesome behaviors either reported or observed in non-West- em studies of children ages 6-12. That is, children do not appear to be nearly as troublesome and/or their parents report far fewer behavioral troubles than do American parents and school or medical personnel. Unfortunately, there has been very little basic research done in cross- cultural samples on the naturally occurring behavioral problems that appear in children ages 6-12 (see Edgerton, 1976~. Similarly, it is startling to discover that there is no systematic account in the comparative literature- of which ~ am aware that compares cross-cultural treatments of children who are identified as troubled in some way. The suggestion that there is relatively less troublesome behavior among children ages 6-12 in non-West- em societies depends in part on the negative evidence that little is reported in the available literature. Some of the problems reported for American children depend on what definition the culture provides for a particular behavior pattern e.g., what do parents mean by poor peer relations? Others depend on cultural concep- tions of what a child is capable of or what is perceived as normal for this period e.g., do Tahitian parents fee! that children between 6 and 12 have a sense of personal, autonomous self-worth? Some Western-defined problems refer to public institutions, such as schools, courts, or welfare agencies, that do not exist in other societies. It is not known, however, which of these or

360 DEVELOPMENT DURING MIDDLE CH1LDH~D other differences in how troublesomeness may be reported by parents pro- duces differing patterns of behavioral problems in children ages 6-12. Basic comparative research is needed on what parents in other ecocultural envi- ronments report in the way of troubles for children in this period. Were we able to replicate Achenbach and Edelbrock's study (1981) in a large, cross- cultural sample, what descriptors of children's troubles would appear' Some items might appear on nearly all lists, some on only one or two, and some in one cluster of societies but not other clusters. In this way, we could begin to disentangle which behavioral problems appear to have some universal recognition and which do not. The appearance of troublesomeness in children's behavior depends in part on whether parents fee} that continuation of the behavior would cause the child to be unable to adapt or survive in his or her niche in the future. The widespread practice of sending children away to other kin or fostering them during the 6-12 age period is sometimes intended to change the child's environment in hopes that the child's troublesomeness will decrease. Al- though, again, empirical studies are needed, it seems that in general, treat- ments like sending a child to other kin are usually effective. The 6-12 age period seems to have relatively few children acting out or seriously troubling families, although covert tensions and difficulties with children are certainly present, as is the possibility of pathology (see Korbin, 1981~. Another reason for this apparently lower incidence of problems is the strong, generalized expectable climate of compliance in non-Westem fam- ilies described earlier in this chapter. Deference to adults is expected, as is submission to their requests and commands. Children ages 6-12 participate in training for [earned helpfulness expectations to act in a responsible and prosocial manner to others. It is possible that expectable compliance in the home and learned helpfulness among children of these ages may inoculate them against many of the behavior problems described in American parents' reports. Werner and Smith (1982) found that ecological (particularly house- hold personnel) features were most important in accounting for children's troubles during middle childhood in their longitudinal study of the children of Kauai; and they also found that nonmatemal and sibling caretaking played an important role in providing supports for resilient children those children who were at earlier risk, but without troublesome outcomes. Every one of these suggestions regarding children's relatively infrequent troublesomeness in non-Westem ecocultural contexts needs testing. None has been systematically studied at the present time. Both direct behavioral observation of children and the collection of parents' folk conceptions of troubles need to be obtained. The ecocultural niche differences that may -reduce troublesomeness should be studied at the same time as the data on

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 361 children's problems and troubles are gathered. The transitions from early childhood into middle childhood and from middle childhood into adoles- cence are certainly not necessarily smooth, and these boundary points also need new, comparative research. SCHOOLING AND LITERACY Each cohort of children ages 6-12 over the past two generations, as well as the one to come, is participating in a transformation unique in the history of our species: the spread of formal schooling and literacy around the world. The United States has nearly universal school attendance of children ages 6-12 and has one of the highest rates of literacy in the world; however, formal school attendance is far from universal in much of the world. Indeed, most nations are still in the transition to widespread literacy. Rogoff (1981) recently published a comprehensive review of the rela- tionship between schooling and the development of cognitive skills, such as perception, memory, classification and concept development, logical problem solving, and Piagetian tasks. When Western task and testing paradigms and materials are used, schooled subjects generally do better on such tests than nonschooled subjects. But Rogoff questioned this research strategy and pat- tern of results on many grounds and pointed out that the natural experiment created by different formal schooling in different societies has not begun to be exploited by basic researchers. First, research is needed to investigate the many threats to the general- izability of school-nonschoo! samples. For instance, parents "who allow or encourage their children to go to school may be wealthier, hew more modern attitudes, or hold different aspirations for their children than parents who do not" (Rogoff, 1981:267~. Children who are already better on skills assessed by Westem tests may have been selected by their parents to attend school. Schooled children may be more familiar with the test materials, testing situations, and the language in which the test is administered than non- schooled children. Tests given in school or based on school-related skills often do not appear to generalize to contexts outside the classroom in any event. Thus the differences between schooled and nonschooled subjects may be, in a variety of ways, an artifact of the tests, selection of children for school attendance, or the context-specificity of school cognitive abilities. A more telling research need and critique of existing research is the "lack of empirical research studying the mechanism for schooling's presumed ef- fect" (Rogoff, 1981:276~. Rogoff suggested four specific aspects of school experience that might be tested for in trying to discover mechanisms un- derlying the schooling effect (p. 286~:

362 DEVELOPMENT DURING MIDDLE CHILDHOOD (a) Schooling's emphasis on searching for general rules; (b) the use of verbal instruction out of context from everyday activities; (c) the teaching of specific skills in school, such as memory strategies, taxonomic categorization, and the treatment of"puzzles" in which the answer is to be derived from information in the problem; and specifically (d) literacy, which may allow the examination of statements for consistency or may simply teach some specific cognitive skills. Scribner and Cole (1981) questioned the generality of literacy effects in particular, and school effects more generally. Their Liberian study of un- schooled but literate Vai literate in an indigenous Liberian Vai script learned in the home indicates specific transfer effects for specific skills but not a generalized cognitive restructuring traceable either to literacy alone or to schooling. The challenges for new basic research in this area are of enormous im- portance. It is only through comparative work with children with different literacy experiences and different formal school experiences that effects of education can be distinguished from maturational and other age-related developmental differences. Educational comparisons (Epps and Smith, in this volume) and cognitive comparisons (Fischer and Bullock, in this vol- ume) between children ages 6-12 need cross-national studies in order to separate the effects of Western mass education and literacy from other in- fluences on development. Literacy and school skills, in this view, are specific cultural tools, aiding the attainment of localized skills reamed in a context in which such skills are needed and valued (see NerIove and Snipper, 1981~. What of other new Western cultural tools looming on the horizon, which go beyond books and literacy, such as the computer? What contextually specific, culturally lo- calized cognitive skills and changes in social-behavioral styles may appear as this new cultural too! continues to spread during the next generation? SOME COMMENTS ON METHODS The well-trained developmentalist prepared to study children ages 6-12 is a scholar with a diverse set of research skills packed into a traveling backpack. Depending on the circumstances, this researcher can do partic- ipant observation) various kinds of informant interviewing; formal controlled observation, using time and event sampling; experimental manipulations; tests and other kinds of structured tasks; and combinations of these as needed. The location of research work a school, a middle-cIass suburb in Chicago, a village in Western Kenya should not by itself determine the methods to be used. Nor should the substantive problem determine the methods. The study of achievement in children, for example, should never be limited to just a single method (Gallimore, 1981~.

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 363 Comparative research in human development has used a mix of qualitative and quantitative methods since Tylor If. Whiting and B. Whiting, 1960~. Qualitative, naturalistic research methods have developed a fairly substantial literature with recognized procedures to validate or compare research using these methods (Agar, 1980;Johnson, 1978; LeVine et al., 1980; Pelto and Pelto, 1978; SpradIey, 1979; Thomas, 1976~. Current research on field- work methods includes: role management in field situations; techniques for notetaking; methods for summarizing and coding field notes; systematic observation of behavior; quantification of field observations; styles and pro- cedures for writing up and presenting ethnographic materials; techniques for informant interviewing; and techniques for analysis and interpretation of texts. Naturalistic field methods will continue to be important in cross- cultural research. The basic research need is for more systematic attention to these procedures. The decision rules for which methods to use, under which circumstances, are particularly in need of attention. Better specification of the units for analysis would assist cross-cultural and Westem work alike. In this chapter, for instance, ~ have suggested the activity unit (Cole, 1981) or behavior setting (B. Whiting, 1980) as the link between the ecocultural niche variables and individual-level data typical of Westem studies. The activity unit consists of an individual, engaged in goal-directed activities, under the constraints of his or her localized niche. Events in such activity settings or units are regulated by others in the setting, by what the actor brings to the situation, and by the environmental cir- cumstances. ~ believe that methods need to be developed that take the activity unit as the unit of analysis not the individual actor alone, nor the thought or language of that actor, nor the localized environment. The goal for new basic research should be the development of methods suitable to a comparative theory of activity units. CONCLUSION The topics selected for more extended discussion in this chapter (the caretaking roles of children ages 6-12 and children's participation in work for the family; the public and nonindividuaiistic nature of the self; the possibly reduced troublesomeness of children ages 6-12 in non-Western cultures; and literacy and schooling) are included because each of these issues is an important developmental issue for American children ages 6- 12. These certainly do not exhaust the important topics that need new basic research using an ecocultural and comparative approach. Additional topics include, at least, the socialization of emotions and affect; beliefs about temperamental differences of children held by parents in other societies; the effects of urbanization and modemization on children ages 6-12; the com

364 DEVELOPMENT DURING MIDDLE CHILDHOOD parative phenomenology of childhood that is, children's own theories of development and accounts of their own behavior; and sex role and gender training. Reviews of these and other topics appear in several recent books (Levine and Shweder, no date; Munroe et al., 1981; Munroe and Munroe, 1975; Triandis and Lambert, 1980; Werner 1979~. Finally, an ecocultural perspective shows not only the marvelous diversity of children's environments in cultures around the world but also how vul- nerable children are to assaults on their safety and subsistence base. Children participate in a world economy; they can be exploited by governments, capitalists, socialists, and terrorists jUst as adults can. They suffer the con- sequences of insecticide poisoning, poor food distribution, distorted govem- ment, and social policies favoring special interests (see Davis, 1977~. The social processes that drive the increasing urbanization, modernization, and exploitation of the weak and the poor in third and fourth world countries are immediate threats to children of all ages. Isolated tribes and regions of great poverty within developed and developing countries deserve special study due to threats to the very survival of some of these peoples. REFERENCES Achenbach, R.M., and Edelbrock, C.S. 1981 Behavioral problems and compeeencies reported by parents of normal and disturbed children aged four through sixteen. Monographs of the Society for Research in Child De- velopment 46(1). Agar, M.H. 1980 The Professional Stranger: An luforrnal Introduction to Ethnography. New York: Academic ~ I Press. Bane, M.J., Lein, L. Stueve, A., Welles, B., and O'Donnell, L. 1978 Child Care in the United States. Working Paper No. 2. Wellesley College Center for Research on Women. Barker, R.G., and Wright, H.F. 1954 Midwest and Its Children: The Psychological Ecology of an American Tourn. Evanston, III.: Row, Peterson. Barry, H., III, and Paxson, L.M. 1971 Infancy and early childhood: Cross-cultural codes 2. Ethnology 10:466-508. Barry, H., Child, I.L., and Bacon, M. 1959 Relation of child Braining to subsistence economy. American Anthropologist 61:51-63. Beekman, D. 1977 The Mechanical Baby: A Popular History of the Theory and Practice of Child Raising. New York: Lawrence Hill. Berry, ].W. 1976 Human Ecology and Cognitive Style: Comparative Studies in Cultural and Psychological Ad- aptation. Beverly Hills: Sage-Halseed. 1979 Culture and cognitive style. In A.]. Marsalla, R.G. Tharp, and T.~. Ciborowski, eds., Perspectives on CToss-Cultural Psychology. New York: Academic Press.

ECOCULI-URAL NICHES OF MIDDLE CHILDHOOD 365 Blurton Jones, N.G., and Konner, M. 1973 Sex differences in the behavior of Bushman and London two- to five-year-olds. In ]. Crook and R. Michael, eds., Comparative Ecology and Behavior of primates. New York: Academic Press. Bronfenbrenner, U. 1979 The Ecology of Human Development: Experiments by Nature and Design. Cambridge, Mass.: Harvard University Press. Cole, M. 1981 Society, Mind and Development. Center for Human Information Processing Report No. 106. University of California, San Diego. Damon, W., and Hart, D. 1982 The development of self-understanding from infancy through adolescence. Child Devel- opment 53:841-864. Davis, S.H. 1977 Victims of the Miracle: Development and the Indians of Brazil. Cambridge, England: Cam- bridge University Press. Dunn, J. 1983 Sibling relationships in early childhood. Child Development August 54(4):787-811. Edgerton, R.B. 1971 The Individual in Cultural Adaptation. A Study of Four East African Peoples. Berkeley: University of California Press. 1976 Deviance: A Cross-Cultural Perspective. Menlo Park, Calif.: Cummings Publishing Co. Edwards, C., and Whiting, B. 1980 Differential socialization of girls and boys in light of cross-cultural research. In C. Super and S. Harkness, eds., Anthropological Perspectives on Child Development. No. 8. San Francisco: ]ossey-Bass. Gallimore, R. 1981 Affiliation, social context, industriousness, and achievement. In R. Munroe, L. Munroe, and B. Whiting, eds., Handbook of Cross-Cultural Human Development. New York: Gar- land Press. Gallimore, R., Boggs, J.W., and Jordan, C. 1974 Culture, Behavior and Education: A Study of Hawaiian-Americans. Beverly Hills: Sage Publications. Geertz, C. 1973 The Interpretation of Cultures. New York: Basic Books. Goody, E.N. 1982 Parenthood and Social Reproduction: Fostering and Occupational Roles in West Africa. Cam- bridge, England: Cambridge University Press. Gottman, ].M. 1983 How children become friends. Monographs of the Society for Research in Child Development 48(3). Chicago: University of Chicago Press. Hallowell, A.I. 1955 Culture and Experience. Philadelphia: University of Pennsylvania Press. Johnson, A.W. 1978 Quantification in Cultural Anthropology. Stanford, Calif.: Stanford University Press. Jordan, C., and Tharp, R. 1979 Culture and education. In A.]. Marsella, R. Tharp, and T.J. Ciborowski, eds., Perspectives on CToss-Cultural Psychology. New York: Academic Press. Katz, M.M., and Konner, M.]. 1981 The role of the father: An anthropological perspective. In M. Lamb, ea., The Role of the Father in Child Development. New York: John Wiley & Sons.

366 DEVELOPMENT DURING MIDDLE CHILDHOOD Konner, M. 1975 Relations among infants and juveniles in comparative perspective. In M. Lewis and L. Rosenblum, eds., The Origins of Behavior (Vol. 3). Friendship and Peer Relations. New York: John Wiley & Sons. 1977 Evolution of human behavior development. In P.H. Leiderman, S.R. Tulkin, and A. Rosenfeld, eds., Culture and Infancy: Variations in the Human Experience. New York: Academic Press. Korbin, I., ed. 1981 Child Abuse and Neglect. Cross-Cultural Perspectives. Berkeley: University of California Press. Lamb, M.E., ed. 1981 The Role of the Father in Child Development. New York: John Wiley & Sons. 1982 Nontraditional Families: Parenting and Cluld Development. Hillsdale, N.~.: Lawrence Erl- baum. Lambert, W. 1971 Cross-cultural backgrounds to personality development and the socialization of aggression: Findings from the six-culture study. In W.W. Lambert and R. Weisbrod, eds., Com- parative Perspectives on Social Psychology. Boston: Little, Brown. Lambert, W., Hamers, J.F., and Frasure-Smith, N. 1979 Child-Rearing Values: A Cross-National Study. New York: Praeger. Langness, L., and Frank, G. 1981 Lives: An Anthropological Approach to Biography. Novato, Calif.: Chandler and Sharp. Lee, R. 1981 Politics, sexual and nonsexual, in an egalitarian society: The !Kung San. In G. Berreman, ea., Social Inequality: Comparative and Developmental Approaches. New York: Academic Press. Leiderman, P.H., and Leiderman, G.F. 1973 Polymatric Infant Care in the East African Highlands: Some Affective and Cognitive Consequences. Paper presented at the Minnesota Symposium on Child Development, Minneapolis, Minn. 1977 Economic change and infant care in an East African agricultural community. In P. '' Leiderman, S. Tulkin, and A. Rosenfeld, eds., Culture and Infancy. Variations in the Human Experience. New York: Academic Press. Levine, H.G., Gallimore, R., et al. 1980 Teaching participant-observation research methods: A skills-building approach. Anthto- pology and Education Quarterly 6(1):38-54. LeVine, R. 1977 Child rearing as cultural adaptation. In P.H. Leiderman, S. Tulkin, and A. Rosenfeld, eds., Culture and Infancy. Variations in the Human Experience. New York: Academic Press. 1980 Anthropology and child development. In C. Super and S. Harkness, eds., Anthropological Perspectives on Cluld Development. San Francisco: Jossey-Bass. Levine, R., and Shweder, R. No The Acquisition of Culture. Manuscript in preparation, Harvard University. date Levine, S. 1979 Mothers and Wives: Gusii Women of East Africa. Chicago: University of Chicago Press. Levy, R.I. 1973 Talutians: Mind and Experience in the Society Islands. Chicago: University of Chicago Press. Madsen, M.C. 1971 Developmental and cross-cultural differences in the cooperative and competitive behavior of young children. Journal of Cross-CultuTal Psychology 2:365-371.

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD 367 Madsen, M.C., and Shapira, A. 1970 Cooperative and competitive behavior of urban Afro-American, Anglo-American, Mex- ican-American, and Mexican village children. Developmental Psychology 3:16-20. Mead, M. 1961 Coming of Age in Samoa. New York: The New American Library of World Literature. Medrich, E.A., Roizen, J.A., Rubin, V., and Buckley, S. 1982 The Serious Business of Growing Up: A Study of Children's IAves Outside School. Berkeley: University of California Press. Miller, D. 1963 Personality and social interaction. In B. Kaplan, ea., Studying Personality Cross-Culturally. New York: Harper & Row. Minturn, L., and Lambert, W. 1964 Mothers of Six Cultures. New York: John Wiley & Sons. Munroe, R.L., and Munroe, R.H. 1975 Cross-Cultural Human Development. Monterey, Calif.: Brooks-Cole. Munroe, R., Munroe, L., and Whiting, B. 1981 Handbook of Cross-Cultural Human Development. New York: Garland Press. Murdock, G., and Provost, C. 1981 Measurement of cultural complexity. In H. Barry and A. Schlogel, eds., Cross-Cultural Samples and Codes. Pittsburgh: Pittsburgh University Press. Nerlove, S., and Snipper, A. 1981 Cognitive consequences of cultural opportunity. In R. Munroe, L. Munroe, and B. Whiting, eds., Handbook of Cross-Cultural Human Development. New York: Garland Press. Nerlove, S., Roberts, J.M., Klein, R.E., Yarbrough, C., and Habicht, J. 1974 Natural indicators of cognitive development: An observational study of rural Guatemalan children. Ethos 2:265-295. Ogbu, I. 1981 Origins of human competence: A cultural-ecological perspective. Child Development 52:413 429. Pelto, Pal., and Pelto, G.H. 1978 Anthropological Research: The Structure of Inquiry. Cambridge, England: Cambridge Uni versity Press. ,' Pleck, ]. 1979 Men's family work: Three perspectives and some new data. Family Coordinator 28:481 488. Radin, N. 1982 Primary caregiving and role-sharing fathers. In M.E. Lamb, ea., Nontraditional Families: Parenting and Child Development. Hillsdale, Ad.: Lawrence Erlbaum. Ritchie, James, and Ritchie, Jane 1979 Growing Up in Polynesia. Sydney: George Allen and Unwin. Rogoff, B. 1981 Schooling and the development of cognitive skills. In H.C. Triandis and A. Heron, eds., Handbook of Cross-Cultural Psychology (Vol. 4). Developmental Psychology. Boston: Allyn & Bacon. Rogoff, B., Newcombe, N., Fox, N., and Ellis, S. 1980 Transitions in children's roles and capabilities. InteTnationalJourrml of Psychology 15: 181- 200. Rogoff, B., et al. 1975 Age of assignment of roles and responsibilities to children. A cross-cultural survey. Human Development 18:353-369.

368 DEVELOPMENT DURING MIDDLE CHILDHOOD Rohner, R. 1975 They Love Me: They Love Me Not. New Haven, Conn.: HRAF Press. Scribner, S., and Cole, M. 1981 The Psychology of Literacy. Cambridge, Mass.: Harvard University Press. Seymour, S. 1981 Cooperation and competition: Some issues and problems in cross-culeural analysis. In R. Munroe, L. Munroe, and B. Whiting, eds., Handbook of Cross-Cultural Human Development. New York: Garland Press. Shostak, M. 1981 Nisa: The Life and Words of a Mung Woman. Cambridge, Mass.: Harvard University Press. Shweder, R. 1981 Beyond Self Constructed Knowledge: The Study of Culture and Morality. Unpublished manuscript, Committee on Human Development, University of Chicago. Shweder, R., and Boume, E. 1981 Does the concept of the person vary cross~culturally? In A.~. Marsalla and G. White, eds., Cultural Conceptions of Mentor Health and Therapy. Boston: D. Reidel. Spradley, J.P. 1979 The Ethnographic Interview. New York: Holt, Rinehart & Winston. Super, C.M. 1981 Behavioral development in infancy. In R. Munroe, L. Munroe, and B. Whiting, eds., Handbook of Cross-Cultural Human Development. New York: Garland Press. Super, C., and Harkness S., eds. 1980 Anthropological Perspectives on Child Development: New Directions for Child Development. No. 8. San Francisco: ]ossey-Bass. Super, C.M., and Harkness, S. 1982 The infant's niche in rural Kenya and metropolitan America. In L.L. Adler, ea., Issues in Cross-Cultural Research. New York: Academic Press. Thomas, D.H. 1976 Figunug Anthropology: First Principles of Probability and Statistics. New York: Holt, Rinehart & Winston. Triandis,, H.C., and Lambert, W.W., eds. 1980 Handbook of Cross-Cultural Psychology. 6 vole. Boston: Allyn & Bacon, Inc. Weisner, T.S. 1976 Urban-rural differences in African children's performance on cognitive and memory tasks. Ethos 4(2):223-250. 1979 Urban-rural differences in sociable and disruptive behavior of Kenya children. Ethnok~gy 18(2): 153- 172. 1982a As we choose: Family life styles, social class, and compliance. Pp. 121-141 in ].G. Kennedy and R.B. Edgerton, eds. ~ culture and Ecology: Eclectic Perspectives. Washington, D.C.: American Anthropological Association. 1982b Fathering in egalitarian American families: Shared support and co-equal management. Paper presented at American Anthropological Association, Washington, D.C., Decem- ber. 1982c Sibling interdependence and child caretaking: A cross-cultural view. In M. Lamb and B. Sutton-Smith, eds., Sibling Relationships: Their Nature and Significance Across the Lifespan. Hillsdale, Ad.: Lawrence Erlbaum. Weisner, T.S., and Gallimore, R. 1977 My brother's keeper: Child and sibling caretaking. Current Anthropology 18(2) :169- 189. Weisner, T.S., Bausano, M., and Komfein, M. 1983 Putting family ideals into practice: Pronaeuralism in conventional and nonconventional California families. Ethos 11 (4):278-304.

ECOCULTURAL NICHES OF MIDDLE CHILDHOOD Wenger, M. 369 1983 Gender Role Socialization in an East African Community: Social Interaction Between 2- to 3-Year-Olds and Older Children in Social Ecological Perspective. Doctoral disser- tation, Graduate School of Education, Harvard University. Werner, E. 1 979 Cross'Cultural Child Development. A View from the Planet Earth. Monterey, Calif.: Brooks- Cole. Werner, E., and Smith, R. 1982 Vulnerable But Invincible: A Study of Resilient Children and Youth. New York: McGraw- Hill. Wertz, R., and Wertz, D. 1977 Lying-In: A History of Childbirth in America. New York: Free Press. Whiting, B. 1976 The problem of the packaged variable. In K. Riegel and J. Meacham, eds., The Developing Individual in a Changing World: Historical and Cultural Issues. Vol. 1. Netherlands: Mouton. 1980 Culture and social behavior: A model for the development calf social behavior. Ethos 8:95-116. The Genesis of Prosocial Behavior. In D. Bridgeman, ea., The Nature of Prosocial De- velopment: Interdisciplinary Theories. New York: Academic Press. Whiting, B., and Whiting, J. 1975 Children of Six Cultures. Cambridge, Mass.: Harvard University Press. Whiting, B., et al. In The Company They Keep: The Genesis of Gender Identity. Cambridge, Mass.: Harvard press University Press. Whiting, ]., and Whiting, B. 1960 Contributions of anthropology to the methods of studying child rearing. In P. Mussen, ea., Handbook of Research Methods in Childhood Development. New York: John Wiley & Sons. Altruistic and egoistic behavior in six cultures. In L. Nader and T. Maretzki, eds., Cultural Illness and Health: Essays in Human Adaptation. Washington, D.C.: American Anthropological Association. 1975 Aloofness and intimacy of husbands and wives: A cross~cultural sturdy. Ethos 3(2):183 207. 1978 A strategy for psychocultural research. In G. Spindler, ea., The Making of Psychological Anthropology. Berkeley: University of California Press. 1983