Families We Choose
Kath Weston

Every Thursday night in the cityscape that framed my experience of "the field," my lover and I had dinner with Liz Andrews. The three of us juggled work schedules, basketball practice, and open‑ended interviews around this weekly event. Occasionally these gatherings meant candlelight dinners, but more often Thursday found us savoring our repast in front of the TV. The first few weeks of gourmet meals gave way to everyday fare with a special touch, like avocado in the salad or Italian sausage in the spaghetti sauce.

Responsibility for planning, preparing, and subsidizing the meals rotated along with their location, which alternated between Liz's home and the apartment I shared with my lover. At only one point did this egalitarian division of labor and resources become the subject of conscious evaluation. Liz offered to pay a proportionately greater share of a high‑ticket meal, reasoning that she had the largest income. In the ensuing discussion, reluctance to complicate "power dynamics" in the group resolved the issue in favor of maintaining equal contributions.

After supper we might play cards, trade anecdotes about mutual acquaintances, describe recent encounters with heterosexism, discuss world politics or the opening of a new lesbian strip show, exchange recipes, explain how we would reorganize the Forty‑Niner offensive lineup, or propose strategies for handling the rising cost of living in San Francisco. Or we might continue watching television, taking advantage of commercial breaks to debate that perennial enigma: "What do heterosexual women see in Tom Selleck, anyway?" While we grew comfortable with argument and with differences in our class backgrounds, age, and experiences, we tended to assume a degree of mutual comprehension as white women who all identified ourselves as lesbians.

After a few months of these dinners we began to apply the terms "family" and "extended family" to one another. Our remarks found a curious counterpart in a series of comments on changes in the behavior of Liz's cat. Once an unsociable creature that took to hiding and groveling from the other room when strangers invaded her realm, now she watched silently from beneath the telephone table and even ventured forth to greet her visitors. Not that she does that for everyone, Liz reminded us: clearly we were being taken into an inner circle.

In retrospect, the incipient trust and solidarity imaged in this depiction of a world viewed through cat's eyes appears as one of several elements that combined to make Thursdays feel like family occasions. The centrality of the meal‐sharing food on a regular basis in a domestic setting‐certainly contributed to our growing sense of relatedness. In the United States, where the household is the normative unit of routinized consumption, many family relationships are also commensal relationships. Although we occupied separate households, we interpreted the option of independent residence as a feature distinguishing gay families from straight, one that qualified "our" kind of family as a creative innovation. In truth, this contrast may have been a bit overdrawn. Moving to the same neighborhood had prompted the routinization of the weekly dinner meetings, and I personally enjoyed walking over to Liz's apartment when it was her turn to cook. These evening strolls underlined the spatial contiguity of our households while allowing me to avoid the seemingly interminable search for a parking space in San Francisco.

Efforts to encourage a low‑key atmosphere framed our interactions during supper as everyday experience rather than a guest‑host relationship. It was not uncommon for any one of us to leave immediately following the meal if we were tired or had other things to do. Conversation, while often lively, seldom felt obligatory. Also facilitating the developing family feeling was a sense of time depth that arose after the arrangement had endured several months, a dimension augmented by a ten‑year friendship between Liz and myself.

On some occasions other people joined our core group for activities, events, and even Thursday night get‑togethers. Once Liz asked two gay male friends to dinner, and another time‐with somewhat more anticipation and formality‐the group extended an invitation to Liz's parents. When her parents arrived a guest‑host relationship prevailed, but Liz, my lover, and I became the collective hosts, preparing and serving the food and making sure that her parents were entertained. One could imagine other possible alignments:  for example, Liz and her parents busy in her kitchen while my lover and I waited to be served. The differentiation of activities and space presented a graphic juxtaposition of the family Liz was creating with the family in which she had been raised. By introducing my lover and me to her parents in the context of a Thursday night meal, Liz hoped to bridge these two domains.

About the time that the three of us began to classify ourselves as family, we also began to provide one another with material assistance that went beyond cooking and cleaning up the dinner dishes. When one of us left on vacation, another volunteered to pick up the mail. After Liz injured her foot and decided to stay at her parents' house, I fed the cat. On street cleaning days Liz and my lover moved each other's vehicles. Liz offered me the use of her apartment for interviews or studying while she was at work. "Emotional support" accompanied this sort of assistance, exemplified by midweek phone calls to discuss problems that could not wait until Thursday. Our joint activities began to expand beyond the kitchen and living room, extending to the beach, the bars, political events, restaurants, a tour of Liz's workplace, and Giants games at Candlestick Park.

Faced with the task of analyzing this type of self‑described family relationship among lesbians and gay men, my inclination while yet in the field was to treat it as an instance of what anthropologists in the past have termed "fictive kin." The concept of fictive kin lost credibility with the advent of symbolic anthropology and the realization that all kinship is in some sense fictional‐that is, meaningfully constituted rather than "out there" in a positivist sense. Viewed in this light, genes and blood appear as symbols implicated in one culturally specific way of demarcating and calculating relationships. Under the influence of Continental philosophy, literary criticism, and an emerging critique of narrative form in ethnographic writing, anthropological monographs‐like the kinship structures they delineated‐came up for review as tales and constructions, inevitably value‑laden and interpretive accounts (Clifford 1988; Clifford and Marcus 1986; Geertz 1973; Marcus and Fischer 1986; Rabinow 1977). Although the category "fictive kin" has fallen from grace in the social sciences, it retains intuitive validity for many people in the United States when applied to chosen families. From coverage in the popular press to child custody suits and legislative initiatives, phrases such as "pretended family relations" and "so‑called family" are recurrently applied to lesbian or gay couples, parents, and families of friends.

The very concept of a substitute or surrogate family suffers from a functionalism that assumes people intrinsically need families (whether for psychological support or material assistance). Commentators who dispute the legitimacy of gay families typically set up a hierarchical relationship in which biogenetic ties constitute a primary domain upon which "fictive kin" relations are metaphorically predicated. Within this secondary domain, relationships are said to be "like" family, that is, similar to and probably imitative of the relations presumed to actually comprise kinship. When anthropologists have discussed the institutionalization of "going for sisters" (or brothers, or cousins) among urban blacks in the United States, for example, they have emphasized that such relationships can be "just as real" as blood ties to the persons involved (Kennedy 1980; Liebow 1967; Schneider and Smith 1978; Stack 1974). While framed as a defense of participants' perspectives, this type of argument implicitly takes blood relations as its point of departure. Insofar as analysis becomes circumscribed by the unvoiced question that asks how authentic these "fictive" relations are, it makes little difference that authenticity refers back to a privileged and apparently unified symbolic system rather than an empirically observable universe.

Theoretically I have adopted a very different approach by treating gay kinship ideologies as historical transformations rather than derivatives of other sorts of kinship relations. Some might contend that these emergent ideologies represent variations modeled on a more generalized "American kinship" to the extent that they utilize familiar symbols such as blood and love, but this terminology of modeling would prove misleading.[1] As Rayna Rapp has convincingly argued,

When we assume male‑headed, nuclear families to be central units of kinship, and all alternative patterns to be extensions or exceptions, we accept an aspect of cultural hegemony instead of studying it. In the process, we miss the contested domain in which symbolic innovation may occur. Even continuity may be the result of innovation (1987:129).

Gay families do not occupy a subsidiary domain that passively reflects or imitates the primary tenets of a coherent "American kinship system." The historical construction of an ideological contrast between chosen (gay) families and blood (straight) family has not left biologistic and procreative conceptions of kinship untouched. But if coming out has supplied gay families with a specific content (the organizing principle of choice) by exposing the selective aspects of blood relations, it remains to be shown how choice became allied with kinship and gay identity to produce a discourse on families we choose.

Building Gay Families

The sign at the 1987 Gay and Lesbian March on Washington read: "Love makes a family‐nothing more, nothing less." From the stage, speakers arguing for domestic partner benefits and gay people's right to parent repeatedly invoked love as both the necessary and the sufficient criterion for defining kinship. Grounding kinship in love deemphasized distinctions between erotic and nonerotic relations while bringing friends, lovers, and children together under a single concept. As such, love offered a symbol well suited to carry the nuances of identity and unity so central to kinship in the United States, yet circumvent the procreative assumption embedded in symbols like heterosexual intercourse and blood ties.

It has become almost a truism that "family" can mean very different things when complicated (as it always is) by class, race, ethnicity, and gender (Flax 1982; Thorne with Yalom 1982). In her studies of kinship among ,Japanese‑Americans, Sylvia Yanagisako (1978, 1985) has demonstrated how the unit used to calculate relatedness ("families" or "persons") may change, and additional meanings adhere to symbols like love, based on variable definitions of context that invoke racial or cultural identities. Determining who is a relative in a context that an individual perceives as "Japanese" may draw on different meanings and categories than determining relationship in a context defined as "American."

In speaking broadly of "gay families," my objective is not to focus on that most impoverished level of analysis, the least common denominator, or to describe symbolic contrasts in pristine seclusion from social relations. Neither do I mean to imply an absence of differences among lesbians and gay men, or that gay families are constructed in isolation from identities of gender, race, or class. Rather, I have situated chosen families in the specific context of an ideological opposition between families defined as straight and gay‑families identified with biology and choice, respectively. On the one hand, this highly generalized opposition oversimplifies the complexities of kinship organization by ignoring other identities while presenting its own categories as timeless and fundamental. On the other hand, the same discourse complicates understandings of kinship in the United States by pairing categories previously believed to be at variance ("gay" and "family").

The families I saw gay men and lesbians creating in the Bay Area tended to have extremely fluid boundaries, not unlike kinship organization among sectors of the African‑American, American Indian, and white working class. David Schneider and Raymond Smith (1978:42) have characterized this type of organization as one that can "create kinship ties out of relationships which are originally ties of friendship." Listen for a moment to Toni Williams' account of the people she called kin:

In my family, all of us kids are godparents to each others' kids, okay? So we're very connected that way. But when I go to have a kid, I'm not gonna have my sisters as godparents. I'm gonna have people that are around me, that are gay. That are straight. I don't have that many straight friends, but certainly I would integrate them in my life. They would help me. They would babysit my child, or . . . like my kitty, I'm not calling up my family and saying, "Hey, Mom, can you watch my cat?" No, I call on my inner family‐my community, or whatever‐to help me with my life.

So there's definitely a family. And you're building it; it keeps getting bigger and bigger. Next thing you know, you have hundreds of people as your family. Me personally, I might not have a hundred, because I'm more of a loner. I don't have a lot of friends, nor do I want that many friends, either. But I see [my lover] as having many, many family members involved in what's going on.

What Toni portrayed was an ego‑centered calculus of relations that pictured family members as a cluster surrounding a single individual, rather than taking couples or groups as units of affiliation. This meant that even the most nuclear of couples would construct theoretically distinguishable families, although an area of overlapping membership generally developed. At the same time, chosen families were not restricted to person‑to‑person ties. Individuals occasionally added entire groups with preexisting, multiplex connections among members. In one such case, a woman reported incorporating a "circle" of her new lover's gay family into her own kinship universe.

In the Bay Area, families we choose resembled networks in the sense that they could cross household lines, and both were based on ties that radiated outward from individuals like spokes on a wheel. However, gay families differed from networks to the extent that they quite consciously incorporated symbolic demonstrations of love, shared history, material or emotional assistance, and other signs of enduring solidarity. Although many gay families included friends, not just any friend would do.[2]

Fluid boundaries and varied membership meant no neatly replicable units, no defined cycles of expansion and contraction, no patterns of dispersal. What might have represented a nightmare to an anthropologist in search of mappable family structures appeared to most participants in a highly positive light as the product of unfettered creativity. The subjective agency implicit in gay kinship surfaced in the very labels developed to describe it: "families we choose," "families we create." In the language of significant others, significance rested in the eye of the beholder. Participants tended to depict their chosen families as thoroughly individualistic affairs, insofar as each and every ego was left to be the chooser. Paradoxically, the very notion of idiosyncratic choice‐originally conceived in opposition to biogenetic givens‐lent structural coherence to what people presented as unique renditions of family.

The variety in the composition of families we choose was readily apparent. At the MCC service described in chapter 2, when the time came for communion, the pastor invited congregants to bring along family members. In groups and in couples, with heads bowed and arms linked, people walking to the front of the church displayed ties of kinship and friendship for all to see. On a different occasion, I joined several people preparing for a birthday party in someone's home. When I asked what, if anything, separated those who came early to help decorate from those who arrived after the time officially set for festivities to begin, the host explained that the helpers were family, closer to her than most of the other guests.

Obituaries provide a relatively overlooked, if somber, source of information about notions of kinship. Death notices in the Bay Area Reporter (a weekly newspaper distributed in bars and outer gay establishments) were sometimes written by lovers, and included references to friends, former lovers, blood or adoptive relatives (usually denominated as "father," "sister," etc.), "community members" present at a death or assisting during an illness, and occasionally coworkers. While I was conducting fieldwork, the San Francisco Chronicle, a major citywide daily, instituted a policy of refusing to list gay lovers as survivors, citing complaints from relatives who could lay claim to genealogical or adoptive ties to the deceased. Although the Chronicle's decision denied recognition to gay families, it also testified to the growing impact of a discourse that refused to cede kinship to relations organized through procreation.

By opening the door to the creation of families different in kind and composition, choice assigned kinship to the realm of free will and inclination. In the tradition of Thoreau's Walden, each gay man and lesbian became responsible for the exemplary act of creating an ideal environment (cf. Couser 1979). People often presented gay families as a foray into uncharted territory, where the lack of cultural guideposts to mark the journey engendered fear and exhilaration.[3]  Indeed, there was a utopian cast to the way many lesbians and gay men talked about the families they were fashioning. Jennifer Bauman maintained that as a gay person, "you're already on the edge, so you've got more room to be whatever you want to be. And to create. There's more space on the edge." What to do with all that "space"? "I create my own traditions," she replied.

"Choice" is an individualistic and, if you will, bourgeois notion that focuses on the subjective power of an "I" to formulate relationships to people and things, untrammeled by worldly constraints. Yet as Karl Marx (1963:15) pointed out in an often quoted passage from The 18th Brumaire, "Men [sic] make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly encountered, given and transmitted from the past." Only after coming out to blood relatives emerged as a historical possibility could the element of selection in kinship become isolated in gay experience and subsequently elevated to a constitutive feature of gay families.

Despite the ideological characterization of gay families as freely chosen, in practice the particular choices made yielded families that were far from randomly selected, much less demographically representative. When I asked people who said they had gay families to list the individuals they included under that rubric, their lists were primarily, though not exclusively, composed of other lesbians and gay men. Not surprisingly, the majority of people listed tended to come from the same gender, class, race, and age cohort as the respondent.

Both men and women consistently counted lovers as family, often placing their partners at the head of a list of relatives. A few believed a lover, or a lover plus children, would be essential in order to have gay family, but the vast majority felt that all gay men and lesbians, including those who are single, can create families of their own. The partner of someone already considered family might or might not be included as kin. "Yeah, they're part of the family, but they're like in-laws," laughed one man. "You know, you love them, and yet there isn't that same closeness."

Former lovers presented a particularly interesting case. Their inclusion in families we choose was far from automatic, but most people hoped to stay connected to ex‑lovers as friends and family (cf. Becker 1988; Clunis and Green 1988).[4] When former lovers remained estranged, the surprise voiced by friends underscored the power of this ideal. "It's been ten years since you two broke up!" one man exclaimed to another. "Hasn't he gotten over it yet?" Of course, when a breakup involved hard feelings or a property dispute, such continuity was not always realizable. After an initial period of separation, many ex‑lovers did in fact reestablish contact, while others continued to strive for this type of reintegration. As Diane Kunin put it, "After you break up, a lot of people sort of become as if they were parents and sisters, and relate to your new lover as if they were the in‑law." I also learned of several men who had renewed ties after a former lover developed AIDS or ARC (AIDS‑Related Complex). This emphasis on making a transition from lover to friend while remaining within the bounds of gay families contrasted with heterosexual partners in the Bay Area, for whom separation or divorce often meant permanent rupture of a kinship tie.

Jo‑Ann Krestan and Claudia Bepko (1930:285) have criticized lesbians' efforts to maintain relationships with former lovers as "triangling" (a no‑no in therapeutic circles). They argue that such relationships "tend to be intrusive and involve inappropriate claims." But notions of appropriateness are culturally constituted and contested. What a person expects from an "ex" may not be what they expect from a friend who is also family. In the context of gay kinship, former lovers can be both.

A lover's biological or adoptive relatives might or might not be classified as kin, contingent upon their "rejecting" or "accepting" attitudes. Gina Pellegrini, for example, found refuge at a lover's house after her parents kicked her out of her own home as an adolescent. She was out to her lover's mother before her own parents, and still considered this woman family. Jorge Quintana claimed that his mother adored his ex‑lover and vice versa, although Jorge had broken up with this man many years earlier. After years of listening to her father attack homosexuality, remembered Roberta Osabe, "My girlfriend Debi and my father shot pool together. And she whipped his ass! . . . That was his way, I think, of trying to make amends." Jerry Freitag and his partner Kurt had made a point of introducing their parents to one another. "My mother and his mother talk on the phone every once in a while and write letters and stuff. Like my grandmother just died. Kurt's mother was one of the first people to call my mom." For Charlyne Harris, however, calling her ex‑lover's mother "family" would have been out of the question. "Her mother didn't like me. Number one, she didn't want her to be in a lesbian relationship; number two, she knew that I was black. So I didn't have a lot of good things to say about her mother . . . . Pam told me, `She can't even say your name!' "

In addition to friendships and relationships with lovers or ex‑lovers, chosen family might also embrace ties to children or people who shared a residence.[5] Gay Community News published a series of letters from gay male prisoners who had united to form "the Del‑Ray Family" (only to be separated by the warden). Back in San Francisco, Rose Ellis told me about the apartment she had shared with several friends. One woman in particular, she said, was "like a big sister to me." When this woman died of cancer, the household split up, and "that kind of broke the family thing." In other circumstances, however, hardship drew people together across household lines. Groups organized to assist individuals who were chronically or terminally ill often incorporated love and persisted through time, characteristics some participants took as signs of kinship. Occasionally a person could catch a glimpse of potential family relationships in the making. When I met Harold Sanders he was making plans to live with someone to prepare for the possibility that he might require physical assistance as he moved into his seventies. Harold explained that he would rather choose that person in advance than be forced to settle for "just anyone" in an emergency.

The relative absence of institutionalization or rituals associated with these emergent gay families sometimes raised problems of definition and mutuality: I may count you as a member of my family, but do you number me in yours? In this context offers of assistance, commitment to "working through" conflicts, and a common history measured by months or years, all became confirming signs of kinship. By symbolically testifying to the presence of intangibles such as solidarity and love, these demonstrations operated to persuade and to concretize, to move a relationship toward reciprocity while seeking recognition for a kin tie.

Like their heterosexual counterparts, most gay men and lesbians insisted that family members are people who are "there for you," people you can count on emotionally and materially. "They take care of me," said one man, "I take care of them." According to Rayna Rapp (1982) the "middle class" in the United States tends to share affective support but not material resources within friendships. In the Bay Area, however, lesbians and gay men from all classes and class backgrounds, regularly rendered both sorts of assistance to one another. Many considered this an important way of demarcating friend from family. Diane Kunin, a writer, described family as people who will care for you when you're sick, get you out of jail, help you fix a flat tire, or drive you to the airport. Edith Motzko, who worked as a carpenter, said of a woman she had known ten years, "There's nothing in the world that [she] would ask of me that I wouldn't do for her." Louise Romero joked that a gay friend "only calls me when he wants something: he wants to borrow the truck 'cause he's moving. So I guess that's family!"

Overall, the interface between property relations and kinship relations among lesbians and gay men who called one another family seemed consistent with such relations elsewhere in the society, with the exception of a somewhat greater expectation for financial independence and self‑sufficiency on the part of each member of a couple. Individuals distributed their own earnings and resources; where pooling occurred, it usually involved an agreement with a lover or a limited common fund with housemates. Some households divided bills evenly, while others negotiated splits proportionate to income. A person might support a lover for a period of time, but this was not the rule for either men or women. Putting a partner through school or taking time off from wage work for childrearing represented the type of short‑term arrangements most commonly associated with substantial financial support.

Across household lines, material aid was less likely to take the form of direct monetary contributions, unless a dependent child was involved. Services exchanged between members of different households who considered themselves kin included everything from walking a dog to preparing meals, running errands, and fixing cars. Lending tools, supplies, videotapes, clothes, books, and almost anything else imaginable was commonplace in some relationships. Many people had extended loans to gay or straight kin at some time. Some had given money to relatives confronted with the high cost of medical care in the United States, and a few from working‑class backgrounds reported contributing to the support of biological or adoptive relatives (either their own or a lover's).

Another frequently cited criterion for separating "just plain" friends from friends who were also family was a shared past. In this case, the years a relationship had persisted could become a measure of closeness, reflecting the presumption that common experiences would lead to common understandings. Jenny Chin explained it this way:

I have, not blood family, but other kind of family. And I think it really takes a lot to get to that point. Like years. Like five years, ten years, or whatever. I think that we're gonna have to do that to survive. That's just a fact of life. Because the whole fact of being gay, you're estranged from your own family. At a certain level, pretty basic level. Unless you're lucky. There are some exceptions.

So to survive, you have to have support networks and all that kind of stuff. And if you're settled enough, I think you do get into a . . . those people become family. If you kind of settle in together. And your work, and your lives, and your house, and your kids or whatever become very intertwined.

While people sometimes depicted the creation of ties to chosen kin as a search for relationships that could carry the burden of family, there are many conceivable ways to move furniture, solicit advice, reminisce, share affection, or find babysitters for your children. All can be accomplished by calling on relationships understood to be something other than family, or by purchasing services if a person has the necessary funds. But allied to the emphasis on survival in Jenny's account was the notion of a cooperative history that emerged as she bent her litany of years to the task of establishing rather than assuming a solidarity that endures.

Relationships that had weathered conflict, like relationships sustained over miles but especially over time, also testified to attachment. Allusions to disagreements, quarrels, and annoyance were often accompanied by laughter. Charlyne Harris named five lesbians she counted as kin "because if they don't see me within a certain amount of time [they check up on me], and they're always in my business! Sometimes they get mad, too.. They're like sisters. I know they care a lot." Another woman chuckled, "I never see these family, so you can tell they're family!" Still others mentioned, as a sign of kinship, hearing from people only when they wanted something. Through reversal and inversion, an ironic humor underscored meanings of intimacy and solidarity carried by the notion of family in the United States (cf. Pratt 1977).

In descriptions of gay families, sentiment and emotion often appeared alongside material aid, conflict resolution, and the narrative encapsulation of a shared past. "Why do you call certain people family?" I asked Frank Maldonado. "Well," he responded,

Some of my friends I've known for fifteen years. You get attached. You stay in one place long enough, you go through seasons and years together, it's like they're part of you, you're part of them. You have fights, you get over them . . . . It's just unconditional love coming through to people that you didn't grow up with.

Though imaged here as the sole defining feature of kinship, love represents as much the product as the symbolic foundation of gay families. Closely associated with the experience of love were the practices through which people established and confirmed mutual, enduring solidarity.

Substitute For Biological Family?

Far from viewing families we choose as imitations or derivatives of family ties created elsewhere in their society, many lesbians and gay men alluded to the difficulty and excitement of constructing kinship in the absence of what they called "models." Others, however, echoed the viewpoint‐popular in this society at large‐that chosen families offer substitutes for blood ties lost through outright rejection or the distance introduced into relationships by remaining in the closet.[6] "There will always be an empty place where the blood family should be," one man told me. "But Tim and I fill for each other some of the emptiness of blood family that aren't there." In Louise Romero's opinion,

A lot of lesbians . . . I think they're just looking for stuff‐maybe the same stuff I am. Like my family ties, before coming out, there was a lot of closeness. I could share stuff with my sisters. You used to talk all your deep dark secrets. You can't any more 'cause they think you're weird. Which is true in my case they really do . . . . I think a lot of women look for that, and you need that.

This theory has a certain appeal, not only because it speaks to the strong impact of coming out on lesbian and gay notions of kinship, but also because it is consistent with the elaboration of chosen families in conceptual opposition to biological family. On a practical level, most of the services that chosen kin provide for one another might otherwise be performed by relatives calculated according to blood, adoption, or marriage.

Although gay families are families a person creates in adult life, this theory portrays them primarily as replacements for, rather than chronological successors to, the families in which individuals came to adulthood. If chosen families simply represent some form of compensation for rejection by heterosexual relatives, however, gay families should logically focus on the establishment of intergenerational relationships. (Remember that the loss of parents, as opposed to other categories of relatives, was the main concern in deciding whether or not to reveal a gay or lesbian identity to straight family.) But when lesbians and gay men in the Bay Area applied kinship terminology to their chosen families, they usually placed themselves in the relationship of sisters and brothers to one another, regardless of their respective ages. In cases where gay families included children, adults who were chosen kin but not coparents to a child sometimes characterized themselves as aunts or uncles.

As with any generalization, this one admits exceptions. Margie Jamison, active in organizing a Christian ministry to lesbians and gay men, described her work with PWAs (persons with AIDS) while tears streamed down her face. "When I have held them in my arms and they were dying, it's like my sons. Like my sons." In this case the intergenerational kinship terminology invoked Margie's pastoral role as well as her experience raising two sons from a previous heterosexual marriage. However, the characterization of most ties to chosen kin as peer relationships brings families we choose closer to so‑called "fictive kin" relations found elsewhere in the United States than to even a moderately faithful reconstruction of the families in which lesbian‐and gay‐identified individuals grew up.

Equally significant, the minority of gay people who had been disowned were not the only ones who participated in the elaboration of gay kinship. Many who classified relations with their biological or adoptive relatives as cordial to excellent employed the opposition between gay and straight family. Among those whose relations with their straight families had gradually unproved over the years, ties to chosen kin generally had not diminished in importance. If laying claim to a gay family in no way depends upon a break with one's family of origin, the theory of chosen family as a surrogate for kinship lost dissolves. A satisfactory explanation for the historical emergence of gay families requires an understanding of the changing relation of friendship to sexual identity among the large numbers of gay people who flocked to urban areas after the Second World War.

Friends And Lovers

"That's the way one builds a good life: a set of friends." At 64, Harold Sanders had no hesitation about indulging his passion for aphorisms, the turn of phrase stretched backward to gather in experiences of a lifetime. His statement reflected a conviction very widely shared by lesbians and gay men of all ages. People from diverse backgrounds depicted themselves as the beneficiaries of better friendships than heterosexuals, or made a case for the greater significance and respect they believed gay people accord to friendship.[7] Most likely such comments reflected a mixture of observation and self‑congratulation, but they also drew attention to the connection many lesbians and gay men made between friendship and sexual identity (as well as race or ethnicity). The same individuals tended to portray heterosexuals as people who place family and friends in an exclusive, even antagonistic, relationship. As a child growing up in a Chinese‑American family, said Jenny Chin,

I had a lot drilled into me about your friends are just your friends. Just friends. Very minimalizing and discounting [of] friendships. Because family was supposed to be all‑important. Everything was done to preserve the family unit. Even if people were killing each other; even if people had twenty‑year‑old grudges and hadn't spoken.

In contrast, discussions of gay families pictured kinship as an extension of friendship, rather than viewing the two as competitors or assimilating friendships to biogenetic relationships regarded as somehow more fundamental. It was not unusual for a gay man or lesbian to speak of another as family in one breath and friend in the next. Yet the solidarity implicit in such statements has not always been a taken-for‑granted feature of gay lives. According to John D'Emilio (1983b), recognition of the possibility of establishing nonerotic ties among homosexuals constituted a key historical development that paved the way for the emergence of lesbian and gay "community"‐and, I might add, for the later appearance of the ideological opposition between biological family and families we choose.

When Harold Sanders was coming out in the 1930s, particularly in the white and relatively wealthy circles where he traveled, same‑sex ties were experiencing a historical devaluation that coincided with a new affirmation of eroticism in relations between women and men embodied in the ideal of companionate marriage. Strong bonds between persons of the same sex became something best left behind with childhood (Pleck and Pleck 1980). By 1982, Lillian Rubin found that two‑thirds of the single men in her sample of 200 could not name a best friend. While the disparagement of same‑sex ties may have had a greater impact on men than women, all same‑sex relationships became subjected to a higher degree of scrutiny. Today many heterosexuals in the United States are quick to judge certain friendships as "too intense," taking intensity as a sign of homosexuality.[8] According to Lourdes Alcantara, who was born in Peru during the 1950s, such associations are no longer confined to North America.

I read an article in the newspaper, and they present two women hugging like friends in the street. Latin friends, right? And I was in love with this woman. We were lovers. And I was in her house. So I brought the Sunday newspaper to her house, and I took that page out, so her mother didn't see that. And then we were so hot, reading that. But the distortion! They put us like sick people. So to be a lesbian, the description was terrible! Even my girlfriend got upset. She said, "We better be friends, just friends, and get married." And we were eleven years old, nine years old! God? Qué terrible! Can you believe that?

In the United States during the twentieth century, sibling ties and friendship have offered some of the few cultural categories available for making sense of powerful feelings toward a person of the same sex. During high school, Peter Ouillette had what he later identified as a "crush" on another boy. "Absolutely under no circumstances would I think about sex," he said. "It was friendship. But real close friendship, that's the way I thought of it. Almost like brothers." Philip Korte remembered thinking, "Wouldn't it be nice to have a big brother. Or wouldn't it be nice to just have a best friend that I could be affectionate with and spend a lot of time with, companionship, those kinds of things. Now I recognize that as gay. But, at the time, what I knew of gay, that didn't fit at all." What did not seem to fit were his fantasies of love and caring for another man, since homosexuality then appeared to him as a matter of sex and sex alone.

Given this alliance between the language of kinship and the language of friendship, which Jonathan Katz (1976) dates to the nineteenth century, one might expect to uncover a direct link between the early interchangeability of these terms and contemporary discourse on gay families. However, historical evidence and day‑to‑day observation suggest otherwise. By mid‑century coming out as a lesbian or gay man entailed learning to discriminate between feelings of erotic and nonerotic love, drawing meaningful contrasts between sexual attraction and friendship. A person could then theoretically sort relationships into two groups: "just friends" (not sexually involved) and "more than friends" (lovers). One day while sitting in a coffeeshop, for example, I overheard a woman in the next booth tell the woman sitting across from her that she was "only" interested in a friendship, since she already had a lover. Coming‑out narratives invoke this distinction when they establish a double time frame, the "before" and "after"' of coming out, effectively reinterpreting relationships previously described with the terminology of blood ties as having been "really" erotic all along.

The years following World War II‐a watershed period for many groups in the United States‐witnessed an unprecedented elaboration of nonerotic solidarities among homosexuals (Berube 1989; D'Emilio 1983a, 1983b, 1989b). During the 1950s and 1960s gay men adapted kinship terminology to the task of distinguishing sexual from nonsexual relationships.[9] At that time the rhetoric of brothers, sisters, and friends applied primarily to nonerotic relationships. In the film version of The Boys in the Band, one character quips, "If they're not lovers, they're sisters." This camp usage of "sister" among gay men coexisted with the institutionalization of mentor relationships in which older men introduced younger men to "the life." Normatively, mentor relationships were intergenerational and emphatically nonsexual. Bob Korkowski, who flouted convention by having sex with his mentor, described the experience as "weird, because a mentor is kind of like a father. [It was] like sleeping with your father." This reservation of kinship terminology for nonsexual relations represents a very different usage from its subsequent deployment to construct gay families that could include both lovers and friends.

The contrast between the sexual and the nonsexual was drawn only to be blurred in later years after the possibility of nonerotic ties among gay people became firmly established. By the 1970s both gay men and lesbians had begun to picture friends and lovers as two ends of a single continuum rather than as oppositional categories. "We women been waiting all our lives for our sisters to be our lovers," announced the lyrics of the song Gay and Proud (Lempke 1977). The contribution of lesbian‑feminism toward codifying this notion of a continuum is evident in Adrienne Rich's (1980) work on "compulsory heterosexuality." Carroll Smith‑Rosenberg's (1975) classic article on relations between women in the nineteenth‑century United States was also widely read in women's studies classes and cited to buttress the contention that sexual and sisterly relations were semantically separable but overlapping in practice‐with little regard for efforts to distinguish precisely these relationships during the intervening decades.

The realignment that linked erotic to nonerotic relations through the device of a continuum was not confined to political activists. As San Francisco moved into the 1980s, "friend" seemed to be overtaking "roommate" in popularity as a euphemistic reference to a lover in situations where lesbians and gay men elected not to reveal their sexual identities. Victoria Vetere's 1982 study of lesbian interpretations of the concepts "lover" and "friend," though based on a small sample, found that most lesbians were uncomfortable with any suggestion of a dichotomy between the two terms. A similar continuity was implicit in coming‑out stories narrated by women who had first claimed a lesbian identity during the 1970s. One said coming out was epitomized for her by the realization that "oh, wow, then I get to keep all my girlfriends!" Elaine Scavone explained with a laugh, "All of a sudden I felt I could be myself. I could be the way I really want to be with women: I could touch them, I could make friends, I could make my girlfriends and I could go home and kiss them." Although women were sometimes said to be more likely to come out by falling in love with a friend and men through an encounter instrumentally focused on sex, both men and women featured early attractions to friends in their coming‑out stories.

The category of mentor, which epitomized one type of nonsexual relationship between gay men, appeared to be losing rather than gaining currency during the same period. In the few cases when the term came up in casual conversation during my fieldwork, its meaning seemed to be changing. One man in his early thirties described himself as a mentor to his lover, based on his claim to have been out longer and to know more about what he called "the gay world." Such a statement would have been a non sequitur not so many years ago.

Given that any continuum is defined by its poles, these changes did not represent a complete collapse of the categories "lover" and "friend" into one another. The phrases "just friends" and "more than friends" remained in common usage to indicate whether two people had incorporated sex into their relationship. A certain unidirectionality also characterized the enterprise of melding sex and friendship. While a lover ideally should become a friend, many believed that sex could ruin a preexisting friendship. People who were single seemed as wont as ever to invoke the old gay adage that friends last, while lovers are simply "passing through."

In a 1956 study (reprinted in 1967) Maurice Leznoff and William Westley found that most gay men looked to friends, not lovers, for security in old age. Yet the dictum that friends rather than lovers endure took on a different cast for a later generation that believed lovers should not only double as friends, but continue as friends and kin following a breakup. New contexts can engender novel interpretations of received wisdom.

In retrospect, this shift from contrast to continuum laid the ground for the rise of a family‑centered discourse that bridged the erotic and the nonerotic, bringing lovers together with friends under a single construct. But the historical development of friendship ties among persons whose shared "sexual" identity was initially defined solely through their sexuality turned out to be merely an introductory episode in a more lengthy tale of community formation.

From Friendship To Community

Among lesbians and gay men the term "community" (like coming out) has become as multifaceted in meaning as it is ubiquitous. In context, community can refer to the historical appearance of gay institutions, the totality of self‑defined lesbians and gay men, or unity and harmony predicated upon a common sexual identity. Older gay people generally considered the term an anachronism when applied to the period before the late 1960s, since "community" came into popular usage only with the rise of a gay movement.[10]

Often contrasted with "isolation," community subsumed one of the earlier senses of coming out: making a public debut at a gay bar. The area of overlap involved locating other gay people, a project that can remain surprisingly difficult in an era when homosexuality makes headline news. Toni Williams, who had grown up in a large metropolitan area and begun identifying as a lesbian only a few years before I interviewed her, insisted, "I didn't think that there was nobody that was going to be like me. But I didn't know where to search for that person. I didn't think that there was a community."

Finding community, as one man very eloquently put it, meant discovering "that your story isn't the only one in the world." Such a discovery need not entail meeting other gay people, but rather becoming convinced of their existence. Sean O'Brien, originally from New York City, used to listen to a weekly gay radio show, "a voice coming through a box once a week," which he said helped him "understand myself as part of a community, even though I was not connected with that community." During the 1970s the concept of community came to embody practical wisdom emerging from the bars, friendship networks, and a spate of new gay organizations: the knowledge that lesbians and gay men, joining together on the basis of a sexual identity, could create enduring social ties. In the process, sexuality was reconstituted as a ground of common experience rather than a quintessentially personal domain.

From its inception, activists pressed the community concept into the service of an identity politics that cast gays in the part of an ethnic minority and a subculture.[11]  Lesbians and gay men represent a constant two percent of the population, they contended, a veritable multitude prepared to claim its own distinct history, culture, and institutions. The basis for these arguments was, of course, laid earlier with the recognition that homosexuals could unite through bonds of friendship as well as sex, and elaborated through analogies with identity‑based movements organized along racial lines.[12] Many social scientists of the period subscribed to a similar paradigm in their studies of "the gay world."[13] Whether describing an aggregate of persons in ongoing interaction (Evelyn Hooker) or a "continuing collectivity" of individuals with common interests and activities (William Simon and John Gagnon), they tended to treat homosexuals as a fairly homogeneous group with concrete, if not readily ascertainable, boundaries. More recently Stephen Murray (1979) has used sociological criteria to argue for the validity of applying the community concept to gay men in urban areas of the U.S. and Canada, dubbing them a "quasi‑ethnic community.[14]

Deborah Wolf's (1979) ethnography of lesbians (actually lesbian-feminists) in the Bay Area falls prey to many of the same traps that have ensnared other investigators who treat lesbians and/or gay men as members of an integrated subculture. Most studies that set out to explore a "gay world" or "gay lifestyle" not only situate their subjects in a historical vacuum, but assume an amazingly uncomplicated relationship between claiming an identity and feeling a sense of belonging or community. With their presumptions of harmonious solidarity and their reduction of varied experience to a single worldview, such approaches have proven far from satisfactory.

Yet the shortcomings of previous research offer no reason to reject the community concept altogether, as Kenneth Read (1980) does in his study of patrons in a gay bar on the West Coast. It is important to understand how gay men and lesbians came to use a category that over time has served as everything from a rallying cry for political unity, to a demographic indicator, to a symbol for a small sector of wealthy white men set apart from the majority of people who call themselves lesbian or gay. Viewed in cultural and historical context, the so‑called minority model appears as part of a series of historical struggles to define and dispute the boundaries of communities based on sexual identity, struggles that in turn paved the way for a discourse an gay families. Gay community can best be understood not as a unified subculture, but rather as a category implicated in the ways lesbians and gay men have developed collective identities, organized urban space, and conceptualized their significant relationships.

My interpretation of community departs in several key respects from the long tradition of community studies in the United States (see Hillery 1955). Conrad Arensberg (1954), for example, treats community primarily as a setting in which to conduct sociology, whereas gay communities are only roughly defined spatially and rest on variable interpretations of identity. In the hands of W. Lloyd Warner (1963), community becomes a microcosm of society at large, yet lesbians and gay men have contested and transformed hegemonic understandings of kinship and sexuality. My approach perhaps comes closest to Robert Lynd's and Helen Lynd's (1937) depiction of community as a vantage point from which to view historical events (in their case, the Great Depression), but again I am not concerned with a bounded entity or with community as locale. To comprehend the historical ascendance of a family‑centered discourse among lesbians and gay men, my analysis focuses on social movements, and on the meanings of togetherness and identity that have shaped community as a cultural category defined in opposition to equally cultural notions of individualism and selfhood (Varenne 1977).

Although lesbian and gay communities cannot be reduced to a territorial definition, this has not prevented San Francisco from becoming a geographical symbol of homosexuality, renowned here and abroad as the "gay capital" of the United States.  With the gay movement came the consolidation of "gay ghettos," neighborhoods featuring a variety of gay‑owned businesses and residential concentrations of gay men (Castells 1983). "At some points I have thought, 'Oh, my life is too gay.' I work in a gay environment, I live in a gay neighborhood, most of my friends are gay," remarked Stephen Richter, who rented an apartment in the Castro district. "But I don't know, you go out in the straight world and you can't wait to get home!" Ronnie Walker agreed: "For all the dishing that people do about Castro Street, whenever I go away to middle America, I'm always glad to kneel down and kiss the earth when I get to Castro Street." Others who lived in outlying areas traveled to gay neighborhoods for the express purpose of "feeling the community." Neighborhood had became another marker of the contrast between gay and straight, signifier of belonging, "home," and things held in common.

During the 1980s gay areas of San Francisco did not escape the restructuring of the urban landscape taking place in cities across the United States.[15]   On Castro and Polk Streets, many small gay businesses gave way to banks, chain stores, and franchises. Residents fought extension of the downtown financial district into the South of Market region. Even under these economic pressures, however, gay neighborhoods retained enough of their character to contribute materially to the formation of gay identity by offering a place to meet and forge ties to other gay people.

Because gay neighborhoods in San Francisco have been formed and populated principally by men, many lesbians looked to the Bay Area at large as a place to make such connections. John D'Emilio (1989b) has pointed out the link between male control of public space and the greater public visibility of gay male (as opposed to lesbian) institutions in the United States. Economic factors are also involved, since rental or ownership in the Bay Area can be prohibitive, and women in general receive lower incomes than men. By the mid‑1980s, however, lesbian institutions and residential concentrations had begun to appear in the less expensive Mission and Bernal Heights districts.

"In terms of meeting people," said Sharon Vitrano,

I feel a bit controlled by [being a lesbian], in that I'd like to at least have the option of living in a small town. One of the reasons that I came out here was that I felt I could meet lesbians in a context that was "normal." Where I could go about my business and meet people that way. I don't like having to hang out with a group of people just because they're gay.

Sharon's juxtaposition of small towns with life in the metropolis echoed the folk wisdom that gay men and lesbians are better off relocating in a big city where they can find others "like" themselves. Almost paradoxically, many people described the urban community they had hoped to discover in terms that incorporated mythical notions of the rural "America" of a bygone era. Expectations of homogeneity based on a common sexual identification lent credence to bids for political power, while depictions of lesbian and gay community as a club or secret society composed of "people who know people" invoked the face‑to‑face relationships supposed to typify small‑town life.

Nonterritorial understandings of community that rest on a sense of belonging with one's "own kind" have numerous antecedents in the United States; those most relevant to a gay context include such unlikely compatriots as religion and the tavern. Long before the first gay activists portrayed lesbians and gay men as sisters and brothers, the Puritans elaborated a notion of brotherhood based on the leveling effect of original sin (Bercovitch 1978; Burke 1941). A concept of "beloved community" ushered in the Civil Rights struggle so instrumental to the emergence of later social movements (Evans 1979). On the secular side, community has been symbolically linked to bars, saloons, and neighborhood in the United. States since the massive urban immigrations of the late nineteenth century (Kingsdale 198o). During that period, the saloon became a locus for the formation of same‑sex (in this case male) solidarity and a proxy for small‑town paradise lost. Although lesbians and gay men are now as likely to "find community" through a softball team, a coming‑out support group, or the Gay Pride Parade as through a bar, bars remain a central symbol of identity, and almost everyone has a story about a first visit to a gay club (see Achilles 1967).

Among political activists and the bar crowd alike, the notion of community voiced during the 1970s resembled nothing so much as a Jeffersonian version of Victor Turner's (1969) communitas: an alternative, nonhierarchical, and undifferentiated experience of harmony and mutuality.[16] Founded on the premise of a shared sexual identity, gay community remained, like friendship, an egalitarian and fundamentally nonerotic concept.

In extending homosexuality beyond the sexual, the notion of identity‑based community opened new possibilities for using kinship terminology to imagine lesbians and gay men as members of a unified totality.[17]  Identity provided the linking concept that lent power to analogies between gay and consanguineal relations. Wasn't this what families in the United States were all about: identity and likeness mediated by the symbolism of blood ties?

Yet the application of kinship terminology to gay community differed from the subsequent discourse on gay families in that it described all lesbians and gay men as kin: no "choice" determined familial relationships. To claim a lesbian or gay identity was sufficient to claim kinship to any and every other gay person. Some people hoped community would replace alienated biological ties (Altman 1979), appealing not to chosen families but to the collectivity: "If I could gain acceptance in the community of lesbians, I would have, I hoped, the loving family I missed" (Larkin 1976:84).[18] In gay bars across the nation, this was the era of circle dances to the popular music hit, We Are Family (Rodgers and Edwards 1979).

While the use of kinship terminology to indicate community membership has fallen into disfavor as the politics of identity have given way to the politics of difference, people still employed it from time to time as a way of hinting at sexual identity. "Don't worry, he's one of the brethren," explained a man I was meeting for lunch when his housemate walked into the room. On another occasion, a woman told me to expect a relatively smooth job interview because the person I would be seeing was "a sister." Marta Rosales, who worked at a hospital, reported one of the nurses asking if a new staff member was "family," and another woman remembered the back door of an East Bay bar being fondly termed the "family entrance." In 1985 a blood drive for persons with AIDS incorporated a unique play on biogenetic notions of kinship and the materialization of identity as shared substance. Leaflets bearing the headline "Our Boys Need Blood" called on lesbians as "blood sisters" to help "our brothers" in a time of need. By all accounts the drive was a great success, and soon became a model for similar events (with similarly styled publicity) across the country.

Tales of "coming home" into community are structured much like the scenes in Victorian novels that depict the recognition of concealed kinship. As metaphor, "home" merges the meanings of coming out and living in a place with a large lesbian and gay population (cf. Dank 1971:189).

[Coming out] was like coming home. I can't explain it. It felt so right. It really felt so right. It was like, you know, keeping your eyes shut and looking around a floor full of shoes and when you put your foot into your shoe you know it fits. You don't have to see it, you just know it.

Portrayals of fitting and belonging became a conventional element in coming‑out stories with reference to which individuals either equated or distinguished their experiences.

I've heard of people's experience, like moving from different parts of the country, moving here, and just like going into a women's bar and feeling, oh, wonderful. They've finally found their home, or something like that. The experience that I wanted, but I just haven't had . . . . I don't feel like I've come home or anything or that I belong here.

Identity and community, so often taken to define the limits of lesbian and gay experience, have become polarized in ways that presuppose culturally specific values of individualism. In the United States, tensions between notions of personhood and collectivity date back to Tocqueville's warnings about a tyranny of the majority. The paradigm that casts lesbians and gay men in the part of a minority (or subculture) interposes community between "the individual" and "society." In this context, it becomes relatively easy to move from a view of community as a comfortable home or unified interest group to a picture of community as a mini‑enforcer, mediator of all the conformity and oppression attributed to Society with a capital "S."

By the late 1970s, signs of disenchantment with the unity implicit in the concept of community began to appear: a popular critique of the look‑alike styles of "Castro clones"; a resurgence of butch/fem relations among lesbians that flew in the face of feminist prescriptions for androgyny; and a heated debate about sadomasochism (s/m), pedophilia, and other marginalized sexualities. Though some dissenters insisted upon their right to be included in the larger collectivity of lesbians and gay then, others did not experience themselves as community members, much less as agents in community formation. "I was just me, in a gay world," explained Kevin Jones.

During the same period, lesbians and gays of color critiqued the simplistic assumption that mutual understanding would flow from a shared identity. Along with Jewish lesbians and gay men, they drew attention to the racism and anti‑Semitism pervading gay communities, and exposed the illusory character of any quest for an encompassing commonality in the face of the crosscutting allegiances produced by an identity politics. Predictably, this recognition of differences, while important and overdue, tended to undermine meanings of harmony and equality carried by "community." Accompanying the positive explorations of what it meant to be black and gay or lesbian and Latina was widespread disillusionment with the failure to attain the unity implicit in the ideal of communitas.

Deliberating Difference

By the 1980s the rhetoric of brotherhood and sisterhood had begun to seem dated and trite. Sherry McCoy and Maureen Hicks (1979:66), attempting to grapple with "disappointment" and "unrealistic demands" among lesbian‑feminists, wrote, "The concept of `sisterhood' at times seemed to evaporate as we watched. "[19] This newfound reluctance to apply kinship terminology to all other lesbians and gay men extended well beyond activist circles. Many gay men and lesbians began to doubt the existence of "the" community or any single gay "lifestyle." Some abandoned the notion of identity‑based communities altogether, attempting to escape social categorization by adopting extreme forms of individualism. "I am who I am," they explained. Others associated community strictly with wealthy white men, who were neither representative of nor identical with the totality of gay people. Along with a recognition of the relative privilege of this sector came the refusal to allow this part to stand for the whole. Seemingly unable to comprehend the inequalities that structure identity‑based difference in the United States (white being privileged over Native-American, men over women, and so on), the concept of community lost credibility.

The most popular alternative was to divest community of its egalitarian associations by using it as a proxy for "population." Dissemination of Alfred Kinsey's (1948, 1953) data on the incidence of homosexual sex in the United States had opened the way for picturing an essential 10 percent who make up the imagined universe ("community") of gay men and lesbians. One indication of the extent of this muddle in the model of community is that, by the time of my fieldwork, most people qualified the term by adding a phrase such as "whatever that means."

The practice of identity politics in the United States has rested upon the cultural configuration of race, ethnicity, class, gender, and sexual identity as categories for organizing subjective experience (Epstein 1987; Omi and Winant 1983). What motivated the transition from "speaking sameness" to a division of community into ever‑narrower circumscriptions of identity?[20] In the first place, perceptions of fragmentation represent a view from the top. Attempts to understand the integration of sexuality with other aspects of identity were not experienced as "splits" by those who had never felt included in community from the start. Paradoxically, however, the very process of building gay community contributed to the emergence and timing of this discourse on difference.

John D'Emilio (1989b) has argued that the political tactic of coming out to others as a means of establishing gay unity had the contradictory effect of making differences among lesbians and gay men more apparent. The distance is considerable from the Chicago of the late 1960s, where Esther Newton (1979) found little social differentiation among gay men and no gay economy to speak of, to the San Francisco of the 1980s, where gay institutions had multiplied and residents were heirs to a social movement for gay pride and liberation. In the Bay Area the sheer size of the relatively "out" gay and lesbian population permitted the recognition and replication of differences found in the society at large.

During the 1980s, categories of identity remained integral to the process of making and breaking social ties among lesbians and gay men. Most gay bars and social or political organizations in San Francisco were segregated by gender. Some of the community institutions that lesbians associated with gay men maintained a nominal lesbian presence. A gay theater, for example, included scripts with lesbian characters in its annual repertoire, and the number of women in attendance grew from two or three to a third of the audience when lesbian plays were performed. Yet the most visible gay institutions, businesses, and public rituals (such as Halloween on Castro Street) remained male‑owned and male‑organized. Even the exceptions seemed to prove the rule. After a crafts fair in the gay South of Market area, the Bay Area Reporter published a picture of two women kissing, over the caption, "It wasn't all men at the Folsom Street Fair either." When gay groups in southern California suggested adding a lambda to the rainbow flag supposed to represent all gay people, lesbians denounced the addition as a noninclusive male symbol. At a benefit for the Gay Games sponsored by the Sisters of Perpetual Indulgence (a group of gay men in nun drag), lesbians cheered the women's softball game and martial arts demonstration, but some voiced impatience with "all the boys parading around in their outfits." Disagreements periodically erupted concerning the proportion of men's to women's coverage in newspapers that attempted to serve "the community" as a whole. It was not uncommon for lesbians and gay men to stereotype one another, building on constructions of identity and difference in the wider culture. Jenny Chin, herself Chinese‑American, combined notions of gender, sexuality, and racial identity with the image of the Castro clone to portray difference and position herself outside "gay community":

I would read the Bay Guardian, and they'd say "gay rap." And I would take all these buses crosstown, through all these parts of town I'd never been at night, and transfer, and wait on bus corners, and go to this huge room that had like 300 gay men . . . . These men were very much talking from their hearts, and they were really needing the support, but it's hard for me to identify with all these tall white guys with moustaches talking about how they're being judged because they're not coming well enough, or something like that.

Joan Nestle (in Gottlieb 1986) has condemned the essentialism implicit in generalizations that assert "lesbians do this, gay men do that." When it comes to something like public sex, Nestle points out, some do and some don't. Like other differences, divisions between lesbians and gay men are not absolute, but socially, historically, and interpretively constructed. After a women's musical troupe was asked to play for a gay male swimsuit contest, group members voiced positions ranging from "support our gay brothers," to "porn is porn," to "who cares, let's take the money!" Several lesbians cited their work with AIDS organizations as an experience that had helped them "feel connected" to gay men. Social contexts defined as heterosexual also fostered expectations of solidarity based on sexual identity. At one of our Thursday evening family dinners, Liz told with dismay the story of fighting with a male coworker at a holiday party given by her employer. "There we were," she explained, "the only two gay people in the place, having it out with each other."

Class differences traced out lines of division within as well as between the men's and women's "communities." Many lesbians attributed the visibility of gay male institutions to the fact that men in general have greater access to money than women. Gay vacation spots at the nearby Russian River proved too expensive for many lesbians (as well as working‑class and unemployed gay men), who tended to stay at campgrounds rather than resorts if they visited the area. Popular categories opposed "bikers" to "professionals" and "bar gays" (presumably working‑class) to "politicals" (stereotyped as "middle class"). People described making painful choices regarding employment, based on their perceptions of how out a person could be in a particular type of job. David Lowry, for example, had dropped out of an MBA program to become a waiter after he experienced pressure from corporate employers to be more "discreet" about his sexual identity.

Individuals who had purposefully sought employment in gay businesses reported their surprise at finding the gay employer‑employee relationship as marked by conflict and difference as any other (cf. Weston and Rofel 1985). In a dispute between the lesbian owner of an apartment building and one of her lesbian tenants, both sides seemed perplexed to discover that their shared sexual identity could not resolve the issue at hand. The hegemony of a managerial and entrepreneurial class within "the community" was also evident in the relative absence of gay owned and operated discount stores. While merchants encouraged people to "buy gay" and pointed with pride to the proliferation of shops that had made it theoretically possible to live without ever leaving the Castro, only a very small segment of lesbians and gay men could have afforded to do so, even if they were so inclined.

Anyone who visits a variety of lesbian and gay households in the Bay Area will come away with an impression of generational depth. Gay organizations and establishments, however, tended to serve a relatively narrow middle age range. Bowling, for instance, is a sport that many people in the United States pursue into their older years.  But gay league nights at bowling alleys across the city found the lanes filled with teams predominantly composed of men in their twenties and thirties. Young lesbians and gay men came to San Francisco expecting to find acceptance and gay mecca but instead experienced trouble getting into bars and often ended up feeling peripheral to "the community" (cf. Hefner and Austin 1978; Heron 1983). Gina Pellegrini had initially gained entrance to one bar with a fake ID, only to encounter hostility from one of the older "regulars":

I just felt like we all should have been the same no matter [what]‐age or not. And she was discriminating against her own quote "kind" unquote. That was very strange to me. I didn't realize that a fifteen‑year‑old could be pretty damn much of a pain in the ass when you want to relax and talk to your friends and have a drink.

For their part, older people mentioned ageist door policies at bars, and complained about feeling "other" when surrounded by younger faces at community events.

Racially discriminatory treatment at gay organizations, white beauty standards, ethnic divisions in the crowds at different bars, and racist door policies were other frequently cited reasons for questioning the community concept. Kevin Jones, an African‑American man, said that when he first came to San Francisco,

I thought that if I was white, it would be a lot different then. Because it seemed like it was hard for me to talk to people in bars. But it didn't seem like other people were having a hard time talking to each other. It almost seemed like they knew each other. And if they didn't know each other, they were gonna go up and talk to each other and meet. But I'd go to the bars, and I could sit there and watch pool, and nobody would ever talk to me. And I couldn't understand that. And I thought, "If I was white, I bet you I would know a lot more of these people."

Something more is involved here than racial identity as a ground for difference and discrimination, or ethnicity as an obstacle to the easy interaction implicit in notions of community. Most people of color claimed membership in communities defined in terms of racial identity, attachments that predated coming out as a lesbian or gay man. Simon Sub, for example, believed that his own coming out was complicated by thinking of gays as "very outside of my own [Korean-American] community." Metaphors like "home" served as we'll for describing race and ethnicity as sexual identity. Because his best friend was also Latino, Rafael Ortiz explained, "it makes it more like home." This is not to deny divisions of class, language, age, national origin, gender, and so forth that cut across communities organized through categories of race or ethnicity. It is simply to note that many, if not most, lesbians and gay men of color did not experience coming out in terms of any one‑to‑one correspondence of identity to community.[21]

Whites without a strong ethnic identification often described coming out as a transition from no community into community, whereas people of color were more likely to focus on conflicts between different identities instead of expressing a sense of relief and arrival. Implicit in the coming‑out narratives of many white people was the belief that whites lack community, culture, and a developed sense of racial identity. As Scott McFarland, a white man, remarked when we were discussing the subject of gay pride day, "There were no other parades that I could march in."

Division of the master trope of community into multiple communities has forced individuals to make difficult choices between mutually exclusive alternatives, like living in an Asian‑American or a gay neighborhood, or working for a gay or an African‑American newspaper. Some political activists have endeavored to fabricate a solidarity capable of spanning "the community" without denying differences that divide its members. The general trend, however, has involved building coalitions composed of autonomous groups that invoke more specialized combinations of identities (cf. Reagon 1983).

To avoid prioritizing identities, a person could integrate them‐seeking out other gay American Indians, joining a group for lesbians over 40, or hanging out in a bar for gays of color‐but this solution is limited in the number of identities and settings it can encompass. A person could move back and forth among communities as an "out" lesbian or gay man, giving up the hope of having all identities accepted in any one context. He or she could pass for heterosexual in situations defined by race or ethnicity, like Kenny Nash, who had decided to remain closeted to other African‑Americans. "I didn't want people to think that I'd left the [black] community," he explained, "so that therefore I had no right to speak about things that were of concern to me." Or that person could turn toward a radical individualism which focused on issues of style and railed against conformity, whether it be as a "lesbian for lipstick" or a gay man who objected to uniforms of jeans, keys, and sculptured muscles.

For some, sexual identity had become a minimal defining feature, all "we" have in common. Scott McFarland told the story of getting on the wrong bus when he first arrived in the city during the 1970s, and finding himself on Castro Street:

It just devastated me. [I thought], this is it! This is the dream of all these people like me moving to somewhere [gay] . . . . Everybody was dressed in these incredibly macho fashions . . . . These weigh‑a‑ton shoes. jeans. The first five years I lived in San Francisco, I refused to wear blue jeans . . . . It took me years to recover from finding out that gay people weren't like me much at all!

"I knew that I didn't fit into the Castro any more than I fit into my family," another man insisted. Whether that sense of difference was based on categorical understandings of self (mediated by race, age, class, gender) or on tensions between the individual and the social, the result has been a generalized rejection of the unity and above all the sameness implicit in the concept of gay community.

In contrast, the family‑centered discourse emerging during this period did not assume identity (in the sense of sameness) based upon sexuality alone. Lesbians and gay men who claimed membership in multiple communities but felt at home in none joined with those who had strategically repositioned themselves outside community in transferring the language of kinship from collective to interpersonal relations. While familial ideologies assumed new prominence in the United States at large during the 1980s, among gay men and lesbians the historical legacy of community‑building and subsequent struggles to comprehend relations of difference mediated a shift in focus from friendship to kinship. Meanwhile the possibility of being rejected by blood relatives for a lesbian or gay identity shaped the specific meanings carried by "family" in gay contexts, undermining the permanence culturally attributed to blood ties while highlighting categories of choice and love.

Defined in opposition to biological family, the concept of families we choose proved attractive in part because it reintroduced agency and a subjective sense of making culture into lesbian and gay social organization. The institutionalized gay community of the 1970s, with its shops and bars and associations, by the 1980s could appear as something prefabricated, an entity over and above individuals into which they might or might not fit. Most understood gay families to be customized, individual creations that need not deny conflict or difference. Family also supplied the face‑to‑face relationships and concrete knowledge of persons promised by the romantic imagery of small‑town community (Mannheim 1952). As a successor to nonerotic ties elaborated in terms of community or friendship, chosen families introduced something rather novel into kinship relations in the United States by grouping friends together with lovers and children within a single cultural domain.