Ambivalence in Gay and Lesbian Family Relationships

Abstract

Intergenerational ambivalence—the simultaneous presence of both positive and negative dimensions of a parent–child tie—is a concept widely used in family studies. Scholars have clarified the measurement of psychological ambivalence, or an individual’s own feelings of ambivalence towards others. Yet research has yet to demonstrate whether—and, if so, how—individuals characterize others as ambivalent. Moreover, relatively little is known about ambivalence in gay and lesbian families. In the present study 60 in-depth interviews were analyzed to identify what the author calls perceived ambivalence in the parent, sibling, extended kin, and “in-law” relationships of gay and lesbian adults. Perceived ambivalence is revealed through gay and lesbian adults’ characterizations of family members’ simultaneous positive and negative overt and covert beliefs and behavior. In addition, the author refines the concept of collective ambivalence, wherein perceived ambivalence typifies an entire family unit. The findings further revealed the importance of broader sociological factors, such as homophobia, in structuring perceived ambivalence.

Over the past decade, intergenerational ambivalence has emerged as a central concept for understanding relationships between adult children and their parents (Lüscher & Pillemer, 1998). Intergenerational ambivalence brings together psychological ambivalence—the simultaneous experience of opposing feelings or emotions (Bleuler, 1922)—and sociological ambivalence—incompatible and conflicting expectations and norms of behavior, beliefs, and attitudes (Connidis, 2015; Merton & Barber, 1963)—to articulate how parents and adult children experience “opposing feelings or emotions that are due in part to countervailing expectations” for how each generation should act (Connidis & McMullin, 2002b, p. 558; Lüscher & Pillemer, 1998). A significant body of work demonstrates that, much like positive and negative parent–child relationships, ambivalent intergenerational relationships are negatively related to psychological well-being (Kiecolt, Blieszner, & Savla, 2011; Suitor, Gilligan, & Pillemer, 2011), which may in turn lead to stress spillover and proliferation into other domains of family life (Pearlin, Aneshensel, & LeBlanc, 1997).

Despite important advances in the ambivalence construct, significant gaps remain. First, the focus has been nearly entirely on individual feelings of psychological ambivalence toward others, with little attention to the presence and operation of sociological ambivalence (Connidis, 2015). Second, a focus on individuals’ own feelings of ambivalence toward others has stunted a view of how family members may construct others as ambivalent. Constructing others as ambivalent, much like constructing others as loving or disapproving, likely has an impact on individual and family well-being. Third, the focus has been nearly entirely on intergenerational ties. Yet ambivalence likely occurs in other lifelong and emotionally close family relationships, such as sibling, extended kin, and in-law ties (Connidis, 2007; Matthews, 2002; Suitor et al., 2009; Ward, Deane, & Spitze, 2008). Finally, previous research has focused on relationships between presumably heterosexual children and heterosexual parents. This absence of non-heterosexuals is notable, as recent work shows that gay and lesbian families have unique dimensions unarticulated in heterosexual families (Cohler, 2004; Connidis, 2007; Ocobock, 2013; Reczek, 2014a). For example, broader institutional forces of homophobia and heterosexism that structure the family relationships of gay and lesbian adults may engender an exceptional view of sociological ambivalence (Connidis, 2012), one that in turn provides a lens into the causes, dynamics, and consequences of family interaction. As such, a study of ambivalence in gay and lesbian families informs a theoretical and empirical account of broader family relationships.

To advance an understanding of ambivalence, gay and lesbian families, and family systems more broadly (Bowen, 1978), in the present study I analyzed qualitative interviews with 60 gays and lesbians to determine the nature of ambivalence in family-of-origin (e.g., parents, siblings, extended kin) and “in-law” (i.e., partners’ family of origin) relationships. Specifically, in this study I aimed to identify how gay and lesbian adults narrate their family members as exhibiting co-occurring positive (e.g., loving, giving of instrumental or emotional support) and negative (e.g., rejecting, disapproving) feelings and actions (Gilligan, Suitor, Feld, & Pillemer, 2015; Willson, Shuey, & Elder, 2003) within the structural conditions of a gay or lesbian family (Connidis, 2015). In doing so, this study moves beyond research focusing on individuals’ reports of their own experiences of ambivalence toward a study of how adults construct others as ambivalent, with specific attention to the intermingling of psychological and sociological ambivalence. This study provides a new lens through which to view how adult gays and lesbians—a marginalized group—experience family ties, in turn revealing new dimensions of family relationships previously undiscovered by heteronormative family research.

Ambivalence in Family Relationships

Intergenerational ambivalence highlights the fact that both solidarity and conflict coexist in parent–child relationships (Lüscher & Pillemer, 1998). Ambivalence operates on both the psychological level—wherein parents and children experience mixed feelings, emotions, and sentiments—and on the sociological level—wherein social-structural conditions engender contradictory expectations between adult children and their parents (Connidis, 2015; Connidis & McMullin, 2002a). Psychological and sociological ambivalence are distinct, but they overlap, wherein the contradictions created by structural norms and institutions are revealed in family interaction (Connidis, 2015). In this way, psychological ambivalence is experienced in relation to—and as a result of—the broader structural contradictions of sociological ambivalence, although few studies have directly addressed the interaction of micro-level psychological ambivalence and macro-level sociological ambivalence (Connidis, 2015). Recent research suggests that intergenerational ambivalence is common in the parent–adult child tie (Peters, Hooker, & Zvonkivoc, 2006; Rappoport & Lowenstein, 2007). About 50% of parents report some degree of ambivalence toward their adult children (Fingerman, Hay, & Birditt, 2004; Pillemer & Suitor, 2002), and about 30% of adult children experience ambivalence toward their parents (van Gaalen & Dykstra, 2006; Willson et al., 2003). This body of work has focused nearly entirely on understanding how one individual feels ambivalent about others and thus has failed to account for how one may perceive others as experiencing ambivalence.

Although intergenerational ambivalence may theoretically occur in any parent–adult child tie, ambivalence is more likely to occur when commonly held and institutionalized beliefs, expectations, and practices regarding parent–child relations are violated (Kiecolt et al., 2011; Pilliemer et al., 2007; Pillemer & Suitor, 2002; Ward et al., 2008; Willson, Shuey, Elder, & Wickrama, 2006). This applies especially to norms of independence and dependence (Birditt, Fingerman, & Zarit, 2010; Hillcoat-Nallétamby & Phillips, 2011; Ingersoll-Dayton et al., 2011). For example, parents are more likely to experience ambivalence when an adult child does not achieve normative adult statuses (e.g., complete college; Pillemer & Suitor, 2002); does not share in parents’ values and opinions (e.g., religious values; Pillemer et al., 2007); has lifestyle–behavioral, emotional, or physical health problems (Birditt et al., 2010; Kiecolt et al., 2011); or requires financial support (Pillemer & Suitor, 2002). In addition, parents experience higher levels of ambivalence when their children are unmarried (Kiecolt et al., 2011; Pillemer et al., 2007; Pillemer & Suitor, 2002), choose romantic partners disapproved of by parents (Peters et al., 2006), or fail to maintain romantic relationships (Birditt et al., 2010; Peters et al., 2006; Pillemer et al., 2007). These circumstances present contradictory structural norms of parenthood wherein parents are expected to encourage independent lives in their adult children while also providing continued, nonnormative assistance.

The parent–child tie has been the primary site for studying ambivalence in family relationships, yet hypothetically any relationship that is long term and emotionally close is a potential site of ambivalence (Bowen, 1978; Connidis, 2007; Merton & Barber, 1963; Rook, 1997). Adolescents report diminishing ambivalence toward their siblings during the transition to adulthood (Fingerman & Hay, 2004). Sibling ambivalence in adulthood occurs as a consequence of perceived parental favoritism (Suitor et al., 2009), competition between siblings, and unequal accumulation of resources (Connidis, 2007; Matthews, 2002; Whiteman, McHale, & Soli, 2011). Fingerman and Hay (2004) did not find high levels of ambivalence toward extended kin (e.g., aunts/uncles, cousins, grandparents); however, Connidis (2003) found that extended kin exhibit ambivalence toward a divorcing family member. Ambivalence may also result from conflicting norms regarding levels of independence and closeness between kin (Mason, May, & Clarke, 2007), or because extend kin relationships tend to be lifelong and relatively close, yet involuntary (Bulanda, 2011; Mason et al., 2007; Silverstein & Giarrusso, 2010). In the same vein, in-laws may also be an object for ambivalence “because these relationships come pre-packaged with other, more intimate social ties” (Fingerman & Hay, 2004, p.135). Fingerman and Hay (2004) did not find that individuals report ambivalence toward in-laws, yet Willson and colleagues (2003) reported higher rates of ambivalence in adult children’s relationships with in-laws than in relationships with one’s own parents. In a move toward broadening the scope of intergenerational ambivalence, Ward (2008; Ward et al., 2008) demonstrated that a mother can feel positively about one child, and negatively about a different child, a concept known as collective ambivalence.

Ambivalence in Gay and Lesbian Family Relationships

Only a handful of studies have empirically examined ambivalence in gay and lesbian families. To date, Connidis (2003), who uses a case study approach, and Reczek (2014a) and Cohler (2004), who use in-depth interviews, have provided some initial evidence that analytically suggest that parents feel ambivalent toward a gay or lesbian child, particularly during the coming-out process. However, these studies do not provide an account of how and why adult children believe their parents are ambivalent in the gay and lesbian structural context. This deficit is glaring because gay and lesbian families provide a unique vantage point of interplay between psychological and sociological ambivalence given widespread structural aspects of gay and lesbian identity (Connidis, 2015). For example, heterosexuality is a central organizing principal of everyday life; the belief that lifelong heterosexuality is a normal, natural, and inevitable course of development is so pervasive that it is assumed that everyone is (i.e., heteronormativity) and should be (i.e., heterosexism) heterosexual and that any sexual identity outside of heterosexuality is undesirable or “spoiled” (i.e., homophobia; Goffman, 1986; Meyer, 2003; Schulman, 2009). Family ties are embedded within these broader social structural factors of heteronormativity, heterosexism, and homophobia (Connidis, 2012; Connidis & McMullan, 2002a; Heatherington & Lavner, 2008).

Research shows that parents have the expectation that their children will be heterosexual and marry someone of a different sex (Schulman, 2009). Failure to achieve heterosexuality has been shown to promote parental feelings of disapproval, distancing, disappointment, disgust, and guilt over a perceived role in raising a gay or lesbian child (Biblarz & Savci, 2010; Cohler, 2004; LaSala, 2000, 2001; Ocobock, 2013; Oswald, 2002a, 2002b). Homophobia and rejection of a gay or lesbian adult is often, but not always, tied to underlying structural notions of religious moral values (Jones, Cox, & Navarro-Rivera, 2013); families with religious members may experience a structural context that is typified by heightened homophobia. At the same time, parents may experience affection, love, and concern for maintaining family solidarity (Cohler, 2004; Connidis, 2012). These contradictions may be created by broader structurally ambivalent expectations wherein parents reject their adult children for failure to adhere to expectations of heterosexuality in a homophobic society while also expressing love and support for their adult child (Cohler, 2004; Connidis, 2012). No studies have examined whether adult children perceive their parents as ambivalent in the context of these processes.

Widespread homophobia, heterosexism, and heteronormativity may also structure ambivalence in other family ties (e.g., extended kin, siblings; Oswald, 2002a; Scherrer, 2010). A gay or lesbian family member may disrupt desired consensus regarding morally “correct” life paths (Cohler, 2004). Although siblings and extended kin may disapprove or perceive loss, they may wish to simultaneously maintain solidarity with their gay or lesbian family member (Miz, Turell, & Meier, 2004). Ambivalence is likely present, and perhaps exacerbated, in relationships with a partner’s family (i.e., “in-laws”). (The term in-law is used most easily demonstrate the nature of relationships between an individual and the intimate partner’s family of origin members. There is not necessarily a legal connection between “in-laws.” This term is used for both ease of discussion and because respondents used this term in their interviews.) Although family members may experience more positive interactions with their own child, sibling, or extended kin, an intimate partner may become the source of blame as the embodiment of a gay or lesbian identity. Virtually no research has examined how gay or lesbian adults’ conceptualizations of their own family of origin can be perceived as ambivalent. .

Method

Data

In the present study I analyzed 60 in-depth interviews with gay- and lesbian-identified adults in 30 long-term cohabiting relationships. Interviews were collected as part of a larger study on family dynamics that included 60 in-depth interviews with heterosexuals not analyzed in the present study. With institutional review board approval, a research team composed of the author and three research team members conducted interviews in a midsized southwestern city in the United States. The sample was restricted to individuals in a self-defined “committed” relationship for 7 years or longer because the goal of the larger project was to capture the family dynamics of individuals in long-term relationships. Including individuals in committed intimate relationships allows for the analysis of retrospective reports of both partners’ families of origin across the life course, enhancing sample richness (Roy, Zvonkovic, Goldberg, Sharp, & LaRossa, 2015). Each partner in a couple was interviewed separately in order to obtain independent accounts (Reczek, 2014b). Interviews lasted, on average, 1.5 hours and were conducted in the respondent’s home or at university offices. Interviews, conducted from 2005 to 2007, were recorded and transcribed.

Sample recruitment was conducted with strict attention to qualitative integrity regarding sample richness and quality as well as data saturation (Roy et al., 2015). This was done from an interpretivist and social constructivist epistemological framework (Reczek, 2014b). Respondents were recruited at a local Pride Festival, which attracts a large percentage of the city’s mid-life gay and lesbian population. After initial contacts were made via the festival, recruitment continued through a variety of methods (e.g., a story in the city newspaper and a local magazine directed at lesbian/gay/bisexual/transgender individuals, flyers, word of mouth). Respondents were chosen with attention to racial and socioeconomic diversity aligned with city income and racial composition (Sandelowski, 1995). Sample richness and saturation were a main priority in determining how many respondents to include in the sample (Roy et al., 2015). Because gay and lesbian family ties are relatively rarely studied, this sample was exploratory. Thus, a goal was to obtain at least 60 interviews with gay and lesbian adults in order to capture a wide range of experiences (Sandelowski, 1995). After the initial 60 interviews were collected, I determined that data had reached theoretical saturation on the topics of intimate and family relationships, wherein clear and repeating—yet rich and multifaceted—patterns emerged during initial readings of the transcripts (Daly, 2007). Forty-nine respondents were White; eight were Hispanic, Latino, or Latina; one was Black; one was Native American/Hispanic; and one was South American. Household income ranged from $40,000 to $120,000. The average age was 49 years (range: 31–72) for gay men and 43 years (range: 29–60) for lesbian women. The average relationship duration for gay couples was 21 years, and for lesbian couples it was 14 years. This is consistent with some research that suggests lesbian women have shorter relationship durations than gay men within the context of mid-life adults (Lau, 2012).

The purpose of the interviews was to obtain narratives that focused on general family dynamics; topics included relationship quality and satisfaction between intimate partners, coming-out experiences, relationships with family of origin, mental and physical health, unemployment, children, sexual behavior, and relationship conflict. All discussions of family relationships were analyzed, including a section of the interview specifically focused on family ties as well as any extemporaneous mention of family ties that occurred ad hoc throughout the interview. To obtain a narrative of family ties, interviewees were asked several initial questions about family life; for example: “What is your/your partner’s relationship like with your/your partner’s family?”; “Have you/your partner disclosed your intimate relationship to your/his/her family?” A series of follow-up questions elicited in-depth accounts of relationships with one’s own and one’s partner family; for example, “How does your relationship with your mom make you feel?” and “Can you think of a time when she made you feel good? Unhappy?” This line of questioning continued until the respondent had nothing further to add about any family member. As such, there is clear integrity in closeness of fit between the unit of observation (respondents with family members) and the unit of analysis (respondents’ perceptions about family members; Roy et al., 2015).

Analysis

To identify and analyze the perception of both psychological and sociological ambivalence in interviews, I took a multistaged, standardized approach that emphasizes the dynamic construction of codes for the purpose of developing analytical and theoretical interpretations of data. Notably, family scholars have paid most attention to psychological ambivalence; few empirical directives for measuring sociological ambivalence exist. Thus, the present study was exploratory in identifying sociological ambivalence, with open-ended and emergent analysis. Inductive reasoning primarily guided the analysis, wherein patterns and conceptual categories were identified as they emerged from the transcripts, not from predetermined categories. Respondents were not given prompts as to whom in their family to discuss in order to allow for a focus on whomever the respondent deemed the most relevant family ties.

Following the initial reading of each interview transcription, I reread all transcripts to conduct line-by-line, data-driven categorization in order to summarize each piece of data (e.g., mother disapproves of gay identity; father is supportive of commitment ceremony). Next, I once again read the transcripts and used “focused” coding, which involved constructing categories by connecting initial codes together for the development of themes of ambivalence. A broad, inductive, and interpretive view of ambivalence was taken in order to account for all possible occurrences that often were discussed in subtle ways that spanned decades. This analytical technique is guided by two quantitative understanding of intergenerational ambivalence that have emerged from previous literature. In quantitative literature indirect ambivalence is measured as two separate survey questions that attain independent measures of solidarity and conflict; a scale is created from these two questions that determines whether there are concurrently high degrees of solidarity and high degrees of strain in that relationship. Direct ambivalence is obtained with the direct question: “Do you have ‘mixed feelings’” about a family member (Suitor et al., 2011; Willson et al., 2003)?

In the present study interviewees were asked to provide general accounts of a family relationship. From these general accounts I interpreted the presence of ambivalence. This most closely aligns with the measurement of an indirect ambivalence wherein independent measures of individuals’ positive and negative feelings are combined by the researcher into an ambivalence index (Fingerman et al., 2004; Gilligan et al., 2015; Willson et al., 2003). Notably, the measure of indirect ambivalence in the present study is different from quantitative work in that a numerical scale cannot be created to measure the degree of ambivalence; instead, a qualitative analysis relies on the author’s subjective identification of ambivalence. For example, two different portions of the same interview may be coded “sister is loving” and “sister disapproves of gay relationship.” In a secondary analytical stage the author re-codes these independent codes as “sister both loving and disapproving” (i.e., ambivalence). Respondents were not asked to identify direct moments of contradiction, mixed feelings, or ambivalence (i.e., direct); no respondents used ambivalence terminology. Nevertheless, some responses to the general indirect questions about a family tie resulted in a respondent articulating that they believe that family member experiences mixed feelings. Thus, this analysis moves beyond a dichotomous understanding of indirect versus direct ambivalence to draw attention to how the interpretation of ambivalence in interview data is completely subjective, driven by the author’s characterization of the broad category of ambivalence regardless of “direct” or “indirect” categorizations.

In the final stage of analysis I analyzed how the categories and subcategories of ambivalence relate to one another on a conceptual level. These conceptual themes are outlined in each theme and subtheme in the Results section. It is of note that the analysis relied on perceptions of a family member, and thus this analysis is a departure from previous work on ambivalence, which characterizes only one’s feelings of ambivalence about others (Lüscher & Pillemer, 1998; Pillemer & Suitor, 2002). As such, the findings reflect interviewees’ own realities and have no bearing on another family member’s feelings. However, perceptions of others’ ambivalence, as shown below, are a critical dimension of family life and thus an empirical and interactional reality in their own right. Note that nearly all findings of ambivalence were related to a gay or lesbian identity or relationship; however, other sites of ambivalence did arise (e.g., conflict with a family member over money, family decisions, divorce). In order to keep the focus on gay and lesbian relationships, only occurrences of ambivalence that directly relate to a gay or lesbian identity or relationship are presented in this article.

Results

The analysis revealed a wide range of relationship dynamics that are consistent with previous research on gay and lesbian families. Most respondents had at least monthly communication with parents and siblings by phone or in person; a small minority had contact with family members less often. About one quarter of respondents lived in the same town as a family member at the time of the interview. Fewer than one quarter of respondents in this study had not disclosed or openly discussed their gay or lesbian identity or partnership with at least one family member for various reasons (e.g., fear of rejection, privacy, protecting family members, religious values). These respondents tended to be about 5 years older than the average age of the sample and were more likely to be men. Respondents who were out to all of their family members disclosed their sexual identity at various stages across the life course. A small minority who had come out did so prior to their current relationship, but most who came out did so within at least 2 years of marking their commitment to their current partner. As is consistent with previous research (Coleman, 1982), siblings were often the first to be told of a gay or lesbian identity, followed by parents and extended kin.

Approximately 85% of the sample described least one family member, or an entire family unit, in ways that align with the concept ambivalence. The findings revealed three unique ways ambivalence was manifest in respondents’ accounts. First, in the theme overt perceived ambivalence across the family history, respondents discussed contradictory accounts of positive and negative exchanges and interactions that accumulate across the family history. Second, in the theme covert perceived ambivalence across the family history, respondents constructed the contradiction of a family members’ perceived feelings of disapproval alongside overt positive family interactions across the family history. Third, respondents situated each individual relationship collectively in an ambivalent family unit across the family history. Nearly all respondents had one family member (or the entire family unit, in the case of collective ambivalence) characterized within one analytical category of ambivalence and additional family members typified in another; an examination of multiple family relationships (e.g., siblings and siblings-in-law, parents and parents-in-law) allowed for the articulation of the multifaceted and diverse family unit. Therefore, the three themes described in the following sections are not mutually exclusive but instead reflect the range of descriptions of family members. Accounts were nearly always consistent across both partners in a couple; discrepancies are noted below.

Overt Perceived Ambivalence Across the Family History

A majority (70%) of respondents described a family relationship—most often parents and parents-in-law but also siblings and siblings-in-law—as explicitly both positive and negative across the family history; these descriptions are characterized as overt perceived ambivalence. Overt perceived ambivalence occurs when a family member is characterized as exhibiting simultaneously outward support and outward disapproval in either one moment in time or across multiple moments throughout the family history.

Gus and Andrew, partners for 23 years, described how overt acceptance and rejection had always been present in the relationship with Gus’s mother. As an illustration, Andrew said:

She has actually gone as far as sending birthday cards to Gus that say, “You will always be my son, I love you even though you are going to burn in hell.” That is quite a birthday card to get.

Andrew described one of the most obvious examples of outward love and explicit disapproval in the same moment in time, a marker of ambivalence present throughout his adult life. Similarly, Andrew described that there are reoccurring moments of both acceptance and nonacceptance of his relationship with Gus. For example, he recounted the events of one Christmas:

My mom was fairly kindhearted but she never really accepted it on religious reasons. But she accepted us, and she kept her point of view to herself. Except on very rare occasions when she would just say something out of the blue. One Christmas, we were all here together celebrating Christmas together at our house. And my mom just out of the blue said, “Well, someone really ought to get y’all separate beds.”

Andrew and Gus articulated how family members exhibit explicit moments of both overt support and overt disapproval, wherein Andrew’s mother both “accepted” them into her home but “never really accepted” their relationship in a sustained way throughout the entire family history.

Other respondents recounted the ways family members oscillate between outward positive interactions and outward negative interactions at different points across the family history; these past and present events intertwined in the interview narrative to fit with the analytical category of perceived ambivalence. Albert, partnered to Larry for 23 years, described how his mother has historically been supportive but recently disapproving. After the death of Albert’s father, Albert’s mother’s relationship to Albert and Larry shifted. Albert says, “My mother suddenly decided, ‘Huh! It’s time to focus on Albert. He’s living in sin!’” Albert recognized his mother’s contradictory actions, wherein she has previously accepted Larry but has recently expressed her condemnation of his intimate relationship with her son. Albert, like others in this theme, was surprised by recent outbursts of negativity in what has been considered a historically positive tie—demonstrating perceived overt ambivalence. Structural changes in the family—such as the death of a parent—are often precedents to contradictory actions and conversations. In this way, transitions and events are catalysts for perceiving a family relationship as outwardly ambivalent, as shown in other contexts outside gay and lesbian families (Connidis, 2015).

Similarly, Ann discussed how there have been clear overt moments when the parents of her partner of 14 years, Jullian, have rejected their relationship in the past—a feeling that has not dissipated but still lingers—but that her parents also exhibit behavior that is simultaneously accepting:

In the first year, her parents weren’t speaking to us, and we got the Jesus talk about how we live in sin and that we can’t come into the[ir] house. They just had to make their point. They still do from time to time. They told me two years ago, “Ann, we tried not to love you but we can’t help it. We do.” They can’t help it. It’s just ingrained in them.

Ann said she believes that although Jullian’s parents overtly demonstrate love for Anna and Jullian today, Ann has past overt evidence of “ingrained” disapproval that also exists in the present. With a view of the whole family history, an ambivalent view of Jullian’s parents emerged. Similarly, Megan described how the parents of her partner of 12 years, Clarissa, have seemed overall accepting in recent years, yet her parents have previously expressed their opposition:

There were a couple times where [Clarissa’s mom] was quite opposed to it and was verbal about it. And actually, I get along well with her family. Probably the only problem was that we did have a relationship, otherwise they would really love us.

Megan said she believes that she, Clarissa, and Clarissa’s mom have a positive relationship, but her previous experiences of rejection reveals the characterization of Clarissa’s as ambivalent. Clarissa, in turn, identified a past overt moment of rejection from Megan’s mother, describing the contradiction of past rejection and current affirmation:

Her mom is so strange. She didn’t come to the commitment ceremony. She said she didn’t agree with us doing that. And of course it hurt Megan very deeply. Yet, she [now tells] me she wants me to call her Mom, and she has referred to me as her daughter-in-law, compared me to her son-in-law. So it is kind of strange, for her to say one thing and then—mixed messages. But, I get along well with her.

Clarissa and Megan both described how interactions that are perceived as both supportive and rejecting accumulate over the long view of a family history, facilitating the presentation of overt evidence of an ambivalent family member in the interview context.

Covert Perceived Ambivalence Across the Family History

The theme above demonstrates the outward and explicit positive and negative interactions characterized as overt ambivalence. More than half (about 55%) of respondents in the sample described a second theme of covert ambivalence, wherein a family member is described as exhibiting overt evidence of positive interactions but are also characterized as having non-explicit covert or clandestine negative feelings. This is described in two primary ways: (a) perceived religious and/or homophobic disapproval, and (b) the “glass closet”—the “open secret” of respondents’ sexual identity (Sedgwick, 1990). These subthemes were described most commonly in regard to parents, parents-in-law, and siblings.

Perceived religious and/or homophobic disapproval

Respondents described that, despite outwardly positive interactions, they perceive that family members secretly have underlying negative feelings, a dynamic I characterized as covert ambivalence. Respondents’ covert ambivalence is a consequence of broader perceptions of homophobia due to the knowledge of a family member’s religious beliefs. Spencer, partnered to Elliott for 25 years, said, “I think with both sets of parents, even though there’s embracing, there’s also an embarrassment factor for them. You can feel that when you’re with them, you know, your son is gay.” Although Spencer and Elliott’s parents do not outwardly reject either partner, both partners said they believed that both sets of parents covertly experience irreconcilably conflicted feelings of disapproval due to a broader social context of homophobia. Stanley, partnered to David for 16 years, discussed this perceived belief regarding his sibling:

One of my siblings works at it harder than the rest of them, and her husband struggles with it quite a bit. Never outwardly. Never disrespectful to us. Don’t treat us differently. But there are things that they will say you can tell that they struggle with it a little bit more. It’s an issue they are dealing with. But like I said, they are still respectful and they are still good to us.

Stanley, like others in this theme, said he assumes that one of his sisters and brothers-in-law internally “struggle” with him being gay—evidence of these negative feelings or discomfort with Stanley and David’s relationship—yet he experiences only positive outward interactions with these family members. Although he did not link this discomfort with religious beliefs in this quote, he discussed this as a possible reason for his siblings’ discomfort later in his interview. This contradiction of assumed disapproval and outward support fall within the category of covert ambivalence.

Like several other respondents, Courtney, Janet’s partner of 15 years, described how Janet’s mother’s religious beliefs reflect her mother’s contradictory feelings and behavior: “They are strongly Catholic, so I think in some ways she would really like to like me even more, but the religion gets in the way a little bit.” Courtney suggested that although Janet’s mother appears supportive she is not wholly comfortable with their relationships because of her religious sentiments about homosexuality and her participation in the Catholic Church—a dynamic characterized as covert ambivalence. Edwin, partnered to Kevin for 13 years, also described how he perceives Kevin’s mother as ambivalent because of her religious beliefs:

His mother is very understanding. I feel very welcome when I go to her house. She’s got her beliefs. Her religious beliefs, they are funny. She will not allow us to sleep in the same room. She says that it is not because we’re gay. It’s because we’re not married. I find it already remarkable that she’s so accepting considering her religious upbringing. She’s very conservative. So it’s fine.

Edwin described his mother in-law as “understanding” but rejecting because of her religious beliefs. Respondents like Edwin recognize social-structural factors that reveal the contradictory presence of his mother-in-law’s love alongside her homophobic religious beliefs and affiliations—characteristics of covert ambivalence.

The “glass closet”

Fewer than one quarter of respondents in this study had not openly discussed their gay or lesbian identity or partnership with at least one family member. However, respondents suggested that these family members do in fact know about their long-term same-sex intimate tie. Visibility of a gay identity despite not being officially out is known as the glass closet—the open secret of one’s sexual identity (Sedgwick, 1990). The glass closet occurs when respondents describe the belief that family members hold strong negative feelings about respondents’ sexual identity, most often due to perceived homophobic and religious values, but also are outwardly accepting of the intimate tie.

Emilia, partnered to Diana for 10 years, described an event that made her recognize that she is accepted and supported by her partner’s family but that she is in the glass closet because there is a simultaneous rejection:

We never talked about it with her dad. But her dad knew. You could tell that he knew and he accepted us as a couple, until the first time he was in the emergency room. We were all standing in the emergency room, behind the curtains. And the doctor said, “Do you know who that is?” “Yeah, that is so and so.” He pointed to me. He said, “She is my outlaw.” He didn’t know quite how to phrase what I was to him, because he was saying, “That is my daughter, that is my son.” “That is my daughter-in-law.” “She is my outlaw.” So it was that famous sentence. And he accepted it in that very bizarre way.

Diana never discussed her and Emilia’s long-term relationship with her father because Diana believed that her father would have rejected her, yet at the same time she experienced her father as supportive across the life course. This “bizarre” acceptance is in line with the concept of covert ambivalence.

Edwin, introduced above, articulated how his gay identity was not discussed because he assumed his family’s reaction would be negative. Yet he characterized his family as generally outwardly positive:

I have never come up to them and said, “Dad and Mom, I’m gay.” No, I have never used the words. I don’t know if it is my parents’ mentality. They know my partner, and we go home every year. We stay in their master bedroom. I remember when they had their 50th anniversary; one of my sisters-in-law did say that the issue of how they were going to list our names in the program came up, because they had the four sons listed and the wives. But then when it came to me, it was like, how do we list Kevin’s name? They never discussed that with me. I don’t know if it is they feel like they’re invading my privacy or if they are uncomfortable about it. They did not put his name down. I don’t know if it’s one of those things that they prefer not to talk openly. It’s unspoken.

Edwin simultaneously said he felt he has supportive family ties due to past and present overt interactions but also that he experiences these family members as unsupportive due to his assumptions of their shame over his sexual identity and relationship. This contradiction is maintained over time and evidenced by the family’s inability to openly discuss Edwin’s gay identity. Similarly, Marcus discussed how he believes his family feels love for his partner of 20 years but are not ready to hear they are in a gay relationship:

They love Austen quite strongly. They are very supportive of him, as much as they can. In fact, all the family is. We just don’t talk about it specifically. My mother couldn’t use the word gay if she had to. It is fine. It is Marcus and Austen. It is Austen in family pictures.

Marcus believes that his parents are supportive as they “can be,” but his mother’s inability to outwardly discuss her son’s gay identity is an indicator of her underlying negative feelings. Thus, even though Austen is in family pictures the intimate nature of their relationship stays hidden. As this theme demonstrates, the glass closet provides a highlighted condition of sociological ambivalence, wherein the visible and supported gay or lesbian identity is obscured and ignored because of the assumption that such disclosure would prompt a negative reaction.

Collective Perceived Ambivalence Across the Family History

The above themes elucidate how respondents’ characterizations of their individual relationships fall within the bounds of ambivalence. The findings discussed below reveal a distinct type of ambivalence that is related not only to perceptions of individuals but instead to a characterization of the “family unit” that encompasses supportive and unsupportive individuals (i.e., collective ambivalence; Ward et al., 2008). Nearly 40% of respondents described their family as a relational group of individuals who create a simultaneously positive and negative family “unit.” This occurs in two primary ways. First, respondents described either their own or their partner’s family as an independent contradictory unit. Raymond illustrated this in his discussion of the family of origin of his partner of 14 years, Christopher:

His family is not so great. When he finally did [come out to his mom], she was accepting. His sister was also accepting. I think she probably cried or something. But her husband, his brother-in-law, total redneck. When he heard about Christopher, the first thing he said, “Well, they are never allowed to see our kids again.” And his sister was like, “You are crazy. They love Christopher.” Uncle Bubba. I just can’t even believe that guy is for real. What a jackass. But he has a half-brother also who is a really great guy and was totally cool with it. So his family has got kind of both sides.

Second, respondents conceptualized both their own family of origin and their partner’s family of origin as a family unit, comparing family relationships across this family unit as a whole—a form of collective ambivalence. Julie, partnered to Amanda for 13 years, said:

We are good with our family. I am very close to her mom. My grandparents, who are still alive, they know it, just nobody is allowed to talk about it. My uncles will sit around and they will say, [sneering] “So, Julie, have you done anything new with your lesbian activities?” And my grandmother—she is Pentecostal too—she will just start praying and covering her ears. “Lord Jesus, no. No.” She doesn’t want to hear about it. But otherwise, our family is pretty accepting. And the ones who aren’t, we just don’t have much to do with them.

Julie constructs “our family”—including both her and her partner’s family of origin—as complex and contradictory—a form of collective ambivalence. Aidan described this dynamic with his and his partner of 10 years, Max’s, contradictory families:

My two brothers were at the commitment ceremony as ushers. I am pretty much estranged from my oldest sister is a very, very born-again Christian. I’ve already gone to hell as far as she’s concerned. My little sister, that’s never been a good relationship. His parents really respect our relationship. I am the most favored daughter-in-law. But his brother’s a Methodist minister. And feels that our relationship is not right. It’s just the relationship is sinful. Max talks to him once a year on birthdays. Not a lot of interaction. So in a lot of things, you know, we have that on both sides of our relationship.

Max, Adian’s partner, similarly said:

Probably stronger on Aidan’s side with extended family. My parents and I are very close. I am not close with my brother. He had major problems coming to our covenant ceremony, and his two daughters did not come. He’s a Methodist minister, very conservative, and has problems with me being gay. And it’s been a few years. And we’ve had a lot of contact with two of the four [siblings]. The other two sisters also steer clear of us because of our sexuality. They think it’s sinful and wrong and don’t want their children exposed to us. So some of his family has close relations.

Max and Aidan, like others in this sample, both typify their family unit comparatively as experiencing a “mixed bag” wherein some family members were perceived as positive and others as negative in comparison to one another; this dynamic is characterized as collective ambivalence.

Discussion

The present study extends family research with one of the first in-depth analyses of ambivalence in gay and lesbian family-of-origin relationships (Lüscher & Pillemer, 1998). These findings move research beyond the view that gay and lesbian families are either supportive or rejecting (LaSala, 2002) by calling attention to how ambivalence is a central dimension of gay and lesbian family life (Connidis, 2012). Beyond revealing an underexamined dimension of gay and lesbian family relationships, this study makes three additional contributions to the field of family studies—and ambivalence theory in particular—which are described below.

First, as a departure from the original conceptualization of ambivalence, which emphasizes one’s own ambivalence toward others (Lüscher & Pillemer 1998), the present findings break new ground by demonstrating the ways in which individuals construct others as exhibiting contradictory behavior, feelings, and actions—what is conceptualized as perceived ambivalence. Symbolic interactionism stresses the primacy of interaction to the human experience, positioning individuals as self-reflexive beings whose self-concepts are forged in interaction with others (Blumer, 1969). This occurs through what Cooley (1902) called the “looking-glass self”—the process of imagining how others see us and internalizing perceived judgments into a self-understanding. Indeed, family scholars have long shown that individuals make appraisals of other family members’ thoughts, behavior, and feelings, wherein adults perceive levels of support, approval, love, strain, and rejection. In turn, these appraisals have consequences for self-concept and well-being (Fingerman et al., 2012; Kiecolt et al., 2011). Similarly, then, characterizing a family member as exhibiting contradictorily positive and negative behavior and/or feelings (i.e., perceived ambivalence) has clear relevance for individuals’ sense of self, the quality of family ties, and overall well-being. These findings suggest that a family member’s perceived ambivalence—regardless of that person’s actual feelings of ambivalence—may increase minority stress and psychological distress when ambivalence is understood as unfair treatment or ambiguous standing in the family relationship (Meyer, 2003).

The theorization of perceived ambivalence is a critical next step in fully articulating how ambivalence operates in family ties (Connidis, 2015). This study further extends our empirical knowledge by demonstrating precisely how individuals discursively characterize family members as exhibiting ambivalence, even as respondents did not use this terminology. First, respondents took stock of what they read as everyday overt contradictory (i.e., both positive and negative) interactions, providing analytical evidence for the presence of ambivalence. Second, respondents perceived overt positive interactions alongside perceptions of covert religious or homophobic beliefs as evidence of ambivalence. Notably, both covert and overt illustrations do not only exist in the present (Suitor et al., 2011) but rather accumulate in nonlinear and dynamic ways across a family history (Bengtson & Allen, 1993; Connidis, 2003; Elder, 1998; Finch, 2007; Kiecolt et al., 2011). These findings push forward the measurement of the broader ambivalence construct, demonstrating the need to capture not only overt affectual, interactional, expressive, and behavioral relationship components (Birditt et al., 2010; Fingerman & Hay, 2004; Gilligan et al., 2015) but also perceptions of others’ internal beliefs. Moreover, the clear commingling of the past and present in constructing family narratives moves the field beyond a static construct of present-day ambivalence toward a view of ambivalence as both cumulative and dynamic over time (Connidis, 2015). Perceived ambivalence as revealed via covert and overt evidence across family history opens up new possibilities for the articulation, operationalization, and effects of ambivalence over the life course.

Second, scholars have lamented that previous research privileges psychological ambivalence, virtually ignoring sociological ambivalence (Bulanda, 2011; Connidis, 2015; Peters et al., 2006; Pillemer et al., 2007). In speaking directly to this critique of the current literature, the present study takes an important step forward in providing new empirical evidence for the co-occurrence of psychological and sociological ambivalence. Previous quantitative research on sociological ambivalence shows that parents experience ambivalence due to children’s failure to meet structural norms (e.g., an unmarried child; Kiecolt et al., 2011; Peters et al., 2006; Pillemer et al., 2007) or when family members experience a disconnect between their own personal beliefs and a family member’s identity and behavior (Bulanda, 2011). In the case of gay and lesbian adults, perceived ambivalence occurs when the structural norms of homophobia, heterosexism, religious beliefs, and an absence of acceptance of federally legalized same-sex marriage intersect with norms of close family relationships (Bulanda, 2011; Pfeffer, 2012; Taylor, Kimport, Van Dyke, & Anderson, 2009). As a result, respondents perceived their family members as having contradictory emotions, beliefs, expectations, and actions—most notably the normative aspects of support and love coupled with rejection of a “spoiled” gay or lesbian identity (Meyer, 2003). Adult gays and lesbians recognize this irreconcilable sociological contradiction as manifesting in their family relationships, providing a concrete illustration of how structural forces are part and parcel of the psychological experiences of ambivalence.

Note that the glass closet, a term that Sedgwick (1990) used to refer to the “open secret” of a person’s sexual identity, provides one highlighted site of the intersection of psychological and sociological ambivalence. Individuals in the glass closet are not officially out but are simultaneously located where others can see their gay or lesbian identity via social markers of homosexuality—in the case of this study, a same-sex intimate tie. The glass closet highlights a central structural inconsistency that reveals ambivalence, wherein respondents see themselves both in and out of the closet in ways that bring them in and out of ambivalence across the family history. Thus, these findings reveal one condition in which ambivalence is likely to occur: when family members virtually ignore controversial aspects of an individual’s life (i.e., being gay or lesbian) coupled with the contradictory relative outness of that controversial dimension (i.e., a gay and lesbian long-term relationship). It may be that this is especially apparent in a cohort of older adults, and it was more common among individuals with older parents in this sample, and increasing acceptance for gay rights may in turn shift the social-structural contexts that engender this perceived ambivalence (Powell, Blozendahl, Geist, & Steelman, 2010). In addition, sociological ambivalence emerged in other ways in the data via the structural events of a commitment ceremony or a death in the family (Kiecolt et al., 2011; Schenk & Dykstra, 2012). It may be that other types of structural shifts—such as larger institutional shifts in same-sex marriage laws, nondiscrimination ordinances, or the birth of a child—alter relationships with family members in ways that reveal or dissipate ambivalence (Ocobock, 2013; Martin, Hutson, Kazyak, & Scherrer, 2010; Scherrer, 2014). Future research should address these points as potential sites of sociological ambivalence in the gay and lesbian context.

Third and finally, these findings confirm previous research that suggests ambivalence is most prevalent in parent–child/parent–child-in-law dyad relationships (Willson et al., 2003). Yet, as family systems theory suggests (Bowen, 1978; Heatherington & Lavner, 2008; Scherrer, 2014), individuals are embedded within a network of ties (Birditt et al., 2010; Pillemer et al., 2007; Ward, 2008; Ward et al., 2008). Examination of contradiction within only one dyad (i.e., parent–child) obscures the larger family landscape (Pillemer et al., 2007; Ward, 2008). This study calls for new attention to the ways in which an entire family unit is characterized with the ambivalence frame, wherein individuals are viewed in relation to one another and as a whole family via collective ambivalence. This finding builds on Ward’s (2008; Ward et al., 2008) collective ambivalence, wherein “multiple relations include the possibility that some are positive but other are negative” (Pillemer et al., 2007, p. 776). Whereas Ward’s collective ambivalence demonstrates how a mother holds positive feelings toward one child and a simultaneously negative feelings toward another child, this study extends this construct by introducing a more holistic account of perceived collective ambivalence. Here, individuals actively construct an ambivalent family unit with members in relation to one another—either their own family of origin or of both partners’ comparative family of origins—based on perceptions of all family members relationally. It may be that respondents in this study contrast supportive and rejecting members of their family as one strategy to cope with highly unsupportive family members, protecting both their own well-being and their intimate relationship (Oswald, 2002a). Viewing parents, in-laws, siblings and, to a lesser extent, extended kin as part of an ambivalent family unit highlights the relational nature of the ambivalence construct, wherein scholars must view multiple family ties together in order to comprehend the full tapestry of ambivalent family life (Pillemer & Suitor, 2008).

Limitations and Conclusion

The goals of this study was to understand the contradictory experiences of simultaneous support and strain in family relationships based on the accounts of gay and lesbian adults. The views of multiple family members may be necessary to fully reveal ambivalent relationship dynamics. It is of note that respondents described ambivalence most commonly in relationships with their parents and parents-in-law (Peters et al., 2006; Rappoport & Lowenstein, 2007), a finding consistent with previous research (Willson et al., 2003). Because in-laws are not as immediately emotionally close and do not constitute lifelong relationships, the structural context of in-law ambivalence may be more clearly and openly revealed as a highly ambivalent relationship. It is of note that few respondents mentioned ambivalence in relationships with extended kin. Extended kin may not be emotionally or proximally close enough to register levels of ambivalence (Fingerman et al., 2004), perhaps especially for gays and lesbians, who are more likely to live farther away from extended kin (Rosenfeld, 2007). It is also possible that when adults experience ambivalence with siblings or extended kin they terminate these relationships because they cause them discomfort. In contrast, respondents may be more reluctant to disown parents for the same types of ambivalent relationships.

Other limitations include an inability to speak to income and race differences with the current data, given that this sample was primarily White and Hispanic, with relatively high socioeconomic indicators. This shortcoming is a function of sampling design in a moderately wealthy town in the Southwest, and thus these findings may underestimate the perception of ambivalence in family relationships. In particular, research shows that Black and White gay and lesbian adults and adolescents experience very different family contexts and consequences (Moore & Stambolis-Ruhstorfer, 2013). Future work should address this issue. Moreover, the study sample included only individuals who self-identified as gay or lesbian in committed relationships; future research should include research on bisexual, queer, and transgender individuals as well as individuals who are not in committed partnerships, in order to capture a wider range of ambivalent experiences. Approximately 15% of respondents did not describe what I characterize as ambivalence in any family-of-origin relationship or in the entire family-of-origin unit. These respondents either described entirely positive or entirely negative family relationships or did not describe relationships in enough detail to characterize them with analytical integrity. These respondents did not differ substantially from other respondents on dimensions of gender, age, race, level of outness, or income; however, future research should address differences across individuals who do and do not perceive ambivalent relationships. Note that same-sex relationships have become increasingly accepted since the data were collected in 2007. Thus, this mid-life sample is of a particular cohort that came of age in the late 20th century—a time when individuals and couples were beginning to come out to family and friends yet low levels of societal acceptance remained, as evidenced by lack of access to all civil rights (e.g., military service, marriage, workplace discrimination). Attitudes toward gay marriage and gay relationships have changed dramatically in the United States since the data for this study were collected, and these social and familial attitudes will likely to continue to shift moving forward. Thus, the current cohort of young adults who are gay and lesbian may experience very different family-of-origin relationships, altering the presence and negotiation of ambivalence. Future work should address this possibility.

Overall, the present findings establish the centrality of perceived ambivalence characterizations of gay and lesbian family relationships. This study demonstrates the specific contexts in which gay and lesbian adults conceptualize independent family relationships and interdependent family units within the analytical category of ambivalence. These findings give new insight into how gay and lesbian adults negotiate their family relationships and further present a new line of inquiry in family research regarding the exploration, measurement, and theorizing of ambivalence within and beyond the intergenerational tie. The consequences for perceived ambivalence beyond those demonstrated in the present study, such as those for general well-being, psychological health (Pillemer & Lüscher, 2004), or intimate partner relationship quality, is a critical area for future inquiry.