CAIRN-INT.INFO : International Edition
Volume
66
format_quote Cite share Share Switch to French

Full-text

1In France, a new form of civil partnership, the Pacte Civil de Solidarité (known as PACS) was voted and then promulgated on 15 November 1999 after a long political battle. While applying equally to homosexual and heterosexual couples, the PACS was initially associated mainly with the recognition of same-sex couples, and it was they who first made use of it. The PACS has since became increasingly popular among heterosexual couples, and the number of new PACS unions now rivals the number of marriages. In this article Wilfried Rault looks at couples who contracted a PACS union during the first few years after its introduction. Using data from the survey on the Context of Sexuality in France (CSF) conducted in late 2005 and early 2006, he examines attitudes and practices with regard to sexuality among heterosexual couples in a PACS, a marriage or a consensual union, and seeks to verify whether PACS partners are characterized by a less conventional, more “liberal” approach to sexuality than the other groups. He shows that they stand out more in terms of their attitudes to sexuality than in their practices.

2Individual behaviours with regard to union formation have changed considerably in France over the past four decades. Marriage, once the only option, has continued its more than 30-year decline, periods of cohabitation have lengthened, ways of marrying have diversified and, for some couples, consensual unions have become a long-term choice (Toulemon, 1997). These changes count among the major contemporary transformations affecting family and private life (de Singly 2008; Déchaux; 2009). In November 1999, this diversity was further increased in France with the creation of the registered civil partnership, the Pacte civil de solidarité (known as PACS). This was the French response to the demand for recognition expressed by same-sex couples, but differed from other European forms of civil union which exclusively targeted same-sex couples and were based on civil marriages (“registered” partnerships in the Nordic countries and, to a lesser extent, in Germany, or more recently, the civil partnership in the United Kingdom). The PACS was also intended for heterosexual couples wishing to benefit from a legal framework for their union without getting married; [1] it was designed to be different from marriage (Box).

3The specific characteristic of the French PACS has been the sustained annual increase in the number of such unions. As awareness of the PACS as an option for heterosexual couples grew and legal changes were implemented (notably concerning taxation), the number of PACS unions increased much more quickly than initially expected, rising from 20,000 per year in the early 2000s, to 77,000 in 2006 and 205,000 just 11 years after its creation. The majority of unions were heterosexual, accounting for 79 % in 2001 and 95 % in 2010 (source: SDSE, French Ministry of Justice). By 2009, there were two heterosexual PACS unions for every three marriages (Prioux and Mazuy, 2009).

4The statistical data for studying PACS unions is scarce and incomplete. The Ministry of Justice records some information, such as the date and place of registration, date of birth and sex of the partners, and the date and reasons for dissolving the partnership, but holds no information on the partners’ sociological profiles. Census surveys do little to fill this knowledge gap since they contain no questions about the PACS, which is not considered to be a marital status. Some general population surveys carried out in the 2000s do provide further information. They reveal that in the early years of the PACS, the contracting parties shared a number of characteristics, first and foremost social ones. PACS partners had a high educational level, frequently came from the higher socioeconomic categories and worked in the public sector (Davie, 2011; Rault and Letrait, 2010). They held a shared system of values, being fairly detached from religion and expressing less support for sexual stereotyping in both the workplace and in private life (Rault and Letrait, 2009a and 2009b). These trends are reflected in the answers sometimes given by PACS partners when interviewed about the reasons for their choice. They claim to prefer a type of union which, unlike marriage, is free of religions connotations or associations with sexual stereotypes (Rault, 2009).

Box. The main characteristics of the Pacte civil de solidarité

Under Law no. 99-944 of 15 November 1999, “A Civil Solidarity Pact is a contract entered into by two natural persons of legal age, of different sex or the same sex, to organize their lives together.” (Art. 515-1 of the French Civil Code).
Registered in a magistrate’s court (Article 515-3), the PACS is a partnership that grants both fiscal and social rights to the co-signatories and organizes mutual material solidarity between the two members of the couple (Art. 515-4). All the articles in the 1999 law reflect the legislator’s intention to differentiate the PACS from marriage. For instance, the modes of dissolution of the PACS contract (Art. 515-7), which include the possibility of unilateral dissolution, the absence of any mention of sexual exclusivity, or of inheritance in the absence of a will, all differ from the marriage contract. Strictly confined to couples, the PACS does not alter filiation rules. Other components, such as tax provisions or measures governing the residency rights of foreign partners, which might have brought the PACS closer to the institution of marriage, were modified and accompanied by a time limit, or excluded.
The Finance Law no. 2004-1484 of 30 December 2004, effective 2005, amended the tax aspect by removing the clause whereby PACS partners were not eligible for joint taxation during the first three years of their union: “Partners joined by a Civil Solidarity Pact as defined under Article 515-1 of the Civil Code, are subject to joint taxation on the income referred to in Paragraph 1. The tax notification shall be established in both their names, separated by the word ‘or’”. The new measure was effective from 1 January 2005.
Other changes were made to the PACS in 2006 and 2007, notably after the survey on the Context of Sexuality in France (hereafter referred to as CSF). Apart from an adjustment to the property regime, the measures all served to bring the PACS closer to marriage.

5In this article we will explore this differentiation from another angle, that of sexuality. In the surveys mentioned above, sexuality was sometimes mentioned by the contracting parties to explain their choice of a PACS, alongside other legal and ideological factors. Specifically, the PACS was sometimes favoured because it was seen to convey a more liberal view of sexuality. The fact that it is available to same-sex couples and makes no mention of fidelity (unlike Article 212 of the Civil Code relating to marriage) is highlighted as a representative aspect of the PACS. But for all that, do PACS couples have a specific attitude to sexuality? And if so, is this attitude merely a question of representations, or does it also concern behaviours?

6In order to understand the specific traits of PACS couples with regard to sexuality, we studied their “intimate orientations” by comparing them with the population who did not choose a PACS union. By “intimate orientations” we understand “ways of defining and using sexuality, expressed in representations and cultural norms and in ways of interacting between partners or in affects linked to sexuality” (Bozon, 2001). Michel Bozon distinguishes three forms of orientation: the “sexual network” model, that of “individual desire” and that of “conjugal sexuality” which, unlike the first two, is constructed with reference to the couple. Our aim here is not to assign any particular model to the respondents in our study, who all expressed their sexuality in a conjugal framework. [2] Rather, we test the hypothesis that the way in which this third model is expressed differs according to individual attitudes to types of union, by examining whether the representations and practices of PACS couples are characteristic of recent transformations in sexuality.

I – A study based on the CSF survey (INSERM-INED, 2006)

PACS couples in the CSF survey

7The survey on the Context of Sexuality in France (CSF) provides a good data source to investigate the hypothesis of a specific attitude towards sexuality among persons in PACS unions. The survey was carried out by the National Institute for Health and Medical Research (INSERM) and the National Institute for Demographic Studies (INED) on 12,364 persons during the autumn and winter of 2005-2006. The PACS group interviewed for the CSF survey comprised contracting parties who had opted for the 1999 law, or the amended version whereby tax measures for PACS couples were aligned with those applicable to married couples (Box 1). The CSF survey comprised 162 respondents who reported being in a PACS union. [3]

8The CSF survey and the Étude des relations familiales et intergénérationnelles (ERFI, the French version of the Generations and Gender Survey), a longitudinal survey carried out in 2005 and 2008, were among the rare general population surveys carried out in the 2000s that collected information about PACS partners. The CSF survey over-represented persons aged 18-39, the age group in which the most civil unions occur (Davie, 2011), and this is an advantage for our study. Because PACS partners were almost exclusively in the under-40s age group in the CSF sample, and in order to limit diversity in a field – that of sexuality – characterized by strong age and cohort effects, our study is restricted to the 18-39 year age group. We also removed individuals in same-sex couples. While it would be very interesting to study such couples, they raise a number of problems. There would be no point in including same-sex couples in the study groups already formed (see below) because they would not to be visible as such, and we would be mixing two population groups in different situations with regard to the law, since only heterosexual couples can marry in France. The only satisfactory approach would be to study gay and lesbian couples and heterosexual couples separately, but here the numbers are too small (there were only 12 same-sex PACS couples in the entire CSF survey). Comparisons with non-PACS couples, who are slightly more numerous, would not produce significant results.

A sociological approach: the advantages of the CSF

9With regard to sexuality, the past few decades have seen a diversification of intimate trajectories that is closely related to the waning influence of the institutions – marriage in particular – that have traditionally governed sexual behaviour. This diversity of individual attitudes to sexuality emerged with the rise of new sexual mores, the broader range of practices, and the fact that the experiences of men and women have grown more similar (Bajos and Bozon, 2008). The first conclusions of the CSF survey revealed the persistence of a normative system largely based on three components: heterosexuality, sexuality within conjugality (especially for women) and the persisting norm of vaginal penetration, all of which form part of the marriage-as-institution model that prescribed them, and that linked marriage to sexuality and procreation. That said, the components of this normative system have been weakened by the changes revealed in the survey. For instance, social perceptions of homosexuality are comparatively less disapproving, although it has yet to become an accepted feature of everyday life (Bajos and Beltzer, 2008). While monogamy is still customary, the increase in the number of partners over a lifetime has weakened the concept of “lifetime fidelity”, but without undermining the norm of fidelity itself. Similarly, representations of a possible dissociation between sex and emotional attachment, while still very gender-based, remain more frequent among the youngest cohorts than among the over-50s. Last, the widening of the repertoire of sexual practices reflects a lesser predominance of vaginal penetration (Bozon, 2008; Andro and Bajos, 2008). The transformations in sexuality can be used as a prism through which to examine whether PACS partners have specific characteristics, and to investigate whether those individuals stand out in any way from the normative system. To this end, we studied their attitudes to three components of the dominant norm in contemporary sexuality: heteronormativity, the link between conjugality and sexuality, and the predominance of vaginal penetration (Bajos, Ferrand and Andro, 2008), by asking questions about representations and practices (see Table 1).

Table 1

Questions used and recoded

Table 1
Components of normative change Question Possible responses and recoding (dichotomization of modalities in bold versus modalities in italics) Attitudes and representations relating to homosexuality(a) Concerning homosexuality, would you say that… It’s a form of sexuality like any other(b) It’s the sign of a liberated sexuality It’s the sign of a psychological problem It’s unnatural Don’t know If one of your children told you that he/she was homosexual … Imagine if one of your children told you that he/she was homosexual… You would have no problem accepting it It would be a problem for you You would not accept it Don’t know A child may be raised by two women A child may be raised by two men Strongly agree Agree Disagree Strongly disagree Don’t know In your lifetime have you been sexually attracted(c) … Only to men Mainly to men but also to women To men and women equally Mainly to women but also to men Only to women Link between conjugality and sexuality When two people are in a couple, which is the attitude you relate to the most? Lifetime fidelity Fidelity as long as the couple is together Fidelity with the possibility of a few affairs Other It is possible to have sexual relations with someone without loving them Strongly agree Agree Disagree Strongly disagree Don’t know In your life, have you had casual sex at least once? Yes No Don’t know In your life, were there times when you had two sexual relationships at the same time? Yes No Attitudes and practices relating to penetrative sex When having sex with a man/woman, what practice gives you the most pleasure? Mutual touching Your partner gives you oral sex You give your partner oral sex Vaginal penetration Anal penetration Do you think that sex without penetration is frustrating … Mainly for men Mainly for women Equally for both For neither men nor women Don’t know Attitudes and practices relating to penetrative sex Has a man/women given you oral sex over the past 12 months? Have you given a man/woman oral sex over the past 12 months(d)? Often Sometimes Rarely Never Have you and this partner had non-penetrative sex, not counting incidents of sexual dysfunction? Often Sometimes Rarely Never (a) In addition to these indicators of representations and attitudes, we asked a question to find out if the respondents had already had a same-sex relationship during their lifetime. (b) Only this first possibility was isolated for dichotomization purposes, for reasons explained by Nathalie Bajos and Nathalie Beltzer (2008). At the outset, the research team considered that the second item referred to a positive attitude, but revised its position after a study of the respondents’ responses: “All in all, the view that homosexuality was “liberated sexuality” appeared to refer to a “depraved” model rather than a positive one for many respondents”. (c) The order of the responses read out by the interviewer was reversed for the men. (d) These two questions were used jointly to construct the indicator for “reciprocal oral sex over the past 12 months”.

Questions used and recoded

Note: In all cases, non-responses were grouped together with the responses in italics (including “Don’t know”).
Source: CSF survey (INSERM-INED, 2006).

Comparing the choice of PACS with preferences for other types of union

10Based on the observation that marriage has evolved since the 1960s and is no longer closely associated with couple formation, and that cohabitation has become a lasting factor in many conjugal trajectories, we designed our analysis groups to reflect distinct individual attitudes to the ways in which couples are institutionalized.

11While the CSF survey is one of the few French general population surveys in the 2000s to mention the PACS in its questionnaire, it is difficult to exploit as it does not include the dates when the PACS or wedding were registered. Only the time since entry into union is recorded. It was thus possible to form groups of people who registered a PACS between the date when the PACS law was passed and the survey date, but this was not possible for people who married during this same period. We therefore opted for a two-stage analysis. In the first section we looked at all the individuals who formed unions in 1999 at the earliest. That was the year the PACS law was adopted and consequently the same types of union were available to all respondents (PACS, marriage and cohabitation). The first two types of union involve the signature of a legal contract. In this new context, the group of PACS couples exhibited relatively distinct behaviour in their attitude to forms of union, either temporary (people may get married after a PACS) or permanent. For the group who married, the marriage took place fairly early on in the couple’s trajectory. We did not conduct an exhaustive study of marriage and PACS for either of the two groups, but rather of a certain type of usage that excluded configurations in which individuals opted for one or the other after a long period of cohabitation. We added a third group that reflected a far more complex reality. These were the individuals who also formed couples in 1999 at the earliest, but who were neither married nor in a PACS at the time of the survey. Some will later opt for marriage, some will remain in a consensual union, while others will choose the PACS (and sometimes marriage afterwards). All these choices could be made in a short- or long-term timeframe. Persons in the “cohabiting” group certainly cannot be considered as advocates of free unions in the sense of an ideological refusal of all forms of institutionalization of the couple. By construction, this group is largely composed of individuals who had not been in a couple for a long time, since the married and PACS couples who had been together since 1999 had “left” that group to join one of the two others (unless there was no initial cohabitation phase). Consequently, possibilities for comparison between this group and the PACS (and married) group are limited.

12This first approach has a major limitation, since it does not allow us to compare the PACS with certain types of couple trajectory, such as late marriage or not marrying in the long term. As there were no marriage dates in the survey, the first comparison was impossible. However, we were able to examine whether PACS partners differ from long-term cohabitants who are also unattracted to marriage as an institution occurring early in the couple trajectory. We then applied a second approach to explore whether PACS partners have specific characteristics with respect to other non-married people. We compared individuals whose union was formed at least four years before the survey, [4] which allowed us to remove from the cohabiting group those individuals who might rapidly opt for marriage but who were in the cohabiting group analysed in the first section.

II – People in PACS unions differ more in their representations of sexuality than in their practices

Characteristics of the study population

13The groups we studied (Table 2) have distinct characteristics, notably regarding their age structure and time since union formation. Their social characteristics vary considerably, with persons in the PACS group more likely to have a degree in higher education than the others, notably the population in the third group, a small portion of whom are still students (classified as economically inactive). In PACS couples, both partners are more frequently in employment (74 % versus 64 % and 62 % in the other two groups). Similarly, the proportion of persons in higher-level occupations is higher among PACS couples. These two characteristics conceal yet another: women in PACS unions are better educated and more frequently in higher-level occupations than the others, since 76 % of them have a degree in higher education (versus 58 % of married women), and 34 % are in higher-level occupations (versus 17 % of married women). Here, the characteristics of persons in PACS unions are consistent with the few general population surveys that included this category. If we look at the educational differentials within couples, we see that a characteristic of heterosexual PACS couples is that the women often have higher educational qualifications than the men compared with the other two groups.

Table 2

Respondents who reported forming a couple between 1999 and 2006. Characteristics of the population(a) (% unweighted)

Table 2
Group 1 (PACS) Group 2 (marriage) Group 3 (cohabitation) Numbers 97 708 1,063 Sex (%) Men 44 328 494 Women 53 380 567 Age group (%) 18-24 12 6 22 25-29 43 36 36 30-34 35 38 25 35-39 10 20 17 At least one child in the household (%) 41 72 40 Occupational category (%) Farmers 0 2 2 Self-employed: business, trade and crafts 2 4 3 Higher-level occupations 29 20 13 Intermediate occupations 29 31 30 Clerical and sales workers 23 27 24 Manual workers 16 14 20 Economically inactive 1 2 8 Educational level (%) Lower secondary 21 24 29 Baccalauréat (high-school exit examination) 16 19 23 Higher education 63 57 48 Monthly household income (%) > €2,500 53 45 38 Time since union formation (%) < 2 years 21 6 29 > 2 years to < 5 years 48 35 41 > 5 years to < 8 years 31 59 30 Had already lived as a couple (%) 32 22 32 (a) In all the categorizations that follow, the results are weighted (for more on weighting and post-stratification of the survey, see Toulemon and Razafindratsima, 2008). The “Overall” column (Tables 3 and 11) is calculated on the basis of results obtained by sex.

Respondents who reported forming a couple between 1999 and 2006. Characteristics of the population(a) (% unweighted)

Reference population: Men and women aged 18-39 who formed a couple between 1999 and the date of the CSF survey (different-sex couples).
Source: CSF Survey (INSERM-INED, 2006).

14Persons in PACS unions also report the highest incomes and much more frequently indicate that the woman earns more than the man (more than one third, compared with one fifth in married couples and one quarter among cohabitants). With regard to age, in married couples the man is older than the woman more often than in PACS couples (62 % versus 50 %, 54 % in the third group).

15The large variations in age and educational level from one group to another were taken into account by performing logistic regressions to see whether the type of union had any influence on the representations and practices of sexuality, with other variables held constant. We added the presence of at least one child in the household to the model, the arrival of a child being an important milestone in the sexual trajectory of couples (Bozon, 2009).

Acceptance in principle of homosexuality, and persistence of heterosexual exclusivity

16The CSF survey enabled us to see whether the universe of values and representations of different-sex PACS partners is more favourable to homosexuality by recording opinions on same-sex parenting, perceptions of homosexuality, and the attitudes respondents would have if they had a homosexual child. Table 3 (register A) shows a split between the PACS group and the others, particularly clear-cut with respect to the married group. Because of the small sample size, the differentials between the PACS and cohabiting groups are not always significant. However, we observe a systematic trend towards a greater acceptance of homosexuality in the PACS group. In all cases, there are considerable disparities of opinion between men and women, the women in all groups being more “gay-friendly” than the men. While the distribution of attitudes by sex varies considerably from one group to another, the gap between men and women remains, and is especially strong for the concept that homosexuality is “a form of sexuality like any other” and for acceptance of a homosexual child. Examining the responses to each question also revealed greater reluctance on the part of men to accept same-sex parenting, which has also been observed in the French population as a whole (Bajos and Beltzer, 2008).

Table 3

Attitudes to transformations in sexuality by type of union (percentage of each category)

Table 3
Group PACS Marriage Cohabitation Overall Register Numbers 97 708 P(a) 1,063 P(a) 1,868 A. Attitudes and representations relating to homo- sexuality Homosexuality “is a form of sexuality like any other” Overall M F 80 68 91 56 49 62 *** ** *** 70 62 77 ** n.s. *** 65 58 72 “No problem” in accepting a child’s homosexuality Overall M F 67 62 72 44 36 52 *** *** *** 57 49 65 ** ** n.s. 53 45 61 A child may be raised by two women Overall M F 64 61 67 48 44 51 *** ** ** 58 51 65 n.s. n.s. n.s. 54 49 60 A child may be raised by two men Overall M F 52 39 65 40 33 47 ** n.s. ** 52 41 63 n.s. n.s. n.s. 49 39 59 B. Link between conjugality and sexuality Agrees with the statement: “fidelity is for life” Overall M F 42 36 48 64 67 61 *** *** * 47 42 52 n.s. n.s. n.s. 53 51 55 It is possible to have sex with someone without loving them Overall M F 57 72 42 34 47 21 *** *** *** 44 54 34 *** *** n.s. 41 52 29 Has already had casual sex with someone Overall M F 43 52 33 29 38 21 *** * * 39 48 31 n.s. n.s. n.s. 35 45 26 Has already had two sexual relationships at the same time Overall M F 18 23 13 20 26 13 n.s. n.s. n.s. 20 32 18 n.s. n.s. n.s. 22 29 16 C. Attitudes and practices relating to penetrative sex Prefers mutual touching Overall M F 42 43 41 31 28 34 ** * n.s. 29 22 36 ** *** n.s. 31 26 36 Non-penetrative sex is not frustrating for “either men or women” Overall M F 55 59 50 40 41 39 *** ** n.s. 42 45 40 ** * n.s. 42 44 40 Had reciprocal oral sex over the past 12 months (often or sometimes) Overall M F 33 34 32 30 27 32 n.s. n.s. n.s. 36 35 38 n.s. n.s. n.s. 33 32 33 Had non-penetrative sex with current partner (often or sometimes) Overall M F 37 38 35 30 32 27 n.s. n.s. n.s. 34 36 32 n.s. n.s. n.s. 33 35 30 Interpretation: 42 % of PACS respondents agree with the statement “fidelity is for life” compared with 64 % of married ones. (a) The results are tested for the significance (p) of the difference between the compared percentages (PACS versus the two other groups).

Attitudes to transformations in sexuality by type of union (percentage of each category)

Significance levels: *** p<0.01; ** p<0.05; * p<0.10, n.s. non-significant.
Reference population: Respondents aged 18-39 who formed a couple between 1999 and the date of the CSF survey (different-sex couples).
Source: CSF Survey (INSERM-INED, 2006).

17The regression analysis of the variable most representative of the attitude towards homosexuality (“a form of sexuality like any other”) confirmed that persons in a PACS union stand out by their acceptance of homosexuality, all other things being equal. This specificity is much more pronounced with respect to married persons (Table 4). This is not surprising since, by construction, group 2 (marriage) over-represents people who married quickly and who are often characterized by hostility towards homosexuality, a corollary of their strong ideological attachment to gender differences (Rault and Letrait, 2009a).

Table 4

Logistic regression on the fact of considering homosexuality to be “a form of sexuality like any other” (logit model, odds ratio)

Table 4
Overall Men Women Sex Male Ref. – – – Female 1.8 *** Age group 18-24 1.1 n.s. 0.9 n.s. 1.3 n.s. 25-29 Ref. – Ref. – Ref. – 30-34 1 n.s. 1.3 n.s. 0.9 n.s. 35-39 1 n.s. 0.9 n.s. 1.1 n.s. Union group 1. PACS Ref. – Ref. – Ref. – 2. Marriage 0.3 *** 0.4 ** 0.2 *** 3. Cohabitation 0.6 ** 0.8 n.s. 0.4 ** Educational level Lower secondary Ref. – Ref. – Ref. – Baccalauréat (high-school exit) 1.4 ** 1.3 n.s. 1.6 ** Higher education 1.5 *** 1.7 *** 1.3 n.s. Children in the household No Ref. – Ref. – Ref. – Yes 0.6 *** 0.7 ** 0.6 *** Time since union formation < 2 years 1 n.s. 1 n.s. 0.9 n.s. > 2 years to < 5 years Ref. – Ref. – Ref. – > 5 years to < 8 years 1.4 *** 1.2 n.s. 1.6 ***

Logistic regression on the fact of considering homosexuality to be “a form of sexuality like any other” (logit model, odds ratio)

Interpretation: An odds ratio that is statistically significant and above 1 indicates that for the category concerned, the factor increases the likelihood of considering homosexuality to be a form of sexuality like any other compared with the reference category of the variable considered. The further the odds ratio is from 1, the greater the influence of the factor to which it is associated.
Significance levels: *** p<0.01; ** p<0.05; * p<0.10, n.s.: non-significant.
Reference population: Respondents aged 18-39 who formed a couple between 1999 and the date of the CSF survey (different-sex couples).
Source: CSF Survey (INSERM-INED, 2006).

18Other factors influence the probability of being favourable to the idea that homosexuality is a form of sexuality like any other. The probability decreases with the presence of a child in the household, but increases for women and persons with a high educational level.

19Regression analysis by sex showed that the effect of the type of union is stronger for women. Being in a PACS union increases the probability of considering homosexuality to be a form of sexuality like any other compared with the two other groups. Among male respondents, the PACS group only stands out with respect to the married group, the results obtained for cohabitants being barely significant.

20The PACS is part of a specific universe of representations and values. In terms of sociability and sexuality, this is reflected in the respondents’ varied individual life event histories. Persons in a PACS union less frequently report knowing no homosexuals in their social network of family, friends and colleagues (22 %) than married persons (34 %) or cohabitants (25 %). The responses are sharply divided by sex, with men more frequently reporting no homosexuals among their acquaintances than women. The CSF survey allowed us to explore the terrain of sexuality more explicitly, to find out if persons in PACS unions are more attracted to people of the same sex over their lifetimes. In the three groups, reporting attraction for a person of the same sex is rare, concerning just one person in ten, and more frequently women than men. The differences between the groups are not significant and the largest gap is between the PACS and the marriage groups, with 9 % of persons in a PACS union reporting having already been attracted to someone of the same sex, and 6 % of married persons. Lastly, very few reported having had at least one same-sex partner in their lifetime: 7 %, 3 % and 4 % of women in the three groups, respectively, and 6 %, 2 % and 3 % of men. These observations concerning homosexuality tend to reflect representations [5] rather than reported practices.

21Viewed in terms of their more favourable attitude to homosexuality, the hypothesis whereby users of the new PACS union are more representative of the current transformations in sexuality appears to be confirmed. In reality, however, their attitudes reflect representations rather than actual practice. The choice of a PACS union reflects a less exclusive attachment to the heterosexual norm. This observation is not surprising, since the PACS was not established on a basis of gender difference (unlike marriage), and therefore represents a challenge to the sexual order based on the sole recognition of heterosexuality. This result probably reflects a particular context, since the survey respondents chose the PACS union shortly after it came into being, when it was still closely associated with the recognition of same-sex unions.

Rejection of prescribed fidelity, persistence of monogamy

22A second component of the dominant normative system concerns the connection between sexuality and conjugality. Two dimensions can be explored to test the hypothesis of a specific attitude of persons in PACS unions in this area. One is the degree to which they adhere to a model of fidelity that values the relationship for itself rather than its durability (Bozon, 2004; Marquet, 2004). In the sample studied, models of “lifetime” fidelity and “as long as we’re together” are represented in similar proportions (Bajos et al., 2008). The former is mentioned more often, however, and 55 % of women in the sample and 51 % of men report agreeing with the “lifetime” fidelity model. PACS partners identify with this model much less frequently than respondents who opted for marriage (42 % versus 64 %, see Table 3, register B), with a particularly strong contrast for the men (36 % versus 67 %).

23A second aspect of this close correlation can also be analysed. It concerns the association between sexuality and emotional attachment. A greater contrast is observed here, and the PACS group is more likely to break with dominant representations. Among all 18-39 year-olds who have formed a couple since 1999, 29 % of the women agree with the possibility of dissociating sexuality and emotional attachment, and 52 % of men, and similar results were obtained in the third group. However, both the men (72 %) and the women (42 %) in the PACS group clearly differ. This is even more manifest with respect to the married group (47 % and 21 %). In all cases, there is a considerable gap between the responses of the men and women.

24The regression performed on the dissociation between sexuality and emotional attachment shows a significant effect of the group studied (Table 5). PACS respondents less frequently reject this dissociation than the others, with PACS men standing out from the two other groups, whereas the strongest rejection of this dissociation is observed among married women. For both men and women, rejection is also much stronger among those with a high educational level.

Table 5

Logistic regression of the fact of agreeing with the idea that it is possible to have sexual intercourse without emotional attachment

Table 5
Overall Men Women Sex Male Ref. – – – Female 0.4 *** Age group 18-24 0.8 n.s. 1 n.s. 0.7 * 25-29 Ref. Ref. – Ref. – 30-34 1.1 n.s. 1.2 n.s. 0.9 n.s. 35-39 1.3 * 1.4 * 1.2 n.s. Union group 1. PACS Ref. – Ref. – Ref. – 2. Marriage 0.4 *** 0.4 *** 0.5 ** 3. Cohabitation 0.6 ** 0.4 ** 0.9 n.s. Educational level Lower secondary Ref. – Ref. – Ref. – Baccalauréat (high-school exit) 1.6 *** 1.9 *** 1.3 n.s. Higher education 1.9 *** 1.6 *** 2.5 *** Children in the household No Ref. – Ref. – Ref. – Yes 0.8 ** 0.7 ** 0.9 n.s. Time since union formation < 2 years 1.1 n.s. 1 n.s. 1.2 n.s. > 2 years to < 5 years Ref. – Ref. – Ref. – > 5 years to < 8 years 0.9 n.s. 0.9 n.s. 0.8 n.s.

Logistic regression of the fact of agreeing with the idea that it is possible to have sexual intercourse without emotional attachment

Interpretation: See Table 4.
Significance levels: *** p<0.01; ** p<0.05; * p<0.10, n.s.: non-significant.
Reference population: Respondents aged 18-39 who formed a couple between 1999 and the date of the CSF survey (different-sex couples).
Source: CSF Survey (INSERM-INED, 2006).

25While the PACS group appears to stand out, we nevertheless observe that these representations of sexuality do not reflect any real transgression of the norm of fidelity. The percentage of individuals who place themselves in the category “possibility of a few affairs” (the question on fidelity) is extremely small, with no significant difference between the groups (2 %, 3 % and 2 % respectively). Nor do persons in the PACS group distinguish themselves from the other groups by mentioning more frequently that they have had two affairs concurrently, but they do stand out in another way, in that the responses of both men and women converge more than in the other groups.

26When we look at the association between sexuality and emotional attachment through actual practice, the fact of having already had “sex with someone who was not important” may be a complementary indicator of a specific attitude to sexuality. As an extension to the question on the possibility of separating sexuality and emotional attachment, we might expect this experience to be more frequent among the PACS group, but in fact the only difference is with the married group (Table 3). Compared with the third group, which has a greater diversity of individual attitudes to ways of institutionalizing couples, the trajectories of PACS respondents are significantly different with regard to these practices, when comparing either the group as a whole or by sex. These results are consistent with the observations relating to homosexuality, with greater acceptance not necessarily signifying different sexual behaviours. PACS respondents often express liberal attitudes in the sense that their opinions and representations break away more from the dominant normative order, even though their practices do not follow suit. While the differences are large with respect to the married group, this does not reflect the diversity of uses of marriage, since the group is largely composed of people who chose marriage early on in their couple trajectories, often in line with their specific values, and notably their religious beliefs (Régnier-Loilier and Prioux, 2009). Nor can we speak of a specificity with respect to free unions, understood as a deliberate choice to reject all forms of conjugal institutionalization. The “cohabiting” group is not exclusively composed of advocates of free unions, but also of people who intend to register a PACS or get married in the future. [6]

Valuing alternative practices?

27The third component of the normative system in sexuality is the importance attributed to vaginal penetration in sex. This has been challenged for a number of decades by the increase in oral practices and the value placed on mutual touching, as observed by the ACSF team in the 1990s (Spira and Bajos, 1993), and later confirmed by the CSF survey (Bozon, 2008; Andro and Bajos, 2008). This allows us to approach the diversification in the sexual repertoire from two main angles: the attitude towards and appreciation of non-penetrative practices on the one hand, and actual practice on the other. The slight difference in representations among the PACS group compared with the dominant standard is also visible in the indicators concerning non-penetrative sex (Table 3, register C). The PACS group differs from the others by more frequently reporting that they prefer “mutual touching” (42 % versus 31 % and 29 %) and that sex without penetration is not frustrating “for either partner” (55 % versus 40 % and 42 %). If we look at the responses by sex, we see that men in PACS unions stand out most clearly from the others. A regression on the question of preferences merely highlights an effect of the respondent’s sex (Table 6).

Table 6

Logistic regression on reporting a preference for mutual touching (odds ratio)

Table 6
Overall Men Women Sex Male Ref. – – – Female 1.8 *** Age group 18-24 1.0 n.s. 0.6 n.s. 1.4 n.s. 25-29 Ref. – Ref. – Ref. – 30-34 1.2 n.s. 1.1 n.s. 1.1 n.s. 35-39 1.2 n.s. 1 n.s. 1.5 ** Union group 1. PACS Ref. – Ref. – Ref. – 2. Marriage 0.7 n.s. 0.5 * 1.0 n.s. 3. Cohabitation 0.7 n.s. 0.4 *** 1.1 n.s. Educational level Lower secondary Ref. – Ref. – Ref. – Baccalauréat (high-school exit) 1.0 n.s. 0.7 n.s. 1.4 * Higher education 0.9 n.s 0.6 *** 1.3 n.s. Children in the household No Ref. – Ref. – Ref. – Yes 1 n.s. 1.0 n.s. 1.0 n.s. Time since union formation < 2 years 1.0 n.s. 1 n.s. 1.0 n.s. > 2 years to < 5 years Ref. – Ref. – Ref. – > 5 years to < 8 years 1.3 ** 1 n.s. 1.6 ***

Logistic regression on reporting a preference for mutual touching (odds ratio)

Interpretation: See Table 4.
Significance levels: *** p<0.01; ** p<0.05; * p<0.10, n.s.: non-significant.
Reference population: Respondents aged 18-39 who formed a couple between 1999 and the date of the CSF survey (different-sex couples).
Source: CSF Survey (INSERM-INED, 2006).

28Women are more likely than men to report a preference for “mutual touching”, the gender effect being statistically highly significant. But regressions by sex provide additional insight: among men, the fact of belonging to any group other than the PACS group significantly reduces this probability. No such effect is observed for the women.

29Shifting from representations to reported practices, we built two composite indicators to see where these alternative forms of sexuality fit into the respondents’ repertoire. The question about, “regular mutual oral sex during the past 12 months”, distinguishes configurations in which respondents had engaged in both fellatio and cunnilingus as two components of the sexual act, during the past year. The second corresponds to the fact of reporting having had non-penetrative sex “often or sometimes” over the past 12 months.

30The observable differences (Table 3, register C) were not significant for these two questions on practices, even though in both cases the PACS respondents differ from the married ones. However, when we looked in detail at the responses for the second indicator, we found that PACS respondents have the same significant specific characteristics as observed earlier. Far fewer of them report never having had non-penetrative sex (27 % versus 45 % for the married group and 39 % for the cohabitants). A regression performed on this response (not presented) using the same variables as in Table 6 confirmed that being married considerably increases the probability of answering “never”. Regression by sex showed that married men differ from the other groups, whereas the women do not.

31The most striking difference, therefore, is the gap between men and women, mainly visible in the variables relating to attitudes. What should we deduce from this observation? In an earlier research project on household chores and perceptions of gender relations, based on the French version of the Generations and Gender survey (Étude des relations familiales et intergénérationelles, ERFI), it was found that men in a PACS union or considering registering a PACS, stand out from other men by their rejection of any ideology promoting distinct sexual roles (Rault and Letrait, 2009a). Their representations vary far more than those of women, and contrast with those of men who rapidly formalized their union through marriage. However, this specificity is less marked for actual practice in relation to household chores. Based on these representations of sexuality, a similar observation can be made. This characteristic of men in a PACS union may reflect a distinct way of perceiving sexual roles, and a less “virile” masculinity, in the sense that it is less strongly anchored in systematic opposition to the representations and practices associated with femininity. In terms of actual practice, however, men in the PACS group stand out far less.

A lesser attachment to the principle of “differences between the sexes” and the persistence of a differentialist view of sexuality

32The survey enabled us to examine this point in greater detail and to identify whether or not PACS partners do indeed express specific representations in this domain (Table 7). In fact, the representations of PACS partners all show a lesser attachment to the principle of a “difference between the sexes” (Table 8). This is reflected in the question about whether it is “a very good thing” to educate boys and girls in the same way: 47 % of men in the PACS group (50 % of women) report agreeing with this idea, compared with 32 % (41 %) and 29 % (45 %) in the other groups. The most striking result lies in the difference by sex.

Table 7

Questions used to study perceptions of gender relations

Table 7
Subject Possible responses and recoding (dichotomization of modalities in bold versus modalities in italics) Representations of gender relations Today, boys and girls are increasingly being raised in the same way. Do you believe that… It’s a very good thing It’s good but shouldn’t be taken too far It’s a mistake Don’t know Overall, with regard to gender equality, would you say that… Equality has been practically achieved There’s still some progress to be made There’s still a lot of progress to be made Don’t know Would you say that feminism… Has enhanced relations between men and women Has made them more difficult Hasn’t changed them at all I don’t know what feminism is No opinion Don’t know Some people believe that there are jobs that can only be done well by men and others only by women. Do you agree with this point of view? Strongly agree Agree Disagree Strongly disagree Don’t know

Questions used to study perceptions of gender relations

Note: In all these cases, non-responses were grouped together with the responses in italics (including “Don’t know”).
Source: CSF Survey (INSERM-INED, 2006).
Table 8

Representations of gender relations by type of union (%)

Table 8
Group PACS Marriage Cohabitation Overall Numbers 97 708 p(a) 1063 p 1868 “It’s a very good thing” to bring up boys and girls in the same way Overall M F 48 47 50 37 32 41 ** * n.s. 37 29 45 ** ** n.s. 37 31 43 “There’s still a lot to do” with regard to gender equality Overall M F 86 83 89 67 65 68 *** *** *** 70 70 70 *** ** *** 70 69 71 Feminism has enhanced relations between men and women Overall M F 63 58 67 47 47 48 *** n.s. *** 48 50 46 *** n.s. *** 48 49 48 There are jobs that can only be done well by men and others only by women “Strongly disagree” Overall M F 33 30 35 28 28 27 n.s. n.s. n.s. 26 22 29 n.s. n.s. n.s. 27 25 29 Interpretation: 48 % of PACS respondents reported agreeing with the statement “it’s a very good thing to bring up girls and boys in the same way”, compared with 37 % of married respondents. (a) The results are tested for the significance (p) of the difference between the compared percentages (PACS versus the two other groups).

Representations of gender relations by type of union (%)

Significance levels: *** p<0.01; ** p<0.05; * p<0.10, n.s. non-significant.
Reference population: Respondents aged 18-39 who formed a couple between 1999 and the date of the CSF survey (different-sex couples).
Source: CSF Survey (INSERM-INED, 2006)

33Men in PACS unions stand out clearly from the other two groups, which corroborates the above hypothesis. The difference is less marked for women. Regressions by sex on this question (Table 9) clearly confirm these observations, since the link with the type of union only appears for men. Being in a PACS union, massively increases the probability of approving the principle of non gender-based education. For women, this attitude is affected by the fact of being highly educated.

Table 9

Logistic regression on the fact of considering that it’s a very good thing to raise boys and girls in the same way (odds ratio)

Table 9
Overall Men Women Sex Male Ref. – – – Female 1.8 *** Age group 18-24 1.2 n.s. 1.3 n.s. 1.2 n.s. 25-29 Ref. – Ref. – Ref. – 30-34 1.0 n.s. 1.2 n.s. 0.9 n.s. 35-39 1.3 * 1.6 ** 1.2 n.s. Union group 1. PACS Ref. – Ref. – Ref. – 2. Marriage 0.6 ** 0.4 ** 0.8 n.s. 3. Cohabitation 0.6 ** 0.4 *** 0.9 n.s. Educational level Lower secondary Ref. – Ref. – Ref. – Baccalauréat (high-school exit) 1.1 n.s. 0.9 n.s. 1.3 n.s. Higher education 1.4 *** 1.2 n.s 1.6 *** Children in the household No Ref. – Ref. – Ref. – Yes 1.0 n.s. 1.0 n.s. 1.0 n.s. Time since union formation < 2 years 0.8 n.s. 0.8 n.s. 0.9 n.s. > 2 years to < 5 years Ref. – Ref. – Ref. – > 5 years to < 8 years 1.1 n.s 1.0 n.s. 1.6 n.s.

Logistic regression on the fact of considering that it’s a very good thing to raise boys and girls in the same way (odds ratio)

Interpretation: See Table 4.
Significance levels: *** p<0.01; ** p<0.05; * p<0.10, n.s.: non-significant.
Reference population: Respondents aged 18-39 who formed a couple between 1999 and the date of the CSF survey (different-sex couples).
Source: CSF Survey (INSERM-INED, 2006).

34The characteristics of the groups studied (Table 2) and the regressions suggest that the PACS appeals more to women with a high level of education and to men who, all other things being equal, show a greater sensitivity to egalitarian principles in their representations. However, it does appear that this relative specificity of PACS partners, characterized by their attachment to practices based on reciprocity, is not associated with a more symmetrical relationship of men and women to sexuality. Compared with other groups, we observe that men and women are characterized by a broader range of attitudes to sexuality, but also that the gap between the two sexes is also moving. The persistence of this gap goes hand in hand with a representation of sexuality that remains couched in terms of physiological differences between men and women. While slightly more PACS partners agree with the idea that “boys begin their sex lives earlier because of their education”, and disagree with the statement that “by nature, men have greater sexual needs than women”, the differences between the three groups are not significant. A difference between men and women is observed, however: the women in the PACS group are significantly more hostile to the idea that “by nature men have greater sexual needs than women”, than the women in the other groups (47 % versus 25 % and 29 %).

III – Smaller disparities between persons in a PACS union and long-term cohabitants

35To complement an approach based on recent unions, which only examined one segment of the total population of persons in PACS unions, we shifted the focus to older unions. This enabled us to examine the types of PACS union from a different angle. The hypothesis of a specific attitude to sexuality of persons in PACS unions was tested by comparing them with couples in long-term cohabiting relationships. [7] All the individuals in this part had been in a couple for at least four years at the time of the survey. They were all in the same position with respect to marriage, which had been held off, at least during the first years of the partnership (more than 4 years) but still remained a possible choice.

36This restriction enabled us to study individuals whose behaviour denotes a certain distancing from matrimonial institutionalization, at least in its most typical form where marriage takes place at an early stage in a couple’s trajectory. We can thus limit the heterogeneity of a “cohabiting” group which does not have this time limitation, as was the case for the group studied in the previous section and which, by construction, conceals a wide diversity of individual relationships to types of union.

Characteristics of the study population

37The two groups studied are quite different (Table 10). We found a social distribution specific to PACS partners, whose educational levels are higher and who are more often in higher-level and intermediate occupations. The disparities also vary by sex, with far more PACS women being highly educated and in higher-level occupations. From this viewpoint, the PACS appears to be the framework for a different arrangement between the two sexes, in the sense that the women have more resources relative to the other configurations. Besides these disparities, already visible in the study of “recent unions”, there are other characteristics more specific to the groups in this section because of the construction criteria used. Individuals are older (especially those who are cohabiting) and the unions began earlier. In this respect, the comparison of these two groups is complementary to that of the previous section. [8]

Table 10

Non-married individuals who formed a couple more than four years before the survey. Characteristics of the study population (% non-weighted)

Table 10
Group 1 (PACS) Group 2 (Cohabitation) Number 83 967 Sex (%) Men 57 43 Women 43 57 Age group (%) 18-24 3 5 25-29 26 22 30-34 48 37 35-39 23 36 At least one child in the household (%) 63 72 Occupational category (%) Farmers 0 3 Self-employed: business, trade and crafts 4 3 Higher-level occupations 24 14 Intermediate occupations 40 28 Clerical and sales workers 17 26 Manual workers 15 24 Economically inactive 0 2 Educational level (%) Lower secondary 19 38 Baccalauréat (high-school exit examination) 17 20 Higher education 64 42 Monthly household income (%) > €2,500 63 39 Time since union formation (%) > 4 years to < 10 years 74 63 10+ years 26 37 Had already lived as a couple (%) 21 20

Non-married individuals who formed a couple more than four years before the survey. Characteristics of the study population (% non-weighted)

Reference population: PACS or cohabiting respondents aged 18-39 (different-sex couples).
Source: CSF Survey (INSERM-INED, 2006).

A less distinctive attitude to sexuality

38All the registers reveal a PACS specificity (Table 11), but this is less clear than in the previous group. PACS partners and long-term cohabitants are relatively similar. Attitudes relating to perceptions of homosexuality and “alternative” practices reflect the transformations in sexuality. Likewise, these attitudes are associated with different representations of gender relations and gender equality. PACS partners express greater sensitivity to this question, together with a stronger attachment to the principle of equality and less male-female differentiation (see the question on raising children).

Table 11

Attitudes of PACS partners and cohabitants in a couple for more than four years with regard to transformations in sexuality

Table 11
PACS Cohabitation p(a) Register Numbers 83 967 A. Attitudes and represen- tations relating to homo- sexuality Homosexuality “is a form of sexuality like any other” 77 66 ** “No problem” in accepting a child’s homosexuality 71 56 *** A child may be raised by two women 65 58 n.s. A child may be raised by two men 58 51 n.s. B. Link between conjugality and sexuality Agrees with the statement that “fidelity is for life” 38 43 n.s. It is possible to have sex with someone one doesn’t love 47 39 n.s. Has already had casual sex with someone 32 33 n.s. Has already had two sexual relationships at the same time 17 24 n.s. C. Attitudes and practices relating to penetrative sex Prefers mutual touching 46 35 * Non-penetrative sex is not frustrating “for either party” 57 45 ** Had reciprocal oral sex over the past 12 months (often or occasionally) 37 31 n.s. Had non-penetrative sex over the past 12 months (often or occasionally) 40 32 n.s. D. Represen- tations of gender relations “It’s a very good thing” to raise boys and girls in the same way 52 40 ** “There’s still a lot of progress to be made” about gender equality 80 72 * Feminism has enhanced relations between men and women 59 50 n.s. Strongly disagrees that there are some jobs that can only be done well by men and others only by women. 31 29 n.s. (a) The results are tested for the significance (p) of the difference between the compared percentages (PACS versus cohabitation).

Attitudes of PACS partners and cohabitants in a couple for more than four years with regard to transformations in sexuality

Significance levels: *** p<0.01; ** p<0.05; * p<0.10, n.s. non-significant.
Reference population: Respondents in different-sex couples, aged 18-39, not married and reporting being in a couple for at least four years.
Source: CSF Survey (INSERM-INED, 2006).

39Here, even more than in the earlier section, the idea of specificity needs to be qualified, however. Persons in PACS unions always stand out primarily by their representations rather than their practices, but some of the representations studied show no difference between PACS and non-PACS respondents. That is the case for the perception of links between conjugality and sexuality, where there is no significant difference between the two groups.

40Given the small numbers studied and the resulting effect on the significance of results, and to avoid any redundancy with respect to the previous section, we decided not to show the responses by sex. However, the results we present do conceal large variations by sex, as already observed in the previous section, which would combine to increase the variations across groups. For instance, the women in the PACS group stand out with respect to their acceptance of homosexuality, in contrast to the attitudes of men, including men in a PACS union, and female cohabitants. Men in a PACS union stand out through representations of sexuality that are more readily dissociated from conjugality than is the case for the cohabiting men and the women in general. The overall percentage of 47 % of PACS respondents who reported that it was possible to have sex without emotional attachment conceals a great deal of asymmetry, since this attitude concerns 65 % of men and 29 % of women (50 % and 28 % respectively in the cohabiting group). In the two other registers studied, which concern valuing alternative sexual practices and gender relations, we found a similar specificity in PACS men. Once again, reporting a preference for “mutual touching” depends on whether the men are in a PACS union (47 %) or not (28 %), whereas there is no clear difference between the women in the two groups. Similarly, with regard to perceptions of feminism, which are far clearer here than in the recent unions, 71 % of men in PACS unions agree with the idea that feminism has enhanced gender relations, compared with 47 % of cohabiting men. This is the case for 47 % and 44 %, respectively, of women in the two groups.

41Logistic regressions performed on the groups studied, confirmed that there is less difference between the non-married individuals who registered their partnership in a PACS and those who stayed outside any legal framework over the past years, than between PACS partners and married one in more recently-formed unions (Table 12). The differences in the indicators used for registers A and B were no longer significant, [9] but they did remain so for the two other registers. The educational level and sex of the respondents systematically influences the responses.

Table 12

Regressions on non-married individuals in a couple for more than four years (odds ratio)

Table 12
Considers homosexuality as a “form of sex like any other” Agrees with the idea that one can have sex without emotional attachment Reports a preference for mutual touching Considers that it’s a good thing to raise girls and boys in the same way Sex Men Ref. – Ref. – Ref. – Ref. – Women 1.9 *** 0.4 *** 1.8 *** 1.9 *** Age group 18-24 0.9 n.s. 1.2 n.s. 1.2 n.s. 1.5 n.s. 25-29 Ref. – Ref. – Ref. – Ref. – 30-34 0.9 n.s. 1.4 * 1.2 n.s. 1.1 n.s. 35-39 1.0 n.s. 1.4 n.s. 1.5 ** 1.2 n.s. Union group 1. PACS Ref. – Ref. – Ref. – Ref. – 2. Cohabitation 0.7 n.s. 0.8 n.s. 0.6 ** 0.6 * Educational level Lower secondary Ref. – Ref. – Ref. – Ref. – Baccalauréat (high-school exit) 1.5 ** 1.7 ** 1.0 n.s. 1.7 ** Higher education 2.0 *** 2.3 *** 0.8 * 2.0 *** Children in the household No Ref. – Ref. – Ref. – Ref. – Yes 0.8 n.s. 0.7 ** 0.8 n.s. 0.9 n.s. Time since union formation > 4 years to < 10 years Ref. – Ref. – Ref. – Ref. – 10+ years 0.8 n.s. 0.9 n.s. 1.2 ** 1.5 **

Regressions on non-married individuals in a couple for more than four years (odds ratio)

Interpretation: See Table 4.
Significance levels: *** p<0.01; ** p<0.05; * p<0.10, n.s. non-significant.
Reference population: Respondents in different-sex couples, aged 18-39, not married and reporting being in a couple for at least four years.
Source: CSF Survey (INSERM-INED, 2006).

Conclusion

42Through their acceptance of homosexuality, their representation of the link between sexuality and conjugality and their attitudes to certain practices, persons in a PACS union value alternatives to the “classic” sexuality associated with the institution of marriage, which many of them consider to be a very prescriptive model (Rault, 2009), both because of its clear-cut gender roles, associated mainly with penetrative sex, and its differentiation between “good” and “bad” practices, which the PACS partners reject in principle – as witnessed in their attitudes to homosexuality and fidelity. It also appears that this rejection of the dominant norm sometimes goes together with a rejection of another injunctions: some PACS partners report being more comfortable with a system detached from any invitation to procreate (Rault, 2009). Questions on representations on this subject in the CSF survey provide a useful complement for analysing this attitude. Persons in a PACS union are more likely to agree with the idea that “one can succeed in life without children” than the others, and the gap between men and women is smaller. In the longer term, by studying the children of PACS partners, it will be possible to find out if this point of view reflects a truly different attitude towards procreation, or, as is more likely, simply a rejection of an injunction still associated with marriage by certain people.

43Choosing the PACS in the first years of its existence can therefore be understood as a combined demand for legal recognition of the couple relationship and for an institutional framework seen as more acceptable because it lacks the connotations associated with marriage. PACS partners are often aware that such connotations do not always find a legal expression today and mainly result from powerful representations associated with the cultural image of marriage. Free from any allusion to fidelity, heterosexuality and procreation, the PACS satisfies this demand as well as conferring rights upon the couple. The fact that a diversification of practices, and more importantly, of representations of sexuality, coincides with the choice of the PACS, is more than just a coincidence, it is part of the same “construction of self” (Bozon, 2001) whereby individuals seek to maintain a distance from perceived normative prescriptions that are incompatible with their own representations of self. Persons in PACS unions stand out both by their social characteristics (Tables 2 and 10) and their intimate orientations, which diverge from the dominant tendencies while not being radically different. The numerous models of intimate orientations in conjugal sexuality may thus exist in a variety if forms, depending on the couple’s institutional framework. As a form of union that diversifies the conjugal norm without challenging it, the PACS is echoed by an attitude to sexuality characterized more by an attachment in principle to diversity than by a radical re-composition of sexual practices. This correspondence between the choice of a new form of union – the PACS – and the attitudes of the contracting parties to sexuality, amply illustrates the ongoing “multiplication of normative legitimacies” (Commaille and Martin, 1998), which involves replacing a holistic organizational model of the private sphere characterized by a set of agreed conventions, notably in marriage, with a more individualistic conception in which individuals have greater freedom to define the terms of their private lives.

44For all that, this conclusion needs to be qualified. This specificity is not the exclusive domain of persons in PACS unions. It may also characterize advocates of free union who are hostile to marriage, or those who marry after a very long period of cohabitation. These hypotheses, as well as the differences observed between PACS and marriage partners in the first section, call for a new approach to the different types of union in sociological surveys, and a better understanding of the diversity of usages to which these unions lend themselves. That was not possible in the CSF survey because of the lack of data on the timing of unions. In this new approach, we must also take account of changes made to the law on the PACS. It has now been modified several times, notably on 1 January 2005 when the tax provisions for PACS couples were brought more closely into line with those of married couples, and in 2006 and 2007. The PACS has thus became more attractive, growing immensely popular in the second half of the 2000s. Its seasonality has also changed (Carrasco, 2007), with more frequent registrations in periods that maximize tax benefits. The social profiles of PACS partners, including their socio-occupational characteristics and their value systems, are likely to have changed too. No doubt the terms of the relationship between choice of a PACS and attitude to sexuality also.

45

46 Acknowledgements: My thanks to the colleagues and Population referees whose helpful remarks and suggestions enabled me to improve the initial version of this text.
I would also like to thank the CSF group for kindly making available the Context of Sexuality in France survey data. The conclusions of this article are my own, however, and not those of the CSF group.

Notes

  • [1]
    The PACS is closer to the Belgian form of legal cohabitation, the Dutch form of partnership and certain Spanish “regional partnerships” (Waaldjick, 2005; Pichardo-Gálan, 2004) adopted in the late 1990s and also available to both same-sex and heterosexual couples. However, the Belgian version is also available to non-conjugal couples. Moreover, the backdrop to these unions differs, in that these three countries have opened marriage to same-sex couples since those partnerships were adopted, whereas France preferred to revise the PACS several times to bring it more into line with marriage.
  • [2]
    Which does not mean that their trajectories do not involve several of the models described by Michel Bozon.
  • [3]
    The question on the legal situation of respondents was as follows: “Currently”, from a legal standpoint, are you: Married/In a PACS union/Single/Divorced/Separated or awaiting a divorce/Widowed/No reply?
  • [4]
    The average period of pre-marital cohabitation for unions formed after 1995, still ongoing and married in 2005, is just under 5 years (Rault, Letrait, 2009a). Here we consider that after 4 years, individuals in unmarried couples are expressing a certain distancing from marriage.
  • [5]
    And not only sociabilities. When we isolated the responses of persons who reported knowing no homosexuals in their social network, the PACS group were more likely to state that, “homosexuality is a form of sexuality like any other”.
  • [6]
    Similarly, the PACS group includes individuals who may later marry and the married group may comprise people who formed a PACS union before marriage. The survey does not enable us to make these distinctions.
  • [7]
    Persons who formed a union in this period and who married were excluded for two reasons relating to the absence of a marriage date in the questionnaire:
    – the absence of a marriage date made it impossible to distinguish two very different contexts, i.e. marriage before or after the existence of the PACS. It seemed wiser to not ignore this difference and to avoid unverifiable hypotheses about the similarity of behaviours in relation to marriage in these two distinct contexts.
    – such an approach would be against the spirit of this article, which seeks to compare the PACS with other individual attitudes to couple institutionalization. We stated in the introduction that one of the main transformations of the couple and the family is the diversification of usages of marriage. Without a wedding date, the introduction of a group of married individuals would involve grouping direct or quasi-direct marriages with those occurring late in a couple’s trajectory – without being able to identify each one’s share – and ultimately ignoring this diversity.
  • [8]
    Note that we used a different approach and not an extension of the preceding one. Some respondents were included in both analyses.
  • [9]
    However, a similar regression performed on the acceptance of a child’s homosexuality revealed a difference between the two groups.

Abstract

English

Abstract

Since its creation in France in 1999, the civil partnership (Pacte civil de solidarité, or PACS), which provides recognition for same-sex couples in terms other than those of marriage, has become increasingly popular among heterosexual couples. While contemporary transformations in types of union and in sexuality are often studied independently, this article breaks with this dissociation to challenge the hypothesis whereby individuals who register a PACS in France present attitudes and behaviours with respect to sexuality that reflect the normative diversification observed by the group “Context of Sexuality in France” (Contexte de la sexualité en France, CSF). Using data from the CSF survey of 12,364 individuals conducted by INED and INSERM in 2005-2006, we studied the relationship of individuals in PACS unions with the three main components of the dominant normative order in sexuality: heterosexuality, interlinking of conjugality and sexuality, and the value placed on “alternative” practices. PACS unions are compared with other types of union. We find that persons who opt for the new type of registered partnership have specific characteristics that relate more to liberal representations of sexuality than to actual differences in practice. Behind this specificity lies a considerable gap between the sexes, even though differences between men and women in PACS unions are sometimes relatively smaller.

Keywords

  • PACS
  • marriage
  • cohabitation
  • sexuality
  • heterosexuality
  • homosexuality
  • conjugality
Français

Les orientations intimes des premier.e.s pacsé.e.s

Résumé

Incarnant la reconnaissance du couple de même sexe dans des termes différents du mariage, le Pacte civil de solidarité (pacs) est, depuis sa création en 1999, de plus en plus souvent choisi par des couples de sexe différent. Si les transformations contemporaines des formes d’union et les changements dans le domaine de la sexualité sont souvent étudiés séparément, cet article rompt avec cette dissociation pour interroger l’hypothèse selon laquelle les contractant.e.s du pacs donnent à voir des attitudes et des comportements vis-à-vis de la sexualité qui font écho à la diversification normative constatée par le groupe Contexte de la sexualité en France (CSF). On étudie, à partir de l’enquête CSF réalisée par l’Inserm et l’Ined en 2005-2006 auprès de 12 364 personnes, le rapport des pacsé.e.s aux trois principales composantes du socle normatif dominant en matière de sexualité : l’hétérosexualité, les modalités de l’intrication de la conjugalité et de la sexualité et la valorisation de pratiques “alternatives”. Les pacsé.e.s sont comparés à d’autres profils d’union. En définitive, les usagers de la nouvelle forme d’union se caractérisent par des spécificités qui tiennent davantage à des représentations plus libérales de la sexualité qu’à des différences en termes de pratiques. Ces spécificités ne doivent pas dissimuler d’importants clivages par sexe, même si hommes et femmes sont parfois relativement plus proches dans le groupe des pacsé.e.s.

Español

Las orientaciones intimas de los primeros “pacsados”

Resumen

Aunque el Pacto civil de solidaridad (Pacs) creado en 1999 constituye, en términos diferentes al matrimonio, el reconocimiento de las parejas del mismo sexo, este tipo de unión es escogido cada vez más frecuentemente por parejas de sexo diferente. Las transformaciones contemporáneas de las formas de unión y los cambios en los comportamientos sexuales son estudiadas las más de las veces de forma separada. Este articulo, al contrario, plantea la hipótesis según la cual los contrayentes del Pacs manifiestan actitudes y comportamientos que corresponden a la diversificación normativa constatada por el grupo “Contexto de la sexualidad en Francia (CSF). A partir de la encuesta CSF realizada por el Inserm y el Ined en 2005-2006 sobre 12 364 personas, se estudian aquí las relaciones de los “pacsados” con los tres principales componentes del basamento normativo dominante en materia de sexualidad: la heterosexualidad, las modalidades de imbricación entre relación conyugal y sexualidad, y la valoración de prácticas alternativas. Los “pacsados” son comparados a otros tipos de unión. En definitiva, los primeros se caracterizan más bien por representaciones más liberales de la sexualidad que por diferencias en términos de prácticas. Esta particularidad se acompaña de diferencias importantes según el sexo, aunque a veces los hombres y las mujeres están relativamente más cercanos en el grupo de “pascados”.

Outline

  1. I - A study based on the CSF survey (INSERM-INED, 2006)
    1. PACS couples in the CSF survey
    2. A sociological approach: the advantages of the CSF
    3. Comparing the choice of PACS with preferences for other types of union
  3. II - People in PACS unions differ more in their representations of sexuality than in their practices
    1. Characteristics of the study population
    2. Acceptance in principle of homosexuality, and persistence of heterosexual exclusivity
    3. Rejection of prescribed fidelity, persistence of monogamy
    4. Valuing alternative practices?
    5. A lesser attachment to the principle of “differences between the sexes” and the persistence of a differentialist view of sexuality
  5. III - Smaller disparities between persons in a PACS union and long-term cohabitants
    1. Characteristics of the study population
    2. A less distinctive attitude to sexuality
  7. Conclusion

References

References

  • OnlineAndro A., Bajos N., 2008, “La sexualité sans pénétration : une réalité oubliée du répertoire sexuel”, in Bajos N., Bozon M. (eds.), Enquête sur la sexualité en France. Pratiques, genre et santé, Paris, La Découverte, p. 297-314.
  • OnlineBajos N., Beltzer N., 2008, “Les sexualités homo-bisexuelles : d’une acceptation de principe aux vulnérabilités sociales et préventives”, in Bajos N., Bozon M. (eds.), Enquête sur la sexualité en France. Pratiques, genre et santé, Paris, La Découverte, pp. 243-272.
  • Bajos N., Bozon M. (eds.), 2008, Enquête sur la sexualité en France. Pratiques, genre et santé, Paris, La Découverte, 601 p.
  • OnlineBajos N., Ferrand M., Andro A., 2008, “La sexualité à l’épreuve de l’égalité”, in Bajos N., Bozon M. (eds.), Enquête sur la sexualité en France. Pratiques, genre et santé, Paris, La Découverte, pp. 545-576.
  • OnlineBozon M., 2001, “Orientations intimes et constructions de soi : pluralité et divergences dans les expressions de la sexualité”, Sociétés contemporaines, 41-42, pp. 11-40.
  • Bozon M., 2004, “La nouvelle normativité des conduites sexuelles, ou comment mettre en cohérence les expériences intimes”, in Marquet J., Normes et conduites sexuelles : approches sociologiques et ouvertures pluridisciplinaires, Louvain-la-Neuve, Bruylant-Academia, 222 p.
  • OnlineBozon M., 2008, “Pratiques et rencontres sexuelles : un répertoire qui s’élargit”, in Bajos N., Bozon M. (eds.), Enquête sur la sexualité en France. Pratiques, genre et santé, Paris, La Découverte, pp. 273-296.
  • Bozon M., 2009, Sociologie de la sexualité, Paris, Armand Colin (1st edition 2002), 126 p.
  • Carrasco V., 2007, “Le pacte civil de solidarité : une forme d’union qui se banalise”, Infostat Justice, 97, 4 p.
  • Commaille J., Martin C., 1998, Les enjeux politiques de la famille, Paris, Bayard, 199 p.
  • Davie E., 2011, “Un million de pacsés début 2010”, Insee première, 1336, 4 p.
  • OnlineDéchaux J.-H., 2009, Sociologie de la famille, Paris, La Découverte (2nd edition updated), 126 p.
  • Marquet J., 2004, “Sexualité consentie, fidélité et performance”, in Marquet J., Normes et conduites sexuelles : approches sociologiques et ouvertures pluridisciplinaires, Louvain-la-Neuve, Bruylant-Academia, pp. 34-49.
  • Pichardo Galán, J.-I., 2004, “Same-sex couples in Spain: Historical, contextual and symbolic factors”, in Digoix M., Festy P. (eds.), Same-sex couples, same-sex partnerships and homosexual marriages. A focus on cross national differentials, Paris, Ined, Document de travail n° 124, pp. 159-174.
  • OnlinePrioux F., Mazuy M., 2009, “Recent demographic developments in France: Tenth anniversary of the PACS civil partnership, and over a million contracting parties”, Population, English Edition, 64(3), pp. 393-442.
  • OnlineRault W., 2009, L’invention du pacs. Pratiques et symboliques d’une nouvelle forme d’union, Paris, Presses de Sciences-Po, 275 p.
  • OnlineRault W., Letrait M., 2009a. “Diversité des formes d’union et ‘ordre sexué’”, in Régnier-Loilier A. (eds.), Portraits de famille. L’enquête Étude des relations familiales et intergénérationnelles, Paris, INED, Grandes enquêtes, pp. 59-85.
  • OnlineRault W., Letrait M., 2009b, “Choix du pacte civil de solidarité et rapport à la religion”, Politiques sociales et familiales, 96, pp. 41-54.
  • OnlineRault W., Letrait M., 2010, “Formes d’unions différentes, profils distincts ? Une comparaison des pacsé.e.s en couple de sexe différent et des marié.e.s”, Sociologie, 1(3), pp. 319-336.
  • OnlineRégnier-Loilier A., Prioux F., 2009, “Comportements familiaux et pratique religieuse en France”, in Régnier-Loilier A. (ed.), Portraits de famille. L’enquête Étude des relations familiales et intergénérationnelles, Paris, INED, Grandes enquêtes, pp. 397-423.
  • Singly F. de, 2008 (1st edition 1993), Sociologie de la famille contemporaine, Paris, Armand Colin, 128 p.
  • Spira A., Bajos N., 1993, Les comportements sexuels en France, rapport officiel au ministre de la Recherche et de l’espace, Paris, La Documentation française, 352 p.
  • OnlineToulemon L., 1997, “Cohabitation is here to stay”, Population, An English Selection, 9, pp. 11-46.
  • OnlineToulemon L., Razafindratsima N., 2008, “Plan de sondage et pondérations de l’enquête”, in Bajos N., Bozon M. (eds.), Enquête sur la sexualité en France. Pratiques, genre et santé, Paris, La Découverte,, pp. 45-60.
  • Waaldjik K. (ed.), 2005, More or Less Together: Levels of Legal Consequences of Marriage, Cohabitation and Registered Partnership for Different-Sex and Same-Sex Partners. A Comparative Study of Nine European countries, Paris, INED, Document de travail n° 125, 192 p.

Authors

Wilfried Rault
Institut national d’études démographiques, Paris.
Correspondence: Wilfried Rault, Institut national d’études démographiques, 133 boulevard Davout, 75980 Paris Cedex 20, tel.: 33 (0)1 56 06 20 02, e-mail: wilfried.rault@ined.fr
CSF group
The CSF group (Context of Sexuality in France) comprises Nathalie Bajos and Michel Bozon (scientifi c leaders), Nathalie Beltzer (coordinator), Armelle Andro, Michèle Ferrand, Véronique Goulet, Anne Laporte, Henri Leridon, Charlotte Le Van, Sharman Levinson, Laurent Toulemon, Nicolas Razafi ndrastsima, Josiane Warszawski, as well as Lucette Aussel and Caroline Laborde.
Translated by
Krystyna Horko
Latest publication on cairn or another partner portal
Abstract Full-textfile_download Download
This is the latest publication of the author on cairn.
Uploaded on Cairn-int.info on 03/03/2014
Cite
Distribution électronique Cairn.info pour Ined Editions © Ined Editions. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
keyboard_arrow_up
Chargement
Loading... Please wait