Gender diversity, gender liminality in French Polynesia

Abstract

Background: The phenomenon of the māhū and raerae of French Polynesia represents a long-standing historical recognition of a third gender status that makes space for a level of respect and integration within a broader society. Raerae is a more recent type of māhū who more overtly express themselves as women in society. Aim: This article describes the gender identity, gender expression, sexual behavior, and sexual orientation of gender diverse individuals in French Polynesia, including the Society, Tuamotus, and the Marquesas Islands. By studying cultures with protective factors, this research seeks to contribute to the development of effective approaches to sexual and gender health promotion, including HIV/STI prevention. Method: Our analysis is based on ethnographic work, field observation, in depth interviews, and survey data. The focus of this study was specifically on the māhū and raerae (gender diverse individuals). Results: We collected data from ten islands in French Polynesia and included 47 participants. The results include demographics, information about self-identification, gender identity development and family and social acceptance, current gender identity, expression, early and current sexual experiences, relationships, involvement in sex work, and health status, including HIV risk and status. Conclusion: While not a utopia for gender variant individuals living in French Polynesia, this cultural recognition serves as a protective factor as compared to other cultures that greatly stigmatize transgender and gender diverse individuals or those who transgress socially defined roles for men.

Introduction

There is a growing field of literature on gender diversity among Pacific Islanders that notes the historical presence of individuals who do not identify by western gender and sexual categories (Alexeyeff & Besnier, 2014; Besnier, 1994; Besnier & Alexeyeff, 2014; Dolgoy, 2014; Dvorak, 2014; Farran, 2014; Ikeda, 2014; Kuwahara, 2014; Mageo, 1998; Nanda, 2014; Presterudstuen, 2014; Schmidt, 2016; Stip, 2016; Tcherkézoff, 2014; Vasey & VanderLaan, 2014). This literature has sparked interest, as well as reflection, on the ways that Pacific Island cultures have formulated different gender identities and sustained these divergent communities based upon a long cultural history of social tolerance in the midst of colonial influence. For example, Samoa also has a culture which has long recognized a third gender variant role for males known as fa’afafine (meaning “in the manner of a woman” in Samoan) (Dolgoy, 2014; Petterson et al., 2018, 2020; Petterson & Vasey, 2022; Schoeffel, 2014; Tcherkézoff, 2014; VanderLaan et al., 2015; Vasey & Bartlett, 2007). Tonga also has these categories, fakaleitī and fakatangata, for gender non-conforming and diverse males and females, respectively (Besnier, 1994, 2004; Besnier & Alexeyeff, 2014). In the Cook Islands, there are gender variant males that are referred to as laelae or ‘akava’ine (Alexeyeff, 2007). In French Polynesia, a third gender/nonbinary role of māhū or raerae has long been recognized within society (Bauer, 2002, Elliston, 2014; Kuwahara, 2014; Matzner, 2001). Across these islands, these individuals are referred to by anthropologists as being “gender liminal” (an anthropological term meaning on the border or outside of social restrictions) (Besnier, 1994; Levy, 1971, 1973).

The study of cultures with natural protective factors may be worthwhile in terms of expanding our understanding of the social roots of thriving, including less stigma around gender expression, less non-affirmation, more family support and cultural connectedness. This might lead to developing effective approaches for the promotion of sexual and gender health, including HIV and STI prevention strategies in populations who are marginalized based on gender and sexuality in different cultural contexts (Anderst et al., 2022; Coleman et al., 2018).

In terms of sexual and gender health promotion and STI and HIV prevention, we are keenly aware that minority stress (Meyer, 1995, 2003, 2015) and gender minority stress (Testa et al., 2015) contribute to poor health outcomes. While minority stress has been seen as a major contributor to health disparities including proximal and distal factors, Testa and colleagues have developed a gender minority stress framework which considers specific minority stressors that are unique to transgender and gender diverse individuals. It has been hypothesized that on top of minority stress factors, gender minority stress adds to minority stressors to explain the health disparities compared to cigender people but also to lesbian, bisexual and gay individuals. gender minority stress emanates from distal stressors that are comprised of gender-based victimization, gender-based rejection, and gender-based discrimination, and non-affirmation (Testa et al., 2015). These distal stressors are conceptualized as independent of personal identification and associated with a minority position in society (Meyer, 2003). Tan et al. (2020) have expanded on the gender minority stress framework and suggest that family support and cultural connectedness are also relevant when considering the various proximal and distal factors, including those factors that are protective.

French Polynesia offers an interesting example of a culture that has been described in ways which would presume that transgender and gender diverse individuals would have protective factors which might mitigate other negative minority stress or gender minority stress factors. French Polynesia has a long tradition of a third gender status that has been institutionalized and has enjoyed a relatively non-stigmatized status within their society (Besnier, 1994; Kuwahara, 2014; Matzner, 2001; Stip, 2016). Early travelers to Tahiti described māhū who dressed, performed the work of, and lived as women (Cook, 1821; Oliver, 1974, pp. 369–374). In French Polynesia, as well as in Hawaii, male-bodied māhū were associated with government advisors, cultural affairs, spiritual functions, and the hula dance (Matzner, 2001). By nature of its sheer geographic size, remoteness, and lack of valuable natural resources, the long-held traditions have been somewhat protected from colonization and imposition of religious beliefs. Thus, French Polynesia presents an opportunity to study a culture where a third gender status has had historical protection and where the phenomenon and traditions are still intact. Such examination sheds light on how certain societies with sanctioned and institutionalized roles for gender diverse people can create a protective factor leading to emotional and physical well-being.

Specifically, this study employs ethnographic work, field observation, in depth interviews, and survey data in an effort to examine the risk and protective factors, including stigma experiences, resilience, family support and health among the māhū and raerae of French Polynesia.

Historical background

French Polynesia had its first record of colonial contact from England and France in the later eighteenth century (e.g., Tahiti in 1767; Society Islands in 1769; Tubuai in 1777) (Besnier & Alexeyeff, 2014; Stip, 2016). French and British Christian missions were established in the late 1700s. Tahiti became a French colony in 1880. The arrival of Christian missionaries signaled a change in the “moral and sexual order” (Stip, 2016). As in much of the world, the missionaries were concerned with the sexual mores of indigenous Polynesians and promoted sexual mores based upon Christian beliefs, which stigmatized public nudity and cross-dressing behavior (Besnier & Alexeyeff, 2014; Stip, 2016). In Hawaii, māhū were also stigmatized and policed through sodomy laws (Zanghellini, 2013). In 1946, the archipelago of the Society Islands (including Tahiti), the Tuamotus, and the Marquesas Islands were reconstituted into a French Overseas Territory and given the official name French Polynesia. French Polynesia gained Administrative Autonomy Status in 2004.

It is significant that before European contact in the eighteenth century, transgender and non-binary roles were socially accepted. Māhū served as advisors to royalty throughout Polynesia and were respected teachers and spiritual healers in the community (Matzner, 2001). Christian missionaries attempted to stamp out the influence of māhū in Polynesian culture, categorizing their existence as morally and sexually deviant (Matzner, 2001). However, these efforts to impose rather strict mores of males having sex with women in committed relationships did not fully succeed. The māhū of French Polynesia, in particular, continued to survive. The blend of Christian and traditional beliefs in contemporary Polynesia society today reflects mixed attitudes toward gender diversity ranging from acceptance to condemnation which varies by context (Besnier & Alexeyeff, 2014). In many cases, gender diverse males are accepted within the culture, integrated into families, and even held in high regard. In other instances, they are stigmatized as sex workers or seen as a source of shame (Bauer, 2002; Matzner, 2001).

Levy (1971, 1973) has provided some of the earliest research on gender liminality and diversity in French Polynesia. He described the māhū as a role for males who dressed and took on the social and occupational roles of Tahitian women. According to Levy, the māhū tradition continued from precolonial times. Levy considered the māhū a role-variant for men, and effeminacy was associated with māhū. Thus, the Polynesian gender paradigm has been one in which sex and gender are separate and distinct categories, with the māhū identity describing a male with a third gender identity. Māhū were regarded as being “natural” (Elliston, 2014). In French Polynesia prior to pre-European contact, the māhū were not believed to practice sodomy, but were fellators of other men. While the māhū engaged in fellatio with cis-gendered male partners, these partners were not considered māhū or androphilic. Māhū were also reported to have been sex partners of male chiefs. According to Levy (1973), māhū were considered as a substitute female.

Because of an emphasis on strengthening the family line through reproduction, exclusive same-sex relations were discouraged in precolonial times (Archer, 2018), although some same-sex relations appear to have been documented among some Polynesian cultures. For example, Europeans writing in the eighteenth century recount that Hawaii noblemen and women often took same-sex lovers (called aikane). These relationships were generally between unequal partners in terms of age and status, although more egalitarian couplings were also possible. Unlike māhū, the aikane role was not associated with gender variance. Aikane fulfilled the gender roles associated with their biological sex and often had spouses and children (Archer, 2018).

While some degree of social acceptance of gender variance predates European influence, over the course of recent centuries, western influences have affected local perceptions and also introduced new identities and behaviors into this region (Bauer, 2002; Besnier & Alexeyeff, 2014). Today, for instance, within the context of French Polynesia, the māhū are generally recognized as a third gender status (Bauer, 2002; Ferdon, 1981; Matzner, 2001; Robertson, 1989; Stip, 2016). Māhū translated from Tahitian (the Polynesian language spoken in French Polynesia) means “in the middle.” Māhū are birth-assigned males with non-binary identity and feminine and gender liminal traits, while raerae are a more recent classification of birth-assigned males whose gender expression is more overtly as a woman (Bauer, 2002; Stip, 2016). Like māhū, they are almost exclusively interested in sexual relationships with cis-gendered men. The term raerae emerged in Tahiti in the 1960s to describe a type of māhū, many of whom began expressing their gender more overtly as women in society (Bauer, 2002; Elliston, 2014; Kuwahara, 2014). Raerae were more likely to transform their bodies with feminizing hormones and gender affirming surgery (Kuwahara, 2014). Originally, the word raerae had a negative connotation such as “whore,” “to suck,” or “cock sucker.” Many became sex workers, although increasingly raerae have a transgender role in society with growing acceptance and greater economic opportunity other than sex work (Bauer, 2002; Kuwahara, 2014).

What is striking to western foreigners in French Polynesia is a nonchalant attitude toward māhū, who are integrated into family structures, social, and cultural life. In contrast, the raerae, while enjoying more recognition as a type of māhū, still stretch the limit of an accepted third gender traditional role (Bauer, 2002; Kuwahara, 2014; Stip, 2016).

Aim

This study aims to contribute to the discourse on gender diverse individuals in the context of French Polynesia and to examine the role of stigma, family and social acceptance, cultural recognition, and risk and protective factors that may impact health outcomes. Overall, these findings have implications for sexual and gender health, including HIV/STI prevention (Anderst et al., 2022). Research has shown that homophobia, transphobia, and stigma negatively impact sexual health and overall health (Hatzenbuehler et al., 2017; Hatzenbuehler & Pachankis, 2016; Meyer, 2003), while acceptance, social support, belonging, and self-esteem protect sexual health (Tan et al., 2020; Testa et al., 2015). French Polynesia provides an important context to study possible factors that combat stigma and give space for gender diverse behavior (Johns et al., 2018).

Method

French Polynesia encompasses a vast area (approximately 1,200 miles) in the South Pacific, with over 100 islands and approximately 275,000 people. We could not sample the more than 100 islands but we sampled some of the major islands in the archipelago. As part of recruitment, we visited various islands throughout the French Polynesia Archipelago, including the Society Islands, the Tuamotus, and the Marquesas Islands.

We identified participants through word of mouth and snowball sampling. Except for the more populated islands, participants were easily identified and accessed, as the number of māhū/raerae were well known to everyone. Potential participants were briefed regarding the nature of the study, the benefits and risks of participation, and their ability to discontinue involvement in the study at any time. They were informed that no incentive would be provided. Participants were given a consent form to read and sign and then proceeded to fill out the survey in French. Most everyone who was contacted agreed to participate. The survey covered demographic information, gender identity development, gender affirmation treatment, family and societal reactions to their gender identity and gender expression, sexual orientation, sexual behavior, general health, and HIV risk and status. All survey data were collected by the last author with assistance and translation from guides. The survey took between 30 and 60 min to complete. A number of participants skipped questions due to time or not wanting to answer some questions. The data were collected in 2004 and then again in 2014. These two time periods of data gathering were governed by opportunities afforded by the last author to travel to French Polynesia during sabbatical leaves. The second visit provided an opportunity to gather more data and spend more time with some of the key informants (more information regarding key informants is provided below). During the second visit, the last author visited all the aforementioned islands except for Raiatea in the Society Islands and none of the Tuamotus, based on limitation of time. Any individual who had completed the survey previously was excluded from repeating the survey. As there was no incentive for completing the survey, this did not cause any consternation among potential participants. In addition to the survey instrument, this study used ethnographic methods, field observation, and extended interviews by the last author that were conducted with key informants who spoke English. Some of the individuals who took the survey and were willing to spend more time with the last author became key informants that provided greater understanding of the situation and context for survey responses. The conversations were not recorded or written in any verbatim manner. For some informants, the interviews took hours of conversation on a variety of occasions. The last author wrote and kept summary notes on these conversations. The ethnographic work included visiting informants in the community and sometimes in their homes, which involved engaging in extended conversations with their families and extended families (DeWalt & DeWalt, 2011). There were plans for a subsequent follow up visit, however, COVID precluded this possibility. We decided to summarize our data to date, which are presented here.

All research was approved by the University of Minnesota Institutional Review Board (IRB – 0409S63829).

Results

Demographics

We collected data from ten islands in French Polynesia. These islands included the following Marquesan Islands (Nuka Hiva, Hiva Oa, Fatu Hiva, Tahuata), Tuamotus (Fakarava, Rangiroa), and the Society Islands (Tahiti, Bora Bora, Raiatea, Huahine). The demographics of the participants can be viewed in Table 1. Our study included 47 participants who ranged in age from 18 to 48 (average age 27.8). In terms of race/ethnicity, 34% identified as French Polynesian and 44% as Marquesan. The remainder identified as a mixed race identity of French Polynesia/European, French Polynesian/Chinese, or French Polynesian/South Pacific. Our sample was over 90% Christian (Catholic, Protestant, or other Christian). Thirty-eight percent had at least a high school education; 24% had a bachelors degree. Eighty-one percent were single but sexually active, 11% were cohabitating with a man, and the remaining participants were celibate or had missing data. None had been married, although two reported having had children.

Table 1.

Demographics.

	Average/N	Range/Percent
Age	27.8	(18–48)
Ethnicity
French Polynesia	16	34%
Marquesan	21	44%
Mixed French Polynesia/European	4	9%
Mixed French Polynesia/Chinese	2	4%
Mixed French Polynesia/South Pacific	4	9%
Religion
Catholic	26	55%
Protestant	14	30%
Other Christian	4	9%
None	2	4%
Missing	1	2%
	47
School
Primary school	9	19%
Some secondary	6	13%
High school/Some College	18	38%
BA+	11	24%
Missing	3	6%
Relationship status
Single, sexually active	38	81%
Cohabitating with a man	5	11%
Celibate	2	4%
Missing	2	4%
Ever married
yes	0	0%
no	47	100%
Has children
Yes	2	4%
No	40	85%
Missing	5	11%
Caretaking
Takes care of non-biological children	18	38%
Yes	25	53%
No	4	9%
Missing
Takes care of older parents/family members
Yes	6	13%
No	30	64%
Missing	11	23%
TOTAL N	47

Terminology and gender identities and self-acceptance

Early in our ethnographic work, it became clear that there was not a universal acceptance of terminology for the māhū throughout the Marquesas, Tuamotu, and Society Islands. As mentioned previously, the two most popular terms were māhū and raerae. Based upon information from key informants, for some islands, one term was preferred over the other—and for others there were individual differences in taste. Frequently, the terms māhū and raerae were used interchangeably or became more or less popular depending on the local context. For example, while raerae is a more preferred term in Nuku Hiva in the Marquesas, māhū is preferred in Hiva Oa—another nearby island in the Marquesas. When asking some of the informants in Nuku Hiva why they preferred the term raerae, they explained that in Marquesan language, words take on different meanings by tonal sounds. The meaning is also determined by who is using the word. It seems the analogy could be with the terms “faggot” or “queer” where their connotation may vary substantially based on intent, as well as the individual using the terms. There were some informants in Nuku Hiva who simply preferred the term “fille” or “fifi,” using the French terms to refer to “lady.” This seemed like a creative attempt to define themselves as women in society with a positive connotation. However, what was apparent is that there was no term on which all could agree that had a contemporary and clearly positive connotation. It is worth noting here that degrees of tolerance also differ based on geographical location, in that acceptance is greater in rural, less “developed” locations, while more discrimination and the potential for trans-negative violence is greater in urbanized areas.

Through our fieldwork, we also learned that the notion of sexual or gender identity was difficult to understand for many of the māhū/raerae in French Polynesia. You could not ask them if they knew what their “biological” or “physical” sex was versus their “gender identity” or “sexual orientation.” These are western constructs that had little or no meaning to most of the participants. One would have to ask whether they were born with male or female genitals, what was the sex of the partner, etc.

While participants were aware of western terminology, such as gay, transgender, trans, transsexual, transvestite, etc., gender identity was best explored if you asked if they felt like a boy, man, woman, or both. Most of our participants believed that they would grow up to be a woman but less than half felt positive (38%) about their gender diverse behavior when they were growing up (Table 2). This improved substantially over time, with 64% feeling positively toward their gender identity now. In order to avoid problems with western terminology, we asked very descriptive questions about gender roles such as asking about toys and tasks that were typical for men or women. In the case of sexual orientation, we explored by asking specific questions about whether they had sex with “men,” “women,” or “both," or who they found themselves attracted to or wanted to be sexual with in the future (see below for more detail). They might describe their sexual behavior with other males simply as having sex with other males. They understood their sex as male but identified their gender as māhū or raerae. They were almost exclusively attracted to males. As they viewed themselves as women who had sex with men, no one identified themselves as “gay.”

Table 2.

Gender identity development.

	Average/N	Range/Percent
Age first felt different	7.97	3–16)
Realized was māhū/raerae	9.84	(4–17)
Believed would grow up to be a woman
Yes	17	36%
No	13	28%
Sometimes	1	2%
Missing	16	34%
How felt about feminine behavior
Positive	18	38%
Neutral	10	21%
Negative	9	19%
Missing	10	21%
How feel about feminine behavior now
Positive	30	64%
Neutral	4	9%
Negative	2	4%
Missing	11	23%
Father felt about feminine behavior
Positive	17	36%
Neutral	6	13%
Negative	18	38%
Missing	6	13%
Father feels now about feminine behavior
Positive	23	49%
Neutral	3	6%
Negative	3	6%
Missing	18	38%
Mother felt about feminine behavior
Positive	24	51%
Neutral	9	19%
Negative	8	17%
Missing	6	13%
Mother feels now about feminine behavior
Positive	29	62%
Neutral	1	2%
Negative	0	0%
Missing	17	36%
How would you describe relationship to parents growing up
Hostile	10	21%
Neutral	12	26%
Loving	12	26%
Missing	13	28%
Growing up, closer to
Mother	22	47%
Father	0	0%
Both	9	19%
Neither	2	4%
Missing	14	30%
Sibling(s) felt about feminine behavior
Positive	25	53%
Neutral	8	17%
Negative	10	21%
Missing	4	9%
Sibling(s) feels now about feminine behavior
Positive	27	57%
Neutral	2	4%
Negative	3	6%
Missing	15	32%
Teachers felt about feminine behavior
Positive	22	47%
Neutral	16	34%
Negative	3	6%
Missing	6	13%
Friends felt about feminine behavior
Positive	31	66%
Neutral	2	4%
Negative	1	2%
Missing	13	28%
Friends feel now about feminine behavior
Positive	30	64%
Neutral	0	0%
Negative	2	4%
Missing	15	32%
Church felt about feminine behavior
Positive	12	26%
Neutral	16	34%
Negative	9	19%
Missing	10	21%
Church feels now about feminine behavior
Positive	15	32%
Neutral	10	21%
Negative	6	13%
Missing	16	34%
Comfortable being māhū/raerae
Comfortable	34	72%
Neutral	1	2%
Uncomfortable	4	9%
Missing	8	17%
Feel like a woman
Yes	35	74%
No	3	6%
Sometimes	9	19%
Missing	0	0%
Do you wear female clothes
yes	35	74%
no	11	23%
Missing	1	2%
How often female clothing
All	20	43%
Most	4	9%
Some	11	23%
Never	12	26%
Women’s undergarments
Yes	22	47%
No	12	26%
Missing	13	28%
Comfort with genitals
Comfortable	13	28%
Neutral	4	9%
Uncomfortable	13	28%
Missing	17	35%
Taken female hormones
Yes	23	49%
No	19	40%
Missing	5	11%
Taking hormones now
Yes	16	34%
No	25	53%
Missing	6	13%
Breast augmentation
Yes	2	4%
No	41	87%
Missing	4	9%
Genital operation
Yes	2	4%
No	40	85%
Missing	5	11%
Ever thought of having gender affirmation surgery?
Yes	23	49%
No	19	40%
Missing	5	11%
TOTAL	47	100%

Gender identity development and social acceptance

At an early age, participants reported that they felt themselves to be different from other boys. Because the third gender status is well known and talked about, many identified as māhū or raerae from an early age. From the survey results (Table 2), the vast majority of participants perceived themselves as different in terms of gender identity at an early age (average age 7.97). They identified as māhū/raerae a few years later (average age 9.84).

Through interviews and field observation, we learned that children were often identified as māhū by adults through clothing preferences or interest in work associated with the women. This is not always greeted by parents as a wanted role for the child but over time the child’s persistent gender non-conformity was more accepted. Given that the child took on household duties, including caring for younger children, the advantages start to outweigh the disadvantages or disappointment that a male role will not be adopted and grandchildren would be unlikely.

What emerged from the data was that there was a certain social advantage to having a māhū in one’s family which was not always appreciated but there was a growing appreciation of this. By not getting married and establishing a new family, māhū are freer to remain devoted to the maintenance of their childhood family. In some instances, the māhū are available to help with the family chores and raising younger children, and then moving into taking care of elderly relatives, including one’s parents. Among participants in our study, 38% told us they currently took care of children that were not their biological children and 13% took care of older family members/parents. Given that our sample was relatively young (average age 27.8), it is likely that elderly caretaking may increase over time as parents age.

Moreover, because of the gender liminality of the māhū who can work in both men’s and women’s activities, such as making tapa (women), jewelry (women), wood and bone carving (primarily men), tattooing (men), housework (women), cooking and cleaning in guest houses, hospitality, government administration (men and women), health care (men and women), fishing (men), and hunting (men)—māhū can make reasonably good money and can help provide for the family through financial contribution. Thus, in some instances, the potential loss of a māhū family member to a new family structure (i.e., marriage) would be perceived as a worse outcome than the fact that a child is transgressing gender norms and/or having same-sex sex. Thus, we speculate that the advantages of gender liminality and family connection might overcome any concerns about transgressing gender norms or non-procreative sex with other males.

Growing up, participants perceived their relationship with their parents as more a mixture of hostile, neutral, or loving (21%; 26%; 26% respectively) (Table 2). The majority reported being closer with their mothers growing up (47%) or closer to both parents (19%). No participants reported being closer to their father. In childhood, 51% reported that their mothers felt positively toward their gender diversity compared to 36% of fathers. As they grew up, perceptions of support from one’s parents improved. Almost half (49%) currently felt that their father felt positively about their behavior and 62% that their mothers felt positively. Participants also reported that their siblings were more supportive of their gender identity/expression (53% felt positively growing up) than their fathers and about as supportive as mothers. This positive perception from siblings slightly increased over time (58% of siblings felt positively now).

Beyond family members, participants reported that childhood friends were more supportive of their gender diversity than family members (66%) and that did not change significantly over time (64%) (Table 2). Teachers were also perceived as supportive overall (47% positive; 34% neutral).

The majority of the participants identified and were raised in Christian households (as noted above in Table 1)—but not all practiced their religion in formal ways. Compared to family members and childhood friends, church officials were perceived to be less supportive of gender diversity while growing up (26% felt positively) and currently (32% felt positively). But, during interviews, participants also told us that, at least for some, going to church, helping out with Sunday school, and organizing activities gave them a legitimate place within the church despite historical condemnations by missionaries in relation to gender and sexual diversity.

Current gender identity/expression

While all participants identified as māhū/raerae, excluding the missing data, 83% felt comfortable with their gender identity currently (6% uncomfortable; 11% neutral) (Table 2). As mentioned previously, most participants felt more like a woman than a man (74%). The majority wore female clothing at times, with 43% wearing female clothing all of the time and 23% wearing it sometimes. Forty-seven percent reported wearing female undergarments. Participants were mixed in their feelings about their genitals, with 28% feeling comfortable, 9% neutral, and 28% uncomfortable (35% missing data).

From our interview data, gender expression, including use of names and pronouns, was flexible given circumstance, time, and space, although differences were found among those who identified as māhū or raerae. The māhū were more likely to dress as men and use their given names even though most had adopted female names, while occasionally dressing as women. Raerae were more likely to dress as women most of the time and use their adopted female names.

In our interview and field observation, we learned that the opportunity for māhū/raerae to dress up in elegant, trendy, and sometimes extravagant ways brings them attention. Our findings align with historical accounts, where this talent of understanding fashion and beauty is seen as an innate trait of māhū. It brings them a sense of pride. They often advise other women on fashion and cosmetics.

Notably, several of the māhū that we encountered were teachers of the Marquesan and Polynesian dances. The classes included mostly other women and other māhū. While the women and māhū practiced their dance, the men and boys were playing volleyball and soccer. We would note in our observations that most children learn the traditional dances today. Boys learn the male role and girls learn the female role in the dance. However, the māhū are more likely to learn both. During interviews, participants explained that the dance classes provided restoration and preservation of culture and affirmed their role in society.

Early sexual experiences

Most of our participants had sexual experiences early in life, often with older male non-māhū partners. A substantial number did not view these as necessarily abusive experiences but in many cases described these experiences as wanted. The average age of sexual debut was 11.8 (Table 3). Participants’ first sexual partner had an average age of 17.2. Many participants reported a first sexual experience with a family member or relative (34% at sexual debut; 45% lifetime). The majority reported their sexual debut as wanted (55% wanted; 15% unwanted; 30% missing data). The first sexual experience involved having been anally penetrated for a portion of participants (28%).

Table 3.

Sexual experiences, relationships and health.

	Average/N	Range/Percent
Age of 1st sexual experience	11.8	(5–20)
Age of 1st sexual partner	17.23	(6–45)
1ST sexual partner gender
Male	32	68%
Female	0	0%
Missing	15	32%
Who was person?
Family/Relative	16	34%
Neighbor	4	9%
Friend/Schoolmate	6	13%
Guest/tourist	3	6%
Missing	18	38%
Involved anal penetration
Yes	13	28%
No	18	38%
Missing	16	34%
Wanted
Yes	26	55%
No	7	0%
Missing	14	1%
Sex with adult relative ever while growing up
Yes	21	45%
No	19	40%
Missing	7	15%
Sex abuse
Yes	15	32%
No	27	57%
Missing	5	11%
Physical abuse
Yes	9	19%
No	21	45%
Missing	17	36%
Lifetime # of male sexual partners	6	(2–8000)*
Ever had sex with women
Yes	6	13%
No	36	77%
Missing	5	10%
Ever had sex for money
Yes	17	36%
No	17	36%
Missing	13	28%
Had sex for money most of the time
Yes	4	9%
No	31	66%
Missing	12	25%
In the future, have sex with:
Men	40	85%
Women	0	0%
Both	5	11%
Missing	2	4%
Ever been in a long-term relationship?
Yes	28	60%
No	15	32%
Missing	4	9%
Average Relationship length	39 months	(<1m–22 yrs)
Desire for long-term sexual relationship in future
Yes	28	60%
No	4	9%
Not sure	11	23%
Missing	4	9%
Condom use
Never	2	4%
Sometimes	7	15%
Most of time	5	11%
Always	14	30%
Missing	19	40%
HIV testing
Yes	31	66%
No	9	19%
	7	15%
HIV Status
Positive	2	4%
Negative	29	62%
Missing	16	34%
Overall health
Excellent/good	40	85%
fair	3	6%
Poor	0	0%
Missing	4	9%
Ever seen therapist
Yes	12	26%
No	34	72%
Missing	1	2%
Treated well by doctors
Yes	38	81%
No	5	11%
Missing	4	9%
TOTAL	47	100%

^*8000 figure is probably an exaggeration.

While 32% of participants reported having been sexually abused at some point in their childhood, many reported in interview data that their early sexual experiences with older cisgender males were wanted and pleasurable.

Current sexual experiences

In interviews, participants told us that premarital sex and infidelity are not exactly condoned but they are widely practiced, and for many, there is no shame in this. Males are not usually assumed to be faithful in French Polynesian culture. The sexuality of French Polynesian women, in contrast, is more regulated and constrained. In fact, our informants suggested that māhū/raerae who were less constrained by family and social structures were “freer” to go outside the home and serve as sex workers, both presently and historically (including for the French military). So, if a man were looking for sex, he might have easier access to sex with a māhū than a cisgendered female. Overall, the māhū with whom we spoke expressed a keen interest in engaging in sexual activity. Many participants also had no problem finding sexual partners among local cisgendered males and young boys who seek out sex with the “uncomplication” of a one-night stand.

In interviews, participants described how females were viewed as seen as “property” of men, as their female partners were highly controlled by their cisgendered male partner while māhū and raerae were considered “community property” and available to all cisgendered men.

Interestingly, many participants described a hesitancy to use hormones because of the known sexual side effects of reduced libido and/or erectile functioning, or undergo operations such as a penectomy or vaginoplasty that might interfere with their sexual pleasure or that of their male cisgendered partners. A number of our informants described how a number of their sexual partners enjoyed receiving anal penetration and performing oral sex that was not possible with cisgendered females given societal mores and body parts. While around half of participants had tried hormones (49%), hormones were not defining for identity. Only 34% of participants were currently taking them. With modern medical advances, nearly half (49%) had thought about having gender affirmation surgery. However, only a few had breast (4%) and genital surgery (4%) to present themselves as cisgendered women.

We would note that this general lack of stigma around sex with a māhū or raerae did vary across age and island context. For participants in Nuku Hiva, for example, cisgendered married males might avoid the māhū during the day but might have sex with them at night (exemplifying a degree of stigmatization). In contrast, for participants from Hiva Oa, cisgendered men and boys who had sex with māhū seemed to encounter little stigmatization. It was reported in fact, at soccer matches, the players would often positively comment and boast without shame of their sexual adventures with various māhū. Their sexual activity with māhū did not seem to be stigmatized.

Our findings revealed that Polynesian sexuality is nevertheless based on a heterosexual model and hence is gendered. For example, sexual relations between a māhū and a cisgendered male is considered normative precisely because māhū are not considered men. Thus, this is not considered sex with another man. This is why the partners of māhū are not necessarily stigmatized; because they are following a sexual script that mimics normative heterosexuality. Ironically, or somewhat defying the normative heterosexual paradigm, key informants reported that there were a number of cisgendered androphilic males who would sometimes take traditionally “passive roles” such as giving oral sex or engaging in receptive anal intercourse.

Relationships

Through our survey and ethnographic work, we learned that long-term sexual relationships are hard to come by for māhū/raerae—even though many participants expressed a desire to have this type of relationship. A māhū and a cisgendered male living together long-term seemed to be rare. However, we encountered a few māhū/raerae who lived together with their cisgendered male partners, and a few of our informants had long-term meaningful and loving relationships that were not clandestine but well known to family and the community.

While some reported having had relationships, they were usually rather brief (mostly a few months but there were some exceptions (the longest relationship lasted 22 years), which skewed the average of 39 months). Most relationships seemed to be ones of convenience—romance with a foreign or a military cisgendered male. It was hard to sustain the relationship if the foreigner left.

Most participants (60%) expressed a desire for a long-term sexual relationship with a cisgendered male in the future. While a few participants had reported sex with women in their lifetime (13%), none had the desire to be sexual with cisgendered females in the future (0%). A desire of many māhū/raerae is not only to live as husband and wife but also to have children through adoption, which is not unrealistic given that adoption is common for the general population in French Polynesia. Eighteen (38%) of the participants reported having had the responsibility for caring for children for their immediate family or relatives. It was unclear from the survey how many of these situations involved doing this on their own, or involved formal adoption. Some informants who were living on their own described taking care of children on their own and referred to them as their adopted children. In French Polynesia, adoption is often an informal process between relatives.

Importantly, sex between two people of the same sex (for example, two females who both identify and present as women) is viewed as being outside of the normative model of sexuality, and therefore generally viewed negatively. For the māhū with whom we spoke, the concept of a māhū/raerae being in a relationship with another māhū/raerae is considered strange and could even evoke disgust. “That would be like being lesbian. I am not a lesbian!” Participants were aware of lesbian relationships among cisgendered females—and in fact, these relationships seem to be better tolerated in society than the notion of two māhū together. In some lesbian relationships, one partner had male characteristics and lived much as a man, suggestive of the female form of māhū. There is no recognized institutionalized identity for gender diverse women. However, for the māhū, the overarching ideal is to be with a cisgendered androphilic male.

Sex work

As noted above, almost all participants had been sexually active with males (range of 2 – up to 8,000 lifetime male sexual partners; one participant reported having been celibate). While a substantial number reported having had sex for money (36%), only a few had done this regularly (9%). Yet, the existence of sex work came up frequently in our fieldwork.

Some of our participants noted they did not want to do sex work because of the loss of sexual desire or pleasure (e.g., “when you do sex work, it is not always pleasurable—it is work”). Of course, there are sex workers that enjoy themselves as they work. However, for others, it can rob them of sexual pleasure—not only in the work but in their lives as well.

It may be difficult for many to understand why so many māhū/raerae may engage in sex work at some point in their lives given the relative social tolerance and integration with family and societal structures (Matzner, 2001). However, as Matzner points out, some māhū have been kicked out of their home and face work dis­crimination that forces them into sex work for survival. Also, because the health care system does not provide hormones, some māhū need a source of income for hormones, electrolysis, and surgeries. While employment is available to māhū, the cost of living in French Polynesia is quite high—and Papeete is like New York in terms of cost of living. While not perceived as legitimate work by society or work that would be accepted by family members, more money can be earned in a short period of time with sex work (Mojola, 2011). Yet, opportunities for higher paying jobs are increasing for māhū/raerae and there is greater tolerance for māhū/raerae to dress as women on the job. A common sight in French Polynesia is māhū/raerae working in managerial or government positions. This has created an opportunity for more people to choose occupations other than sex work and afford the costs of gender affirming treatments.

When young māhū/raerae do engage in sex work, some are “taken in” by older māhū, who teach them survival skills and offer guidance. Because māhū/raerae are not controlled by pimps, as is often the case of female street prostitution, there is sometimes a community formed whereby sex work provides a source of social support. Participants told us that men finding them attractive and wanting sex with them could be affirming. This dynamic not only affirmed their sexual attraction to men but their identities as women.

There are obvious public health implications to sex work, including HIV and other sexually transmitted diseases (STDs). In addition, there is alcohol and drug abuse, particularly in the larger towns and cities. In Papeete, for instance, access to highly addictive drugs like cocaine, crack, “ice,” and various types of amphetamines is common. There is also a higher risk of violence and sexual abuse.

General health and HIV risk and status

Nearly all participants reported that their health was good or excellent (85%); only one participant reported that they were in fair health. The vast majority of participants (81%) felt that doctors had treated them in a knowledgeable, non-judgmental, and professional manner. One quarter of participants (26%) had ever seen a therapist.

In terms of sexual health, all participants seemed to be aware of the risks of HIV, although the prevalence was perceived to be very low. Participants understood they would be at risk for HIV infection if they did not use condoms for penetrative sex. The majority reported condom use, especially with anonymous sexual partners, but only 30% reported always using them; 4% never used one.

Many had been tested for HIV (66%) (Table 3). Two participants reported on the survey that they were HIV-positive. However, in conversations many people did not fully understand the meaning of the results of their HIV test. If you asked if they knew their HIV status, they might say it was positive. In further exploration, they explained that they were HIV negative and interpreted that as a “positive” result. What was more disturbing is that they seemed to receive little explanation of their test results from providers. Some said the doctors would simply say, “Everything is fine.” Others reported that they were shamed for having a positive gonorrhea test, indicating that they were engaging in unsafe sex.

Discussion

In numerous non-Western cultures around the world, a meaningful proportion of individuals occupy alternative gender categories beyond the man/woman binary, often grouped as transgender and gender diverse people. This article explores gender identity, gender expression, and sexual orientation of gender diverse individuals in French Polynesia. Building upon existing literature (Besnier, 1994; Kuwahara, 2014; Matzner, 2001; Stip, 2016), we find that the phenomenon of the māhū/raerae of French Polynesia represents a long-standing historical recognition of a third gender status that creates space for a level of respect and integration within families and broader society (Kuwahara, 2014). Our findings suggest it may be less stigmatizing to be gender diverse in French Polynesia than in many parts of the world despite there being a range of social tolerance and discrimination. Further research would be needed to confirm this hypothesis.

Globally, transgender individuals are disproportionately burdened by chronic (e.g., interpersonal rejection) and acute (victimization) discrimination. According to minority stress theory (Meyer, 2003), these forms of discrimination are associated with adverse mental health outcomes, including PTSD, a stress sensitive disorder that may arise from exposure to life-threatening events, serious injury, or sexual assault. However, in French Polynesia, the māhū/raerae are recognized and accepted as part of the culture and long-standing cultural traditions and, as such, are less stigmatized and in many cases are respected. Many māhū/raerae participants with whom we spoke described feeling comfortable with their identities and receiving support. Such experiences of acceptance generally improved across time.

This finding suggests that the māhū and raerae might be protected to some degree from negative proximal and distal factors as described in Minority Stress Theory (Meyer, 1995, 2003, 2015) and gender minority stress (Testa et al., 2015). While our subjects described gender-based rejection and discrimination, and non-affirmation, there was a fair degree of acceptance and affirmation afforded by the recognition of their third gender status, which has had a long and respected role in French Polynesian society. Their ability to find family support and a role within their families (and some of that was gained over time) created an important family connectedness. Polynesian identity is largely determined through one’s family relationships and family acceptance is very important in Polynesian society (Kuwahara, 2014). Family acceptance while growing up was more likely to come from mothers and siblings than fathers. There was an increase in acceptance as they grew older (and particularly among fathers) and this seemed to correlate with their own increased self-acceptance. This seemed to be reinforced by their emerging roles of caring for aging parents, relatives and, in some cases, adopting young relatives in need of care. This is consistent with Kuwahara’s observations of the mahu/raerae in French Polynesia (2014). This phenomenon has also been described by Ikeda (2014) among the māhū in Hawaii.

Thus, their ability to be recognized as part of the social structure as a third gender status and one which has value to society also seems to mitigate some of the negative effects of minority stress and gender minority stress that they might experience. Certainly, their ability to find employment and even incorporate outward feminine expression added to their self and other acceptance.

Not all is paradise however, as Vasey (2022) notes; there is danger in glamorizing and idealizing cultures where third gender status is institutionalized while missing their experiences of violence and trauma as a result of social oppression, stigma, structural inequality, poverty, and family rejection. We have tried to present our findings in a balanced way, recognizing the stigma that still exists on some level based on gender non-conformity with the social and family acceptance that many experience and the ability to be integrated into the social structure in a positive way. Are French Polynesian māhū and raerae more well off and experience less health disparities than other cultures? Our results can only support the possibility. Further, and more systematic, research will be needed to answer this question more definitively.

These findings open up areas for ongoing inquiry. First, we would emphasize that the observations reported on here are by no means exhaustive. Our most recent data were collected in 2014 (see limitations for more discussion on this). Thus, we envision this work as a baseline upon which to build future work. Beyond being descriptive, this research highlights questions related to the structure of sexuality.

While we were interested in examining risk and protective factors related to health disparities, we were also interested in simply describing the constructions of gender, sexuality and sexual relationships in French Polynesia. We would note the following concluding observations.

First, as noted in our results, in discussing gender diverse individuals in French Polynesia, western-based terminology, such as transsexual, transvestite, gender dysphoria, gender incongruence, and homosexuality did not resonate with participants. These terms tended to be inappropriate since they refer to twentieth-century Western psychiatrically derived categories (Alexeyeff & Besnier, 2014). We found, as noted by others, that the terms within French Polynesia differ in popularity within different islands (Alexeyeff & Besnier, 2014). For a number of māhū/raerae—they simply preferred to label themselves as a woman or a girl. Many adopted female names but could be addressed by either their given name or their adopted name given the context of the situation. The concept of māhū/raerae seemed closest to the term transgender—in that it seemed to be used irregularly by many who had various expressions of gender variance. The concept of māhū/raerae allows them to define themselves in a broader identity encompassing diverse gender identities and expressions rather than to live in western categories of cross dresser, trans, transsexual, transfeminine, pre-op. These findings suggest that the presence of the māhū/raerae people therefore needs to be understood in its own sociocultural context and not viewed from the Western perspective as “deviant” behavior (Besnier, 1994; Farran, 2014). In addition, we found that the terms within French Polynesia are not so clearly defined and shift over time and space as others have noted (Kuwahara, 2014, p. 94) is true in other Pacific islands.

Second, we found that māhū/raerae do not restrict themselves exclusively to passive/receptive sexual roles nor do their cisgendered gynophilic male partners restrict themselves to only active/insertive roles. This finding opens up questions about the way gender is constructed during sex. For instance, how do gynophilic men experience sex with māhū/raerae? How are bodies and genitals interpreted and navigated? Is reality suspended for the fantasy of being with a woman? Or, is sex with a man intriguing? In line with research suggesting that technical skills are important for pleasure (Armstrong et al., 2012), are māhū/raerae more skilled at oral sex because they also have penises? Or, are these practices functional and exemplary of a “second best” opportunity because women’s sexuality is highly regulated in French Polynesia? While subjective experiences no doubt differ, our findings suggest the need for more research with māhū/raerae, as well as their partners. This is a similar phenomenon which has been found in certain cultures where cis-gendered androphilic men can have sex with gender non-conforming men without stigma (e.g., Besnier, 1994; Coleman et al., 2018; Schmidt, 2016)

Third, our findings indicate that there is still a lack of institutional support in French Polynesia. As Christian churches are one of the main social structures in French Polynesia, our research suggests that the attitudes of these churches are still less supportive than other social structures (e.g., family, friends, teachers). Churches may be an important place where acceptance and creating roles for the māhū/raerae within congregations can contribute to self-esteem and well-being. Likewise, the role of the internet and social media remains unexplored as a possible venue through which to build social support, connection, and possible interventions for māhū and raerae people.

Fourth, our findings have implications for public health and HIV/STI prevention. While most participants reported being in good health, substance use, interpersonal violence, and unprotected sex were not uncommon. Existing research shows that epidemics of HIV/AIDS, interpersonal violence, and mental health and substance use problems are concurrent and mutually reinforcing, and contribute to inequities in HIV prevention outcomes. As we face public health challenges, it is important to study the complexity of human sexual behavior which often defies oversimplified approaches to HIV prevention and sexual health promotion. Our results indicated a need for more counseling and sex-positivity around HIV/STI testing in particular. At present, HIV/AIDS rates remain higher among sex workers and vulnerable populations in French Polynesia. In terms of global public health, we would point out that there remains a dearth of culturally congruent interventions designed to meet the mental health and HIV prevention needs of gender diverse people in lower and middle income countries, including French Polynesia.

Fifth, historical context matters. Contrary to one report suggesting that māhū/raerae were becoming less common and/or tolerated (Kirkpatrick, 1983), we found that as of 2014, they were a visible part of the fabric of society in the French Polynesia Archipelago (Marquesas, Tuamotus, and Society Islands). This is consistent with other more recent research in this area and consistent with other research on French Polynesia (Bauer, 2002; Elliston, 2014; Kuwahara, 2014; Matzner, 2001). Our data suggest that part of this tolerance results from historical acceptance of a third gender. In other words, even though there can be stigma attached to being māhū/raerae, there is also a historical tradition of gender diversity that people can fall back upon and perhaps feel less shame. This cultural recognition serves as a protective factor as compared to other cultures that greatly stigmatize transgender and gender diverse individuals or those who transgress socially defined roles for males. Yet, we also find evidence that same-sex sex among the māhū/raerae is normalized in part because they are seen as women who are participating in and reproducing heterosexual sex. As noted above, most, if not all, gender diverse males in Polynesia today are attracted to and have sex with cisgendered gynophilic men.

Finally, French Polynesia represents a culture in rapid transition with the process of globalization taking place. Importing and exporting constructs (e.g., gay identity) have their risks and benefits to the developing world and especially in societies with more collectivistic versus individualistic social structures (Coleman, 1996). The developed world also has an opportunity to learn about risk protection factors from indigenous cultures in the developing world. It is our hope that by exploring the less stigmatized phenomenon of the māhū/raerae presented here, we can ignite a more fluid sharing of ideas and possibility to support gender diverse individuals throughout the world.

Limitations

This study was limited by time constraints spent in each island, trying to cover the great expanse of French Polynesia, language barriers, and the Western stance of the investigators. Clearly, the sample was a convenience sample and not representative of all māhū and raerae in French Polynesia. The methodology did allow for follow-up visits to some islands and collecting data from various parts of French Polynesia, which allowed for some validation of findings mitigating the brief time that was spent in the two visits (2004 and 2014) across multiple islands. Further, this report is constrained by the inability to re-visit more recently due to COVID-19. This study also suffers from the lack of a co-French Polynesian scholar to investigate and interpret these results. Further, the last author, who conducted interviews and collected survey data, was not a trained anthropologist. While the methods employed may not be consistent with advanced anthropological methodologies, the first author is a trained sociologist who helped frame the study and contextualize it, and helped with review of the literature, data, and discussion. Nevertheless, this study reports on observations of behavior and cultural phenomenon that exist in French Polynesia and are understudied. Hence, the goal was to report on these observations within the French Polynesian context and their relevance for broader work on gender, gender diversity, and sexuality. As such, we adopted an inductive ethnographic approach rather than a theoretically grounded deductive approach. Finally, we felt it important to publish these findings now and use this as baseline and preliminary data that can inform future research in this area.

Conclusion

While French Polynesia is not a utopia for gender variant individuals, this cultural recognition serves as a protective factor as compared to other cultures that greatly stigmatize transgender and gender diverse individuals or those who transgress socially defined roles for males.

It is our hope that this article provides a discursive space for French Polynesian and other scholars to continue to explore this phenomenon of gender diversity, third-gendered, and trans identities. In addition, we hope this article adds to the extant literature on this phenomenon and stimulates further research to, and in reaction to, these observations. There is a great opportunity to more systematically study minority stress and gender minority stress risk and protective factors and examine these in light of health disparities. Ideally, this work can lead to more effective sexual and gender health promotion strategies, including HIV/STI prevention not only within French Polynesia but throughout the Pacific islands.

Funding Statement

This study was partially supported by funds from the Tawani Foundation.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. This study protocol was reviewed and approved by the Institutional Review Board for the Protection of Human Subjects of the University of Minnesota in 2006 (IRB – 0409S63829). Confidentiality of the survey data and the qualitative interviews and the field notes were maintained at all stages of data collection and analysis. Informed consent was obtained from all participants included in this study.

Disclosure statement

No potential conflict of interest was reported by the authors.