Interviewing Women about Friendship

My exploratory aims and interest in friendship directed me to a method that would produce rich detail and adapt to evolving theoretical concerns: a focused interview, malleable enough to follow emergent leads and standardized enough to register strong patterns. Unobtrusive observation or participant observation in private friendships was, of course, impossible. Appendix A details this method of research and the associated problems of inquiry and inference. My interview elicited two kinds of information about friendship and marriage: women's self-descriptions of their friendships and their views on friendship and marriage; and more general information on the formal correspondence between friendship and marriage. To construct analyses of friendship independent of the women's own subjective understandings, I elicited consider-

able information about patterns and practices of friendship and prepared to interpret statements about reasons, needs, beliefs, and desires.

I directed most of my attention to middle- and working-class married, employed mothers, although I interviewed a wider variety of women. Because companionate marriage is a theoretical concern here, I might have focused on urban, educated professional women, the sector with which companionate marriage is most strongly identified in the literature. Yet the recent outpouring of writing on friendship—most of it about women in this stratum—suggests that under the influence of feminism, many of these women are reviving the ideology of romantic friendship; this posed a problem for me. As I explain in chapter l, I believe that romantic friendship has always been more than just stylistic convention; it is rather a self-conscious culture of friendship. Still, like romantic marriage, romantic friendship is a realm where ideal and reality merge easily. I decided to learn about friendship from women who were less likely to idealize it.

Family scholarship since the sixties has amply documented the cross-class diffusion of the ideals of companionate marriage, at least among wives.^[4] Given this evidence of widespread support for companionate ideals, I chose to focus on "average American" women and investigate friendships that should broadly represent contemporary women. (I did interview a few very low-income women. But in slighting their representation in my sample, I acknowledged the structural limitations of my propositions: under conditions of chronic unemployment and poverty and when a modicum of welfare benefits are available, poor single mothers may not find their long-term economic interests in marriage.)^[5]

I focused on mothers for a few reasons. For one, a previous analysis of survey data on friendship suggested that motherhood constricts women's networks of friends considerably more than fatherhood does men's. Because mothers with small networks were unhappier than those with large ones, Claude Fischer and I took those contracted networks to reflect constraints on mothers' ability to conduct friendships rather than their preferences.^[6] I wanted a closer look at that process. In recent decades, 90 percent of American women have married; 95 percent of all married women have expected to have children and, at least up to now, virtually all of

these eventually have.^[7] Thus, most women spend a fair portion of their adult life in that very involving role. I believed that examining women's friendships during a common period of constraint would uncover some of the stronger motives in friendship, and some of the persistent themes of community. In other words, a commonly experienced period of constraint might reveal—along with the particularities of friendship at that specific stage—the values and exchanges in their friendships that women would strive against odds to maintain.

I collected a varied sample of women to interview by contacting associates in various cities, towns, and suburbs in California (see appendix A for sampling details). All but a few of the women agreed to be interviewed. Although not statistically representative, the sample included women of various social classes, ages, races, and stages of life. Before describing this sample by categories, I offer sketches of a few of the women I interviewed. As I do throughout this book, I alter enough facts to preserve confidentiality.

Louise and her family live in a comfortable working-class neighborhood in a small city. She is in the first year of a very close friendship—her first since she and Gary married six years ago. Both Louise and her husband were born nearby, but her high school friendships faded—partly, she says, because her jealous husband discouraged them. Louise met Jan in a night class at the community college; their friendship seems to have added to Louise's growing sense of power in her marriage. School, work, and friendship have all added personal satisfaction as well as marital conflict to Louise's life. She feels that she and Gary are "working it out." He is now cooking meals for the children when Louise works, and he enjoys the company of Jan and her husband, the first couple Louise ever recruited into socializing as couples.

Cass had the help and encouragement of her large close-knit family in throwing out her husband, who beat her and tried to keep her away from her family. Lacking education, she works long hours, often at lonely, home piecework or household work, supporting her children on poverty-level wages. Her own family has provided friendship and support. One sister has always been her favorite confidante, and Cass has had few opportunities and little time to find friendships in other contexts. Cass and her children

live in a very poor, ethnically diverse white neighborhood. Neighbors say hello; but only one has ever been invited into her tiny, impeccably tended home. Cass is reluctant to become friendlier with this one neighbor, who is also a single mother. Although Cass very much likes the woman, she is not eager to care for her children since Cass's own children are older and more independent.

Nancy lives in a comfortable suburban ranch home with her husband and teenage children. She is nearing forty and has worked part-time, off and on, at service jobs accessible to this articulate and personable woman with a high school education. She and her husband lead a very active social life with other couples, many of them friends she and her husband met in children's recreation. Nancy has two very close friends, including a longtime neighbor whom she calls her best friend. Nancy and her husband are also "best friends," and yet she keeps a very large sphere of interests and confidences for her close women friends alone.

Janine, her husband, and young children live in a basement fiat in a deteriorating neighborhood of single-family homes in a small city. Like Janine, most of her neighbors are black. Many are formerly well paid industrial workers, now unemployed by industry shutdowns. Janine was recently laid off a kitchen job at a fist-food restaurant. Her husband is marginally employed; the job that brought them to this city ended when the company closed. Janine is still closest to one of her six sisters, and to the rest of her own family. "We're all friends. When we throw a party, we don't need to invite nobody, because everybody's already there." She also has a good friend in the neighborhood and spends a lot of time with her when their husbands are not at home.

Mary and Hal lead parallel lives. They have raised two children and have remained together through a thirty-year marriage, mostly by going their separate ways. Mary is bitter about her husband's lack of feeling for family, and his preference for his own friends. Hal does skilled industrial work, but Mary professes only the vaguest knowledge of his job of twenty years. For several years Mary has worked full-time at a large insurance company in an office with a stable staff of skilled clericals. She takes great pride and enjoyment in her work, and in sociable relations with her diverse office peers. Her best friend, Vera, is a divorced mother of grown children. The two friends socialize frequently during weekends

and times that most of the other women regard as strictly "family time."

Thea is an urban mother of young teenagers. She is educated, very accomplished in and devoted to her full-time career. Work and family are her most important investments, although she and her husband lead an active upper-middle-class social life. Thea's marriage is the stereotypical companionate marriage; her "friendship" with her husband is still vital after many years of marriage. In recent years she has formed a very ardent friendship with another professional woman who also has strong family bonds. They consistently manage to find time together amid very busy lives.

The employed women in my sample worked in various parts of the "pink-collar ghetto." The household, clerical, service, and "women's professional" jobs (in fields where women dominate and educational requirements far outstrip salaries: teaching, nursing, librarianship, social work) were distributed among the women in a hierarchy of salary, working conditions, and prestige that roughly corresponded to the hierarchy of their husbands' jobs. The husbands' incomes and occupations varied much more than those of the women. Some husbands were low-paid unskilled service workers, others high double-digit professionals. Represented in between were blue-collar skilled workers, white-collar sales workers, small business owners, and corporate administrators. Mirroring patterns in society at large, these men were married to women whose salaries ranged by less than half the spread of their husbands'. The women's salaries constituted just over one-third of their family income, in the cases of the married women who worked full-time, and generally less than one-sixth of the household income, in the cases of those working part-time. Education among the women in my sample was more evenly distributed, because a majority had secured some community college education. Even the very low-income women had high school diplomas; Cass had just been awarded hers after several years of evening school.

Family cultural styles also only roughly corresponded to husbands' occupational status and income. Consumption patterns— evident in dress, household furnishings, organizational membership, and socializing and recreation patterns—were influenced also by wives' education and occupation and, it appeared, by their

friendships as well. For example, Louise, a working-class woman by any standard, has upwardly mobile career aspirations; and her best friend is more prosperous and middle class. In her home, Louise has decorated one room—a living room—with the woodsy hues, nubby textures, and eclectic wall hangings that mark California's contemporary young middle class.

The bulk of my sample fell between stable working class and solid middle class. A few eases each were upper middle class and working and unemployed poor. Predictably, divorced women and black women were disproportionately represented in the latter categories. Since class categories for this sample were complex and particularly unpredictive for the relationships I studied, I use class designations loosely and attempt to specify which elements of stratification appeared to determine the behavior I am describing.

The women I interviewed range in age from early twenties to late fifties. Nearly half are in their thirties. All but two of the twenty-one women are mothers. Fourteen have children at home. Of these, nine have at least one child under six. Mirroring the national rate of employment, half the women with children under six currently do paid work: four work part-time, one full-time (and two others are on layoff or brief leave from full-time work). All six women whose children are older than six work, four women full-time.

Each interview took place in the woman's home and usually lasted two and a half hours. I kept the potentially long-ranging interview as compact as possible, because most women were carving out time for me amid a full day's responsibilities, and nearly half had young children at home with them. They were eager and willing to talk openly and at great length, with only one exception. Penny was a willing participant but said, "I'm the kind of person [who] can't explain myself," We both would have benefited from a longer and less structured exchange; in future exploratory work, I would probably include some multisession interviews.

The interview questions are reproduced in appendix B. Each interview entered a woman's close social world with introductory open-ended questions about self, family, and friends: Tell me a little about yourself. Then I asked for names and descriptions of those people with whom each woman felt most involved and closest. Eventually, I asked specific questions about the conduct and

content of close friendships, probing for their cognitive, affective, and moral orientations. Most questions were open-ended, although often very narrowly focused; a good many invoked short or scaled forced-choice responses.

How should one read my conclusions? If I analyze individual conduct in terms of a group and report incidences of certain behavior among group members, I imply generalization. If a hypothesis generates rich detail, if the subject quickly strikes a resonant chord in a reader, if a theoretical argument is persuasive or uses the terms of a current discourse—then a reader may take an exploratory project as conclusive. I emphasize that I do not generalize the material I have gathered or confirm the propositions it generates or corroborates.

Perhaps the best way to remind a reader that I offer a hypothesis-generating exploratory study is to note its biases and limitations. I do so in appendix A. Assuming the reader is thus prepared, I curb my inclination to use very tentative language and frequent notes of caution. I can use bold interpretation to provoke theoretical refinement and confirmatory replication. As for the data this research method generated, I have presented a great deal—in the form of quotations and response counts—and I have discussed negative cases, so that the reader can judge the conclusions I have drawn. If I am successful, this book should stimulate new thinking and scholarship linking friendship to the social structures of gender.