Sex-Partner Roles in Homoerotic Relations: An Attempt of Classification

ABSTRACT The attempt is made to initially arrange in the terms of methodology the area of the research on partner roles in homoerotic relations. These issues have been noticed very early in human history (e.g., different roles performed or bimodal polarization), but only recently has science become interested in this subject. It is suggested to cover all such roles by the term sex-partner roles (SPR) instead of various other terms used and to classify them according to the ethological evolutionary approach into the appetitive and consummatory SPR. Further details of the division are discussed, and the utility of such classification is marked.


A history of the issue
The issue of the roles of sex-partners in same-sex relationships is rarely undertaken in the scientific literature, although it is of great importance for understanding human sexuality and deserves more attention. However, several publications discussing the topic have already been published, and it seems that there is a need to attempt to initially arrange this field of research in the methodological context.
It should be noted that although only recently have scholars started to discuss this issue, it must have certainly been noticed much earlier by humans in general. The awareness of the existence of different partner roles in homosexual behavior was certainly present from the moment that this type of behavior appeared among men. From some of the laws recorded in a Middle Assyrian code from the second millennium BC, we can learn, for example, that a man who actively acted in such encounters was not condemned, but it was the passive partner who was treated by the community with contempt because his action was seen as not natural, and his "feminine" approach excluded him from among the males (see, e.g., Nissinen, 1998). The awareness of the existence of different sexual roles is also clearly visible in the Assyrian divination manual Summa alu, where it was inter alia stated that "If a man has sex per anion with his social peer, that man will become foremost among his brothers and colleagues" (Nissinen, 2010, p. 75). Interestingly, similar attitudes toward homosexuality have survived until the present day. In some countries, the active partner is very often not considered a homosexual, but only the passive one is (see Faubion, 1993;Prieur, 1996;Carrier, 1985;Alonso, Koreck, 1988;Phellas, 2005;Carrara, Simões, 2008; for more references, see Jeffries, 2009). A similar approach is common in prison subcultures (see, e.g., Gagnon & Simon, 2005;Hensley, 2001) or among American adolescent boys (Pascoe, 2005). In many traditional cultures, homosexual behavior has also been differentiated in various ways due to the roles that the sexual partners played. The most prominent example of this, known in the literature and repeatedly cited, is the male initiation rites of some tribes in which young people make a ritual fellatio for older natives or they are penetrated per anum (cf. Carrier, 1977;Herdt, 1981Herdt, , 1984Kimmel, 2006;Kowalski, 2011;Lidz et al., 1989). Another example from ancient Israel was reported by Burns (2000).
In ancient Athens, various names were used to describe the roles in homosexual intercourse; most often the passive partner was called pais, a boy, or erastes, a beloved, while the active partner was called paidika, of boys, or eromenes, a lover (Dover, 1978). In the book Problems, attributed to Aristotle, and at least coming from his school, the reasons were considered why some men take pleasure from a homosexual act (Aristotle, 1970, IV, 26). It is supposed that this may be the result of the abnormal body composition of some men, which makes them slightly alike to the female body and experiencing pleasure not through genital stimulation, but rather by anal stimulation. The issue considered by Pseudo-Aristotle is an important question that the science still cannot finally answer (as to the hypothesis on the role the prostate plays in this phenomenon, see Hoppe, 2011;Morin, 1998). While in the case of heterosexual behavior, the roles of women and men are clearly defined by the biological conditions of both partners, in the case of homosexual relations, we deal with individuals of essentially the same anatomical features in the instances of both partners, so, it raises a question, what causes that in each homosexual act one partner adopts one role, and the second partner adopts another role, and in addition these both roles complement each other. In some cases, of course, it might be a matter of a fully free, unlimited agreement between the partners who are determined for sexual contact with the same-sex person in any way, but in many cases, a person has a more or less clearly defined range of preferences as possible for his sexual role, and the formation process of such, but no other, predispositions of these persons requires an explanation. It should be noted that this is not so much a question about the causes of homosexuality as such, but rather a question of psychophysiological mechanisms that govern its internal diversity, although it is possible that these two issues are related to each other more closely than it was previously thought.
Michel Foucault captured the question of roles of sexual partners as an important and clearly separate component of any sexual relations. In his view, the diversity of the roles has been noticed much earlier than the separateness of the phenomenon of homosexuality itself. "Role" or "polarity specific," as he defines it, in sexual behavior, has already been distinguished in ancient times (Foucault, 1984(Foucault, /1985. He writes about the ancient Greek verb aphrodisiazein (to copulate): but the verb can also be employed in its active sense, in which case it relates specifically to the so-called 'masculine' role in intercourse, and to the active function defined by penetration. And inversely, one can use it in its passive form -aphrodisiasth enai-designating in this case the other role in sexual union: the 'passive' role of the object partner. This role is the one that nature had set aside for women; (. . .) it is the role that could be imposed by force on someone who was thus reduced to being the object of the other's pleasure; it is also the role accepted by the boy or man who let himself be penetrated by his partner. (p. 46) However, the problematization of homosexuality as a separate phenomenon has become a fact, according to Foucault, only from the moment of the scientific sexology appearance, the event that was even dated accurately by the scholar as the year 1870 (Foucault, 1976(Foucault, /1978. Discussing the nineteenth-century stereotypical image of a homosexual, he indicates two different phenomena: in the deeply negative intensity of this stereotype, one might read the age-old difficulty, for our societies, of integrating these two phenomena-different phenomena at that-of the inversion of sexual roles and intercourse between individuals of the same sex. (Foucault, 1984(Foucault, /1985 A similar problem in sociological terms is discussed by Bourdieu (1998Bourdieu ( / 2001. According to him, the dichotomous active-passive symbolism is very important to understand both the concept of gender and the foundation of male domination. This symbolism is so deeply embedded in patterns of social perception and value systems that the existence of such a dichotomy may be seen as something completely natural and obvious. For a long time, sexology did not pay enough attention to this issue. In the second half of the 19th century, homosexuality was understood mostly as role reversal, so-called sexual inversion (see, e.g., Bem, 1993;Greenberg, 1988). Although, as mentioned by Wilson (2001, p. 191), in 1889 French journalist Ali Coffignon published the book in which he made an amateur classification of homosexuals. First, he divided them into two categories-the active and the passive-and within these two groups, he distinguished three further subgroups. The psychoanalysis, in its various sections and branches, devoted much attention to homosexual inclinations, but, initially, including Freud, it failed to perceive the issue of sexual partner roles, and it often confused this and mixed homosexuality as such with mental femininity-masculinity (Friedman, 1986, pp. 74-75).
The contemporary approach Brown (1957Brown ( , 1958 was one of the first scholars who noticed the necessity to differentiate between sex-role inversion and homosexuality. While the term inversion refers to the adoption of a sex role and the introjection of the psychological identity of the opposite sex, the concept of homosexuality refers to sexual desire and activity between two members of the same sex. Brown stated (p. 613): while certain forms of homosexuality (passive male and active female) are expressions of personality inversion, other forms of homosexuality have nothing to do with inversion. Failure to observe this basic distinction results in confusion, both in theory and research.
At the same time, Grygier (1957) suggested that scores on the femininitymasculinity scales in personality tests vary due to the active or passive role in homosexual activity. Bieber et al. (1962) attempted to establish a neutral, scientific terminology instead of the previously used colloquial and ambiguous names. They coined the neologism insertor to denote a person who is active in anal intercourse and passive at fellatio, while the insertee means the passive partner in anal intercourse and the receptive one in oral sex. Hooker (1965) essentially upheld the distinction proposed by Brown and the terminology proposed by Bieber et al. She wrote about the passive male homosexual or invert and the active or masculine homosexual (p. 25). She also observed the terms that were being used by a City Health personnel attending the study: the dominant and the receptive roles as equivalent terms for the insertor and insertee (p. 34). However, Haist and Hewitt (1974) pointed out that the majority of homosexuals do not conform to the commonly used labels.
The study on the need to distinguish between sex roles and sex orientation was undertaken by Ross (1975). He also empirically investigated the accuracy of the assumption that male homosexuals are either active or passive in sexual practices, and he found that about one half of the studied group exhibited equal preference for both, and many of them preferred oral and other mutual activities. Also, Harry (1977) arrived at similar conclusions. Carrier (1977) introduced the term sex-role preference and pointed out that it is an important explanatory variable. He followed Hooker (1965) in making a distinction between sex role and gender role. The first refers to typical sexual performance only, without assuming the gender connotations (masculinity-femininity) in these performances, while the second refers to "expected attitudes and behaviors that distinguish males from females" (Hooker, 1965, p. 24). Carrier provided the names for the roles that partners may play in sexual encounters: active or passive, dominant or submissive, insertor or insertee, as well as the types of sexual practices: the mutual manipulation of genitals, general body friction, and friction within anal or oral orifice.
In the second half of the 1980s, students started to be interested in sexual roles in same-sex intercourse in the context of the risk of HIV infection. It was quickly noticed that the image of homoerotic behaviors as bipolar is oversimplified. In fact, most often only the intercourse is bipolar, while the underlying predispositions of a substantial part of partners in this regard present a considerable degree of flexibility. Wiley and Herschkorn (1989) investigated the effect of differentiation of roles in homosexual intercourse on the size of the AIDS epidemic, taking into account various sex-role subgroups, and they proved that epidemic intensity increases with increasing size of the dual-role (both insertive and receptive) subpopulation. Weinrich et al. (1992) adopted the previously proposed terms inserter and insertee, and they named the roles performed in sexual intercourses genitoerotic roles. They also distinguished the preferences for an insertive or receptive approach in anal intercourse from the similar preferences in oral-genital and oral-anal contact. Coxon et al. (1993) investigated the prevalence of types of sex roles in behaviors among gay men and found a certain degree of role rigidity, but this rigidity is not large enough to conclude that gay men exclusively engage in either an active or a passive role. Van Druten, Van Griensven, and Hendriks (1994) used the term homosexual behavior roles in the context of the risk of the spread of HIV infection. They distinguished between receptive and insertive anal intercourse and found that changes in the roles of behavior distributions influenced the spread of HIV. Wegesin and Meyer-Bahlburg (2000) drew attention to the phenomenon of self-labeling occurring in the gay community that involves naming partners by some hackneyed, colloquial phrases according to the role played by them in sexual relations. Under this nomenclature, an inserter is called the "top," and an insertee is called the "bottom." These authors have also studied the interplay between the way of performing the roles that were declared using these labels and real behavior, and they found that there is a high degree of compatibility between them. Hart, Wolitski, Purcell, Gómez, and Halkitis (2003) studied the phenomenon of self-labeling, too, and found that 88% of people were using the labels "top," "bottom," and "versatile" and that there was a high degree of consistency between labeled and real sexual roles. Researchers have developed an interesting original scale of preferences for receptive or insertive intercourse, in which the subjects separately assess their predispositions for anal and oral sex on 5-point Likert-type scales. Carballo-Diéguez et al. (2004) found that adoption of the insertive or receptive roles partly depends on partners' characteristics. Ridge (2004) argued that the roles are organized around the partners' meanings of masculinity. Moskowitz, Rieger, and Roloff (2008) established that self-labels regarding partners' roles (i.e., the top, bottom, and versatile) were correlated with corresponding roles within sexual behavior, aside from intercourse. In a study based on a large sample, Jeffries (2009) analyzed the differences in preferences for receptive or insertive sex roles between Latino and non-Latino gay men, whereas in another large survey Grov, Parsons, and Bimbi (2010) investigated differences in gay men's sexual positioning with their partners depending on their penis size. Moskowitz and Hart (2011) conducted a survey on a sample of 429 gay men, in which they found the existence of objective, and not only subjective, premises that determine the intra-individual preferences toward the role in homosexual relations. Wei and Raymond (2011) highlighted that sex roles are an important aspect of identities and cultures among gay men, and they explored sociodemographic and behavioral correlates of preferences for such roles. Lately, Johns, Pingel, Eisenberg, Santana, and Bauermeister (2012) studied the influence of gender roles and interpersonal factors on positioning in homosexual acts. Similarly, Zheng, Hart, and Zheng (2012) found meaningful differences between homosexual self-label groups in gendered traits and interest. Most recently, there can be seen a tendency to use again the ambiguous term sex role or a periphrasis such as role during sexual intercourse (e.g., Armbruster, Roy, Kapur, Schneider, & Graham, 2013;Doerner et al., 2013;Valentova, Stulp, Třebický, Havlíček, & Fink, 2014).
The discussed issue has also gained significant attention of social constructionists. In part, they follow the distinctions and senses that Michel Foucault, their well-known precursor, pointed out and that were mentioned here. The dichotomy of active and passive male homosexuals received more recent interpretation of sex behavior entirely in terms of rigid ideas of masculinity and femininity, and it also reinforced the distinction between "true" (feminine) and "pseudo" (masculine) homosexuality (Plummer, 1981, p. 147). However, there is another vast current of the research pertaining to the dynamics of interpersonal relations in intimate couples. These relations are recognized as a result of the negotiation of day-to-day life in couples or in the community (Plummer, 1981, p. 124). How it is pointed, in such relations, societal values and beliefs are reflected that could be called heteropatriarchal, and-internalized by homosexuals-lead to the inequality of partners and sexual abuses within the couples (Cruz, 2000;Hearn, 2004). At the ground of these disturbed relations is laid a stereotypical, asymmetrical perception of the two parts where the one is perceived as "hyper-masculine" or "masculineacting," and the second, viewed as "feminine" or "feminine-acting," is often "othering." So the dominant partners seek to mirror the traditional heterosexual gender order with men's hegemony and try to subordinate the feminine-acting counterparts (Connell & Messerschmidt, 2005;Hearn, 2004). Moreover, the "feminine" part is sometimes, in extreme instances, recognized as "sissy," which may impose a "sissyphobia" (see, e.g., Eguchi, 2011). Two interesting studies addressed specifically the issue of the gay partners' roles in sex within this perspective. Kippax and Smith (2001) considered the roles as a relation of power between men where power is thought as a domination or negotiable dynamics. In turn, Hoppe (2011) termed the role performing as the positional identity, and he argued that the identities are construed in the frame of culturally structurized sexual scripts.
The issue of women's sexual roles in homoerotic relations is probably even less investigated than those regarding men. Singh, Vidaurri, Zambarano, and Dabbs (1999) studied psychobiological correlates of "butch" and "femme" erotic roles, and they found the validity of such classification. Brown, Finn, Cooke, and Breedlove (2002) explored possible biological differences between "butch" and "femme" lesbians. Rosario et al. (2009) examined whether any considerable differences exist in the sexual identity formation and integration between the butch and femme women. Zheng and Zheng (2011) noted that lesbians use the same sexual self-labels as gay men but with some different sense of those terms. Among Chinese lesbians, the following labels are common: the "T" for those who prefer the active role during sexual intercourse, the "P" for those who prefer the receptive role, and the "H" for the ones enjoying both roles equally. In a survey by Walker, Golub, Bimbi, and Parsons (2012), in turn, the scholars explored the intersection between lesbian gender labels (i.e., butch, soft butch, butch/femme, and femme) and attraction to sexual behaviors, and they did not confirm that the sexual behavior of the participants significantly differs according to the labels.
The phenomenon of self-labeling refers primarily to the phrases that are commonly used within the gay community. As it has been shown by the studies cited, the labels do not always have to be understood in the same way by all gay people, especially those with a rather loose contact with the community or with no such contact at all. However, the manner in which the communities define the roles of partners in sex may be interesting for students of sexual behavior and meaningful (e.g., when formulating questions in a questionnaire survey). To look more closely at this issue, for the article's purposes, there has been performed an overview of some of the most popular and most trafficked Internet sites for gay people seeking partners, the socalled dating Web sites. The data from five different such dating sites has been taken into account-that is, those sites that, according to Wikipedia, are amid the largest undertakings of this type: (1) adam4adam.com, 1 (2) manhunt.net, 2 (3) gay.com, 3 (4) gayromeo.com/planetromeo.com, 4 (5) gaydar.co. uk. 5 Such portals require users to register and to provide a certain amount of information characterizing the person (i.e., the creation of the so-called profile), which content is available to other users. Among the data collected, there is also a request to enter the preferred role that the user would like to play in sexual relations. Table 1 contains a list of phrases used in this context on those Web sites. A strong need to use such labels in all observed gay community sites proves that the categorization has significant practical usefulness.

An attempt of classification
The variety of terminology and the number of ways the issue has been discussed make it very difficult to advance research in this area, including the comparisons of the results of research and the use of the earlier findings. It seems justified, therefore, to try to bring a preliminary methodological order to these issues. One of the first steps, as stipulated by the methodology of science, is to make a typology or classification of these phenomena or facts on which our judgments are formulated, wherein the typology is considered to be a form of less comprehensive, complete, and unambiguous classification (Bailey, 1994).
Classifications are performed relatively rarely in the field of sex research that do not concern clinical aspects, probably due to the extremely complex nature of sexual behaviors, making it difficult to elaborate strict definitions. One of the more significant attempts was the taxonomy, which, in the context of gender-oriented sexual behavior, was offered by Devor (1994). In the field of research on homosexuality, a typology of male bar patrons was proposed in terms of their degree of self-disclosure of sexual orientation (Myrick, 1974), a typology of non-heterosexual male collegiate identities (Dilley, 2005), or a typology of effeminacies in a broader cultural context (Hennen, 2011).
First of all, we should distinguish between the issue of the roles of partners in the sexual act and the much broader issue of the individual's sex role in the society, known in the literature as sex roles (social sex roles) or gender roles (cf. Bem, 1993;Bem & Lewis, 1975;Hochschild, 1973). These roles are performed by individuals even if they have no current sexual partner, and largely they are publicly performed, while the roles of sexual partner are played mostly in isolation and have the interpersonal, dyadic character. The role of discourse on gender in the disentanglement of the discourse on homosexuality was discussed, for example, by Oudshoorn (1995); Rosenzweig and Lebow (1992); Bailey, Kim, Hills, and Linsenmeier (1997); Levitt and Hiestand (2005); or Sandfort (2005). Since the tradition of using the term sex roles in a social context had already been firmly established in the science, it was necessary to adopt another, distinctive name for these roles in the strictly sexual context. Many different phrases have appeared, and the list based on the reviewed literature is presented in Table 2. How it is seen, it is difficult to find even two identical names to specify the same phenomenon. From the analysis of the list, it is possible to deduce that all the terms are related to the roles that sexual partners mutually play, so it seems that they could be covered in one inclusive, general name: sex-partner roles (SPR). In the case of conventional and colloquial expressions, linguistic misunderstandings can easily occur in relation to non-anal intercourse, because the terms active and passive usually have the opposite meaning than in the case of anal intercourse. The term insertor has the advantage over the remaining types of terms because in both cases it suggests the person who is experiencing sexual gratification thanks to his genitals.
When attempting to classify, we are starting with a distinction that is often made in the literature between different variations of sexual acts, such as genito-genital, oral-genital, oral-anal, or masturbatory. In each of these types of behavior, each role listed in Table 1 can virtually occur. As it seems, different types of homosexual acts not only have different behavioral characteristics but also unequal importance from the other points of view-for example, from the ethological and evolutionary viewpoint. It is known that not all sexual behaviors originate in a straight line from our evolutionary ancestors, and not all find their evolutionary counterparts in other organisms (e.g., mammals); some of these behaviors are specific and unique to humans (see, e.g., Bagemihl, 1999). From this point of view, it is possible to make quite clear arrangement of homosexual acts with those that are an approximate equivalent, an analogue of the classical heterosexual genitogenital acts, appropriate for most organisms reproducing by sex, and with all those that may not be included there. The former should include anogenito-anogenital homosexual acts, while the others cover oral-genital, oralanal acts, masturbatory techniques, and others. Subsequently, it is worth noticing that all the sexual behaviors that we have previously taken into account refer to such actions that provide sexual gratification for at least one of the partners and can thus be defined as consummatory behavior. In regard to the behavior based on instinctual ground, the distinction between the consummatory phase and the preceding appetence phase has yet a long tradition in ethology (see, e.g., Manning, 1972) and is also being widely accepted in psychology (see, e.g., Kappas, 2011). Hence, it seems reasonable to identify all the SPR discussed so far as the consummatory sexual partner roles and to presume that it is possible to indicate such roles of sexual partners that are performed in the appetence phase of sexual behavior. It seems that the homoerotic appetence sexual partner roles also deserve the researchers' attention (i.e., their attempts to describe and understand them). For example, the possible parallels between similar roles in homosexual and heterosexual relationships may prove interesting. One can expect that quite different relationships exist between consummatory and appetence SPR in both types of sexual behaviors. It seems that a person, passive in the homosexual act, does not need to show passivity in the phase of courtship, seduction, and foreplay.
Finally, therefore, the proposed classification of SPR is exposed in following way: In supplementation to the classification, there could be developed an auxiliary typology of the SPRs. In studies that are included in our review, a certain degree of variability has been shown (i.e., flexibility in choosing the roles possible to perform). Such flexibility may be present at any given time (in regard to sexual orientation, see, e.g., Epstein, McKinney, Fox, & Garcia, 2012;LeVay, 1996;McConaghy, 1999), but it can also evolve over a person's lifetime (in regard to sexual orientation, see, e.g., Baumeister, 2000;Kinnish, Strassberg, & Turner, 2005;Subhi et al., 2011). The concept of role fluidity in relation to a given person raises a question of consistency between the role that is the object of the person's desires, plans, imagination, and the role that is performed in the reality. All manifestations of SPR may be therefore analyzed on two levels: • intentional SPR (desires, plans, imagination, etc.) • behavioral SPR (real conduct). LeVay (1996) discussed some essential classification problems in sexual research, and he proposed the two other levels of classification in regard to people's sexual orientation-specifically, the criterion of the physiological response for erotic stimuli and the criterion of subject's self-identification with a sexual orientation. For the research of SPR, the psychophysiological methods of measure (LeVay, 1996;McConaghy, 1999) seem to have a little, if any, usefulness. As to self-identification, it is the question of a choice (or a need) whether to treat this one as the part of the intentional states of the person or as a separate level of analysis. A schematic and rough character of typology allows each way.
Each of the roles may obviously occur on each of the levels. Additionally, one can examine the occurrence of various roles at both levels in the past, the present, and the future. In this regard, we follow similar assumptions as the authors of the Klein Grid (Klein, Sepekoff, & Wolf, 1985) adopted in respect to the assessment of sexual orientation.
An overwhelming majority of the studies that have been already discussed concerns male homosexuality; also, the proposed classification in the first place applies to this phenomenon. It is a matter of further research determining to what extent the roles adopted in female homosexual behaviors are possible to arrange in the same way, and to what extent they are completely different than those included in the presented classification.

Conclusions
The proposed classification might prove to be cognitively useful. The distinctions between the different subgroups within the population of homosexuals, which are facilitated by the classification, should be taken into account wherever possible in the course of surveys and when formulating theories. In each of these instances, one should indicate a population or its part to which his findings shall be applied. As was aptly observed in 1957 by Trevor C. N. Gibbens, the reason for the most common misconceptions about homosexuality is that there are many types of homosexuality and "people talking about different types as if they were one" (the end note in Scott, 1957, p. 659). A good empirical example of the desirability of such distinctions was provided by one of the formerly performed surveys, carried out by Olivier and Mosher (1968), in which significant differences were found between subgroups of homosexual insertees and inserters in regard to the results achieved by them on the MMPI and the Mosher Forced-Choice Guilt Inventory. Some of the reviewed studies have shown similar examples. Thus, as it seems, the findings of the research could be more precise and relevant, if the scope of their validity in relation to a population range will be described more accurately. This article is to provide a tool for it.