Alain Giami and Marie-Ange SCHILTZ
Representations of sexuality and relations between partners
Sex Research in France in the era of AIDS
Alain Giami is from INSERM (Institut National de la Santé et de la Recherche Médicale) U. 292, Bicêtre, France, and Marie-Ange Schiltz is from CAMS/ CNRS (Centre National de la Recherche Scientifique), Paris.
reprints to Alain Giami, INSERM U. 292, Hôpital de Bicêtre, 82 rue du
Général-Leclerc, 94276 Le Kremlin- Bicêtre Cedex, France. EMail : email@example.com
Keywords: sexual activity, sexual relations, social representations, HIV/AIDS prevention
This article presents a review of research carried out on sexual activity in France in the context of the HIV/AIDS epidemic. First, it aims at identifying the changes that have occurred in lines of research and in research tools under the impact of HIV/AIDS as compared to research performed before the emergence of AIDS. Research in behavioral and social sciences is thus considered both as a form of social response aimed at producing knowledge that can be used by public health officials and actors in prevention, and as a new form of the social construction of sexuality. Second, the article aims at giving a global overview and an integrative interpretation of results obtained through quantitative and qualitative research conducted in the general population and in specific groups. This interpretation is guided by the hypothesis that representations of sexuality and representations of partners held by individuals influence the specific sexual practices enacted according to the type of partner in each relationship. Condom use is studied in particular and, when related to strategies of partner selection, is used as an indicator of the status attributed by each individual to each of his or her partners, and the nature of relationships that are developed between partners in the era of AIDS.
At the present time, HIV continues to be transmitted principally through some specific sexual practices in most countries. This explains the view of numerous authors who have written that in the absence of an effective vaccine, prevention remains, now and for a long time to come, the only way to fight the propagation of the virus through sexual transmission (e.g. Lévy, 1994). This discourse, which has become a leitmotiv of AIDS research and "the fight against AIDS", has given a new legitimacy to research on sexuality in the general population and in those groups considered to be the most exposed to the risk of infection through sexual transmission. The link between "AIDS" (used in the sense of both a biomedical and a social phenomenon) and sexuality (defined as the physical activity, its accompanying mental activity, and the relations between the partners involved) has elicited a reassessment of lines and methods of research on sexuality.
Our general hypothesis is as follows: AIDS strongly inflects scientific and lay representations of sexuality. These representations can only be understood if examined in relation to representations of sexuality that existed prior to the emergence of AIDS. Our analysis of the research conducted in France in this context aims therefore at determining the new representations of sexuality that are taking shape under the impact of AIDS by examining the scientific representations that have been produced by new research studies and the lay representations that have been observed by both quantitative and qualitative studies. Thus, our analysis is undertaken from a double perspective, which is both historical (prior/since the era of AIDS) and comparative (quantitative versus qualitative research findings).
From a social-constructionist perspective (Treichler, 1992), conceptual and methodological aspects of scientific research constitute, at least as much as the results they generate, one of the facets of "the experience of sexuality" (Foucault, 1976, 1984a and b) unfolding within the AIDS context. Concepts of sexuality, the methods employed, the themes studied and those not studied, the specific groups subjected to research, and, last but not least, the practical recommendations on prevention developed in the AIDS context are well anchored in a specific social and historical situation and produce new configurations of sexuality (Gagnon, 1988). As a consequence it is important to describe these new representations with a certain social and historical detachment. In other words, research initiatives on sexuality constitute the social scripts of sexuality, be it explicitly or implicitly (Gagnon, 1989). The results obtained are comprehensible and assessable only if they are considered in the historical perspective of research on sexuality, and in relation to the concepts, tools and methods constructed to generate them (Giami, 1996a; Parker, Gagnon, 1995).
Research on sexual partnering, from an interactionist perspective, usually focuses on processes external to the individuals involved. Our perspective is different. We adhere the approach developed by M. Godelier who considers that "social relations are not only between individuals, they are within them (...) Ideology constitutes the inner framework of social relations and this framework is found as much within the individual as within his or her relationships" (Godelier, 1996, p. 28). This perspective gives a theoretical foundation to criticisms of the objectivism that has developed from surveys; such surveys are usually based on reports by individuals, and not on direct observations of the relations that may be found between two or more individuals (Bolton, 1992). Is it more important to know with whom individuals have a relationship, or to know with whom they think, believe or imagine they have a sexual relationship?
The analyses presented in this review are based on the postulate of a structural opposition and a complementarity between two models of sexuality: the moral opposition between "good" and "evil" underlying "good" sex and "bad" sex. Throughout history, this opposition has taken different guises: "healthy" versus "unhealthy" (Vigarello, 1993) or "normal" versus "abnormal" (Laumann, Gagnon, 1995). This structural opposition was conceptualized in the historical and theoretical works of S. Gilman (1985, 1988) which analyzed different aspects of representations and stereotypes by employing M. Klein’s psychoanalytic model of the cleavage between the "good" and the "bad" object (Klein, Heimann, Isaac, Riviere, 1952). In Gilman’s work, the double model of sexuality constructed from the analysis of different cultural documents (scientific and medical documents, artistic productions and fictions, media documents) states that cultural representations of sexuality include representations of different kinds of individuals. According to this author, different partners embody models or stereotypes of sexuality, and these models may influence the behaviors they elicit. We have applied this approach to research based on reports by individuals in response to questionnaires or interviews, in quantitative surveys and qualitative studies.
We have centered this review generally on the theme of sexual partners and multiple partners in particular, a theme eliciting strong interest in certain lines of research and methods of analysis. The study of the characteristics of sexual partners, be it their self-described characteristics or those attributed to their partners by survey respondents and the study of the interactions taking place between them constitutes one of the principal innovations in terms of surveys of sexual behavior in the AIDS era (Bastard, Cardia-Vonèche, Peto and Van Campenhoudt, 1996). This interest in sexual partners and especially in multiple partners has its foundations in the epidemiology-oriented approach associated with the study of the spread of the disease (Oppenheimer, 1988). We must also take into account the broader societal change in sexual behavior: an increase in matrimonial mobility (Léridon, Villeneuve-Gokalp, 1994), a slight decrease in the average age at first intercourse (Bozon, 1996), and within the last 20 years for women, an increase in number of sexual partners over the course of a lifetime (Spira, Bajos and ACSF Group, 1994). The partner-oriented approach is henceforth made possible by the changing sexual mores and by the liberalization of juridical statutes governing its expression that allow the appearance of new forms of institutionalized and non-institutionalized relations (Mossuz-Lavau, 1991). Studies of sexual practices no longer focus solely on the individual engaging in them, but take into account the characteristics of the partner and the type of relationship existing between the two individuals. It is important to know who does what and with whom since practices vary according to the type of partner involved. In this context, condom use constitutes a central theme within the analysis of sexual practices. In a great deal of research, this theme is used primarily to analyze risk-taking in regard to HIV. In our analyses, we have used this theme as an indicator of the representations that individuals attribute to their partner or partners.
In the following sections, we first outline the principal trends in research on sexuality in France prior to the emergence of AIDS. Secondly, we give an overview of the principal quantitative and qualitative research undertaken since 1985 and financed by the ANRS . Thirdly, we interpret the results of this research following the approach described above.
In France over the last 40 years, the study of sexuality has been primarily conducted by historians. Historians have investigated various aspects of sexuality, ranging from male homosexuality in Roman and Greek antiquity (Veyne, 1982) to the contemporary history of sexology (Béjin, 1982). Other topics related to sexuality that have captured the interest of historians include sexual morality in medieval and modern times (Flandrin, 1981, 1983), male impotence and its juridical treatment (Darmon, 1979) history of contraceptive practices (Ariès, 1984), prostitution (Corbin, 1982; Rossiaud, 1984) and venereal disease (Quétel, 1986). The majority of these works are in keeping with, on the one hand, the "Ecole des Annales" that favored the history of ideologies ("mentalités") and their long-term evolution through the themes of love and types of relationships between individuals, and, on the other hand, the French school of historical demography which examined the issues of marriage rate and natality. Some researchers have sought to establish a link between these two domains by attempting to evaluate the prevalence of sexual relations outside marriage through studies of the rates of illegitimate births. However, this endeavor was not fully implemented (Corbin, 1984). The global framework of these studies was based on the assumption that "all social organization bases itself on relations of kinship, which is to say, on the codification of sexual exchanges" (Duby, 1984).
For anthropologists, despite the heritage of Marcel Mauss (1966) who studied "body techniques" in detail, the sexual act stands out as the only human activity not to have been studied "in and of itself" (Gaignebet, 1991).
French sociologists have "repressed" the theme of sexuality from their field of study. Sexuality was considered by the Positivist School (Auguste Comte) and by Durkheim as "a passion endangering social organization" (Enriquez, 1996). However, following the example of historians and anthropologists, sociologists have studied sexuality in the framework of kinship, marriage, family and love relationships (Dayan-Herzbrun, 1991). From this point of view, institutions and the institutionalized forms of sexualized relationships they generate have become the focus of sociological research. Following the research of Girard (1964) who has studied the social factors influencing the "choice of the spouse", the work of Bozon and Héran (1987, 1988) falls in this tradition. The analysis of factors determining the "choice of sexual partner" however remains to be done since factors that contribute to choosing a lifelong partner differ markedly from those that come into play when one looks for a sexual encounter. Sexuality only seems to be understandable when constructed within the framework of institutions that channel and organize its expression, that determine its finalities, and that give it meaning. Historians, anthropologists and sociologists converge, therefore, in their emphasis on institutionalized sexual relationships, particularly those that occur within the framework of marriage.
The topics of sexual repression has been studied in a specific fashion. Work conducted in the 1960s and 1970s influenced by Wilhelm Reich, a major scholar in Freudian-Marxist tradition aknowledged and condemned the sexual repression of young people, women and homosexuals (Partisans, 1966, 1972; Recherches, 1973). The work of Michel Foucault, who was unable to bring to term his project on the history of sexuality (1976), occupies a place apart from the others and today exercises a stronger influence in Anglo-Saxon countries than in France. After demonstrating how societal power generates sexuality through the implementation of apparently repressive discursive codes, Foucault attempted to describe sexuality as a paradigm of the experience of the subject (Foucault, 1984a and b).
In this context, sociological research in the field of sexuality is rare. Dominique Wolton (1974), studied the institutions and actors who have contributed to the liberalization of access to contraception and abortion. He developed the thesis of the extension and renewal of social control over sexuality by "modernist" groups: doctors, social workers, and pharmaceutical companies (Wolton, 1974). This research explained the ways in which members of these professions, allied with feminists, have acquired legitimacy in broaching problems of sexuality by developing a medical approach to sexuality and by appearing in the public sphere as harbingers of sexual liberalization and modernity.
André Béjin and Michael Pollak (1977) took up the Weberian tradition of rationality (which strongly influenced European sociology) in order to establish the basis of a sociological approach to sexual behavior, free of psychologizing components. This approach aims at considering sexuality 1) as an autonomous sphere of social activity, 2) as an activity subject to accountability, 3) as being the target of a market of experts, 4) as an activity that can be the object of exchanges between free individuals in a market framework, and 5) as a central element of communication between individuals. Pollak abandoned previously developed approaches to the causes of homosexuality and to the psychological well-being and sociability of homosexuals, in order to develop a conception of homosexuality as a "lifestyle". The Weberian hypothesis regarding the rationality of human behavior, and sexual behavior in particular, led Pollak to study the ways of life of individuals belonging to homosexual groups and "ghetto"  movements "claiming a social space of their own allowing them the free rein of their sexuality, even at the cost of segregation" (Pollak, 1982, p. 48).
In the field of social psychology, the work of Alain Giami on the ideologies and representations of sexuality in reference to sex education (Giami, 1981, 1983) converged with the work of J. Gagnon and S. Gilman, which was relatively unknown in France at that time. Giami conducted a research program on representations of sexuality aimed at accounting for levels of society, group identity, and the individual as constituents of these representations. In that work, the individual emerges as the depository and interpreter of social representations. Furthermore, Giami formulated the hypothesis of a bipolar construction of representations, structuring oppositions between different significations attached to sexuality. Work on the representations of sexuality of handicapped individuals established the opposition between the "Angel" and the "Beast" as main components of the opposition underlying the structure of representations (Giami, Humbert-Viveret, Laval, 1983).
The "Rapport sur le Comportement Sexuel des Français" (The Report on Sexual Behavior of the French) (Simon, Gondonneau, Mironer and Dourlen-Rollier, 1972) holds a special place in French sexuality research. Undertaken by one of the leaders of the Mouvement Français pour le Planning Familial (the French Movement for Family Planning), it can be situated in the modernist and medical trend of thought described by D. Wolton (1974). In the tradition of sexual behavior surveys established by A. Kinsey (1948), the Simon Report proposes a naturalistic approach to sexual behavior, enriched by taking into account sociological and social-psychological factors. It constitutes the first systematic French survey of sexual behavior based on a representative sample of the general population, using the quota method (n = 2625). The questionnaire is based on a perspective of sexual health that assesses knowledge of contraception and biological mechanisms of sexuality. Following examples of the main body of quantitative surveys conducted in other industrialized countries during the same period, the Simon Report produced a construction of sexuality centered on heterosexuality, particularly on the evaluation of oral contraceptives use and sex education. It aims at describing and measuring the frequency of certain sexual practices, and attempts to explain these practices as being determined by modes of education, and as being correlated with sexual attitudes (Feldman, 1975; Clement, 1991; Giami, 1991).
During the same period, Michel Bon and Antoine d’Arc (1974) conducted a cross-sectional survey of homosexual men sampled from the readership of the principal "homophiliac" magazine at that time, Arcadie. Out of 15,000 subscribers to the magazine 1,034 individuals responded the questionnaire (6.7%). The sample therefore has no representative value on a national scale. The survey provides information of a sociological nature and uses a psychological approach to explore the familial psychogenesis of the behavior and affectivity of homosexual men. In addition, it provides information on the frequency of sexual practices, and on the influence of social conditions, such as repression and exclusion, on living conditions and sexual expression. In reaction to the Arcadie study which they judged to be "too timid" and preaching integration, begging for tolerance, and "reproducing bourgeois society", Jean Cavailhes, Pierre Dutey, and Gérard Bach-Ignace (1984) undertook a new investigation. Their project had a more explicitly political scope: to examine marginal sexualities, in particular homosexuality and gay lifestyles. The use of the term "gay" by homosexual men to designate themselves was considered at that time a subversive act. The survey was carried out in 1983, primarily in organized homosexual milieux. Press inserts were used to widen the base of recruitment. One thousand six hundred homosexuals replied, including 259 lesbians. The quantitative study was comprised of an open-ended questionnaire and supplemented by qualitative interviews of 300 individuals. Questions addressed different spheres of activity such as participation in the "gay world" (bars, nightclubs, etc.) and in the "pick-up scene", as well as homosexual integration and social interaction through work, religion, and especially politics, and specific risks such as homophobic aggression and sexually transmitted diseases.
The various sociological and social-psychological research interests that we have briefly presented  do not have today an equivalent influence on research developed since the emergence of the HIV-AIDS epidemic: the issues of sexual repression and of the extension of social control over sexuality are not the focus of study at the present time. However, two of the previously mentioned research interests are still influencing sexuality studies currently being conducted: Work founded on the hypothesis of the rationalization of sexuality is being pursued in the form of the study of management and adaptation to the risk of HIV. In addition, research on sexuality-related representations has continued to develop through the study of meanings attached to sexuality and sexual partnership.
Behavioral changes in reaction to the HIV-AIDS epidemic that are recommended by prevention agencies are seen as being legitimate and "rational ". This situation is linked, on the one hand, to the prevalent financing of research by public government and by institutions created in response to the AIDS epidemic (Bajos, Bozon, Giami, Doré and Souteyrand, 1995); and on the other hand, to the alliance of gay activists involved in the fight against AIDS with doctors, epidemiologists, and certain public health officials (Pollak et al., 1991). The contemporary AIDS-related research carried out, and which is presented further in this review, centers on the general population or on members of the principal groups exposed to the risk of HIV infection and constitutes in itself a form of social response to the fight against AIDS (Moatti, 1996). Given the urgent and recent nature of the social construction of this situation, it might appear difficult and premature to criticize the epidemiological paradigm structuring this research work (Giami and Dowsett, 1996). However, a retrospective reading of work carried out prior to the emergence of AIDS makes obvious the separation between research conducted on the general population (e.g. the Simon Report), and research conducted specifically on homosexual men. The questionnaire used in the Simon Report focused on reproduction and addressed the issue of homosexuality superficially, with only one question regarding lifetime same-gender experiences. However, sexual practices and psychosocial factors that enable researchers to understand sexual behavior have only been studied systematically in this medically-oriented research and in surveys on homosexuality.
Since 1985, there has been an unprecedented development of quantitative sexuality research concerning diversified aspects of sexual activity. Most surveys address sexual practices and certain psychosocial aspects considered to be determining factors of risk-taking in regard to HIV infection. Thus, much of this research aims at studying sexuality within the context of AIDS (Gagnon, 1988).
Table No.1 : Surveys of gay men
Surveys on gay men began in 1985 (Pollak and Schiltz, 1991) and have been replicated regularly by the same and other investigators (e.g. Schiltz and Adam, 1996). Centered at first on the study of homosexual lifestyles, this research has progressively shifted to the study of risk-taking and prevention strategies used by individuals or couples. These studies are generally not based on the use of representative samples. Subsequently, the ACSF (Analyse des Comportements sexuels en France: Analysis of sexual behavior in France), based on a random sample of the general French population, enabled researchers to specify the recruitment biases of these spontaneous samples (Messiah and Mouret-Fourme, 1996). The surveys of gay men used questionnaires distributed through the gay press, and were complemented two years in a row by samples of non-readers of the gay press who were recruited using a "snowball" sampling, and interviewed face-to-face (Pollak and Schiltz, 1991). Furthering his first studies of the most institutionalized forms of homosexual "cruising", M. Pollak (in collaboration with G. Pelé), examined new forms of collective and safe gay sexuality, using a quantitative method (Pollak, 1993).
Table No.2 : KABP surveys
Various KABP (Knowledge, Attitudes, Beliefs, and Practices) surveys have also been repeated regularly (Pollak, Dab and Moatti, 1989; Moatti, Dab and Pollak, 1990). These surveys were conducted face-to-face. The 1992 KABP survey and the 1994 KABP/ACSF survey were conducted by phone (Moatti, Dab, Pollak, and Groupe Kabp France, 1992; Moatti, Grémy, Obadia, Bajos, Doré and Groupe KABP/ACSF France, 1995). The main purpose of these surveys is to monitor the evolution of sexual behavior and social attitudes in regard to persons infected with HIV. The surveys aim especially at evaluating the impact of public prevention policies, but also provide information on the evolution of certain sexual behaviors and condom use, in particular.
Table No.3 : ACSF surveys
The national ACSF survey (Spira, Bajos and ACSF group, 1994) was conducted with the dual purpose of gathering information on the sexual behavior of the general population in metropolitan France (n= 20,055), and of providing useful information to improve social responses and prevention campaigns in the face of the HIV epidemic. The ACSF initiative maintains a multidisciplinary perspective that brings together epidemiologists, and a variety of social science researchers. The ACSF survey was conducted, in metropolitan France, by phone questionnaire and was specifically designed to compare the reliability of phone interviews with face-to-face interviews methods (ACSF Investigators, 1992). Moreover, the ACSF project included a survey of sexual behavior in the French Antilles and in French Guyana (n= 2,634) (Analyse des Comportements Sexuels aux Antilles et en Guyane: ACSAG) (Giraud, Gilloire, Halphen and de Colomby, 1996), and a survey of a representative sample of 15 to 18 year olds (n= 6,182) (Analyse des Comportements Sexuels des Jeunes: ACSJ) (Lagrange et al., 1996). These two surveys were conducted face-to-face.
A survey of sexual behavior and AIDS prevention among students in the Paris area was conducted in 1993 (Jaspard, 1994). This study sampled 1,806 students and was conducted face-to-face by a group of their peers.
An epidemiological representation of sexuality influences both the choice of populations to be subjected to in-depth questioning, and the construction of research initiatives and questionnaires (Giami, 1996a). The reproductive dimension of sexuality is rarely examined in these studies. However, the quantitative surveys of the general population (ACSF, ACSAG, ACSJ) and the survey of students in the Paris area do address the theme of contraception, but not that of abortion. Sexual practices defined principally as genital, anal, or oral penetration, but also including other forms of sexual activity, are treated exclusively in quantitative surveys by questionnaire. This body of research abandons the study of sexual positions and sexual dysfunctions that was the focus of sexologists in the 1970’s. Desire, pleasure and satisfaction are studied principally in quantitative surveys, notably in the ACSF while qualitative surveys do not address this aspect of sexual activity. The approach of sexual practices is considered unrealizable through qualitative methods, and especially through semi-structured interviewing methods (Bozon, 1995).
The question of change in sexual behavior is a central research issue. In the ACSF, the questionnaire is constructed so as to give prominence to changes in type of lifetime or short-term sexual activity and to changes in the type of partner with whom the individual had his last and second-to-last sexual intercourse. The repetition of the KABP surveys and of the surveys of gay men aims at measuring changes in a prospective fashion as well. This new approach brings into question the historical works that studied behavior or attitudes over long periods and concluded that changes in these areas can only be observed over historical periods (Elias, 1978). For example, in France, the widespread use of oral contraceptives constitutes a major societal change over the last 30 years, having profound consequences on relations between genders (Léridon, Charbit, Collomb, Sardon, Toulemon, 1987). Research on change in condom use assumes implicitely that condom use is likely to increase significantly every 2 years.
In the Simon Report, the questionnaire referred only to a single partner of the opposite gender when exploring sexual activity over the preceding year. This partner was defined as the "husband", the "wife" or the "boy/girl friend". The possibility of having had other sexual partners - of the opposite gender alone - was mentioned only when exploring lifetime sexual experience and in reference to marriage (before, during or after marriage). In the ACSF, however, marital status is no longer the principal variable used to define sexual partners. The questionnaire’s first objective is to obtain information on the persons with whom the individual has had sexual contacts. A sexual partner is defined as any person with whom the respondent reports having had at least one sexual contact, regardless of partnership status and gender. In the key questions related to partnership (i.e. last intercourse or numbers of partners in different periods of time) the respondent was asked to specify the gender of the partner. Thus the ACSF implicitly addresses the heterosexual bias inherent in previous surveys (Lhomond, 1993). The ACSF questionnaire also allows one to specify whether or not the spouse is a partner with whom the respondent has had sexual contacts in the preceding 12 months. The ACSF questionnaire and especially the ACSAG and ACSJ questionnaires study, in a non-normative manner, sexual relations taking place in non-statutory or very short-term relationships between individuals (i.e. those engaging in commercial sex and teenage relations). This situation may reflect an evolution in contemporary representations that fully recognize the high proportion of non-statutory sexual relations and that also more openly acknowledge the possibility of same gender relationships. In France at the present time, unlike countries such as the Netherlands, same gender relationships take place in a non-statutory framework.
The acknowledgement of non-institutionalized sexual activities also pertains to adolescents. The questionnaire directed at 15 to 18 year olds conducted in school settings required parental authorization for legal reasons. The fact that less than 5% of parents refused to allow their teenagers to participate indicates a recognition of the existence of teen sexual activity and an acceptance of the ability of teenagers to discuss sexual issues with adults.
Thus taking into account the relational nature of sexual activity modifies the individualistic point of view that predominated in earlier eras. Sexual practices, and especially condom use, can be studied as the object of negotiations between partners. Communication between partners, the balance of power underlying relationships, and the norms and representations that orient behaviors are now the focus of numerous analyses (Spencer, 1996).
The comparison of the Simon Report with the ACSF survey reveals differing patterns of change for men and women concerning the average number of lifetime sexual partners. For men, this number was 11.8 partners in 1972 and 12.1 in 1993; however, a more significant increase is observed for women : from 1.8 partners in 1972 to 3.2 partners in 1993 (Bozon, Léridon and Riandey, 1993). Other results provide information on sexual partnership configurations and especially the relation between having new partners with certain cultural characteristics of individuals. In 1992, over the preceding 12 months, although 18% of men and 10.5% of women reported having had a new partner, only 11.4% of men and 4.7% of women reported having had more than one partner over the same period. Over the preceding 5 years, 31% of men and 18% of women reported having had more than one partner; individuals under the age of 30 represent higher percentages than those in older age brackets. Moreover, the number of sexual partners over the lifetime appears to be correlated with the age at first intercourse: men aged 25 to 34 years old who had their first intercourse before the age of 15 average 24 sexual partners while men of the same age bracket who had their first intercourse after the age of 19 have had only 6 partners. The length of time exposed to sexual activity cannot entirely explain such a difference. The age at first intercourse can thus be considered as a major predictor of sexual lifestyle (Bozon, 1996).
If we consider the subgroups of men corresponding to particular cultural, social, geographic situations or sexual orientations, the differences between the numbers of sexual partners or the numbers of new partners in recent time periods are even larger as opposed to men overall. For example the 5.7% of men who declared using erotic magazines "often" are more often engaged in multiple partnership over the preceding 12 months than men overall (20.3% versus 13.5%). Furthermore, among men having multiple partners, readers of erotic magazines reported having had nearly 2 times as many partners than non-readers over the same period: 5.6 partners versus 3.1 partners (Giami, de Colomby and ACSF group, 1995). In the French Antilles, men who reported having had at least two partners over the preceding 12 months were two times more numerous in Martinique, and three times more numerous in Guadeloupe and in Guyana than in metropolitan France (Giraud et al., 1996). In metropolitan France, gay men reported an average of 14 same gender partners over the preceding 12 months (Schiltz and Adam, 1996).
Concerning sexual practices, the results of the ACSF survey show an increase in oral, anal and auto-erotic practices, for both men and women in the general population. Anal penetration was reportedly practiced "at least once over lifecourse" by 30% of men and 24% of women in the 1993 ACSF survey, compared to only 19% of men and 14% of women in the 1972 Simon Report. The structure of the Simon questionnaire did not allow researchers to determine the frequency of different sexual practices as a function of the type of partner involved. Secondary analyses conducted by the ACSF (Messiah, Pelletier, ACSF Group, 1996) showed that heterosexuals having multiple partners engage in sexual practices that vary according to type of partner, and that these practices are more diversified with casual partners than with cohabiting partners (including married partners) . Among gay men, a different picture has been observed: the sexual repertoire appears to be more restricted to safer-sex, between casual partners than between stable partners. In 1995, 16.7 % of gay men surveyed said they did not enact anal penetration with a casual partner compared with only 10.6% who refrained from this activity with a stable partner (Schiltz and Adam, 1996).
The 16% of individuals who reported having "changed behavior since the onset of AIDS" reflect the importance accorded to the characteristics attributed to others during partner selection. More than 70% of individuals in this subgroup reported having, since the emergence of AIDS, "sexual relations with people I know only" or "with whom I am in love only". The partner representations - "loved" or "known" - seem to dictate the decision of whether or not to have sexual relations with a new person. Partners who are "known" or "loved" are perceived as being "less risky" than other types of partners (Spira, Bajos and ACSF group, 1994). This self-reported change can be interpreted as a reinforcement of a "relational" type of sexuality to the detriment of "recreational" sexuality (DeLamater, 1987).
Concerning condom use, the ACSF survey has shown it to be unsystematic, and to vary according to whether the respondent has a single or multiple partners and according to age. The analyses presented in the survey report indicate that, during the last sexual intercourse, condom use varies as a function of the respondent’s characteristics, according to whether he or she has a single partner or multiple partners, and as a function of the respondent's representations of the partner. Thus, for example, more men and women having multiple partners were likely to report having used a condom during their last sexual intercourse when they "knew" or "did not know" whether their partner had had other sexual partners than when they "knew" that their partner had not had any other sexual partners (Ducot et al., 1994 ). These results indicate that when an individual "does not know" whether a partner has had other partners, he or she has a tendency to behave as if the partner is "known" to have had other partners (Giami, 1996b). Other analyses reveal a difference between men and women concerning "acquaintance" with the partner before the first sexual intercourse. On average, heterosexual men reported having known their partner for 82 days prior to the first sexual intercourse, whereas for heterosexual women, the average delay was 148 days. Men protect themselves less frequently when they are in love. Women do not protect themselves more when they are not in love. It appears that women select their partners and decide whether or not to use a condom before progressing to the sexual act. For women, being in love does not signify greater confidence in the partner, accompanied by less protection (Bajos, Ducot, Spencer, Spira, ACSF group 1996).
This male attitude towards partner selection, feelings of love, and condom use is confirmed in the French Antilles. For heterosexual men, not knowing well the person with whom one makes love, ignoring the serological status of this person, feeling that she could be unfaithful, not being in love with her, and the intention to stop engaging in sexual relations with her, often lead to a greater use of condoms than in cases to the contrary. These factors are not linked to greater condom use in heterosexual women however (Giraud et al., 1996). Furthermore, these authors formulate the hypothesis that condom use can also be linked to a contraceptive purpose.
Among French metropolitan adolescents, condom use varies considerably according to whether the female partner uses oral contraceptives: 83.1% of boys used a condom when their partner was not on the pill versus only 46.1% when their partner was on it. Seventy nine percent of girls not on the pill used a condom as compared to 18.8% of girls on the pill (Lagrange et al., 1996). Gay men have different practices and attitudes that vary according to whether they are with a stable partner of unknown or differing serological status, whether they are with a casual but not permanent partner, or with an unknown partner. In 1995, 8,6% of the respondents of the French Gay Press Survey had declared at least an unprotected anal intercourse with a stable partner of unknown or different serological status. In this specific type of relationship, this risky behavior occured on regular basis, for the majority of these men: 2/3 of these men declared that unprotected anal intercourse occured at least once a month. On the other hand, for the 9,5% of respondents who had declared such behavior with a casual partner of unknown or different serological status, this behavior is less frequent: 3/4 of these men declared unprotected anal penetration less than 4 times a year (Schiltz, Adam, 1996). These results indicate that risk perception and management depend on different criteria contingent upon the type of partner and nature of the sexual relationship.
Quantitative research brings to the fore changes in sexual practices or in relations between partners since the emergence of AIDS (over the last 10 years) and over shorter periods. The principal changes explored in these surveys concern changing patterns of condom use, on the one hand, and the strategies of sexual partner selection, on the other. From an epidemiological standpoint, the changes that lead to an increase in condom use and to a reduction or a "better" selection of partners are considered to be legitimate and consistent with increased rationality. This situation reflects the influence of a sanitary ideology regarding the construction of approaches to research and reflects the finality of this research as a support for prevention policies. Condom use in itself is universally considered as a rational practice whatever the risk of exposure (Moatti, 1996).
From the standpoint of the analysis of change in sexuality-related and partners representations, the results obtained from the quantitative surveys give an overview of the relative importance and diversity of forms of multiple partnership as well as some of its associated determinants. These surveys demonstrate that having multiple partners is not a generalized mode of relating to others common to the general population: certain groups have many more partners than others. These results also show that the social, cultural and geographic context, the sexual orientation, and involvement with a erotic culture can be associated with having many partners successively or simultaneously. Having, or having had, multiple partners implies a process of selection of potential and new partners. This in turn suggests a hierarchy, or as will be considered below, a classification and an opposition of types of partners according to various characteristics attributed to them by the respondent. This picture reveals, moreover, correlations between types of partner and more or less diversified repertoires of sexual practices. In addition, it shows a declared systematic or non-systematic use of the condom as a function of the type of partner and the representations he or she evokes. This hypothesis will be considered by presenting and analyzing qualitative research and the results obtained.
A number of qualitative studies concerning specific aspects of sexual activity and its context have been conducted on groups considered to be most at risk. These studies make use of qualitative methodology (e.g. semi-structured interviews, ethnographic observations) and are conducted under the auspices of various social and psychological sciences (anthropology, social psychology, psychoanalysis, sociology). The majority of this qualitative research does not address sexual practices as such, but integrates their analysis in the study of representations of sexuality, relations between partners and representations of sexual partners.
Table No.4 : Qualitative studies
Several important groups of individuals among those most at risk have been overlooked in these studies. It is of particular note that lesbians are completely absent from this field of research. As yet, there is no research done on the sexuality of prison inmates either. Information is also lacking on immigrants to France and haemophiliacs. In addition, the sexual experience and activity of People Living With HIV-AIDS has been addressed in a rather haphazard manner through a variety of research initiatives such as Coppel, Boullenger and Bouhnik's (1993) on youths in at-risk neighborhoods and the research conducted by Laurindo da Silva and Bilal (1992) on male sex workers. Women have not been the focus of specific research from the point of view of gendered sexuality, with the exception of the research of Plaza, Nguyen, Revault d'Allonnes and Schiffman (1993) treating the issue of women in the context of gynecological discourse and practices. Women are usually included in the study groups previously mentioned, but issues pertaining specifically to them are rarely addressed, thus making it difficult to understand their specific behavior and exposure to risk.
Qualitative research has identified different forms of multiple partnership, and has helped researchers to understand the processes inherent in relations between partners of differing status, the representations of different partners that each individual carries, and the varying degrees of influence from the social network and social context. Social networks function as "reservoirs ” of potential sexual partners and as matrices of normative influence and tension (Ferrand, Snijders, 1996). Representations of sexuality influence relations between partners, and the behaviors in which they engage. The analysis of these relations and of the representations underlying them is often linked to the implementation of AIDS prevention policies. Such analyses focus on bringing to the fore the relative proportion of psychosocial components (representations of partners and representations of sexuality) in the adoption or non-adoption of preventive strategies. In this review, the analysis of prevention strategies - in particular the partner selection strategies that are the most frequently adopted by individuals in the groups studied and condom use - sheds new light on partner relationships and representations of sexuality in the era of AIDS.
In a comparative study of youths in France and Greece, T. Apostolidis (1993) observed a distinction between "sexual" practices, labeled as at-risk behaviors, and "love" practices, considered not to incur risk. He paraphrases Greek mythology: "Agape does not contract Eros’ syphilis" to structure his argument. Furthermore, Apostolidis demonstrates that this dual representation of sexuality traces a distinction between men and women. Women are more often proponents of the "love-sexuality ” model with its commitment to relationships, whereas men maintain more of a dual model dependent upon the partner with whom they find themselves, and the type of commitment they expect in the relationship. This dichotomy of representations of sexuality guides practices and relationships between partners, and at times serves as a source of conflict and misunderstanding.
R. Mendès-Leite (1995) shows, in a socioanthropological study on "men practicing homosexual activities", how representations of sexuality and of partners - actual or potential - are structured in a system of "imaginary protections" according to three principal points of reference : "Me", "Us" (the identification group), and "Them" (the Others). These points of reference structure a distinction between categories of gay men. Oppositions between relations of a "conjugal" nature and casual relationships, and a hierarchization of partners based in part on "visible signs" like stigma revealing the serological status of individuals, and in part on geographical origin, exist within this system. For men living in provincial areas, AIDS and anal intercourse are Parisian attributes. The "big city" is seen as the place where "they" live (i.e. men with whom they are unable to identify and who represent a danger from which they must protect themselves).
In a psychoanalytic study of gay men and safer sex, H. Lisandre (1996) examines the issue of an unconscious identification with the gay community and homosexuality. The conscious and unconscious association of AIDS with homosexuality is widespread in the overall population as well as among gay men. Among gay men AIDS is perceived as "the divine punishment of homosexuality" and the AIDS victim is seen as a "martyr and a hero with a paradoxically enviable fate". In the "heat of the moment", issues become clouded, and condom use does not always follow the injunctions of public health discourse (Lisandre, 1996).
The research of Coppel, Boullenger and Bouhnik (1993) examines relations between partners according to a geographical logic that reveals the presence of a social logic. Young people in "at-risk neighborhoods" do not get involved as intensely in relations with partners from the same "housing project ” as with partners coming from "somewhere else". Moreover, within the same territory, individuals may have a different status based on their "reputation". A "good" or "bad" reputation affects young women in particular who are categorized either as "sluts who don’t know how to keep their distance" and who most often accept sexual relations with several young men in the gang, or as "serious" young women with whom a love relationship having a certain future can be envisaged. For their part, many young women from these "housing projects" prefer to meet sexual partners or boyfriends away from their home turf in order to preserve their reputation in their peer group. Moreover, some of these young people use the adjective "clean" to designate a partner presenting guarantees of morality. Inversely, the "rich" and "fashionable" Parisian milieux with which some of these young people come into contact, sometimes through prostitution, are thought of as a social circle where "diseases and sexual sadism run rampant". The choice of condom use seems to be quite complex. Young men sometimes use condoms during group sexual relations or with women they "don’t know well".
In research conducted on individuals in social situations characterized by higher or lower levels of group affiliation and by the assignation or lack of assignation of precise social roles, M. Calvez showed how risk selection operates as a function of what he termed "social experience". For individuals in a situation with a strong group affiliation but lacking a precise definition of their social role (as seen in young men in a housing project and in male homosexuals living in a large city), risk is rejected at the border of the group. For individuals lacking a support group but finding themselves in a strict social role (as seen in young women with social difficulties taken into shelters, and in certain intravenous drug users), risk is perceived to be an inevitable fatality in a globally uncontrolled environment (Calvez, 1992). This research sheds light on social factors providing individuals with, or denying them, the possibility of choosing their way of life and of selecting their partners. It suggests that the selection of sexual partners and condom use varies according to one's social circles.
Although situated outside the logic of HIV prevention, B. Bastard, L. Cardia-Vonèche and M.A. Mazoyer (1992) used "ways of communicating about AIDS" to analyze and understand relationship dynamics among separated or divorced persons who were thereby free of commitments. These authors studied the decision-making processes concerning the evaluation of risks in sexual and affective commitments. The analysis revealed the "strategic" and complex nature of communication about AIDS and of condom use by the couple. According to each case, communication about AIDS can signify a bringing together, or on the contrary, a pushing away of the partner, as well as trust or distrust of the partner. In any case, "condom use is never independent of relational strategies". Therefore instead of being a univocal characteristic of an already structured relationship, communication about AIDS seems to be an element in the strategies of construction of the relationship.
Among commercial female sex workers, there is a distinction made between various partners. S. Day (1990) has already shown the difference between "paying and non-paying sex" and "sex with a regular or casual partners". The type of sexual relation most often associated with condom use is the paying relationship with a casual partner wherein sex is defined as "business". Non-paying relations with a regular partner considered to be "intimate" or "love" relationships usually take place without condom use. The research of L. Laurindo da Silva and S. Bilal (1992) on male sex workers confirms this mode of management of relationships and sexual practices according to the type of partner involved. Male sex workers distinguish the client "they know" from the one "they do not know", between the "very good-looking one" and the one who is not, between clients in general and their companion, and between men and women (contamination can only come from contact with a man). Measures of protection are not limited to condom use, and are taken with partners judged to be "bad" or "ugly ".
A study of pornographic film stars has shown the difficulties faced by actresses who are more aware of the risks of contamination than their male counterparts. Protection strategies are based on the selection of professional partners according to whether they are "known" to be part of the milieu, and on whether they can provide proof of a recent negative HIV test. Only those few actresses who are aware of the risk-taking and successful in their profession are in the position to impose the use of a condom on their partners or on film producers. Some actresses abandon unprotected insertive practices with men and participate instead in "lesbian" films, judged on the X-rated market to be "hotter" and risk free. In this way, prevention strategies become linked to professional strategies (Giami, de Colomby and Paterson, 1995). Furthermore, A. Coppel, N. Boullenger and P. Bouhnik (1993) showed how a young couple of swingers established the safety of other swingers by using the supposedly known serological status of members of a well known network, and unknown couples recruited anonymously on the Minitel (a type of electronic telephone messagery service), to decide upon condom use.
Qualitative research has led to a better understanding of the variety of processes and representations already explored by quantitative surveys. Preventive recommendations to "protect oneself" from the virus are transformed into the necessity to "protect oneself" from certain potential or actual partners. This is manifested during the selection of partners either during sexual relations with new partners or when deciding to have safer sex or use a condom. This transformation is interpreted differently according to the context and situation in which the individual finds himself/herself.
The ACSF examined the issue of multiple partners by constructing a typology of different types of partner - cohabiting, regular, new, or casual partners - as a function of: length of relationship, characteristics of cohabitation, and frequency of sexual relations. This typology enabled researchers to analyze certain aspects of relations between partners, their sexual practices, and their adoption of specific prevention strategies. The ACSF and certain qualitative studies have established that partner selection is the most frequently adopted prevention strategy within the general population and in different subgroups. Condom use also enters into this logic of selection, in either a concurrent or complementary fashion. The condom is not used systematically with all partners: with some partners, it is not used at all; with others, it is used sometimes or often. In contrast, the situations examined in qualitative studies are found in specific and differentiated social contexts. Most of these contexts are supposedly closer to risk exposure than the general population. The qualitative studies have, for their part, developed in greater detail and articulated the analysis of representations of sexuality and of different sexual partners, and have brought to the fore some of the criteria called upon in the process of selection and choice of partners, and decision to use condoms. Beyond the knowledge obtained about these particular situations, the examples taken from qualitative research can function as "enlarging mirrors" to construct the global structure that organizes relations between partners at the present time in the context of the HIV-AIDS epidemic.
The analyses discussed herein have led to the identification of criteria used by individuals in various contexts to establish a classification and a hierarchy of different potential or actual partners. These criteria are as follows: sentiment (love), knowledge of the partner, social distance, appearance of the partner, type of relationship.
Table n° 5 : Psychosocial criteria and sexual practices used by groups studied to select and classify sexual partners
Table 5 is a classification of the criteria used by individuals in various contexts to select partners with who they wish to have sexual relations, and then to choose whether or not to use a condom of have safer sex. It shows how the relationship between the criteria used and the intended (or performed) sexual practices contribute to reveal representations of different types of partner. These representations appear to be structured according to a classification into "good" and "bad" partner consistent with the model described by Gilman (1985,1988). For most groups studied, "feelings of love" and "knowledge" about the partner seem to be of major importance. "Protection" as concerns a potential or actual partner takes on a symbolic dimension (Douglas, 1966) which goes beyond protection from HIV infection. As concerns HIV infection, it is perceived to be not necessary to protect oneself from a "good" partner, whereas the individual looks to protect himself/herself from a "bad" partner. This structural opposition seems to be a pretext for unconscious and conscious reframing of the signification attributed to actual or potential partners.
In the light of available epidemiological data (Réseau National de Santé Publique, 1996), all situations studied within this review do not concern individuals exposed to the same degree of risk of HIV infection. Strategies - especially partner selection strategies - adopted by individuals belonging to transmission groups exposed to low risk seem to be relatively rational in regard to risk, even if they are guided by a rationale other than protection against exposure to HIV. The same strategies adopted by individuals belonging to groups with higher exposure to risk seem less rational and less effective due to the much higher probability of encountering a partner infected with HIV. The German sociologist M. Bochow formulated the following hypothesis concerning this state of affairs: "During the 1980’s, male homosexuals were confronted with the idea - often loaded with moralism - that their terrible promiscuity could be the 'driving force of the epidemic'. This idea must be fought throughout the 1990’s in a polemical manner with the idea that it is love relationships that are becoming the 'driving force of the epidemic'. (....) It is neither the excessive consumption of alcohol nor the use of drugs that change the states of conscience having a dominant influence on at-risk behaviors among male homosexuals. It is rather the psychic dynamics of love relationships and/or highly erotic sexual encounters that lead to risk taking" . Concerning representations of sexual partners, M. Bochow explicitly opposes "the young romantic" to "the dirty old man". The representations that lead to the adoption of protected sexual practices - more with casual partners than with regular partners, and rather more with recent regular partners than with long-standing regular partners - are seen to be problematic in prevention among male homosexuals, although they can contribute to a reduction of risk in less exposed groups (Bochow, 1995).
Changes in sexual activity linked to representations of the AIDS epidemic take place within a structural framework that opposes in turn representations of sexuality and representations of partners, and that redefines itself progressively as a function of new criteria of partner desirability. The criteria implemented by individuals are not always determined by a rationale aimed at HIV prevention. In many cases, they reflect a readjustment of traditional partner selection criteria, reformulated in the language of prevention. The social-psychological criteria elaborated to select the most desirable partners - on a sexual, affective, or social level - do not always include criteria capable of determining the serological status - often unknown - of a partner.
In this way, partner selection, at first perceived as a prevention strategy, allows researchers to understand how the concern over AIDS is interpreted within the representations and relational strategies of individuals, and, inversely, how relationships between partners are recognized in the context of AIDS, through the use of criteria that may stem from considerations other than those of disease prevention.
The understanding of representations of sexuality and of partners, independently from the context of prevention, is essential to the development of prevention policies. AIDS, when considered as a cultural phenomenon, integrates pre-existing representations of sexuality that give meaning to the disease, structure its social role and define interactions between partners. The multiple discourses on AIDS are also the object of "reinterpretations" that are a function of biographies of individuals, their social and relational environment, and the needs they encounter.
We must thank John Gagnon and the anonymous reviewers for their comments on an earlier version of this paper.
Acsf Investigators. (1992). Analysis of sexual behavior in France : A comparison between two modes of investigation, telephone survey and face-to-face survey. AIDS , 6, 315-323.
Apostolidis, T. (1993). Pratiques sexuelles versus pratiques amoureuses : fragment sur la division socio-culturelle du comportement sexuel. Sociétés, 39, 39-46.
Bajos, N., Bozon, M., Giami, A., Doré, V., & Souteyrand, Y. (Eds). (1995). Sexualité et sida : recherches en sciences sociales . Paris: ANRS.
Bajos, N., Ducot, B., Spencer, B., Spira, A., & ACSF group. (1996). Sexual Risk Taking, Socio-Sexual Biographies And Sexual Interaction : Elements of The French National Survey on Sexual Behaviour. Social Science and Medicine (forthcoming).
Bastard, B., Cardia-Vonèche, L., & Mazoyer, M.A. (1992). Les choix et les comportements affectifs et sexuels face au sida. Une étude auprès de personnes séparées ou divorcées. Centre de Sociologie des Organisations, Paris, Rapport ANRS.
Bastard, B., Cardia-Vonèche, L. , Peto, D., & Van Campenhoudt, L. (1996). Relationships between Sexual Partners and Ways of Adapting to the Risk of AIDS: Landmarks for a Relationship-Oriented Conceptual Framework. In L. Van Campenhoudt, M. Cohen, G. Guizzardi, D. Hausser, (Eds.). Sexual Interactions and HIV Risk. New Conceptual Perspectives in European Research. London, Taylor & Francis, Social Aspects of AIDS, forthcoming.
Béjin, A. (1982). Le pouvoir des sexologues et la démocratie sexuelle. Communications , 35, 178-191.
Béjin, A., & Pollak, M. (1977). La rationalisation de la sexualité. Cahiers Internationaux de Sociologie. 62, 105-125.
Bochow, M. (1994). Data deserts and poverty of interpretation - notes on deficiencies in prevention-oriented research, taking gay men as an example. In D. Friedrich, W. Heckmann (Eds.). Aids in Europe - The Behavioural aspect. Vol. 4 : Determinants of Behaviour Change (pp. 249-256). Berlin: Sigma.
Bolton, R. (1992). Mapping Terra Incognita : Sex Research for AIDS Prevention - An Urgent Agenda for the 1990s. In G. Herdt, & S. Lindenbaum (eds.). The Time of AIDS -, Social Analysis, Theory, and Method (pp.124-158). London: Sage.
Bon, M., & d'Arc, A. (1974). Rapport sur l'homosexualité de l'homme. Paris: éditions universitaires.
Bozon, M. (1995). Observer l'inobservable : La description et l'analyse de l'activité sexuelle. In N.Bajos, M. Bozon, A. Giami, V. Doré, & Y. Souteyrand (Eds.). Sexualité et sida. Recherches en sciences sociales (pp. 39-56). Paris: ANRS.
Bozon, M. (1996). Reaching adult sexuality: first intercourse and its sequel, from timing to attitude. In M. Bozon, & H. Leridon, (Eds.). Sexuality and social sciences. An analysis of the French Social Behavior Survey. London: Dartmouth, forthcoming.
Bozon, M., & Héran, F. (1987). La découverte du conjoint I. Population, 6, 943-986.
Bozon, M., & Héran, F. (1988). La découverte du conjoint II. Population, 1, 121-150.
Bozon, M., Léridon, H., & Riandey, B. (1993). Les comportements sexuels en France : d'un Rapport à l'autre. Population et Sociétés , 276, 1-4.
Calvez, M. (1992). La sélection culturelle des risques du sida. Université de Bretagne. Rapport ANRS.
Cavailhes, J., Dutey, P., & Bach-Ignasse, G. (1984). Rapport Gai, enquête sur les modes de vie homosexuels . Paris: Persona.
Clement, U. (1991) : Surveys of Heterosexual Behaviour. Annual Review of Sex Research , 1, 45-74.
Coppel, A., Boullenger, N., & Bouhnik, P. (1993). Les réseaux d'échange sexuels et de circulation de l'information en matière de sexualité chez les jeunes des quartiers à risque. Paris: Rapport ANRS.
Corbin, A. (1982). Les filles de noce. Misère sexuelle et prostitution au XIX° et XX° siècles. Paris: Aubier.
Corbin, A. (1984). La fascination de l'adultère. In G. Duby (Ed.), Amour et sexualité en Occident (pp. 133-140). Paris: Seuil.
Darmon, P. (1979). Le tribunal de l'impuissance. Paris: Seuil.
Day, S. (1991). Anthropological perspectives on STD's. In N. Job-Spira, B. Spencer, J-P Moatti, & E. Bouvet (Eds.), Public Health and The Sexual Transmission of Diseases. Direction for future research and Health policy (pp. 88-97). Paris: John Libbey Eurotext.
Dayan-Herzbrun, S. (1991). La sexualité au regard des sciences sociales. Sciences Sociales et Santé, 9, 4, 7-22.
DeLamater, J. (1987). A sociological approach. In J. Geer, W. O'Donohue (Eds.), Theories of human Sexuality (pp. 237-256). New York: Plenum.
Douglas, M. (1966). Purity and danger, An analysis of conceptions of pollution and taboo. London: Routledge and Kegan Paul.
Duby, G. (1984). Introduction. In G. Duby (Ed.), Amour et sexualité en Occident (pp. 8-12). Paris: Seuil.
Ducot, B., Bajos, N., Belzer, N., Moatti, JP, & Spira, A. (1994). Condom use. In A. Spira, N. Bajos and ACSF group. Sexual Behaviour and AIDS (pp. 202-226). Aldershot: Avebury.
Elias, N. (1978). The Civilizing Process: The Development of Manners. New York: Urizen.
Enriquez, E. (1996). L'objet étrange de la sociologie: la sexualité, un objet étrange pour la sociologie. In J. Cournut (Ed.). Psychanalyse et sexualité. Questions aux sciences humaines (pp. 57-78). Paris: Dunod.
Feldman, J. (1975). Les rapports nationaux sur les comportements sexuels : un exemple de deux types d'interaction science - société. Archives Européennes de Sociologie , 16, 95-110.
Ferrand, A., & Snijders, T. (1996). Social networks and normative tensions. In L. Van Campenhoudt, M. Cohen, G. Guizzardi, & D. Hausser, (Eds.). Sexual Interactions and HIV Risk. New Conceptual Perspectives in European Research. London: Taylor & Francis, Social Aspects of AIDS, forthcoming
Flandrin, J.L. (1981). Le sexe et l'occident : évolution des attitudes et des comportements. Paris: Seuil.
Flandrin, J.L. (1983). Un temps pour embrasser : aux origines de la morale sexuelle occidentale (VI°-XI° siècles). Paris: Seuil.
Foucault, M. (1976). Histoire de la sexualité : La volonté de savoir. Paris: Gallimard. Tome 1.
Foucault, M. (1984a). Histoire de la sexualité : L'usage des plaisirs. Paris: Gallimard. Tome 2 .
Foucault, M. (1984b). Histoire de la sexualité : Le souci de soi. Paris: Gallimard. Tome 3.
Gagnon, J. H. (1988). Sex research and sexual conduct in the era of Aids. Journal of AIDS , 1, 593-601.
Gagnon, J.H. (1989). Disease and Desire. Daedalus , 118, 3, 47-78.
Gaignebet, C. (1991). Y a-t-il un acte sexuel pour l'anthropologie ? In J. Poirier (Ed.). Histoire des moeurs (pp. 875-889) tome 2. Paris: Gallimard, Collection la Pléïade.
Giami, A. (1981). Pour une approche sociale clinique des situations de changements. Bulletin de Psychologie, 349, 407- 414.
Giami, A. (1983). Une approche psycho-sociale clinique des idéologies de la sexualité. Bulletin de Psychologie , 360, 571-574.
Giami, A. (1996a). The influence of an epidemiological representation of sexuality: The ACSF survey questionnaire. In M. Bozon, & H. Leridon (Eds.). Sexuality and Social Sciences -An analysis of the French Sexual Behavior Survey. London: Dartmouth, in print, 1996.
Giami, A. (1996b). Partial non-response and "don't know" responses in surveys on sexual behaviour. Social Science Information, 35, 1, 93-109.
Giami, A., de Colomby, P., & ACSF group. (1995). Men using Porn magazines : Sexual activity and HIV prevention Data from the French National Survey on Sexual Behavior (ACSF). Poster presented at the XXIst Annual Conference of the International Academy of Sex Research. Provincetown Mass. USA.
Giami, A., de Colomby, P., & Paterson, F. (1995). Eroticizing condoms : Representations of sex and AIDS among sex workers of the porn industry, and consumers of porn magazines in France. A preliminary study. In D. Friedrich, W. Heckmann (Eds.). Aids in Europe - The Behavioural aspect. Vol. 2 : Risk Behaviour and its determinants (pp. 235-246). Berlin: Sigma.
Giami, A., & Dowsett, G. (1996). Social Research on Sexuality : contextual and interpersonal approaches, AIDS , 10 suppl. A, S191-S196.
Giami, A., Humbert-Viveret, C., & Laval ,D. (1983). L'Ange et la Bête - Représentations de la sexualité des handicapés mentaux par les parents et les éducateurs. Paris: Ed. du C.T.N.E.R.H.I.
Gilman, S. (1985). Difference and Pathology - Stereotypes of sexuality, race and madness. Ithaca, N.Y: Cornell University Press.
Gilman, S. (1988). Disease and Representation. Ithaca, N.Y.: Cornell University Press.
Giraud, M., Gilloire, A., Halfen, S., & de Colomby, P. (1996). Sexualité et sida aux Antilles et en Guyane. ANRS Information, 17, 28-36.
Godelier, M. (1996). Sexualité et société. In J. Cournut (Ed.), Psychanalyse et sexualité. Questions aux sciences humaines (pp. 27-40). Paris: Dunod.
Jaspard, M. (1994). Amour, sexualité et sida. Réflexions autour des résultats d'une enquête en milieu étudiant parisien. Paris: IUDP.
Kinsey, A., Pomeroy, W., & Martin, C. (1948). Sexual Behavior in the Human Male. Philadelphia: Saunders.
Klein, M., Heimann, P., Isaacs, S., & Riviere, J. (1952). Developments in Psycho-analysis. London: Hogarth Press.
Lagrange, H., Lhomond, B., Calvez, M., Darsch, C., Favier, C., Fierrp, F., Levinson, S., Maillochon, F., Mogoutov, A., Roché, S., & Warszawski, J. (1996). Enquête sur les comportements sexuels des jeunes de 15 à 18 ans. ANRS Information, 17, 14-27.
Laumann, E., & Gagnon, J. H.(1995). A sociological approach of sexual action. In R. Parker, & J. Gagnon (Eds.). Conceiving Sexuality . Approaches to Sex Research in a Postmodern World (pp. 183-214). London: Routledge.
Laumann, E., Gagnon, J.H., Michael, R., & Michaels, S. (1994). The Social organization of Sexuality. Sexual Practices in the United States. Chicago: The University of Chicago Press.
Laurindo da Silva, L, & Bilal , S. (1992). Prostitution masculine et prévention du VIH à Paris. Paris: ANRS.
Léridon, H., Charbit, Y., Sardon, J.P., & Toulemon, L. (1987). La seconde révolution contraceptive. La régulation des naissances en France de 1950 à 1985. Paris: INED-PUF.
Léridon, H., & Villeneuve-Gokalp, C. (1994). Constance et inconstances de la famille. Biographies familiales des couples et des enfants. Paris: INED, Cahiers de l'INED n° 134.
Lévy, J.P. (1994). Le problème d'un vaccin contre le sida. Médecine/Science, 11, 407-419.
Lhomond, B. (1993). Les enquêtes sur les comportements sexuels de Kinsey au Rapport Gai. Sociétés, 39, 29-38.
Lisandre, H. (1996). Les facteurs inconscients de la contamination par voie sexuelle In M. Calvez, M.-A. Schiltz, & Y. Souteyrand (Eds.). Les homosexuels face au sida. Rationalité et gestions des risques (pp. 77-88). Paris: ANRS.
Mauss, M. (1966). Sociologie et anthropologie. Paris: P.U.F.
Mendès-Leite, R. (1995). The meaning of otherness : Male Homosexualities and "Imaginary Protections" In D. Friedrich, W. Heckmann (Eds.). Aids in Europe - The Behavioural aspect. Vol. 2 : Risk Behaviour and its determinants (pp. 97-106). Berlin: Sigma.
Messiah, A., & Mouret-Fourme, E. (1996). Homosexuality, bisexuality: Elements of sexual socio-biography. In M. Bozon, & H. Leridon (Eds.). Sexuality and Social Sciences-An analysis of the French Sexual Behavior Survey. London: Dartmouth (forthcoming)
Messiah, A., Pelletier, A. & ACSF Group. (1996). Partner-specific sexual practices among heterosexual men and women with multiple partners : Results from the french national survey, ACSF. Archives of Sexual Behavior , 25, (3), 233 - 247.
Moatti J.-P. (1996). A quoi servent les enquêtes sur les connaissances, attitudes, croyances et pratiques de la population générale face au sida ? ANRS Information, 17, 8-13.
Moatti, J.P., Dab, W., & Pollak, M. (1990). Les attitudes et comportements des français face au sida. La Recherche, 233, 888-895.
Moatti J.P., Dab W., Pollak M. & Groupe KABP France. (1992). Les Français et le SIDA. La Recherche, 23, 1202-1211.
Moatti, J.-P., Grémy, I., Obadia, Y., Bajos, N., Doré, V. & Groupe KABP/ACSF France (1995). Sida: Dernière enquête nationale. Succés et risques des campagnes de prévention. La Recherche, 282, 30-34.
Mossuz-Lavau J. (1991). Les lois de l'amour. Les politiques de la sexualité en France (1950-1990). Paris: Payot.
Oppenheimer, G. (1988). In The Eye of The Storm: The Epidemiological Construction of Aids, In E. Fee, & D. Fox (eds.) Aids: The Burden of History. (pp. 267-300). Berkeley: University of California Press.
Parker, R.G., & Gagnon, J.H. eds (1995). Conceiving Sexuality Approaches to Sex Research in a Postmodern World. London: Routledge.
Partisans. (1966). Sexualité et répression , I, 32-33.
Partisans. (1972). Sexualité et répression , II, 66-67.
Pollak, M. (1982). L'homosexualité masculine ou le bonheur dans le ghetto. Communications, 35, 37-55.
Pollak, M., (1993). Homosexual rituals and safer sex. In R. Mendès Leite, & P.O. de Busscher (Eds.). Gay Studies from the French Culture. Voices from France, Belgium, Brazil, Canada and the Netherlands (pp. 307-317). New York: Harrington Park Press.
Pollak, M., Dab, W., & Moatti, J.-P. (1989). Systèmes de réaction au sida et action préventive. Sciences Sociales et Santé, 7, 1, 111-140.
Pollak, M., Dur, W., Vincineau, M. et al. (1994). Evaluating AIDS prevention for men having sex with men : The West European Experience. pp. S47-S60 in : Social and Preventive Medicine, "Assessing AIDS/HIV Prevention : What do we know in Europe ? " Vol 39, Suppl. 1.
Pollak, M, & Schiltz, M.A. (1991). Six années d’enquêtes sur les homos- et bisexuels masculins face au sida, Questions de méthode. Bulletin de Méthodologique Sociologique, 31, 32-48
Quétel, C. (1986). Le mal de Naples. Paris: Médecine et histoire.
Recherches (1973). Trois Milliards de Pervers. Grande Encyclopédie des Homosexualités. Paris. CERFI.
Réseau National de Santé Publique (1996). Surveillance du sida en France, situation au 31 Mars 1996. Bulletin épidémiologique Hebdomadaire, 24, 107-109.
Rossiaud, J. (1982). La prostitution dans les villes françaises au XV° siècle. Communications, 35, 68-84.
Schiltz, M.-A., & Adam, P. (1996). Relapse et cantonnement du risque aux marges de la "communauté" : deux idées reçues à l'épreuve de l'enquête "presse gaie" In M. Calvez, M.-A. Schiltz, & Y. Souteyrand (Eds.). Les homosexels face au sida. Rationalité et gestions des risques (pp. 11-24) Paris: ANRS.
Schiltz, M.-A., & Adam, P. (1996). Enquête Presse Gaie 1995. forthcoming, 1996.
Simon, P., Gondonneau, J., Mironer, L., & Dourlen-Rollier, A.-M. (1972). Rapport sur le comportement sexuel des Français. Paris: Julliard, Charron.
Spira, A., Bajos, N., & ACSF Group. (1994). Sexual Behaviour and AIDS. Aldershot: Avebury.
Spencer, B. (1996). Normative context of sexual behaviour and choice of prevention strategies. In M. Bozon, & H. Leridon (Eds.). Sexuality and Social Sciences -An analysis of the French Social Behavior Survey. London: Dartmouth, forthcoming.
Treichler, P. (1992). Aids, HIV and the Cultural Construction of Reality. In G. Herdt, & S. Lindenbaum (eds.). The time of AIDS -, Social Analysis, Theory, and Method (pp. 65-100). London: Sage.
Veyne, P. (1982). L'homosexualité à Rome. Communications, 35, 26-33.
Vigarello, G. (1993). Le sain et le malsain. Santé et mieux-être depuis le Moyen-Age. Paris: Seuil.
Wolton, D. (1974). Le nouvel ordre sexuel. Paris: Seuil.
 This theoretical definition also includes masturbation , in so far as this practice often involves imaginary partners with determinable characteristics.
 Agence Nationale de Recherches sur le Sida (National Agency of Research on AIDS). A public research organism created in 1989 as part of the public plan of action in the fight against AIDS which included the Conseil National du Sida (National Council of AIDS) to address ethical questions, and the Agence Française de lutte contre le Sida (French Agency of Fight against AIDS) in charge of coordinating prevention policies and assistance to People Living With HIV-AIDS.
 A ghetto is defined in urban sociology as a population that frequents a delimited territory wherein a specific concentration of groups leading relatively autonomous lives and developing their own culture can be observed.
 We did not include research on contraception, abortion and human reproduction and sexological and psychoanalytic research performed during the 70' and the 80'.
 These results are confirmed by the US national sex survey (Laumann, Gagnon, Michael, Michaels, 1994) that shows moreover that the longer the relationship between a couple continues, the less diversified the sexual practices.