10
A Wife is Clothes
Family Politics, Cultural Organization, and Social Structure

Mke ni nguo, mgomba kupalilia: [A] wife is clothes, [a] banana plant [is] weeding.

We have seen that the relations between people, as guided by the general expectations in their statuses, provide a basis for those who consider themselves ill to choose a practitioner to treat them. In a similar vein, I will show that women's ability to get their husbands to spend money in ways the wives, but not the husbands, consider useful is mainly the result of a complex of role relationships including, but not limited to, that between spouses. The expectations involved in a number of those roles will be shown to play a central part in family politics generally and the power of wives in particular.

In fact, the central hypothesis here is that the power of Swahili wives in their relations with their husbands is mainly due to the distribution of culture by statuses and to the cultural organization that results from this.^[1] The key element in this distribution is the unique set of expectations for men in the husband-wife role. According to the hypothesis advanced here, it is this distribution that explains the substantial power wives have with respect to their husbands, despite widespread sharing of explicit understandings, viewed as divine in origin, that specifically bar such power.

As seen earlier, the larger-scale structures of social relations as they existed in the various taifa ("tribes") and in the Nine Tribes-Three Tribes dual organization (see chap. 3) have declined in importance. Even neighborhoods have mainly stopped providing a framework for joint or cooperative activity despite their continued salience in personal identification and as a base for multiplex relationships outside the nuclear family.

The household and, more specifically, the nuclear family retains its position as the main center of social participation, division of labor, and the sharing of resources. It has taken on added significance with the decline of

larger-scale groupings in a way characteristic of "modernizing" societies (Bott 1971:124 passim). The distribution of power within this vital family grouping is of substantial interest in itself and takes on additional significance in providing an important demonstration of how Swahili culture works in guiding the behavior of the community's members.

In the last chapter, we saw that social structure operated to shape people's plans regarding medical care through a series of linkages dependent on the general expectations the patients had of advisers in statuses such as mother, father, spouse, and neighbor. These "linkages" were in the form of acceptable ("attractive" would not be too strong a term) advice that made it possible for individuals to formulate plans of action aimed at dealing with their understanding that they are ill. Important as the giving and taking of advice is in transmitting the effects of culture to those who do not share the elements in question, I show here that there are other ways social structure affects culture's operation.

Power in Marriage

Mombasa Swahili informants of both sexes agree that with the exception of such strictly domestic activities as cooking, cleaning, and caring for young children, husbands have, as shared understandings hold they ought to, complete control of all affairs involving their wives and households. Observations and reports of specific incidents across a considerable spectrum of life's activities bear out these reports. Despite this, however, in an area of life the Swahili consider highly important to the household, it is the wives who get what they want even though the husbands say they do not share their spouses' goals in this area.

What is of central concern here is wives' power, with "power" being understood in the Weberian way as "the probability that one actor within a social relationship will be in a position to carry out his will despite resistance" (Weber 1947). The examination of wives' power will include an examination of the forces that lead them to seek ends they know their husbands find unattractive. Following this, the sources of wives' success in getting what they want will be considered, with particular attention to the fact that a substantial body of shared understandings holds that wives can never bring husbands to do what they do not want to do. Finally, attention will be directed to how wives and husbands explain and justify the wives' control in important contexts despite the broad and strong cultural foundation for husbands' monopoly on power.

In all of this, the partly autonomous social lives of men and women (see chap. 4) is a central fact. Based on the Swahili understanding of the separation of the sexes (gawa in Swahili) called for by Islam, men are mainly occupied

in the neighborhood in its widest sense and, beyond that, in the city of Mombasa where they work, while women's main activities focus on the household and intensive interaction with other women, most of whom are either closely related and/or live in houses immediately adjacent to the woman's own.

Women's Social Relations and Prestige

The Koran requires every man to provide his wife with adequate clothing as well as food and shelter, but save for the small group of very poor families among the Mombasa Swahili, "adequacy" is not what women seek.^[2] Western dresses or traditional women's garments, leso , are worn under the all-encompassing traditional veils, so that they are seen only by other women and by men of the wearer's immediate family. Nevertheless, this clothing plays a crucial part in the women's lives. "Allahu akbar!" (God is great!) women sometimes exclaim when they see a particularly desirable new fashion in a shop window. For many women, the intensity of their joy when they get such a garment is matched only by the intensity of their despondency should they fail to get it.

Women say that normally they buy one new dress or leso a month and that they try to get particularly fine ones for major social events such as weddings. When a woman arrives at the area set aside for females at a wedding celebration, she is closely examined by the others there after she removes her veil. What she wears under the veil will be fully discussed by her fellows over the next few days or weeks. Everyday clothing is important to the standing a person occupies in the sense that it provides a basis for demonstrating good taste and, especially, the ability to pay for attractive clothing. It is the special dresses and leso worn at festive events, however, that lend real distinction in relations with other women. Similarly, women who have an armful of the 22-carat gold bracelets (bangili, pl. mabangili ) Swahili women often wear are the object of favorable comment, while those who have few or none of the expensive bangles are dismissed as paupers or as women who have failed to win and keep their husbands' love.

Having many bangles to wear at the weddings and parties when "best clothes" are called for will not by itself make a woman prestigious, nor will the absence of a respectable number of them totally destroy a woman's standing among other women. Character, piety, skill as a mother and household manager, reputation, and similar considerations are also sources of prestige. As with desirable clothing, however, the wearing of a substantial amount of gold jewelry is important, and failure to do so is a serious detriment to getting or keeping prestige.

Prestige is also connected to staging the rituals at which the elegant clothing and jewelry are worn. The marriage of a child, either a son or a daughter, can be the occasion for a celebration on a genuinely large scale. If a woman

decides to hold such an event, and it is extremely rare for a man to make this decision, she will be occupied cooking great quantities of the most elaborate foods for many days and all night the final night before the ceremony. She will be aided in her efforts by most of the Swahili women in her neighborhood who will dance and sing as they work. The celebration traditionally lasts for a full week, and even now it sometimes continues for two or three days. The food, clothing, decorations, and orchestra involved can cost so much that few families giving them fail to exhaust their surplus funds. Sometimes debts are incurred to the extent that homes have to be mortgaged and, occasionally, ultimately lost. Funerals are not, of course, occasions for festivities, but women do stage very expensive ones providing the best sorts of food for virtually the whole community for an entire week.

Both men and women come to the weddings and funerals, but women phrase their interest in staging them with respect to other women. Women say, "I have gone to their weddings (funerals) for many years and now I must pay them back." A woman who gives an outstanding ritual will be remembered as someone with a real accomplishment to her credit, and a woman who fails to give any is likely to be consigned to the category of the impoverished or incompetent.

Both finery and rituals depend on money, which women do not normally control. Occasionally, an heiress will have sufficient funds of her own to outfit herself as she wishes and to stage rituals as she sees fit, but the vast majority of women get the funds they need for these things from their husbands. The crucial fact for present purposes is that although men do not share women's enthusiasm for spending money on clothing and rituals, it is only through them that most women can get the money necessary for the things that will protect or enhance their prestige among other women.

Men do not usually talk together about women, but when they do so, a good part of the time is spent railing against their extravagances, especially as concerns the women's interest in finery and rituals. No one who has heard Swahili men discussing expenditures for their wives' clothing and ceremonies can doubt that men are, at best, unenthusiastic about using money for those purposes, at least in the quantities the wives seek. The wives themselves are perfectly well aware that their husbands are not at all inclined to spend money on clothing, marriages, and funerals and that they must be brought to do so. This in the face of the fact that the women agree with the men that husbands are ideally and actually quite powerful and have every right to do just as they wish in their households without regard to what their wives may want.

The importance of admirable clothing and expensive jewelry rests in other women's response to it, but at least part of this response derives from what women understand about men's expressions of love. When women are very angry at one another and want to hurl the most telling insult, they say, "No man could love you." A woman who is unloved by her husband is taken to

be a pitiful creature without the physical or character traits that are most prized by both sexes. Men's love for their wives, several different informants report, is shown in a number of ways, of which the most publicly visible is giving expensive jewelry and clothing.

It is perfectly well understood that men of modest means cannot give as much as wealthier men can, but, as other women understand it, women who are loved will manage to get at least an acceptable minimum of visible signs of their husbands' love. Thus, at least part of the reason that a woman's standing among other women is influenced by the clothing and jewelry she wears is that they are taken to indicate her standing with her own husband. The fact that the women themselves view the men as well within their rights in refusing the needed money is part of the reason women seek the money, since its availability to them shows they are loved, with all that entails.

Gender Statuses and Salience Understandings

Before going further, it is useful to examine the statuses that are involved in the relationship between the husband and the wife. While the wife status is unique among all his relationships in the expectations it provides the man in the husband status, wives' power cannot be traced to this alone.

A key element in that power derives from the status of woman. The usefulness of positing the presence of a status is closely related to the observable effects of the understandings that make up that status. On this basis, gender statuses have undoubted usefulness in understanding Swahili behavior. Women are expected to behave differently from men when they are categorized in a variety of statuses, and the differences have a consistency that indicates clearly formulated expectations for each of the gender statuses even when they are occupied simultaneously with other statuses.

These gender statuses only rarely guide behavior by themselves, but their salience understandings lead to their expectations being joined with those from a considerable range of other statuses to guide behavior in a variety of circumstances. The only other statuses that appear to be even roughly equivalent in the extent to which their expectations occur together with others are the age statuses and the status, community member.

The joint influence of several statuses can be inferred when an individual's behavior meets the different expectations of several statuses in a single situation and relationship. A woman customer, for example, is expected to behave differently from a man customer, and the same is true of such broader statuses as neighbor, friend, or rival. The gender statuses bring similar expectations of their own to the other statuses with which they jointly occur and show their influence in cross-cutting similarities in otherwise quite different nongender

statuses. Clerks and neighbors are different, but men and women clerks differ in some ways similarly to the differences between men and women neighbors.

This is not to suggest that the salience understandings in the gender statuses give them equal importance when they combine with all other statuses. Taking only statuses I know to be actually occupied by community members, women bank tellers are not expected to behave in a markedly different way from men bank tellers, and the differences between both religious and secular teachers on the basis of their gender statuses are real but limited. However, women neighbors are expected to act quite differently from men neighbors, despite similarities inherent in the expectations of anyone classed as a neighbor, and, in a simplex relationship, women customers are expected to act differently from men customers.

The expectations associated with the woman status involve openness, affectivity, and engagement. I had no opportunity, of course, to spend time with groups of women in "natural" settings, but in talking with women, it was striking how outgoing, responsive, and emotionally active they were as compared to men in a similar situation. The same differences I observed are noted explicitly by members of both gender groups who report themselves and the differences between themselves and the other gender group to be much as I observed. The gender statuses in this community are not only influential across a quite wide range of other statuses but they are explicitly recognized as being so. Further, informants are uniform in asserting that the differences between the genders in their various statuses are aesthetically and morally important. Women behaving in accord with a variety of different statuses are not only recognized to be different from men in many of the same statuses but the differences are explicitly valued.^[3]

Gender Statuses and Differences Between Them

Virtually all adult women maintain close social relationships with their neighbors, especially those, who may or may not be kin, from houses only a few meters from theirs. For some purposes—weddings and funerals are the most frequent and notable—relations with all the women in the mtaa (see chap. 3 for a discussion of the neighborhoods of Old Town) become quite active and, often, intense, and this sometimes extends to all the women of the community in an attenuated way.^[4] The expectations in the women's neighbor-neighbor role are notable for their supportive and expressive character, but they also contain a great deal of open competition for prestige and power.

There are a variety of means whereby women attain precedence among their fellows, but the wearing of new and fashionable clothing and the display of an impressive collection of gold jewelry at weddings are very high among

these. I will show that the competition among women and the importance of finery in that competition are crucial in leading women to made demands on their husbands which both spouses see as burdensome and unwelcome to the husband.

The man status and the expectations in that status are sharply different from those in the woman status. Men's relations with one another are ideally and often characterized by reserve and careful following of rules for proper behavior. Men of the higher classes have adopted what other community men see as quite Arabized manners involving elaborate etiquette and reserve. In relations among men of the other social classes, there is also concern for propriety and restraint, even if the expression of that concern shows less Arab influence. Regardless of class, men avoid talking about "delicate" issues that might bring shame to anyone present and there is a pervasive concern with honor (fakhri). As the expectations in the woman status involve affectivity, openness, and engagement, the man status's expectations emphasize calm, reserve, and detachment.

The reserve and avoidance of things of personal and possibly emotional significance (for that is what makes things "delicate") is characteristic of men's behavior across a wide range of situations. When men are together at their barazas, the talk is quiet, the topics are impersonal, and the tone is affectively neutral. Male neighbors spend a good deal of time with one another, but the laughing is muted if present at all and the conversations are far more likely to concern politics, soccer teams, or religion than family affairs, scandal, or experience of a personal sort.

Brothers greet one another much as they greet any other man, and even close friends spend most of their time talking about neutral topics of general interest. A man who is displeased with a shopkeeper most often states his objections quietly and in a matter-of-fact tone, and the stinging insults mothers hurl at their daughters and, less, other women (see Swartz 1990a ) are more rarely heard from fathers directed at their children and almost never at other men.

The competition between men, when present, is implicit. Men do not try to outdo one another in any observable way and, unlike women, make no comparisons based on any traits, whether physical, characterological, or in possessions. Men may feel competitive, and there are hints that having more honor than another—or at least not having less—is important, but there is none of the open rivalry said to be characteristic of women.

The Swahili saying, "Women are not freeborn nobles" (Wanawake si wangwane ) is taken by men, the only ones who use it, as meaning that women do not have proper regard for honor and the avoidance of shame. There are women who agree that the members of their sex are less obsessed with shame and honor than men, and they agree with the men that the difference is due to God having created men and women differently. The difference in concern

about shame is based in broader differences with women being understood as more emotional, less logical, and less able to plan effectively than men are. These differences manifest themselves in a variety of ways, including greater freedom and emotional lability in relationships between women as compared to those involving men.

The God-given differences, as community members see them, between men and women find expression in gender status expectations. These operate jointly with the expectations in a wide range of other statuses to guide behavior across a broad spectrum of social life. Both men and women are expected to be helpful as neighbors, for example, but men's help is mainly limited to that of a practical and unemotional sort, while women's includes hugs, tears, and laughter.

Comparing Men's and Women's Social Relations

From what has been said about the man status, it will be clear that the kind of emotional expression and freedom broadly expected of women is quite improper for men. But the differences between social relationships involving men and those involving women are not limited to differences in affective display, as can be seen in considering their development in individuals' lives.

The social lives of males and females begin to diverge sharply as soon as the boys are considered old enough to go outside the house to spend the daylight hours playing with other boys or attending school. Most girls nowadays go to secular schools as their brothers do, but they still spend their out-of-school hours in their homes studying, doing housework, and cooking under the supervision of their mothers.

Unmarried girls were traditionally kept from the company of married women not closely related to them, explicitly to shield them from talk of sex. Thus, when their mothers' friends came to visit, the girls stayed in another part of the house. This is not strictly enforced in most homes now, but it is still true that only after marriage do women begin the rich social life to be described below. One of the main reasons commonly given by adolescent girls for saying that the most important thing in their lives is marriage is that it is essential for full acceptance into the company of adult women.

The situation for boys is quite different. From the time they begin to spend the daylight hours outside the home, they are encouraged to be friendly with other neighborhood boys from different households. They are urged to invite their friends home for lunch (the main meal), and boys who do this too little were traditionally, but only rarely currently, punished. Unlike females, it was desirable that males be known through the community and even beyond it. A man whose name is not known by every other Swahili in the city is thought

a poor specimen. A traditional phrase in praise of a man was, "He who is not his brother is his slave." The phrase has not been wholly appropriate for most of this century, but the wide scope of approved social relations it refers to is understood as being as desirable in men now as it ever was.

Wedding Ceremonies and the Gender Statuses

Men's social relationships cross a wider range of the city's population and are more public, but they are also more restrained than women's are. Many of the differences associated with the differences between the two statuses are exemplified by the behavior that can be seen in participation in wedding ceremonies.

These ceremonies are generally held outside, near the house of the parents (either the groom's or the bride's) who are undertaking the responsibility for staging them. A large curtain, generally hung between houses, separates the women's festivities from the men's. From the women's side of the curtain comes the sounds of a band, laughter, shouting, and ululation. The women dance in a rather abandoned way with, one is told, their faces beaming and loud pleasantries freely exchanged.

On the men's side, the atmosphere is totally different. In the contemporary wedding celebration, rented wooden chairs are arranged into precise rows and the men on them quietly chat with those nearby. Boys and young men of the sponsoring family distribute soft drinks and snacks (often elaborate ones prepared by the women in a boisterous all-night session). When the groom, dressed in a turban and robes with an Arab-style dagger at his belt, is led to a seat of honor among the men, everyone turns to look at him during the brief period before he is led away to join his bride in the family house. During this period, there is a slight rise in the volume of the men's conversation, which can, nevertheless, hardly be heard above the din coming from the other side of the curtain. The commentary on the groom's fine costume, however, is short-lived and the men return to polite conversations about ordinary matters. After an hour or two, the men begin to drift away toward their homes, but the women's dancing and noise-making goes on far into the night despite the fact that many of them have been up celebrating and cooking all the night before.

The differences between relations among men and those among women seen at weddings are of the same sort found in everyday life. During their leisure time, the men come together at each other's houses, often not going inside but staying on the benches built into the front of houses that give their name, baraza, to the regular men's gatherings. At some of these gatherings, they sip coffee, although others view this as too Arabized. At all of them

they discuss the affairs of the day. The groups form on the basis of a variety of common interests. Thus, one group is composed of men who are concerned with world affairs and politics; another of men with a greater than usual interest in sports; several of men who are fond of discussing religion; and so on.^[5] All of the gatherings have one thing in common: they share an avoidance of personal topics. Deaths and hospitalizations are freely discussed, but nothing else personal is mentioned.

Talking about someone's wife, daughter, or sister—whether the man is present or not—is entirely unacceptable, and any man who did so, unless the context is illness or death, would be characterized as Hana mizani (without a sense of balance or propriety, see chap. 9). The general reluctance to discuss any specific member of the community is redoubled when the person is female and related by blood or marriage to someone present. The most valued topics for baraza discussion are all impersonal ones, and in a number of barazas "elevated" issues such as religion and philosophy are the most prestigious. Talking about specific people is considered unworthy of freeborn nobles and, especially, of men. For men, such talk is always bad, but they believe it often occurs among women and the low born.

The avoidance of gossip is not the only propriety in barazas. Each man is greeted on arrival by everyone already there with a handshake and a greeting consisting minimally of "Salaam Aleikhum" and generally of a considerably more elaborate sort. This is true even when the gathering is very large. Thirty or forty men were the most I ever observed at a baraza, with five to eight being more usual, but whatever its size, greetings occupy a good deal of the group's time. The tone of barazas I attended, whatever their particular content might be, was decorous, dignified, and restrained. The participants were invariably good humored and agreeable in their relations with one another, but reserve is the most notable trait.

Women's gatherings are less formal and structured than men's are. Unlike men's gatherings, which almost always involve the same men meeting at the same place, at the same hour, and lasting the same period of time, women come together whenever and wherever it is convenient and desirable. A woman may pop over to a neighbor's house to borrow a cup of sugar, and while there the two may chat for a long or a short period depending on how busy the two of them are and how interesting their conversation becomes.^[6] Other women, if they are free and so inclined, may join the conversation and a casual group of four or five women might thus assemble. The same group of women could come together again within a short period of time, or no more than two of them might talk together again for weeks. Like the men, the women's groupings each draws on a limited roster of participants, but unlike the men's, the women's gatherings are spontaneous, casual, and irregular.

Another difference between men's and women's gatherings is that in the latter, I am told, there is little or no reluctance to discuss particular people

and events concerning them. Tales of who is doing what and with or to whom are as common among the women's groups as they are rare among the men's. Also unlike the men's groups, among the women interaction rather often includes heated statements, arguments, and personal remarks including compliments and insults. Women share their joys, sorrows, and angers with one another rather freely; men hardly do so at all.

Relations with Close Kin: Warmth for Women, Restrictions for Men

The general differences between relations involving men and those involving women are as characteristic of relations among close kin as they are elsewhere. Relations among brothers are generally polite but in most cases distant. Informants explain that this is due to the fact that each may be reluctant to discuss the details of his private life with the other lest there be some shame involved. Also brothers may and often do see one another as potential rivals for their father's estate or, should one die with minor children, as the steward of the other's estate who may use it to benefit his own children rather than his nephews and nieces.

Women are less concerned with honor, shame, and matters of inheritance, and it is not unusual for women, especially sisters, to be mutually supportive confidantes. Women compete with one another for prestige, but sisters are generally allies rather than rivals. The concern for honor and the avoidance of shame that restricts relations even between brothers is mainly absent between women in general and especially between sisters. It is fairly common, women report, for women neighbors to be like sisters, including having the closeness and mutual support characteristic of that relationship.

As with siblings, men's relationships with their children do not involve the warmth and mutual support women often have in theirs. A father can express love and warmth for his very young children, but it is difficult for him to do this when the children are older. Daughters are said to occupy a special place in their fathers' affections, but after the girl begins to show physical signs of puberty, it is difficult for the father and daughter to be alone together. In conservative families, the girl does not even stay in the same room with her father, going so far as to speak to him from a hallway or adjoining room rather than face to face.

Although sons are not expected to avoid being in the same room as their mothers, the segregation of the sexes does divide them. The mother and daughter are both assigned to the home by the segregation of the sexes, and they spend most of every day cooking and doing household tasks together. The mother's relations with her unmarried daughters are close and emotionally labile (see Swartz 1990b ), and the tie between them is a strong one in many families.

The mother's realtions with her sons are quite different. She sees them only at mealtimes and in the evenings when the men return to the home, and the freedom of emotional expression in relations with daughters is mostly absent with sons. Although there can be little doubt that the survey data are accurate in reporting the tie between mothers and sons as closer than that between fathers and sons, the mother-son relationship is nevertheless far more restricted than that between mothers and daughters. It is important to note that widows live with their daughters rather more often than with their sons, despite the fact that this puts them in the houses of their sons-in-law who will be mainly responsible for their support even though they have sons with houses of their own.

The restrictions in the mother-son relationship that stem from the segregation of the sexes obviously do not apply to fathers and sons, but the father-son relationship is said to be the most tense and full of conflict in the family. Although some boys and young men obviously admire and even like their fathers and some fathers show considerable love for their sons, the majority of the boys and young men willing to discuss such things freely reported strained relations with their fathers.^[7] In a group of eight young men not yet married but living away from their parents, four were living elsewhere because of quarrels with their fathers and three reported trouble living with their stepfathers. In the more usual situation where the son remains in the family home until marriage, the respect in which a father is to be held dampens emotional expression even when the relationship between father and son is relatively free of conflict.

Further data on the emotional character of relations involving fathers, mothers, and their children can be seen in table 16. It is notable that more than a third of both sons and daughters say that children should love their mothers more than their fathers, while none of the sons and less than 10 percent of the daughters say that the father should be loved more. Twenty percent of the mothers say that mothers should be loved more than fathers, but none of the fathers say that fathers should be loved more. All of the fathers say that children should love both parents the same.

One interpretation of this surprisingly unanimous response by fathers is that they really believe that children should love both parents equally. Another, and equally plausible one, is that the real alternatives fathers see for themselves are limited to two: being loved the same as mothers or being loved less. In either interpretation, there is no basis for believing that fathers see themselves as being the rightful or likely favored parent as contrasted with a fifth of the mothers who seem to. That both parents are realistic in this is indicated by interviews with young people about their actual family showing that fathers are, in fact, loved less than mothers in a majority of families.

Given fathers' common and expected conflict with their sons and their distant relations with their daughters as contrasted with the loving and close relations between mothers and children, it is clear that fathers' relations with

the children are cooler and more isolated than the mothers'. Add to this the fact that a third of the offspring interviewed believe that mothers should be loved more than fathers and the effects of father's emotional limitations in relations with his children become quite clear.

The Spouse Relationship: Feelings and Funds

Marriage provides the one relationship in which men can express themselves freely and in which they can hope for warmth and emotional support. Women can also properly have close and unrestrained relations with spouses, but, unlike the situation for men, the marital relationship is by no means the only one where this is commonly true. Swahili unanimously report that love (upendo) is the most important single consideration in marriage. Even though first marriages were and mostly still are arranged by the parents of the couple, if love and the valued intimate mutually satisfying relationship does not develop, it is thought best for the couple to be divorced. Since one rather often hears Swahili men and women saying that "of the things God allows, He hates none more than divorce," this is a strong indication of love's importance.

Subsequent remarriages—being single is rare for men and only somewhat more common for women—are according to the wishes of the partners themselves, and if the love that ideally and usually led to the marriage should wane, again it is thought best to divorce. Love among the Swahili is very similar to Western love, with a romantic variety common among younger men and women and a calmer but more lasting type generally valued—and seemingly rather often present—for older couples.

The important fact for the present discussion is that the marital relationship is culturally constituted as one in which both partners are expected to be emotionally engaged and in which the male, as well as the female, is allowed to

give and receive support and intimacy. The husband is ideally expected to make his wife happy by giving her his attention when he is not occupied in other ways and by showing her kindness and regard. The wife ideally devotes herself to her husband and indulges him with well-prepared meals, lavish attention, and such luxuries as massages with sandalwood. Informants report that these ideals are generally met most fully in the first year or two of marriage but that they set desirable goals for both spouses throughout life. It is surely true that not all Swahili share all the understandings about love and marriage, but it is equally true that there is no other relationship that so many community members view in this way.

The emphasis on the values and beliefs concerning love and the fact that men are uniquely able to give and receive emotional support in the spouse relationship should not obscure the contribution that economic and social considerations make in determining the nature of that relationship. Husbands and wives retain their own property on marriage, and what one earns or inherits during the marriage remains under his or her direct control. However, the legalities of the marital relationship—and they are given here according to Islamic law as the Swahili view it—fail to account for the social unity of the married couple.

The wife's standing in the women's groups is importantly dependent on her husband's social and economic position. We have already seen the importance of the husband's material support for his wife in the form of finery and ceremonies, but it is also worth noting that any blemish on his reputation, any shame attached to him, will redound to his wife's discredit and will be used against her by her rivals. This is what was meant by women's "indirect concern" with honor.

The wife's standing among women does not directly affect her husband's standing in the community, but if she exposes herself to shame by improper behavior—especially by being suspected of association with other men—his reputation and standing will suffer. More than the parties to any other relationship in this society, spouses are a social unit rather than quite separate individuals. This commonality of interest does not make their relations close and emotionally warm; it imposes obvious stresses resulting from the close interdependence. However, since divorce is easy, marriages that work have partners who are brought together by these social and economic pressures and couples that cannot accept them usually part.

Divorce, Death, and Gender Differences in the Significance of Marriage

Divorce itself can contribute to the prevalence of warm and supportive marital relationships among those that endure. Men are empowered to end

their marriages simply by saying, "You are not my wife" (Si mke wangu ) three times.^[8] The most common reason for their doing so is the absence of love. Women have no formal right to end a marriage, but, in fact, they are usually able to do so by insisting that their husbands divorce them. A husband who refuses to listen to his wife's demand for a divorce finds himself living with a woman who will not speak to him, contribute to the functioning of the household, or participate in sex.

There are pressures other than personal satisfaction that affect marriages and their termination, of course, but one of the pieces of advice sometimes given the groom at the marriage ritual in the mosque is, "You have married peacefully; if it is necessary, then dirorce peacefully as well." Divorce terminates about one in three Swahili marriages,^[9] with more than sixty percent occurring in the first five years. There is sometimes an active effort by kin to try to convince married couples, especially young ones, not to divorce, but when it becomes clear that either partner's dissatisfaction is not temporary or trivial, the effort usually stops.

The marital relationship is crucial to both partners but on different grounds. For the wife, it generally provides the sole means of material support and of assurance she will have her children with her. Without a husband, or such rarely present alternative source of funds as an outside job or an inheritance, a divorced woman is forced to appeal to her male consanguines. Grown sons or brothers willing to contribute to her support and having means beyond what is needed for their own wives and children can provide a reasonably comfortable life, but this is rare. If, as is most common for divorced women, she lacks these alternatives and is unable to remarry, she is destined to a life of poverty.

A divorced woman has the right to support from her ex-husband for four months and ten days (the period deemed necessary to establish whether or not she left the union pregnant), and support for the children is due her until they are seven years old. After that, she has no clear right to assistance from her divorced spouse. He has the right to take the children to raise, and if he does, he supports them. If he allows his ex-wife to keep the children, he may provide funds for their care, but he has no legal obligation either in civil or Koranic law to do so. His support for the children is a matter open to negotiation and depends, among other things, on whether the woman has remarried. Whatever is done about the children, a divorced woman often finds herself in difficult financial circumstances. If she does remarry, her situation is rather often as good or better than it was before divorce even if, as will be seen, she has children in need of support.

Women are less able to initiate a new marriage than man are, partly because women have less freedom to seek a new spouse, and older divorced women often do not find a new husband since many men prefer younger

women. It is very rare for a divorced person to marry someone who has never married before, but a few cases of men doing so were recorded.

Men encounter few if any material difficulties when they are divorced. In most cases, men remarry without delay and their lives go on as before with, perhaps, whatever caused the breakup of their marriage behind them. Even if they do not remarry, they can get servants or female kin to provide most domestic services. Materially, their condition is little changed since, almost always, they provided the income for the first marriage and, when they remarry, provide it in the second as well.

There can be little question that men are generally not adversely affected by divorce in any lasting way. It is important to remember, however, that they most often bring about the divorce themselves, or, less frequently, finally allow a wife who is making them miserable to leave them. For men, it is rarely rewarding marriages that end in divorce, and given their ability to find a new and perhaps more satisfactory wife, men generally benefit from divorce. Divorced women do not fare as well as men unless they remarry. When death rather than divorce is the cause of a marriage's end, however, women usually suffer decidedly less than men.

The traditional practice, aida, of women dressing in special mourning clothes and spending four months and ten days on their curtained beds is still followed. This long period of isolation is very hard on the women, their children, and any others who depend on their participation in household affairs. When women finally emerge from aida, they often suffer from the practical consequences of losing their husbands' income, but they do not usually show signs of serious and lasting emotional damage. Young widows often remarry, and older ones commonly find what some of them say is a reasonably satisfying life caring for their grandchildren in one of their children's houses.

Men are frequently more adversely affected by the death of a spouse than women are. A number of men seen shuffling feebly down the streets were pointed out to me by other men as widowers whose wives' deaths were said to have precipitated their current sorry state. At the baraza I usually attended, one of the regular attendee's wives died following a long illness. The man was in his middle fifties and had been married to the dead woman for decades, but almost a year after her death, this otherwise dignified and reserved man—or so he had been before his wife's death—would burst into tears without obvious reason.

Other men told me after we had witnessed an instance of the man's seemingly uncaused weeping that it is not uncommon for men to react that way on the loss of their wives. In the same discussion, I was told of several men who had been utterly ruined by the loss of their wives, and my informants said that mental breakdown or, even, drunkenness were familiar, if not common, reactions to this loss. Evidence is rather strong, in short, that despite

the fact that women have to endure a prolonged postmortem isolation, it is husbands who most frequently suffer a lastingly adverse reaction to the loss of their spouses. At least part of the reason for this is that the spouse relationship has a significance for men that it does not have for women.

Husbands, Wives, Love, and Marriage

As their reaction to their death shows, men are very closely attached to their wives. There is, in fact, a sharp contrast between the spouse relationship and all the others the men participate in. It is only in marriage that men can express and expect intimacy and emotional support. For women, the spouse relationship is distinctive in its economic significance, but from the standpoint of intimacy and emotional support, the relationship is similar to a number of others including the women's relations with their children. Table 17 shows the quite substantial differences between wife/mothers and husband/fathers in the emotional experience they have in marriage and other nuclear family relationships.

A majority of informants say that there is more love between parents and children than between spouses, but it is notable that fathers and sons lead female family members in choosing the alternative statement that there is more love between spouses. The tense relations between fathers and sons may well play a part in the occupants of these statuses indicating that there is more love between spouses, since they both experience something different from love in their relations with each other. Since, however, it is true that sons (and fathers, after all, were and are sons as well) generally experience a relatively warm, if restricted, relationship with their mothers and that daughters (including those who are also mothers) have distant relations with their fathers, it seems warranted to look at the spouse relationship itself as the basis for the greater frequency with which males choose it as having more love.

We have seen that men's relations with everyone but their wives are restricted by various values and beliefs. There is tension between fathers and sons and competition between brothers. Considerations of honor (fakhri) and the avoidance of shame (aibu) prevent most men from having close relationships with peers. Relationships with daughters are circumscribed by respect and, like relations with the mother, attenuated by the separation of the sexes.

For men, the spouse relationship is unique. This is the one relationship in which men can expect emotional warmth and support and in which they can express themselves. Table 17 shows that the spouse relationship is chosen as the one with the most love by a larger group of sons and fathers than by mothers and daughters, with the women more often choosing the parent-child relationship. Although the same expectations in the spouse relationship apply

to women, for them the relationship is one of a number in which affectivity, a prominent expectation in the woman status, is freely expressed and in which warmth and support are common.

The Real Power of the Husband/Father

The fact that the expectations in the spouse relationship are unique for men in allowing affective expression but that this is common for women's relationships will be shown to be a central element in women's ability to get their husbands to do things they, but often not their husbands, want. It still remains to be shown, however, that wives getting things they want despite their husband's negative views is due to men being reluctant to use power they have rather than lacking the needed power and further, that this reluctance is connected to the unique character of the marital relationship for men and the potent, if unavowed and at least sometimes unrecognized, resource this provides for their wives.

It could be, of course, that women get the expensive things they want because the husbands simply do not have the power to prevent them from doing so. However, if "power" refers to the ability to control, there can be little question of men having insufficient power to prevent their wives from spending money if they wish to. The fact that husbands do have the power needed to deny their wives' demands is indicated by several types of evidence, including the unanimous views of all the Swahili who were asked about the distribution of family power or were heard to volunteer comments about it.

All of them indicate that the husband and father is the final authority in the family and that although it is desirable that he consult with other affected family members, all family decisions—save those concerning cooking, cleaning, and child care—are properly made by him alone. This applies quite unambiguously to decisions concerning money, since the husband not only controls it as the head of the family but also, in most cases, as the person who

earned it. The latter fact is itself a powerful basis for the power to control the use of the money according to most community members' understandings. Adding this to the generally powerful position of divinely selected family head proclaimed in the holy Koran gives the father's authority a broad base.

Further evidence for the husband and father's power to make family decisions is seen in table 18, which shows that 90 percent of all fathers, mothers, and sons say that fathers are the ones with most influence on important family decisions. Only 63 percent of those in the daughter status make the same response, but inquiry reveals that that relatively low figure is due to several of the young, unmarried women in the sample whose fathers are dead having brothers (in the table, the latter are counted as in the son status) who act toward the women as fathers do when they are alive. The table is unambiguous in indicating that Swahili of both sexes say that fathers are powerful, and anyone who has discussed family relations with members of this community would agree that this view is nearly universally held. Still, attributing power to fathers may be a token that is not also a guide, to use the concept developed in chapter 6. It could be that saying fathers are powerful is the accepted and proper statement but that fathers have no real ability to get anyone to do anything the person does not personally want to do.

In fact, this is not so. The statements about fathers' and husbands' power do reflect social reality in the sense that husbands and fathers rarely fail when they actually exercise their power. This is particularly true as concerns their wives and, slightly less, their daughters, but least often so as concerns their sons. Even in dealings with the son, however, when the father/husband makes a decision and insists on it, that decision quite often determines what happens. Decisions concerning wives are even more uniformly successfully implemented, in large part because, unlike sons, wives do not usually eventually attain financial independence.

One man, for example, told his wife on their marriage that she was to associate with no one: not her relatives, not her neighbors, no one. Limiting women's associations with men to their husbands and close kin is, of course, universally supported by members of the Swahili group, but this husband was forbidding his wife to associate with anyone regardless of sex and without respect to relationship. The wife was said to be very miserable about her husband's decision, but she dutifully followed it throughout their long marriage. Only after his death did she begin to involve herself in the usual round of visiting central to the lives of all other wives. No one who told me this story, and several people did, thought the husband's restricting of his wife's activity was commendable, but it was equally clear that everyone understood that it was within his rights.

As a further approach to the issue of men's power, I showed the table, presented here as table 18, to informants of both sexes including some who had told me in general about, or who had cited instances of, women spending

money in ways that were at the limit of their husband's ability to spend—or beyond it. After discussing the table's meaning with the informants, I asked them to explain how it was that men often gave in to their wives' demands for costly things when he was chosen by most informants as being the one with "most influence on decisions." Both men and women said there was no inconsistency between the husband/fathers having the most influence on decisions and important sums of money nevertheless being spent on things they did not support. If the husbands had insisted on smaller weddings, less expensive clothing, or less jewelry, then, the informants said, that is what would have been. In other words, men have power, but they choose to allow their wives to do things they want to do even if the men do not themselves agree with the wives about the desirability of those things.

This leaves little doubt that consistent with cultural understandings of their general power, husbands do actually have the ability to block their wives' access to the funds the wives seek for the things needed for full participation in the women's groups. The husbands, however, do not always or, even, frequently use that ability. In fact, although men have the cultural resources needed to control their wives' behavior, actually doing so requires an additional resource, emotional independence from the wife.

The Husband/Father's Real Need for Warmth and Emotional Support

Because of the nature of Swahili social relations, few Swahili men have the additional resource just mentioned and are consequently reluctant to employ their well-grounded power to block their wives' demands. Looking at the wives' cultural bases for power, it is difficult to see that they have many in the marital relationship other than the emotional dependence of their husbands. This, however, is often sufficient.

Despite their importance in family power use as well as their personal importance for those involved, men do not mention their emotional needs if they

are aware of them, and the same is true of their wives and community members generally. According to very widely held ideal understandings, men should be self-sufficient and strong. Depending on their wives, or anyone else, for emotional support would be a direct contradiction of the ideal.

The reluctance to consider men's emotional needs and the part these needs may play in the spouse relationship leaves people with a difficult situation to explain: why do the powerful men give their wives costly things the men say they do not want to pay for? The most common explanation given by men concerns their wives' mental states. They say they could stop their wives from spending money in ways or amounts they do not agree with but that they sometimes withhold the use of that power because, and various forms of this phrase occur a number of times in my notes, they do not want to make the wives unhappy.

Women's views are quite similar to mens'. They agree that the husbands have the right to forbid them spending money on clothing, jewelry, and ceremonies, but they say that the husbands often allow them to do this because the husbands are "good." Their husbands, they report, buy them things even if they are not really in favor of spending the money, or spending so much money, because they "understand what it means to women" to have what the wife wants or because of the husbands' "love" for them.

Women's Happiness and Wives' Power

We have already seen that women are understood as having been created by God with greater emotionality and less planning ability than men. As part of this, people of both sexes generally imply or directly state that women cannot be happy if they are denied what they want no matter what the difficulties may be in giving it to them. It is not, of course, the view of anyone I talked to that women must or can be happy all the time. Nevertheless, given that it is only in his relations with his wife that a man can hope to express his own emotions more or less freely and only in relations with her that he can expect to get emotional support, this relationship takes on a very special standing for him. It becomes clear, then, why a husband would be willing to take considerable effort to see that his wife is not "unhappy" and why it is to his benefit to have his wife believe that he is "good," "understands what it (i.e., what she wants) means to women," and to show her that he "loves" her.

Men are not equal in their desire to have the opportunity to express themselves with some emotional freedom and/or in their wish to be given emotional support. Similarly, men's assessments of what they have to do to get the support and expression they want surely varies from person to person. Nevertheless, for most husbands, withholding the use of their power to deny

their wives' deeply rooted demands and allowing the wives to have what they want is a readily available and widely effective approach for them to promote their getting whatever emotional satisfaction they may seek.

Understandings about Women, Social Structure, and Wives' Power

Men's emotional dependence on their wives is a key element in explaining wives' ability to get what they want from their husbands. Another element, and it is the one that makes the first effective, is to be found in the social structure. Out of the whole array of relationships involving adult men, all have expectations precluding emotional expression and support with the sole exception of the spouse relationship. That relationship takes on unique importance for the man because of its unique expectations, and this is the major factor in men's acquiescence to their wives' requests for money.

This explanation is not, however, the same as the one offered by the participants in the marital relationship. Their explanation of the ability of wives to get what they want has a dual foundation: the already discussed value on love as a basis for the marital relationship and the understandings concerning the nature of women, especially women's mental and emotional qualities. These latter understandings are important in that they not only play a key role in explaining why wives are able to get what they want but in that they also provide what Malinowski called a "charter" (1944:52, 111, 141 passim) for women's demands on their husbands.

It is basic to these understandings that women are seen as fundamentally different from men. Women are understood as being quite as intelligent as men and at least as likely to show such highly admired traits as compassion and consideration of others. However, the Swahili believe—and they buttress their views with references to the Koran—that women are less able to plan for the future than men are.

Women, however intelligent they may be, are believed to be less logical than men and to be unable to control their emotions as men can. Women's emotional lability, as seen at weddings, funerals, and everyday life, is taken by Swahili of both sexes as undeniable evidence of their deeply emotional natures. The widely shared understandings is that women cannot, as a result of their God-given natures, curb their desires on the basis of a logical assessment of what is possible and practical. There is some implicit belief that in areas where the women's own interests and emotions are not directly engaged (in matters of science, scholarship, or business, for example), women can be as logical as men, but there is general agreement that in their personal lives, it is not realistic to expect women to behave with control and on the basis of logic.

Women, including educated ones, accept—even embrace—this view. I was interviewing eight Swahili high school girls when I mentioned, with what was clearly a disclaimer in my voice, that a number of men had told me that women were highly emotional, poor at planning, and illogical. The girls were quite forceful in their response. "Do you think that is not true?" one of them asked. "Women are not at all like men," another said, "the Koran itself tells us that." I replied that perhaps women were different but asked whether it was not possible nevertheless that they be able to plan well and to control their emotions. All those present denied this vigorously and several of them picked up the theme—not uncommonly heard in a range of discussions with Swahili—that the Koran supported their views. They held that God had made women the way they were, and it was worse than ignorant to deny that this was so.

In discussing the allocation of power in marital relations with several male informants, I asked why husbands give in to their wives if, as they had been telling me, sometimes it is financially harmful for them to do so. One of them answered me with an analogy: "In some ways women are like children. It may be bad for them to have candy, but they want it and you can't explain to them that it is bad. They just become unhappy if they do not get the candy, so, because you love them, you give it to them even if it is not the best thing."

This was heartily agreed to by the other men present and is as close as I ever came to a direct, emic explanation of the ability of wives to get what they want. The men had told me, in sum, that there is no more point in trying to reason with women than with children and that if the women want something it is often best to give it to them since they cannot be brought to understand that it may not be really desirable. There may be some point in withholding desired things from (male) children since they may learn from the experience, but no such result is likely with women because of their God-given nature. The husband's love leads him to avoid causing his wife unhappiness, as the participants see it, and because of her nature, this involves giving in to her demands.

The shared understanding that women are unable to think logically or to control their emotions is not only used as a key element in community members' explanations of women's ability to get what they want but it is also important as a self-justification. So, women are at least as interested in the family budget as are men, yet some of them sometimes use substantial parts of their husbands' income—generally the family's only source of funds—on things they want but that contribute nothing to meet the family expenses for housing, food, clothing, and education. Some women explained that the use of even substantial sums of family funds in ways they particularly desire is not really frivolous. The gold jewelry, especially, is explained as a way to store wealth to be used in times of crises, which is particularly attractive given the Islamic prohibition on interest from bank savings.

Even here, however, when they are asked why it is that many women spend very considerable sums having goldsmiths rework their old jewelry into new and more fashionable designs—as is fairly common—the basic answer is that this is the way women are. Women, female informants told me, like to have things that are beautiful and they do everything they can to get them, especially if other women have them, even if doing so presents their families with financial difficulties.

In addition to comments and observations gleaned from talks with a variety of informants of both sexes, I discussed women's desires for finery and ceremonies intensively with four female informants. They all agreed that women were "not good at planning" and that they often spent money on things that did not really provide for their families', or their own, welfare. The informants reported that to some extent, at least, the many women who they believed used money this way could not help themselves. It was the way women were, and for many, there was nothing they could do about it. Three of the women said they knew of cases in which houses had been mortgaged to pay for elaborate weddings, with the wives and mothers being the principal ones responsible for the mortgaging. Although all of them condemned this as "dangerous" since the family home could be lost in this way, they also agreed that, as one of them said, "women don't think about the future very much . . . they just know what they want and they follow that."

Malinowski reserved the concept "charter" for the explicit justification of such organized sets of activity as those carried out by families, age groups, and religious congregations. These charters, often in myths, provided a statement of the "value, purpose, and importance" (1944:111 et seq.) of the groups' distinctive activities and contributions. The demands women make on their husbands are not institutions in the sense that families or religious congregations are, of course, but they are patterned and recurrent activities and the idea that women cannot avoid making them provides them with a culturally constituted explanation and justification. It gives the wives' demands a stature they would not otherwise have, because they are rooted in the belief in women's unique mental processes, which are understood to be as God wishes them to be.

Social Structure's Strictures: More on Unshared Culture

As with medical treatment, wives' power needs to be looked at as the result of the operation of cultural elements, some of which are not shared—at least not in any active sense—by those involved in the phenomena they produce. Here, wives' power derives, in important part, from a social structure all of whose other statuses include expectations preventing their roles with men from

including emotional support. Wives' power, in other words, is partly due to the restrictions in other statuses.

Understandings shared by many group members maintain that men need no emotional support, so the affectivity of the husband-wife role cannot be culturally based in the sense that people understand that role to be the proper venue for men's emotional expression. Women expressly deny the emotional importance of the spouse role to their husbands, since, as they say, their husbands are "too strong" to need such support. Similarly, the absence of affectivity in other relationships involving men—with their friends, parents, siblings, and children—is not a marked characteristic of those relationships. For those who share the understanding that adult men are without emotional needs, the absence of attention to such needs in specific relationships, if noted, cannot be difficult to countenance.

It is the nature of the other statuses in the social structure as much as the wife status that makes the latter unique for men. This uniqueness, that is, having expectations allowing affectivity in the role relationship with the husband, is mainly a product of Swahili culture rather than a part of that culture. The uniqueness is not the result of shared understandings overtly assigning emotionality to the spouse role or barring it in the other roles in which men are involved. Since men's emotionality is denied, its expression is not subject to explicit regulation. In fact, it is the generally close and private nature of the spouse relationship and the importance of love (upendo) in it that encourages such expression between spouses rather than explicit understandings calling for it. What is most unusual about the expression of affect in this relationship is less that it is prescribed than that, unlike the other relationships involving adult men, nothing interferes with it.

The spouse relationship thus makes a unique contribution to husbands. Its significance varies from man to man, but it seems rarely to be nugatory and provides a powerful, if unrecognized, base for wives' power in dealing with their husbands.

The expectations in the woman status are also important to the wives' power in the spouse role. Since women are expected to be emotional and alogical by their God-given natures, there is no advantage in denying them what they want in the hope that they will learn from the denial. Such educational prohibition is worth trying with sons but not with wives who, as women, will only become unhappy. Their unhappiness, through the strictures on all the other relationships the husband has, is easily transmitted to their husbands who need not have an understanding of why it affects them in order for it to do so.

As noted elsewhere (Swartz 1983), changes in the Swahili community offer the prospect that women may become more emotionally dependent on their husbands than they are now, but thus far this has not happened. It is, however, true that weddings are less important than they were just a few years

ago (see chap. 3), so that wives' opportunities to turn their husbands' expenditures into prestige among women are somewhat reduced. So long, however, as the women in a neighborhood spend much of their time with one another, it will take substantial changes in a number of the statuses occupied by women (including "woman," "Muslim," "wife," and "neighbor") and in other parts of the social structure to diminish their wish to gain prestige with their peers and the importance of their husbands in their ability to get the means for doing so.