Understanding is Like Hair
Limited Cultural Sharing and the Inappropriateness of "All by All" and "Some by Some" Models for Swahili Culture
Akili ni nywele, kila mmoja ana zake: Understanding is [like] hair, each one has his [own] (i.e., people's views of everything differ).
There can be no question that nuclear family life works among the Swahili. Some families are broken by divorce or death. Some are hindered by endemic conflict. But most members of this community spend their lives within the framework of the nuclear family, beginning with the one founded by their parents and continuing with the one they found themselves. The nuclear family is not all there is in the lives of the members of this group, but it plays a central role for almost all of them.
Given this importance, the cultural elements concerned with nuclear family relationships and operation would seem at least as likely to be shared among its members as most other sets of understandings would in this or other sorts of groupings. If culture serves as the basis for group life—and if it does not, it is more than difficult to think what does—it unquestionably serves as the basis for the Swahili nuclear family. To the extent that culture's role in social life depends on sharing, it seems justifiable to expect that as much sharing would be found in the Swahili nuclear family as would be found in most other groups.
In this chapter, I examine the extent of sharing in Swahili nuclear families. This examination assesses the extent of sharing of cultural elements concerned with nuclear family life and relations among all Swahili ("all by all" sharing) and the extent of sharing among individuals belonging to the same nuclear families and among members of the same statuses. The theoretical foundation
for this has been adumbrated in chapter 1 and will become clearer as the data are presented.
It will be shown that three-eights of the beliefs, values, and procedures concerning nuclear family members, their relationships, and the group as a whole are not shared among members of the same nuclear family and that almost half of these cultural elements are not shared among community members who belong to different families. To simplify a bit, the operative (if often disavowed) view of culture and its operation used by social scientists is that people get along with one another and take advantage of the traditional understandings and values that make life possible through sharing these cultural elements with those around them. This sharing was once viewed as greater in "traditional" and "small-scale" societies than in urban and industrial societies, but everywhere it was sufficient to allow explanation to be based on the undoubted similarity in belief and values group members hold. In "complex" societies, the similarity was obscured by variation having limited effect on such groups as families, but in "simple" societies, the similarity is held to be manifest and easily seen. Durkheim's (1961:18) view regarding religion was not different from many others views of culture in general:
. . . the variations of ritual, the multiplicity of groups and the diversity of individuals [makes] the fundamental states characteristic of religious mentality . . . [difficult to find]. . . . Things are quite different in the lower societies. The slighter development of individuality, the small extension of the group, the homogeneity of external circumstances, all contributes to reducing the differences and variations to a minimum. The group has an intellectual and moral conformity of which we find but rare examples in the more advanced societies. Everything is common to all.
The classical and still powerfully influential view is that social life must be understood in the light of all, or at least all relevant , beliefs and values being uniformly shared among all group members. As was noted in chapter 1, a number of students of culture have shown that culture is not , in fact, universally shared. These findings have not, however, displaced the general explanation of culture's effectiveness as due to "shared beliefs and values."
As Holland (1987a :234) points out, some of what appears to be lack of sharing may not really be that but rather contextual differences, differences in expression, or simply errors. But the existence of these false appearances of variation does not gainsay the results of the growing number of studies that show real differences in the beliefs and values held by members of the same group. As seen in the proverb that opens this chapter, the Swahili themselves clearly recognize differences in the understandings held by individuals even if not all social scientists do.
The fact that sharing is limited is true not only of the Swahili but also of other groups, including four directly compared with the Swahili as concerns
the culture of nuclear family life (Swartz 1982a ). These limitations, moreover, are not limited to the complete inventory of a group's culture but also occur as concerns sharing among members of the same status categories. This fact presents difficulties to the position, first advanced by Ralph Linton (1936) in his formulation of "status" and "role," that regardless of limits in general sharing, sharing within social categories was sufficient to account for culture's effectiveness.
Individuals may not, according to this view, share everything with everyone else, but those with the same rights and responsibilities (i.e., the members of the same status) share the cultural elements concerning those rights and responsibilities. Put otherwise, this position holds that the members of the same status do share the cultural elements concerned with that status with one another even if these elements are not completely shared with those in different statuses. To be concrete, this view holds that mothers may not share all the beliefs and values concerning being a father, but the mothers do share with one another the cultural elements concerning being a mother.
Rather startlingly, the data presented here and elsewhere (Swartz 1982a ) suggest that even this modified view of sharing is inaccurate and, therefore, cannot be used as the basis for understanding the ways culture works. It would not be startling to find that mothers share the cultural elements concerned with the father status less than fathers do, but it is contrary to the Linton view of status to find that sharing among mothers is a good deal less than complete even as concerns the understandings directly involving and concerning the mother status itself. In fact, in many cases, sharing among individuals not belonging to a status about issues concerning the status is greater than sharing by status members.
This suggests that a sound view of culture's operation needs to proceed not only from an understanding of the limits in cultural sharing generally but also with attention to the incomplete sharing even among members of the same statuses.
"Status" is a key concept in the understanding of culture's operation. As noted earlier, a status is taken to be a collection of three distinguishable sorts of shared understandings (i.e., cultural elements). The sort concerning the distinguishing characteristics of category members is called "identifying understandings." The understandings about how category members are expected to act and how they expect others to act toward them in their capacity as category members are called "expectations," and it is important to note that these may concern quite specific behaviors (e.g., mothers cook food at mealtimes) as well as very broad ones (e.g., mothers are concerned with the welfare of their sons and daughters and act accordingly).
A final sort of status component can be called "salience understandings." These concern which status or statuses (vis-à-vis others) properly serve as a guide for behavior in particular situations and, when more than one status is involved, which ones are appropriate with what relative importance in guiding behavior.
It is important that "status" refers to nothing but a complex of cultural elements. As will be seen, it is through statuses that culture's constituent parts come to bear on the problems and opportunities of life (personal as well as social) of those who share the parts. Later, it will be shown that despite their purely cultural contents, statuses have an effect on behavior separate from that of the culture that constitutes them.
Measuring Cultural Sharing
There are two serious problems involved in the assessment of the extent of cultural sharing. They are doubtlessly part of the reason for the fact that only limited attempts to make this assessment have been carried out since Roberts's (1951) pioneering effort.
First, what it is, exactly, that actors do or do not share needs to be specified. If this difficulty is overcome, there remains the problem of determining whether this sharing is present. Put otherwise, how can we isolate the units that make up culture, and how can we determine whether these units are shared or not? Cognitive anthropologists have made progress in dealing with these two related problems (e.g., Berlin and Kay 1969). I follow one of their leads by keeping the immediate scope of investigation narrow and mainly limited to data that are readily quantified.
In the study focused on this issue (Swartz 1982a ), I took, of necessity, a provisional and partial approach to the two problems just noted. In this approach, the units of culture are identified with responses to items on a questionnaire (see Appendix and ibid., 335–338). Without doubt there are serious problems in taking responses to questionnaire items as equivalent to cultural elements. In my 1982 study, I used the same multiple-choice questionnaire (appropriately translated) for the Swahili and the other four European and American communities where the interviewing was done.
To take an obvious problem, it is quite possible that what I intended as different items are, in fact, reflections of a single cultural element and that what was a separate element of culture in one of the groups where we worked was not in another. The questionnaire was devoted entirely to questions concerned with nuclear family life. It asked informants to choose among alternatives concerned with issues—61 in all—such as who the family peacemaker was, how children should treat their aged parents, and where most children live after they marry. The status of these questions as cultural elements has
to remain in doubt, and the only basis for treating them as elements is pragmatic: if the problem of cultural sharing and cultural dynamics is to be studied directly, the effort must begin even if all the problems in making the study have not been solved.
A similar rationale was used to justify the approach to cultural sharing that was used. This approach is to take cultural sharing as present when two informants chose the same response to a question. A sharing coefficient was calculated based on the number of questionnaire items on which two informants made the same response as compared to the total number of responses from the two minus the number of identical responses according to the following formula:
So, for example, if two informants answering sixty-one questions each chose the same responses on all of them (61/122–61), they would have a sharing coefficient of 1.00.
It is true, of course, that people can have some—even a good deal—of cultural sharing concerning the subject of a question without choosing the same response and that they can choose the same response without much cultural sharing. Nevertheless, the sharing coefficient as defined for the study gives some approximation of what two actors share on the issues they are asked about, and there is no reason to believe that the responses are systematically biased toward either error.
For the Swahili (and the other four groups studied), we interviewed only people who belonged to families that met the following conditions: (1) they contained a wife and a husband who lived together at the time of the interview; (2) the spouses considered themselves and were considered by their child or children as the natural parents of the child or children included in the study; (3) the spouses had at least one child twelve years of age or older who lived with them and who had never been married. We inteviewed three members of each family meeting the above conditions in each society. We chose this number because it was the largest number of interviews we found it practical to get from a particular family while at the same time being large enough to allow us some measure of intrafamilial sharing across generational and sex, as well as status, divisions. The five-society study aimed at dividing each community's families equally between families represented by two children (siblings) and a parent and families represented by two parents (spouses) and a child (see table 3).
The coefficient of sharing for the total sample was obtained by comparing the responses of each member of the sample from that society to those of every other and taking the mean of all those comparisons. This latter coefficient is not really representative of the whole Mombasa Swahili community,
even allowing for the nonrandom sample (see below), because only individuals belonging to the sorts of families meeting our criteria are included. Thus there is not representation of adults who live alone or with others who are neither parents nor spouses or married people having no children or whose children are all either below twelve or married.
Because of these exclusions, the total sample as constituted is probably more similar to the selected family groups than to a more fully representative sample of the society as a whole. Therefore, the differences in sharing found in the sample as a whole and sharing found within families are probably understatements of the true differences in sharing within families and sharing among community members from different families (i.e., community as a whole). Since the objective is to examine the hypothesis that the level of sharing of all cultural elements is high enough to serve as a (perhaps the ) main source of culture's ability to function as the basis for social life in functioning groups, an overstatement of sharing is more acceptable than an understatement that might wrongly falsify the hypothesis.
As indicated above, the families studied were not randomly selected. A sample based on such selection was not possible because of the demands interviewing made on the families and because of the widespread reluctance to discuss even the most distant family matters with nonmembers. We compensated for this as much as possible by choosing families so that there were representatives of the various mitaa (neighborhoods) and the range of socioeconomic and educational standings.
It should be noted that there was a difficulty in identifying families that considered themselves and were considered Swahili which was not properly resolved until the data had been collected. Because of complexities concerning the affiliations of the occupants of statuses of members of families whose rather recent forebears came from the Persian Gulf area as well as of members of families who are suspected of having slave ancestors, the initial decision (made before I had discovered the extent and profundity of the status problems as concerns group membership) to include members of families whom Swahili interviewers decided belonged to the Swahili community proved unacceptable. Among the families included by interviewers were some, I discov-
ered, that undoubted community members did not accept as "true" or "full" members.
To try to deal with this problem, I assembled a panel of four middle-aged Swahili informants well acquainted with all sections of the group. Each of these men (mature women are difficult to employ for private sessions with other male informants and a male researcher) was, without doubt, viewed by others as a member of the community and of one of its long-established constituent families. These informants examined the names and other demographic data (but nothing else) collected from families interviewed and eliminated families they considered noncommunity members (i.e., considered members of other ethnic groups) with the result that the Mombasa Swahili are represented by only seventeen families despite data having been originally collected from thirty. All results of interviews here draw only on the seventeen undoubted member families.
It may also be that community members descended from fairly recent Arab immigrants are underrepresented if they told interviewers they were not Swahili. This is unlikely, however, since the interviewing was done by four young Swahili (three men, one woman) who consistently erred on the side of including doubtful community members. Moreover, some of those included are known by me to claim descent from Omani and Yemeni forebears who emigrated relatively recently.
The four societies compared with the Swahili in the original study need not be examined or discussed at length here. The characteristics of these families and the data collected from them are reported in the original paper, which presents the findings regarding cultural sharing from all five societies, including the Swahili, and compares the results along the dimensions to be examined here (see Swartz 1982a ).
The Limits in the Amount of Family Culture Shared by Family Members and Community Members from Different Families
Despite the difficulties in this approach to studying cultural sharing, it produces findings that can provide a starting point for further investigation in a largely neglected area of empirical research. It is not that the results of this study are beyond doubt but, rather, that they give indications of having some validity.
Some of the findings are what would be expected: members of the same family share more of the understandings concerned with family life and relations than those belonging to different families (table 4A below). Rather less expected is the size of the difference between sharing among same family
members and sharing in the community generally. Table 4 shows that within the family about five-eighths of the items (0.454 would be 5/8, 0.600 would be 3/4) are shared, while among members of different families in this long-enduring, well-integrated, and traditional community, less than half of all the items (0.333 would be 1/2) are shared.
The fact that cultural sharing is a good deal less than complete among members of the same families and also among community members is probably in accord with the assumption, now widely held, that "all by all" sharing does not occur. It may be, however, that its absence in so small and highly integrated a group as the nuclear family may carry this a bit farther than is generally envisaged.
The nuclear family was chosen as the venue for studying the role of cultural sharing in Swahili social life because it is a crucial element in this society and an undoubtedly effective group as concerns retaining its members and meeting an acceptable (to members) proportion of their social, emotional, and material needs. Given the stability and endurance of the families studied and the broad array and significance for members of the tasks taken care of within it, it is clearly warranted to view the culture of this group (i.e., the beliefs, values, and procedural understandings concerned with its tasks and relation-
ships) as effective. If the effectiveness of culture is taken to depend on the extent to which its constituent elements are shared, there would be every reason to expect a level of sharing here about as high as would be found among the members of any group whose activities cover a broad scope of life.
It is striking, therefore, that three-eighths of the cultural elements directly concerned with family life and relationships (as represented by our questionnaire) are nevertheless unshared among members of functioning, continuing families. This limited sharing may be less than surprising, but the fact that the sample includes only members of active, continuing families, all of whom are concerned with family life and relationships on a daily basis, makes it somewhat more interesting.
The idea that there is a Swahili family culture is not supported by the findings in table 4, if what is meant by "family culture" is a single set of generally shared cultural elements concerned with family life and relations. The ability to explain culture's effectiveness as the basis for Swahili family life (and the results from the other four societies are quite similar in this; see Swartz 1982a ) as a consequence of "shared beliefs and values" is put into question by these findings and made more doubtful yet by findings reported elsewhere here, especially in chapter 9.
Are the Swahili a "Homogeneous Society"?
Before turning to more detailed examinations of who shares at higher and lower levels with whom, it is useful to consider an issue of broad importance both theoretically and ethnographically: do the Swahili comprise a "homogeneous group" with its culture evenly shared among group members regardless of membership in such subgroups as families? If this were so, it could be argued that a distinction between "homogeneous" and "heterogeneous" (or at least less homogeneous) societies is an important consideration in how culture operates. Whatever may be found about culture's operation for the Swahili would quite possibly be different from what would be found in non-homogeneous societies.
The question, then, is whether or not the Swahili give evidence of being more homogeneous in cultural sharing than other societies that differ in composition from the Swahili. Some indication of the answer to this can be found by comparing Swahili cultural sharing to that of two other groups examined as part of my earlier study (Swartz 1982a ). Unlike the Swahili, these two groups are made up of ethnically diverse families, at least some of whose members had migrated to their current homes from other areas. Also unlike the Swahili, most family members in these other two groups associated with nonnuclear family kin only occasionally or rarely. Compared to the other two
groups that experience the isolation of families common to the urban life in the Euro-American setting, the Swahili community is much more nearly a "traditional society" whose culture might be expected to be more "homogeneous."
Despite having the traits of a "traditional society," table 5A shows that as concerns cultural sharing, the Swahili are not as "homogeneous" as might be expected. Although the Swahili family members do share with one another more than same family members do in the other two groups, La Jolla, one of the "heterogeneous" groups, shows more sharing among people from different families ("Total Sample"). This hints that viewing the Swahili as distinctly more homogeneous culturally than the other two groups is not fully warranted. The evidence becomes stronger with examination of table 5B.
Here we see the difference between family and group-wide sharing is substantially greater for the Mombasa Swahili than for either La Jolla or Kahl. This is just the reverse of what would be expected if Mombasa's homogeneity were reflected in a unformity of cultural sharing greater than that found in the other, more heterogeneous groups.
Extent of Sharing within the Family Versus Extent of Total Group Sharing
Table 5C and 5D shows that the level of cultural sharing within the family is more distinctive of the Swahili than the level of cultural sharing within the community at large. That is, the differences between the Swahili and the other two groups are larger with respect to internal family sharing than with respect to total group sharing. Further, if these two sets of differences are compared to the differences between internal family and total group sharing in table 5B, it can be seen that the level of sharing for the Swahili community is more similar to the levels of sharing in the two other societies than Swahili communitywide sharing is to Swahili family sharing.
This finding may be counterintuitive in that interaction, both within families and in the wider community, seemingly would act to increase levels of sharing. Yet the data here show that groups with no interaction among their members (i.e., groups in three widely separated parts of the globe) are more similar in levels of sharing than either coresident families or whole communities whose members interact. A hypothesis that might explain this is that the range of sharing associated with working societies (i.e., those still in existence and whose members are not all leaving in the immediate future) is narrower than the range associated with working families.
Some families, this hypothesis suggests, get along with quite limited sharing while others have substantially more, and this wide range is present in all the societies included in this study. Direct experience with Swahili families
suggests that some families get along quite well (carry out household activities, distribute money and goods, have members who seem reasonably satisfied with one another, etc.) on the basis of rather limited—both in scope and intensity—interaction. The members of these families talk to one another less, stay around one another less, and restrict interaction when it occurs. Other Swahili families are less restrained: they interact more and do so more unrestrictedly. But the latter families "work" quite as well as the former do in keeping together and accomplishing their members' ends.
At the community level, however, there is a narrower variation brought about by the relatively low limit on the level of sharing that can be attained in so large and diverse a group and the relatively high "floor" on what must be present if a group is to continue operating. The minimum needed for social continuation is, I suspect, largely made up of "tokens," which will be discussed in chapter 6, and status identifiers, which will discussed in chapter 7.
Members of different families have little need, or for that matter, opportunity, to deal with one another in nuclear family statuses. There is some pressure to be seen in conversations and to use relationship terms to praise or condemn (see chap. 7) people belonging to different families, and this may represent the basis for some part of the rather consistent, and low, level of sharing of family culture found among people from different Swahili families (and the comparison groups as well). Another part of societywide sharing is found in the agreement about who is what in the family, that is, what categories of people are found in households and who fits in them.
Less Sharing Among Members of the Same Named Statuses than Among Fellow Family Members with Different Family Statuses
As noted earlier, one of the first—and still one of the few—alternatives to the view that everyone in a society shares all culture with everyone else is the view that culture is distributed according to status membership. Linton (1936:113–115) holds that members of the same status category share more with one another—especially as concerns matters directly affecting the status—than they do with members of their society who do not belong to the status category in question. This view is probably more nearly in accord with observations of behavior than is the one that holds that all culture is shared by everyone. Nevertheless, there are a number of questions about sharing within statuses that can usefully be examined even if only with the limited data obtaining with questionnaires. Table 6 gives the sharing coefficients for the sample as a whole, within the families, and for the four basic family status categories.
By making comparisons among these, it is possible to consider a question of basic importance to the place of statuses as foci of cultural distribution: whether members of a given status share more with one another than with fellow community members in different statuses. For the family, the question is, do, for example, mothers share more of the understandings concerning family life with other members than they do with members of their own families given that the latter are not mothers?
Table 6 shows that for the Swahili and the comparison groups, the coefficients for sharing within the designated status categories (i.e., mother, father, etc.) are uniformly lower than those for sharing within families without regard to internal status differences. That is, members of the same family share more of the total family culture despite belonging to different statuses than do members of the same statuses who belong to different families. According to the sign test, this lesser sharing among members of the same status category as compared to sharing within the family is significant for the Swahili, and the other two groups, at the 0.01 level.
"Family Member" as a Status
This lesser sharing among members of the same named status is not obviously consistent with views of culture's operation that depend, as Linton's does, on the assumption that statuses are always and uniformly the centers of sharing of the cultural elements that concern the activities involving status category members as such. The findings here suggest that for family affairs, sharing is greatest among family group members without regard to status dif-
ferences within the group. This suggests that for the family, at least the part played by internal, differentiated statuses has been overemphasized and the part played by the status "family member" underestimated. This point is an important one because the data in table 6 indicate that the statuses that both group members and anthropologists think of as making up the family (i.e., mother, father, daughter, son) are less influential as concerns sharing family culture than the less obviously marked status, "family member."
The fact that family members share more of family culture with one another than they do with members of their particular statuses within the family does not mean that these latter statuses play no part in the distribution of culture. What it means is that the nature of the part played by statuses in bringing particular cultural elements into different situations cannot be assumed and may be different in uniformity for different statuses. Named statuses are surely involved in cultural distribution through greater sharing of certain understandings within the status than in other statuses, but this is not necessarily so for all understandings with all statuses.
Even with respect to total family culture, there are some differences in sharing associated with statuses other than family member. Table 6 indicates that for the Swahili, sharing among mothers and also among fathers is significantly greater (at the 0.01 level according to the sign test) than sharing among unrelated members of the same society but that other statuses do not exhibit this same distinctive level of sharing. The fact that this is true for mothers but not fathers in the two comparison societies is suggestive of the variation in the part particular statuses play as foci of sharing in different groups. Mothers and fathers from different families do not share as much with one another as members of the same family do, but they do share more within their statuses than unrelated people from different families ("Total Sample") do.
This may indicate that these statuses play a distinctive part in Swahili family and community life in that they are particularly important in making family life more similar in different families than it would otherwise be. In simply understanding things more as their counterparts in other families do, the parents exert a homogenizing influence quite apart from whatever their specific behaviors guided by those understandings may be.
This implies that the basis for Swahili society is better understood if, in addition to the distinctive expectations and saliences of statuses, there is also information about the extent of sharing within those statuses. Currently used views of cultural distribution seem to assume that statuses are highly similar in the uniformity of their members' sharing of the elements associated with them, but evidence here suggests that some statuses involve more uniformity (i.e., sharing) in members' understandings and others involve much less.
It may be that statuses have levels of sharing quite as characteristic of them as are the particular elements of culture they share. Mothers may be the main
or only actors in a society who have the understandings needed, for example, to deal with distraught children in ways that will be generally approved, but it may be equally important to the way the society's family culture works that it is mothers who share more of it across family lines rather than do daughters or fathers. Similarly, if there were a society in which no internally differentiated status had members who shared more with their counterparts in other families (and none of the three here are like this), that would probably be associated with a family culture working quite differently from one in which there was at least one status with greater cross-family sharing regardless of what that status was.
The clearest way in which sharing beyond the family's boundaries can influence the culture of the broader group is through the sharers exerting similar influences in their different families and thereby bringing about some pressure toward general uniformity. Sharing culture with fellow status category members in other families need not lead to pressures for homogeneity, but it does provide a necessary base for such pressures.
It also provides a possible base for conflict. This would be found in a family where the father and mother share more with fathers and mothers in other families than they do with group members generally, while sons and daughters do not have the same higher sharing with their counterparts in other families. In many Swahili families, the children, especially sons, label their parents as "old-fashioned" and "too strict." In some part, this may be because the parents are likeliest to share what is identified with tradition and, given the substantial sharing between spouses, bring the children to feel subject to old-fashioned treatment. The children, often characterized by parents as "difficult," need share little with one another or the parents to gain that characterization and to oppose the parents' views.
Status Membership and the Sharing of Status Culture
Members of the various named statuses in the family have been seen to share less of total family culture as represented by all the items on the questionnaire than do those in what might be called the "same family member" status. It would seem, however, that if what is measured is sharing of the cultural elements directly concerned with their own status and its relationships, the members of any status category could hardly fail to share more than others share those same elements.
Since fathers and mothers from different families share more across family boundaries than members of other family statuses do, an examination of their sharing provides a useful test of the hypothesis that understandings concerning statuses are shared more fully by those who are classified in those statuses than by those who are not.
Tables 7 and 8 present the results of comparing informants' responses on questionnaire items (all available in Swartz 1982a :335–338) concerned with the mother/wife and the father/husband statuses, respectively. Table 7 concerns the elements that make up what can be called "the mother/wife scale." This scale is simply a list of questions concerning beliefs and values bearing on the mother/wife's behavior. It has five component questions, and the comparable scale for father/husbands has the same number.
The items in the two scales differ only in their subject, with each focusing on the status concerned. Thus, one question in each scale asks whether or not wives advise husbands (for the mother/wife scale) or husbands advise wives (for the father/husband scale) about the work the husband or wife (respectively) does; another asks who wives (or husbands) consult when something is worrying them; another asks whether or not most wives (or husbands) are happy in their marriages; and the final one asks how much fathers (or mothers) take into account the interests and wants of other family members in what the fathers (or mothers) do.
Obviously, the questions in these scales are removed from behavioral reality, but taking the scales as wholes, it seems reasonable to argue that if members of a status do share more of the cultural elements concerning their status with one another than they do with members of other statuses, this will be seen in the results of tabulating responses to these questions. In fact, such status-centered sharing is not what is found for either the father or the mother status.
In Part A of tables 7 and 8 are the coefficients of sharing for the mother/wife and father/husband scales among pairs outside the family (i.e., "Total Sample"). If it is true that sharing of cultural elements concerned with a status is at a higher level within that status, table 7 would be expected to show that the pairs of mothers (Mo-Mo) have the highest coefficients and table 8 would show that the pairs of fathers (Fa-Fa) have the highest coefficient.
In Part A of table 7, it can be seen that neither for the Swahili nor for the comparison groups do mothers have the highest sharing coefficient for the mother/wife scale. Part A of table 8 shows that fathers have the highest sharing coefficients for the father/husband scale in one of the comparison groups but the lowest coefficient in the other and somewhere in the middle for the Swahili. In neither case does the data strongly support the Lintonderived hypothesis that those classified in a status category share items directly concerned with their own status at a substantially higher level than do those in other categories.
Thus, for both mothers and fathers, our data drawn from comparisons among members of the same societies but different families fail to show a higher level of sharing among those actors classified in the same status category even when the elements in question are all directly concerned with that status.
Family statuses may well be quite different from such technically focused statuses as surgeon, potter, or navigator, among whose members we would expect far more sharing of the cultural elements directly concered with the status than among nonmembers. The fact that there are any statuses whose members do not clearly share cultural elements concerning the statuses at a higher level than nonmembers is striking and the additional fact that these statuses are part of so broadly important a group as the nuclear family makes the finding even more challenging.
However distinctive the mother and father statuses may (or may not) be as concerns sharing, the data collected for this study do seem to raise serious doubts about the fruitfulness of continuing to assume that all statuses are equally uniform foci for cultural sharing with all that entails for the way culture operates as a guide to social life. The role of statuses in cultural distribution has long been taken as a central one and as the main alternative to an all by all model of cultural sharing. This alternative model, some by some, has received little empirical attention since Linton introduced it, although work by Holland (1985, 1987a ) provides a notable contribution. The indications from this study, and they are in accord with Holland's findings, are that, useful as it is, it requires a good deal of elaboration and modification if it is to serve as an adequate basis for understanding culture's operation.
It is not that statuses have no role in the distribution of culture but rather what role they have and how they can function with only limited sharing among those who occupy them requires examination. Even granting that all cultural items shared among some members of a group are not necessarily shared among all group members, we still are called on to account for the finding that classification together in a status does not ensure a high level of sharing of even the cultural elements concerned with that status.
As noted in chapter 1, the all by all model of cultural sharing is no longer respectable, and Holland's work together with the findings here suggest that even the some by some model (i.e., the portion of the group's total cultural inventory which concerns the operation of a status is shared by those who occupy the status) cannot be used with confidence.
The fact that the understandings concerning statuses are by no means uniformly shared among the occupants of the statuses in question throws into doubt some aspects of the view formulated by Goodenough (1965) and elaborated by Keesing (1970) that if we fully grasp the schemata in the minds of status occupants, we will have most or all of the basis for understanding how culture operates as the basis for social life.
Important as the distribution of culture by statuses is, recognizing its existence by no means provides all—or even most—of the answers to the question, how does culture work as the basis for social life if it is only partially shared? The data just considered suggest that not only is there substantial variation in the general sharing of culture but there are also important differences
in sharing among the occupants of the same status even as concerns those statuses directly.
Relationships: Do Participants Share Their Culture More?
One area of culture that has not yet been considered and that might provide the expected support for the idea that status members share more among themselves than with others remains to be examined: relationships between members of different statuses. The scales for mothers and for fathers concern elements dealing with what the status category member says he or she does, values, or believes, but examining elements concerned with joint or reciprocal activities involving relationships between members of different status categories might produce results more in accord with conventional theoretical assumptions concerning status membership and cultural sharing.
Table 9 reports the sharing of answers to questions concerning the relations between parents and children. There are ten such questions, and together they form the parent-child relationship scale. The questions in the scale ask about such things as informants' beliefs about how jointly owning property affects parent-child relations; how best to avoid parent-child conflict and deal with difficult problems; whether parents and children love one another; and what sorts of things parents and children do together (Swartz 1982a :336–338).
Part A reports sharing on the parent-child scale among informants from different families. This sharing outside of the family shows that for the Swahili, same generation pairs are about as high in sharing as cross-generation pairs are so that the relationships seemingly subject to understandings concerning parent-child relations fail to show consistently higher sharing than other relationships do.
These findings are particularly noteworthy since, as our earlier examinations showed, those assigned to status categories failed to share at distinctively high levels either general family culture or the cultural elements concerning their own statuses. The absence of particularly high-level sharing in statuses and relationships where it would be expected given the Linton view of how statuses operate will be discussed below. For now, the point is simply that unrelated members of the participant statuses in the parent-child relationship do not share our scale for that relationship at a notably high level.
Part B of table 9 presents data on sharing within the family, and it, too, offers little support for the idea that the participants in a relationship share culture concerned with that relationship at a uniquely high level. For the Swahili, and the comparison groups, a parent-child pair is the one that shares the relationship scale at the highest level, but for one of the comparison groups, it is a parent-child pair that shares at the lowest level.
This can hardly be taken as strong support for the assumption that members of the two statuses participating in a relationship share more of the culture concerning that relationship than do any others. Once again, then, the evidence indicates that for the nuclear family, the relationship between status membership and cultural sharing appears to be a good deal less clear than would be expected according to the Linton view.
Table 10 offers a somewhat different perspective on the relationship between being party to a relationship and sharing the cultural elements concerned with that relationship. This table reports sharing of the items in the spouse relationship scale. There are six items in this scale concerning such jointly or reciprocally relevant issues as who wins arguments between spouses, whether spouses share friends or not, and which could get along best without the other (see Swartz 1982a :337–338).
Part A reports sharing among actors belonging to different families. Only among the Swahili is the mother/wife and father/husband pair the one that shares at the highest level for members of different families. Part B, which reports sharing among members of the same family, gives different results, however. Here we see the statistically nonsignificant result that the mother/wife and father/husband pair is first not only for the Swahili but for both comparison groups as well.
This last offers some support for the intuitively appealing notion that parties to a relationship share cultural elements concerned with that relationship more than others do. More broadly, however, the results are quite mixed. We saw that for both of the relationships examined, members of the directly involved status categories in the society at large (i.e., informants from different families) did not share at a consistently higher level than did members of other status categories. For the parent-child relationship, the same lack of higher-level sharing among members of the directly involved statuses was also seen for members of the same family. These findings also fail to support a close association between status occupancy and high levels of cultural sharing.
However, the findings regarding sharing of the spouse relationship scale by members of the same family (table 10B) support the traditional view of statuses despite the fact that they are just short of statistical significance. It appears that spouses who are actually married to each other may have more sharing of the spouse scale than do spouses who are not married to each other (i.e., husbands and wives from different families) and also more than do other members of their families. If this is so, it suggests that spouses in the same family do not bring the cultural elements concerned with their relationship into that relationship—spouses in different families would show more sharing than they do if that were so—but, rather, that they developed shared elements in the course of their marriage. This would be what Wallace's "equivalence structure" (1970:27–36) view of cultural distribution would predict. That is,
through associating with one another, people work out understandings of what is happening (including what the partners in interaction will do) that are equivalent but not necessarily even similar in different families.
It seems very likely that in some family relationships, the participants do develop the sort of "structures" (i.e., collections of distinctively shared cultural elements) that Wallace alerted us to. It is my distinct impression that this takes place in developing relationships in the Swahili community and that, for example, the men in the same baraza (see chap. 4) over time come to share understandings about, at least, what topics will be discussed, when to say prayers, and what refreshments will be served. Men join the group with some understandings brought from earlier experience with other groups (including those involving members of their households which took place when the men were quite young) and develop new ones specific to the group they now participate in through their experience with it.
It is important to note that the understandings vital to groups and relationships are not always just those that adult participants bring to them when they begin their participation but also may well include understandings developed through experience. It is less than startling that such development of cultural elements is important in some relations and groups, with the processes of becoming accustomed (zoea) to people and groups a recognized occurrence among the Swahili and in many other societies. That it should occur for children and the relationships involving them in the nuclear family is what would be expected given the family's role in enculturation. That it should be so central in the spouse relationship in an ancient and relatively stable group like the Swahili is, perhaps, less obviously in accord with how culture is often thought to operate in a "traditional society." More generally, the statistical data support the general proposition that culture, the set of understandings shared among those in interaction, is fluid even in such well-established groups as the nuclear family and even in societies as traditional as the Swahili.
Neither "All by All" Nor, without Modification, "Some by Some"
The findings of this comparison of survey interviewing data indicate that substantial revisions in traditional views of cultural distribution (i.e., who shares what with whom) are called for. The data here show that the once-prevalent view that all culture is shared by all members of a society is, as expected, without factual foundation even in a small and long-established group like the Swahili. This is worth noting because the all by all view of sharing, despite its universal disavowal, has by no means lost its place as the basis for formulations explaining how culture, either that of particular societies or generally, actually operates. What is called for is a useful alterna-
tive to an approach to cultural dynamics that does nothing more than invoke "shared beliefs and values." The processes whereby beliefs and values affect those who do not share them have received less attention than they require, and giving this attention begins by demonstrating the inadequacy of formulations based, explicitly or implicitly, on all by all sharing.
This is made more challenging by the findings that show the main alternative to "homogeneous" cultural sharing, Linton's status-centered approach (the "some by some" view), is also far from well supported by the facts. The limited sharing within statuses does not gainsay the part statuses play in cultural dynamics, but it does raise new questions about how they operate.
In chapter 6, where attention moves beyond the family into the general community, we will see that as concerns quite different sorts of understandings and quite different statuses, the sharing by (and about) members of mainly age- and gender-based statuses is extensive despite the fact that these statuses do not have the linguistic marking internal family statuses do. In that discussion, it will be suggested that "specific expectations" regarding statuses play a quite different part in social life than "general expectations" do and that "identifying understandings" (the cultural elements that serve as the basis for assignment to cultural categories) have a unique role in cultural operation.
These results indicate that it is as important to refine and develop the "some by some" model of cultural sharing (that is, Linton's basic view as he formulated it) as it is to reject the all by all model. Statuses are crucial to the distribution of understandings, but among the Swahili (and elsewhere), sharing within statuses is quite incomplete so that members of a given status cannot be assumed to share with one another all understandings seemingly relevant to that status in the various contexts and situations in which the status is involved.
If this last is so, as the data in this chapter suggest, an approach to culture's operation that is based wholly on finding the schemata in the minds of status occupants cannot explain culture's ability to serve as the foundation for social life. This is a consequence of the finding that some of the components of these schemata, including those directly concerned with the statuses, are different for different status occupants; that is, they are not shared.
This limited sharing of culture does not prevent statuses from operating, as will be seen in the following chapters. This is vital in a number of ways that go beyond the functioning of social relationships. As this discussion develops (and especially in chap. 10), it will become clear that a crucial aspect of culture's operation can be understood as a result of the "organization of culture" (a concept to be introduced later but concerned with how the various understandings group members share are related to one another). This organization is not entirely the result of shared understandings that put different cultural elements in relationships with others as some understandings (e.g., "it is better to be liked than wise") do. A vital part of cultural organization
comes from the ways statuses operate to make available the results of the guidance of understandings to those who may not themselves share those elements.
Dealing with the Fact of Diversity
But before addressing the problems of cultural organization, it is useful to consider how people deal with the fact that many of the understandings that are basic to life are not shared by all those around them. The recognition that each person is different from every other in what he believes, in what he values, and in what he wants is probably universal. A Swahili proverb reminds those who are pleased with achieving a desired end, Kizuri kwako, kibaya kwa mwenzako : Your good thing [can be] a bad thing for your companion.
This proverb and the recognition of differences it affirms is like the one that gives this chapter its name: Intelligence is [like] hair, each one has his [own]. Another that makes the same point says, Penye wengi pana mengi : [Where] there are many people, there are many views.
The diversity seen in the data on cultural sharing is not only a finding of research but it is explicitly familiar to the community members themselves. Yet group members have to be able to predict, and to believe they can predict, within broad limits what those around them will do if social life is to continue (e.g., Parsons 1964 :10, 27–29, 36–37). This suggests that people, like anthropologists, do not allow their knowledge of diversity to interfere with their belief in uniformity as concerns important values and beliefs. The next chapter examines a main device Swahili use to obscure the absence of general sharing.