The present research was designed to investigate women’s friendships across sexual orientation identities, specifically lesbian and bisexual women’s friendships with heterosexual women. Research on the experience of lesbian and bisexual women has de-emphasized friendships in favor of a focus on “primary” or romantic relationships (Weinstock, 2000) and on the consequences of being a sexual minority within a heterosexist culture (Herek, 1995, 1998; Katz, 1995). Friendships, when they have been considered, have largely been characterized as social networks that comprise lesbian communities (Esterberg, 1997) and function to provide a type of familial support (Nardi, 1992; Stanley, 1996; Weinstock, 2000).

Research on friendships in general indicates that individuals choose friends who are similar to themselves on the basis of age, race/ethnicity, and socioeconomic status (Duck, 1991). Likewise lesbian women are most likely to develop friendships with other lesbians (O’Boyle & Thomas, 1996; Rubin, 1985; Stanley, 1996). Past research has established that friendships with other lesbians can provide a buffer against being socially devalued as a sexual minority, and can provide an outlet for sharing daily aspects of life, especially for women who are not open about their sexual orientation identity. In addition, lesbian friends can serve as important role models in a culture where lesbian experience is not widely visible (Stanley, 1996).

Friendships enjoy a unique role in the lives of lesbians as they defy traditional heterosexist definitions of friendships (Rose, 2000). For example, Rose and Zand (2000) found that lesbians do not routinely report dating scripts similar to those of heterosexual individuals. Rather, a friendship script for dating tends to be both the most common and the most preferred way to enter into a romantic relationship for lesbians. Diamond (2002) suggested that, especially among young sexual minority women, friendships can lead to intense emotional interactions that incorporate elements of both friendships and romantic relationships. These “passionate friendships” may or may not include a sexual dimension. In addition, lesbian women, when asked about their friendships, count lovers and ex-lovers among their close friends (Clunis & Green, 2000; Nardi & Sherrod, 1994; Weinstock, 1998).

Given that friendships in general are more likely to develop between individuals with similar characteristics (Duck, 1991), an interesting research approach has been to study friendships that defy this trend—to study friendships across categories of difference. Researchers of lesbian friendships have also considered cross category differences: across race (Hall & Rose, 1996), across generations (Stanley, 2002), and across relationship status (Stanley, 1996). That research, however, emphasized lesbians’ friendships with other lesbians. Another way to consider lesbian friendship experiences across difference is to study friendships among individuals who have different sexual orientation identities.

Friendships Across Sexual Orientation Identities

Although the study of friendships provides a rich research area to explore interactions between individuals with different sexual orientation identities, the study of cross-orientation friendships has been limited. Existing studies suggest that cross-orientation friendships are rare and that, when they do exist, they do so despite a number of barriers (O’Boyle & Thomas, 1996; Price, 1999). In addition, much of existing research has focused on the negotiation of difference to the exclusion of other aspects of friendship, and has employed methodologies that limit understanding of the intimate dynamics that exist in close friendships.

Cross-orientation friendships among women

O’Boyle and Thomas (1996) were the first to attempt to characterize friendships between lesbian and heterosexual women. Data were collected from four focus groups: two groups composed exclusively of heterosexual women, and two composed exclusively of lesbian women. Although the focus group discussions revealed barriers to the development of cross-orientation friendships, the researchers concluded that meaningful friendships between lesbian and heterosexual women are possible. Although some issues regarding lesbian-heterosexual friendships were raised in their research, an understanding of close friendships was not possible because of the way in which participants were recruited for the study. Many of the heterosexual participants had never had a lesbian friend, and so they were more likely to describe how they felt about the possibility of such a friendship. Even for those participants who said that they had a close friend who differed in sexual orientation identity, the focus group discussions were more likely to elicit general feelings about cross-sexual orientation friendships than to allow for deep reflection of experiences within an existing close friendship.

Positive aspects of cross-orientation friendships: a focus on close friendships

Although the findings of O’Boyle and Thomas (1996) provide a context for researching issues relevant to cross-orientation friendships among women, their recruitment strategy ensured that the analysis remained general and less focused on the dynamics of existing close friendships. When close friendships are considered, positive experiences in cross-orientation friendships become apparent (Galupo & St. John, 2001; Weinstock & Bond, 2002). Weinstock and Bond (2002) surveyed heterosexual and lesbian women who had at least one cross-orientation friendship. Analysis of the participants’ responses revealed that sexual orientation differences led to unique friendship experiences that have both positive and negative aspects. In addition, lesbian and heterosexual participants highlighted different friendship experiences as salient.

Galupo and St. John (2001) examined the benefits of cross-orientation friendships by recruiting bisexual and lesbian participants along with their close heterosexual friends. The participants reported that benefits of cross-orientation friendships included an increased closeness and trust within the friendship that accompanied sexual orientation disclosure, provided objectivity in life, and broke down stereotypes. In addition, heterosexual women reported an increased sensitivity to sexual minority perspectives and an increased flexibility in understanding their own personal sexual identity. Through their friendships with heterosexual women, lesbian and bisexual participants gained an understanding that acceptance from others is possible, and they reported increased self-acceptance and self-esteem.

Focusing on the friendships of young women ranging in age from 19 to 25 years at the time of interview, Galupo and St. John (2001) provided the first data on cross-orientation friendships that included bisexual women. However, their analysis of ten friendships pairs did not distinguish between lesbian-heterosexual and bisexual-heterosexual friendships, as data for lesbian and bisexual women were analyzed together as “sexual minorities.” Because bisexual and lesbian identities lead to distinct experiences (Rust, 2000, 2001), it follows that the friendships lesbian and bisexual women form with heterosexual women may have distinct characteristics.

Galupo, Sailer, and St. John (2004) analyzed a larger data set of 14 friendship pairs including women ranging in age from 18 to 34 years with the goal of elucidating the unique experiences of bisexual women in cross-orientation friendships. Unique aspects of bisexual-heterosexual friendships were found to include a focus on the friends’ mutual attraction to men, the relative invisibility of bisexual identity within the friendship, and a general emphasis on the similarities between friends. The data indicated a perceived shift in friendship dynamic based on sex of the bisexual friend’s partner. That is, the ways in which the friends disclosed relationship experiences, discussed present and future events, perceived themselves to be similar, and experienced benefits of the friendship were contingent upon whether the bisexual friend’s current partner was female or male. Both friends perceived this shift even though they each maintained a stable sexual orientation identity.

Galupo et al. (2004) characterized bisexual women’s friendship experiences by comparing bisexual-heterosexual and lesbian-heterosexual friendships. Although we alluded to lesbian experience in these friendships, it was only discussed as a point of reference in understanding the experiences of bisexual women. A direct statistical comparison of friendship experiences was not done. Further research is needed to allow a distinction between how friendship dynamics are affected by a difference in sexual orientation in general, and how friendship dynamics are shaped by specific sexual orientation identities.

The present study

The present study is the first to provide a systematic comparison of friendship experiences across lesbian-heterosexual and bisexual-heterosexual pairs. The methodology is similar to that used in my earlier studies (Galupo & St. John, 2001; Galupo et al., 2004) in that participants were interviewed in close friendship pairs. Some interviews included in the present analysis were also included in previous studies. Additional interviews were conducted for the present analysis to allow a large enough sample size to permit comparisons across sexual orientation (lesbians and bisexual women) and across racial identity (same- and cross-race friendships). Although a developmental analysis was not conducted, the present study included a broader age range of participants than the previous studies did, and this allows a greater generalization of results.

A first set of analyses was conducted to characterize how participants’ friendship profiles were similar and different across sexual orientation identity. The following research questions related to participants’ friendship profiles were considered: Do bisexual, lesbian, and heterosexual women differ in (1) the total number of reported close friendships, (2) the number of reported same-orientation friendships; and (3) the number of reported cross-orientation friendships?

A second set of analyses considered whether heterosexual women’s friendships with lesbians or bisexual women are similar or different. Specifically, do lesbian-heterosexual and bisexual-heterosexual friendships exhibit the same characteristics in terms of (1) function of the friendship; (2) shift in friendship dynamic; (3) acknowledgement of non-heterosexual identity; (4) integration of friendship into the social lives of the friends; and (5) political or feminist dimensions to friendship.

A third set of analyses was conducted to determine whether racial differences within women’s cross-orientation friendships are present. Specifically, do same- and cross-race friendships differ on the following dimensions?: (1) acknowledgement of non-heterosexual identity; (2) integration of friendship into the social lives of the friends; and (3) political or feminist dimensions to the friendship. For cross-race friendships an additional analysis considered whether the difference in racial identity was explicitly addressed within the friendship.

Method

Participants

Recruitment into the study was based on the following inclusion criteria: both individuals in the friendship pair agreed to participate in the study, one friend identified as either lesbian or bisexual, one friend identified as heterosexual, participants were unrelated, and both regarded the friendship to have been close for at least 1 year prior to the interview.

Participants were recruited from an east coast U.S. metropolitan area. Recruitment announcements were posted in a variety of areas including academic and non-academic settings on a university campus and commercial locales both within and outside the lesbian/bisexual community. Newsletter advertisements and e-mail postings also announced the study to community organizations with and without feminist affiliations. Upon completion of the interview, participants were provided recruitment announcements to disseminate to individuals who might fit the inclusion criteria but might not otherwise have heard about the study. The goal of this recruitment strategy was to obtain a diverse participant pool in order to broaden the current research on lesbian friendships, which has disproportionately included White, professional women who identify as feminist and are active in the lesbian community (Weinstock, 2000).

Participants learned of the study from a variety of sources, which indicates that the recruitment strategy was somewhat successful. The following percentages characterize the participant pool on the basis of a number of recruitment indices. As these are not mutually exclusive categories, these numbers do not total 100%: 60% of the participant pairs were recruited from campus sources, 15% from lesbian/bisexual community sources, 5% from feminist organizations, 35% were referred by an acquaintance, and 25% of the interviews were initiated by the heterosexual friend.

The sample included a total of 40 volunteer participants (i.e., 20 cross-orientation friendship pairs). Participants ranged in age from 18 to 52 years at the time of interview, and friendship pairs reported friendship duration of 1–17 years. Age of participants at the onset of the friendship ranged from 10 to 40 years. All participants had had some college experience.

The present sample included 13 lesbian, seven bisexual, and 20 heterosexual women. Bisexual and lesbian participants had comparable experiences on a number of descriptive indices. Bisexual participants ranged in age from 18–30 years (M = 22.8), and lesbian participants ranged in age from 19 to 54 years (M = 25.4). Friendship durations for friendship pairs were 1–12 years (M = 4.3) for lesbian-heterosexual dyads and 2–17 years (M = 5.1) for bisexual-heterosexual dyads. Lesbian participants reported having identified as such for 1–11 years (M = 5.4), and bisexual participants reported having identified as such for 1–9 years (M = 4.3).

There was considerable diversity among the 40 participants, and more than one-third of them identified as racial/ethnic minorities. Participants described themselves in the following way: 26 White and 15 Women of Color (six African Americans, two Asian Americans, two Pacific Islanders, two Latinas, one Jamaican, one Afghan, and one Native American). In addition, three participants identified as Jewish (two White, one African American). Ten of the 20 cross-orientation friendship pairs were also cross-race friendships.

Measures

Prior to the interview participants completed a standard demographic sheet to answer questions about themselves and to provide background information on their friendships. Participants also completed a friendship questionnaire developed by the author to report basic demographic information (sex, race, age, education, sexual orientation) about themselves and their close friends. From the friendship questionnaire, a friendship profile was developed for each participant. A friendship profile describes the number of total close friendships as well as the number of same- and cross-orientation friendships for each participant.

Interview questions for the dyad interview covered six basic areas of inquiry, including a characterization of the positive and negative aspects of the friendship, similarities and differences across individuals in the friendship pair, friendship and the social life of friends, management of the friendship, conflicts and potential conflicts in the friendship, and how issues of sexual orientation affect friendship. Individual interviews were constructed to include six different categories of questions: individual definitions of friendship and friendship profile, initiation of friendship, function of friendship in life, perception of friendship by others, sexual tension or activity in friendship, and issues of sexual orientation. When friends differed in racial identity, an additional set of questions was asked during both the dyad and individual interviews regarding the negotiation of difference in racial identity within the friendship.

A semi-structured interview method was used, and each category of inquiry began with an open-ended question. Probe questions were then used to prompt more specificity in participants’ responses to ensure that the topics covered were comparable across participants.

Coding procedures

Based on past research (Galupo & St. John, 2001; Galupo et al., 2004; Hall & Rose, 1996; Rose, 1996) and upon emergent themes in the narratives, broad thematic dimensions were identified to form the basis of comparison across friendship types. Although many topics were covered in the interviews, the current analysis elucidated themes about general friendship dynamics and themes related to the acknowledgement of minority identities based on sexual orientation and race/ethnicity. Data from each participant pair were rated on the presence of the following themes: (1) the friendship functions just as any other friendship; (2) the friend is there when needed; (3) I can talk to the friend about anything; (4) shifts in friendship dynamic; (5) shifts in friendship dynamic based on the sex of the bisexual friend’s partner; (6) explicit acknowledgement of non-heterosexual identity; (7) friendship is integrated into the social lives of the friends; and (8) political or feminist dimensions to friendship. Cross-race friendships were also rated based on whether the difference in racial identity was explicitly acknowledged.

Each interview was coded in binary terms on the basis of the presence of the following themes. Because the interviews were fairly fluid in structure, responses to all interview questions were considered when the presence of the themes was coded. However, sometimes themes emerged in response to specific questions. Table 1 includes the questions typically associated with each of the themes, as well as sample responses that exemplified each of the themes.

Table 1 Thematic analysis: Example of interview questions and participant responses.
Full size table

The author and another researcher rated transcript records independently. The thematic analysis yielded 160 data points, for which there was an initial 93.1% inter-rater reliability. Discrepancies in ratings across researchers were discussed, and a final coding decision was agreed upon.

Procedure

Participants were recruited in close friendship pairs. Prior to the interview, each participant individually completed a demographic sheet and a friendship questionnaire. This approach ensured that participants were answering questions about cross-orientation friendships in the context of their experiences in a current stable friendship. Each participant was interviewed both individually and in the friendship pair to allow for deeper reflection on the friendship and ensure privacy while discussing potentially sensitive topics.

Three interviews were conducted for each friendship pair. Friends were interviewed first together and then individually; all three interviews were conducted on the same day. Two researchers were present at the dyad interview. Following the dyad interview, each researcher conducted an individual interview with one of the participants. Interviews were audio recorded, and the content was later transcribed. Individual interviews ranged in length from 29 to 60 min (M = 41.6), and dyad interviews ranged in length from 35 to 80 min (M = 53.6).

Results

Friendship profile data across sexual orientation identity

An initial analysis was conducted to assess the effect of sexual orientation identity on the friendship profiles of participants. This analysis considered only friendships with women. The mean number of friendships across sexual orientation identity is summarized in Table 2.

Table 2 Mean number of friendships across sexual orientation identity.
Full size table

A MANOVA revealed no significant main effect of sexual orientation on the total number of close friendships, F (2, 37) = .23, p > .05. However, there were significant main effects for sexual orientation on the number of cross-orientation friendships, F (2, 37) = 36.04, p < .01, and the number of same-orientation friendships, F(2, 37) = 10.35, p < .01. Post-hoc Bonferroni pair wise comparisons revealed that bisexual participants had significantly more cross-orientation friendships and fewer same-orientation friendships than lesbian and heterosexual participants did (p < .01). Lesbian and heterosexual participants did not significantly differ in their number of same and cross-orientation friendships.

Comparison of lesbian-heterosexual and bisexual-heterosexual friendship experiences

A comparison of lesbian-heterosexual and bisexual-heterosexual friendship experiences revealed both similarities and differences in these friendships.

Function of the friendship

Three analyses considered the function of close cross-orientation friendships in the lives of the friends. All 20 friendship pairs emphasized that their friendships were similar in function to any other close friendship. Participants described the importance of their friends in terms that were semantically equivalent to “being there when needed” and “having someone to talk to.”

When participants discussed their friendship, it was clear that sexual orientation was an ancillary focus. These friendships were experienced, first and foremost, as any other close friendship. No pattern of difference emerged in the friendship narratives between lesbian-heterosexual and bisexual-heterosexual pairs.

Friendship dynamics of lesbian-heterosexual and bisexual-heterosexual pairs

Friendship interviews were coded based on whether there was a general shift in the friendship for any reason. Results indicated that 100% of the bisexual-heterosexual pairs and 0% of lesbian-heterosexual had experienced a shift or perceived shift in their friendship. The basis for the shift in bisexual-heterosexual friendships was explored, and it was found that for 100% of the bisexual-heterosexual friendships the shift in friendship dynamic was attributed to the sex of the bisexual friend’s current partner.

Three additional analyses were conducted in order to explore the similarities and differences across lesbian-heterosexual and bisexual-heterosexual friendships. First, the effect of friendship type on the explicit acknowledgement of the non-heterosexual identity was considered. Chi-Square analyses revealed that an explicit acknowledgement of non-heterosexual identity was significantly more likely to be found in lesbian-heterosexual friendships than in bisexual-heterosexual friendships, X 2 (1, N = 20) = 13.65, p < .01. Lesbian identity was acknowledged in 100% of the lesbian-heterosexual friendships. In contrast, bisexual identity was explicitly acknowledged in 14.3% of the bisexual-heterosexual friendships.

A second analysis was conducted to examine the effect of friendship type on the way the friendship was integrated into the social lives of the friends. A Chi Square analysis revealed a significant main effect of friendship type, X 2 (1, N = 20) = 8.03, p < .05; bisexual-heterosexual friends were more likely (100%) than lesbian-heterosexual friends (57.14%) to be integrated into each other’s social lives. A third analysis was conducted to consider the extent to which a political and/or feminist aspect of cross-orientation friendships emerged. Overall, only 30% of the friendships pairs discussed their friendship in a political context, and there was no significant difference between lesbian-heterosexual and bisexual-heterosexual friendships on this dimension, X 2 (1, N = 20) = .29, p > .01.

Racial differences and women’s cross-orientation friendships

Chi square analyses were conducted to investigate the effect of racial differences across friends on three aspects of the friendship experience. Data from lesbian-heterosexual and bisexual-heterosexual friendships were combined.

No significant pattern of difference emerged between same- and cross-race friendships with regard to the explicit acknowledgement of sexual orientation identity within the friendship X 2 (1, N = 20) = .02, p > .01. Same- and cross-race pairs also did not differ in terms of whether the friendship encompassed a limited sphere within the social lives of friends, X 2 (1, N = 20) = 1.01, p > .01. However, a Chi-Square analysis did reveal a significant main effect of racial difference on the political and/or feminist framework of the friendship, X 2 (1, N = 20) = 10.77, p < .01. Cross-orientation friends who differed in racial identity (60%) were more likely than friends who had the same racial identity (0%) to conceptualize their friendship in a political and/or feminist framework.

Interview data from cross-race friendships were also coded based on whether the difference in racial identity was explicitly addressed in the friendship. Eighty percent of those friendships incorporated an explicit acknowledgement of the racial identity of the friends. The exceptions included a Native American-White and Asian-White friendship pair.

Discussion

Although the majority of analyses reported here focus on unique aspects of cross-orientation friendships among women, it is important to note that bisexual-heterosexual and lesbian-heterosexual friends consistently emphasized that their friendships function just like any other close friendship. Rawlings (1992) identified the two primary functions of friendships as “being there when needed” and “having someone to talk to.” Close cross-orientation friendships serve the common friendship function of mutual support, and all participants described their friendships using terms that were semantically equivalent to the primary functions described by Rawlings.

Friendship profiles

Women’s friendship profiles differed across participants of different sexual orientation identities. Although lesbian, bisexual, and heterosexual women did not differ in their total number of close friendships, bisexual women had significantly more cross-orientation friendships and significantly fewer same-orientation friendships than did either lesbian or heterosexual women. The present findings may be difficult to interpret because the participants in this study were recruited on the basis of at least one cross-orientation friendship. It is possible that the friendship profiles of women in this study are not representative of the friendship patterns of women in the general population because these participants may be more open to developing friendships across difference. However, the results of my recent study (Galupo, 2007) of the friendship profiles of sexual minorities clarify the present findings. In this study 405 self-identified lesbian, gay, and bisexual participants were asked to describe themselves and their closest friends on a number of demographic dimensions. Sexual minorities were significantly more likely to report cross-orientation as compared to same-orientation friendships. Bisexual women, in particular, had few friendships with individuals of the same-orientation identity, and their most frequent friendship type was with heterosexual women. The unique friendship profile for bisexual women is worthy of additional notice. Even though it may be expected from population demographics that bisexual women have few friendships with other bisexual women, it is likely that a combination of sexism, homophobia, bi-phobia, and racism create a context for the way in which friendship is experienced (Galupo, 2006).

Friendships across sexual orientation: notions of sexual identity and sexual experience

The results of the current study emphasize the difference in dynamic between bisexual-heterosexual and lesbian-heterosexual friendships. Friendship dynamics in bisexual-heterosexual pairs shifted based on the sex of the bisexual friend’s partner. Bisexual identity was rarely explicitly acknowledged within the bisexual-heterosexual friendships, and friends were more likely than lesbian-heterosexual friends to be integrated into each other’s social lives. Lesbian-heterosexual friendships incorporated an explicit acknowledgement of lesbian identity. The majority of cross-orientation friendships did not include a feminist/political dimension. However, cross-race friends were more likely than same-race friends to describe a feminist/political dimension to their friendship. Also, differences in racial identity were explicitly acknowledged in cross-race friendships even when sexual orientation differences were not.

The findings of the present research reflect the way in which sexual orientation identity is conceptualized. Rust (2000) outlined the late twentieth century reconstruction of identity around the notion of sexual eroticism in which lesbianism and heterosexual identities are viewed as polar opposites and bisexuality as a middle point along the same continuum. Within this framework, women’s bisexual experience, then, is distinguished from both lesbianism and heterosexuality on the basis of sexual expression. The present findings suggest that a difference in sexual orientation identity does not have universal implications for friendship dynamics in women. Rather, a difference in sexual orientation leads to distinct experiences for lesbian-heterosexual and bisexual-heterosexual friendships. Given the “opposite” way in which lesbianism and heterosexuality are conceptualized, it makes sense that when close friendships exist between lesbian-heterosexual women, differences in sexual orientation are highlighted, and lesbian identity is explicitly acknowledged. Close friendships between bisexual-heterosexual women, in contrast, do not incorporate explicit acknowledgement of the sexual orientation difference in the same way. Heterosexual women perceive their bisexual friends to be more like themselves than their lesbian counterparts (Galupo et al., 2004). This focus on perceived similarities between friends, however, renders bisexual identity invisible within the friendship.

An understanding of sexual identity does not completely explain the present findings, however. A shift in friendship dynamic occurs in bisexual-heterosexual friendships despite the stable sexual orientation identity of friends. This shift occurs based on the sex of the bisexual friend’s current partner, and it has important implications for disclosure within the friendship, acceptance of the friend’s partner, and validation of bisexual identity (Galupo et al., 2004). The shift in friendship dynamic is dependent, then, upon a behaviorally based criterion rather than upon sexual identity. When the bisexual friend is single or partnered with a man, there is a focus on a shared attraction to men, and the difference in sexual orientation is not explicitly acknowledged within the friendship. In contrast, when the bisexual friend is partnered with a woman, bisexual identity becomes visible, and the friendship operates in a way that is similar to the lesbian-heterosexual friendships described here. Friendship dynamics in cross-orientation friendships among women appear to be mediated by notions of both sexual identity and sexual behavior.

Golden (1996) emphasized that predictions about whether a woman considers herself heterosexual, bisexual, or lesbian cannot be made solely on the basis of past or present sexual attraction and involvement. The dichotomy between sexual experience and sexual identity was not lost on the women in the present sample. Participants selected themselves into the present study based on their sexual orientation self-identification. Consistent with the findings of Fox (1995), Golden 1996, and Rust (2001), the identities of our participants did not always match the range of the sexual attractions and relationships they had experienced. One of our bisexually identified participants had never acted on her sexual interest in women. One of the heterosexually identified women had previously engaged in a sexual relationship with a woman, yet maintained a heterosexual identity, and another had indicated a strong sexual interest in women although she had not acted on it. Many of the lesbian identified participants had previously engaged in sexual and/or romantic relationships with men.

Impact of race and sexual orientation on friendship dynamics

Cross-orientation friendships were generally not conceptualized within a feminist or political framework, unless racial differences were present. Initially these findings appear to contradict those of past research (Hall & Rose, 1996; Stanley, 1996) that showed lesbian friendships to include more political conceptualizations of friendship. That participants from the present study generally did not discuss their friendships in feminist or political terms should be interpreted in light of the specific demographics of the study and in terms of generational differences regarding the way in which women identify with the women’s movement. Schnittker, Freese, and Powell (2003) documented the recent decline in women (and men) who identify as feminists. This may account for the fact the majority of the participants did not describe their friendship experiences within a feminist or political context. In addition, past research on lesbian friendships has disproportionately represented White women with feminist perspectives and with ties to the lesbian community (Weinstock, 2000). Past research has also rarely included individuals who identified as bisexual. I intentionally sought participants that had diverse experiences; only 5% were recruited from feminist organizations, and only 15% were recruited from sources within the lesbian/bisexual community. In the interviews, many of these participants indicated that they were not comfortable with or interested in being part of a larger feminist or lesbian/bisexual community. Participants also reflected a broader age range and more racial diversity than was the case in past research. That the cross-orientation friendships reported here might be less conceptualized in feminist or political terms, then, is not surprising.

It is noteworthy, however, that a different pattern of friendship dynamics emerged based on whether friends also differed in racial identity. When friends differed in racial identity, minority identities (based on race) were explicitly acknowledged in the friendship. And within those friendships the discussion of feminism or politics in broader terms incorporated a discussion of racial inequity and White privilege. Although those discussions sometimes addressed sexual orientation and heterosexual privilege in a similar way, cross-race friends were not more likely than same-race friends to incorporate an explicit acknowledgement of sexual minority identities. With an acknowledgement of racial identity came a more politicized friendship dynamic that more closely parallels lesbian friendships described elsewhere (Weinstock & Rothblum, 1996). Rose (1996) suggested that the development of cross-race friendships depends upon more than just the racial awareness of a single individual. Rather, for a cross-race friendship to be initiated, both friends must demonstrate similar levels of racial awareness. The heightened criterion for cross-race friendships could explain the political/feminist dimension that occurs with differences in racial identity in cross-orientation friendships.

Awareness of sexual orientation identity does not seem to play a central role in cross-orientation friendships in a way that would parallel awareness of race in cross-race friendships. The negotiation of sexual orientation appears to be more subtle, a negotiation that has the potential to undermine the identities of sexual minority friends, as in the case of bisexual women. Sexual orientation identity has unique characteristics that distinguish its negotiation within a social context. For example, sexual orientation identity is fluid. The present research included some close friendships that developed prior to one of the friends “coming out” as either lesbian or bisexual. In addition, sexual minorities have some ability to decide how sexual orientation will be presented in a social context. For example, in a society where heterosexual identity is assumed, sexual minorities can choose to conceal, highlight, or downplay their sexual orientation to some degree. An individual, then, could negotiate sexual orientation identity up front, before it factors explicitly into the friendship dynamic.

This ability (and burden) to negotiate sexual orientation suggests that heterosexist norms within a social context (Katz, 1995) are not lost in close friendship dynamics. In lesbian-heterosexual friendships where lesbian identity is explicitly acknowledged, this acknowledgement is accompanied by a distancing of the friendship from the social lives of friends. The shifting of the friendship dynamic in bisexual-heterosexual pairs illustrates not only a heterosexual norm—where the bisexual friend’s interest and experience with women is not acknowledged on equal par with her interest and experience with men—but also demonstrates a unique bi-phobic response where bisexuality is not acknowledged as a valid and stable identity. Although bisexual-heterosexual friendships are more integrated into the social lives of friends, this integration occurs at the expense of the visibility of bisexual identity.

Limitations of the present study and directions for future research

The present research is a comparative analysis of lesbian-heterosexual and bisexual-heterosexual friendships among women. Participants were recruited and interviewed in close friendship pairs, which provided an in-depth analysis crucial for understanding the complex nature of this topic. However, because of the rather strict criteria for inclusion where both friends had to agree to be interviewed, and because participants represented a convenience sample from the eastern U.S., generalizations from these data should be made cautiously and within the context of the noted demographics of this sample. For example, it is important to note that participants represented a narrow range of educational experiences. And, although the friendship experiences discussed by participants spanned from adolescence to middle adulthood, a life-span analysis was not done.

The present study employed a unique approach for women’s cross-orientation friendship research, both in recruitment strategy and data collection. The recruitment strategy yielded a more diverse participant pool than in past studies on a number of dimensions including racial identity and inclusion of bisexual women. Both of these factors had direct implications for the results. The inclusion of a comparison of lesbian-heterosexual and bisexual-heterosexual friendships allowed an understanding of how friendships are affected by a general difference in sexual orientation and by what is unique to specific orientation identities. In order to complement this research, additional studies should be focused on the dynamics of bisexual-lesbian friendships. Of specific interest is whether bisexual-lesbian friendships exhibit a comparable shift in friendship dynamics that follows the sex of the bisexual friend’s partner. As sexual orientation identity is fluid, future research is also needed to understand better how changes in sexual orientation identity and sexual experience affect the intimate social dynamics between friends.

Continued research on friendship dynamics could allow a better understanding of the way in which women define themselves within a social context in order to establish and maintain the meaning of their sexual orientation identities. Continued research in the area of cross-orientation friendships may ultimately provide a model for understanding the ways in which individuals negotiate a larger social world dependent upon interacting with individuals with different sexual orientation identities.