Changing attitudes to sexual morality: a cross-national comparison.

How many sexual revolutions can there be, before a certain scepticism arises about whether the use of 'revolution' is apt? The so-called sexual revolution that occurred in the 1960s was by no means the first and doubtless will not be the last. Indeed, the term "sexual revolution" was coined at least forty years earlier to describe the changing sexual mores and behaviour of the 1920s and has been used to describe different periods throughout the intervening years (Martin 1996).(1) The term 'revolution' is charged with meaning and implies a purposive overthrow of traditional sexual morality. Yet, as Martin persuasively argues, the reasons that are evoked to explain the second sexual revolution of the 1960s are remarkably reminiscent of those used to explain the first. In short, they involve changes in economic imperatives, the emancipation of women due to labour force participation, new sexual knowledge and contraception, the emancipation of youth due to technological change and independence from adult authority. The various references change to meet the different period: it is women's entry into the workforce since the Second World War, rather than the First; it is the invention of the pill, not access to contraception; it is Woodstock, not the jazz age. Perhaps there has been one century-long revolution involving a constant set of causes (Seidman 1992:21). Or perhaps, like reports of the death of Mark Twain, rumours of sexual revolution have been grossly exaggerated (Smith 1990).

The claim that there was a 'sexual revolution' in attitudes and behaviour during the 1960s and early 1970s was certainly promoted by the media at the time and has become a common metaphor in public discourse. The way sexual values and attitudes towards intimacy have been freed from traditional constraint has also been a dominant theme of some influential strands of social theory. These emphasise that the weakening of rules governing sexual behaviour is intrinsic to life in post-traditional society, with the results that the individual is faced with a far greater range of acceptable choice (Thornton 1989; Weeks 1995; Giddens 1996; Beck 1992; Castells 1997). According to Beck, as modernisation proceeds, the decisions and constraints to decide multiply in all fields of social action, but especially with regards to sexuality and the family. He writes: 'With a bit of exaggeration one could say: "anything goes" ... Marriage can be subtracted from sexuality, and that in turn from parenthood; parenthood can be multiplied by divorce, and the whole thing can be divided by living together or apart' (Beck 1992:116). For Beck, new modernity is equated with the risk society and the risk society, in turn, is a catastrophic society.

With regards sexual behaviour, risk has taken on a new and deadly meaning with the spread of AIDS. It is not surprising then, if the 1980s saw a public backlash against the greater freedoms associated with the sexual revolution. Moreover, both in America and Britain the political agendas of the New Right were in vogue, under the respective regimes of President Reagan and Margaret Thatcher. Both promised to push back the wave of permissiveness. yet, according to Weeks, it is clear in both countries that the loosening of the bonds of sexual authoritarianism associated with the 1960s was continuing, even accelerating through to the 1990s, despite haphazard attempts at moral rearmament (Weeks 1995:28-9). He also notes that it is ironic that the two political leaders, who were both keen proponents of traditional families, presided over 'probably the greatest revolution in sexual mores in the twentieth century, despite their best endeavour'. His explanation for this apparent conundrum is that there is a link between radical individualism in economics and in sexual and ethical values: 'Individual freedom cannot stop at the market; if you have an absolute freedom to buy and sell, there seems no logic in blocking a freedom to choose your sexual partners, your sexual lifestyle, your identity or your fantasies' (Weeks 1995:29). This theme is echoed by Giddens: 'The present day conservative still want to conserve- to protect the traditional family, traditional symbols of state legitimacy, religion and the identity of the nation. Yet these are being eroded, smashed open even, by the very market forces modern conservatism fosters' (Giddens 1996:241). Thus, it is claimed, the market ethic for the economy is a powerful corrosive factor that undermines the traditional equation of sex with procreation and its containment in marriage that was fundamental to the ethical teaching of the Judaeo-Christian religions (Russell 1929).

In many countries, changes in attitudes towards sexual morality and decline in traditional religious authority have been linked (Wilson 1982; Thornton 1985; Greeley 1989; Ingelhart 1990; Braun et al. 1995; Scott et al. 1995; Hayes 1995). In traditional society, morality was a matter for the public domain and there was consensus about what was right and wrong, even if private morality sometimes failed to live up to austere expectations. Religion in the modern industrial world, according to Luckmann (1967) has become invisible, in the sense that sexuality and family, in particular, are now deemed as being located in the essentially private sphere. Religion has also become more pluralistic and potential adherents can choose between any number of brands of faith and accompanying ethical codes. Yet clearly the organised churches hold far more sway in some countries than others, and one major source of variation in cross-national attitudes is likely to be the extent to which the organised churches have maintained their hold as a primary source of moral authority. Similarly, religiosity is likely to be an important factor in within-country differences. In particular, the trend towards secularisation is most clearly manifest among the young. The importance of Christian upbringing to teaching moral values to the next generation is a reoccurring debate, even if the evidence suggests that, as far as church attendance goes, it is a losing battle.(2) Certainly, previous research shows clearly that young people are far more likely than their elders to reject traditional codes of sexual morality (Harding 1988; Thornton 1989).

It has been claimed that the chief 'beneficiaries' of the 1960s sexual revolution have been young people and women (Martin 1996). Rising affluence had given the young greater access to privacy and the pill had made pre-marital intercourse less risky. The combined effect was to give young people sufficient freedom to reject the 'old-fashioned' values of their parents and grand-parent's generations. In addition, women were gaining increased economic autonomy through their participation in the labour force and could demand an end to a double standard concerning sexual expression.(3) As Ira Reiss states: 'Economic autonomy reduces the dependence on others and makes sexual assertiveness a much less risky procedure' (Reiss 1990:88). Certainly, the feminist movement of the 1970s embraced the ideal of sexual equality. Moreover, contraceptive technologies and legal abortion made such sexual expression possible, without a result that would undermine the new-found independence of women (Scott 1998). These new attitudes towards sex allowed for the growth of casual relationships and non-marital cohabitation that often have no implication of life-long commitment. In so far as commitment is no longer necessary for the enjoyment of sexual intimacy, men have also been represented as 'beneficiaries' of the new sexual climate (Ehrenreich 1983).

Contradictory claims have appeared in the literature about whether or not the sexual revolution has continued through into the 1990s. According to Weeks (1995:29) the changes of the 1960s have continued and even accelerated. According to the American researchers Laumann et al. (1994), the revolution appeared to be over by 1984 and perhaps there may even have been a reversal. One possibility is that there is a cross-national difference, with trends towards more permissive attitudes continuing in Britain, but ceasing or reversing in the United States. Another possibility is that the sexual revolution is more nuanced and the trajectories and sources of change may vary depending on what aspect of sexual morality is under consideration (Smith 1990).

Thus one basic question motivating this research is: 'To what extent has there been a revolution in sexual attitudes and is the revolution over?' To answer this question, monitoring trends is not sufficient. It is also necessary to examine the underlying sources of attitudinal change. In this paper, I examine changes in men and women's attitudes to sexual behaviour across nations and time. First, using time-series data from the United States and Britain to track changing attitudes towards pre-marital sex, extra-marital sex and homosexual relations, I investigate whether changes in permissiveness are mainly due to the slow but usually fairly stable process of cohort replacement or the more 'revolutionary' process of period effects which cause attitudinal change among individuals of all ages. I also examine how the trajectory and pace of change differs in the two countries. Second, using data from the International Social Survey Programme, I compare British and American attitudes to sexual morality with those of four other nations with very different socio-political and religious traditions - Ireland, Germany, Sweden and Poland.

Attitudes are important indicators of people's latent tendencies to respond to the opportunities and constraints that are posed by the structural conditions of life. Yet politicians are not the only people who say one thing and do another. Certainly, when it comes to sexual morality, there is likely to be quite a gulf between attitude and practice and even when the two are related, it is not clear which causes which. The primary reason, however, why attitudes are important is not because they are indicators of behaviour, but rather because they help constitute the climate of opinion against which behaviour is judged. In this way shifts in public attitudes undoubtedly facilitate as well as reflect social change (see also Scott et al. 1996). Public opinion is important not just because it is an important mediating factor for the acceptability of different public and private behaviours, but also because it is an important factor that politicians have to consider when weighing policy and legislative decisions. Changes in laws regarding the age of heterosexual or homosexual consent, or whether guilt or fault is a relevant concept in divorce proceedings, are issues where public stance is likely to have some impact on private choices and vice versa.

Sexual norms, however, are not only subject to change over time, but they also vary greatly across different cultures. Cross-national research is essential if we are to determine what attitudinal changes are common throughout the industrial Western world and what is distinctive about each individual nation. This is particularly important in an age when worries about changing sexual behaviour and gender and sexual identities have become the explicit focus for debates about the current shape and desirable future of society (Weeks 1995:5). Sex is regulated in all societies and one measure of the strength of regulation is the extent to which social rules governing sexual conduct are internalised into public attitudes and opinions (Wellings et al. 1994). In this paper, by providing some insight into the relativity of sexual norms and also by providing evidence of the patterns and processes that underlie the changes in sexual attitudes, I hope to shed more light on the degree of moral consensus and conflict evident in this age of uncertainty. Before presenting the findings on changing attitudes in Britain and the United States, I discuss the existing literature on attitudinal change and I also describe the data that are used for the cross-national comparison of attitudes.

Attitudinal Change: Previous Research Findings

Much of the research on attitudes to pre-marital sex and extra-marital sex has been conducted against the backcloth of dramatic demographic change, that includes the extraordinary rise in pre-marital cohabitation, the rising proportion of children born outside of marriage especially to teenagers, and the dramatic increase in divorce. The rise in cohabitation began in the 1970s, both in the United States and Britain. In the United States, the proportion of people who cohabited before first marriage quadrupled from 11 per cent for marriages in 1965-74 to 44 per cent for marriages in 1980-84 (Bumpass and Sweet 1989). A similar picture is found in Britain with just 24 per cent who cohabited before marriages contracted in 1965-74 rising to 63 per cent for marriages in 1985-92 (Buck and Scott 1994). Cohabitation before first marriage is now the norm. The experience of pre-marital sexual activity is even more widespread and, according to the 1990 British National Survey of Sexual Attitudes and Lifestyles, fewer than 1 per cent of men and women aged 16-24 were married at the time of their first sexual intercourse (Wellings et al. 1994). A similar pattern is found in the United States and, according to the 1992 National Health and Life Style Survey, the majority of the cohort born since 1963 were no longer virgins by age 18, and half had first intercourse between the ages of 15 and 17 (Laumann et al. 1994).

Attitudinal and value shifts have been generally consistent with these behavioural trends. People struggle for consistency in their lives and are likely to bring their attitudes into line with their daily experience (Ingelhart 1990). In particular, those who have had personal experience of non-traditional family circumstances, such as cohabitation or divorce, are less likely to condemn the associated sexual behaviour. Both Americans and the British have become far more liberal about sex outside of marriage (Glenn and Weaver 1979; Harding 1988; Singh 1980; Thornton 1989; Smith 1990; Welling et al. 1994). In the United States, attitudes condemning pre-marital sex dramatically declined during the 1960s and early 1970s and then levelled off after the mid-1970s among young people, but the decline extended through to the mid-1980s for older people (Thornton 1989). By contrast, disapproval of extra-marital sex was strong and stable from 1970 to 1987 and then showed a conservative shift at the end of the decade (Saunders and Edwards 1984; Smith 1990). British research has also shown that extra-marital sex is regarded in an altogether more serious light than sex before marriage (Jowell and Park 1996), with the vast majority of all age-groups upholding fidelity.

Historically, in Western societies, homosexuality has been viewed as a sin, a disease, or an aberration (Laumann et al. 1994:284). These notions are still widespread. The American General Social Survey shows consistent condemnation of homosexuality over the 1970s and 1980s. Yet, in spite of the apparent stability in public opinion over a long period, the past twenty-five years have seen a notable increase in the legitimation and visibility of homosexuality, in part the result of the gaining political momentum of the gay movement. The gay liberation movement exploded with great energy in America in 1969 and reached Britain by the end of 1970. Although, in Britain, the 1967 Sexual Offences Act had decriminalised certain aspects of adult male sexual activity carried out in private, it did not legalise homosexuality as such (Weeks 1989:267). By the 1990s, the climate of opinion had changed quite substantially and the age of consent for homosexuality was lowered to eighteen.

however, that homosexuality came to the forefront of the political agenda both because of the AIDS epidemic and because of the political agendas of the New Right. In the United States, the new Christian Right harnessed fundamentalist fervour against what was seen as a rising tide of permissiveness. In Britain, the moral conservatism was more secular, but the political intentions of the Thatcher regime were very similar to the Reagan-Bush agenda.(4) The AIDS crisis provided the fear factor that gave the conservative cause an immense boost, although, by the 1990s, the fear seems to have somewhat dispelled. The AIDS crisis has clearly affected attitudes towards homosexuality, both in the United States and Britain, but it is less clear whether homosexual relations were singled out for censure, or whether there was a much broader conservative backlash against sexual permissiveness (Wellings and Wadsworth 1990).

Previous research indicates that women take a less censorious view of homosexuality than do men (Wellings and Wadsworth 1990). Possibly, this is because the hegemonic model of masculinity has traditionally been bound up with heterosexual prowess and dominance over women (Carrigan et al. 1987), whereas concepts of femininity have more to do with caring and emotion-work in inter-personal relationships (Duncombe and Marsden 1993). Cohort differences are likely to be especially marked in attitudes towards homosexuality, both because of increasing secularisation and also the general liberalisation in moral values (Weeks 1995:25). While pre-marital sex is now accepted as 'normal' behaviour, there is no such acceptance of sexual relations among same sex adults. Indeed, there is considerable dispute about whether homosexuality is a biological aberration (an essentialist view) or one particular manifestation among the range of sexual categories and behaviours that display historical and cultural variability over time (the social constructionist view). If such conceptual differences are influential in determining attitudes towards homosexuality, then education is likely to be an important force in encouraging attitudes that are more supportive of different sexual orientations. In contrast, religion, and especially Catholicism with its emphasis on the procreative purpose of sex, will tend to reinforce the more traditional stance that is censorious of homosexual practice.

Data and Methods

To examine attitude change in the United States and Britain I use data from national probability samples interviewed by the General Social Survey (GSS) of the National Opinion Research Centre (Davis and Smith 1994) and the British Social Attitudes surveys (BSA) of the Social and Community Planning Research Centre (SCPR 1992). The GSS is an annual independently drawn probability sample of English-speaking persons, 18 years or older, living in non-institutional accommodation within the continental United States and interviewed in person. Response rates for the GSS are generally about 75 per cent. The BSA is a nation-wide probability sample of adults (aged 18 and over) and response rates for the face-to-face surveys are generally about 70 per cent, with the lowest at 65 per cent in 1994.

An identical series of questions stretches back to 1965 in the United States and 1983 in Britain, which can be used to monitor social change.(5) This repeated cross-sectional data can also be used to disaggregate the relative importance of period or life-cycle effects and cohort replacement. Interpretations need to rely on prior theory, as there is no known way of empirically separating the influences of cohort, age and period effects with repeated cross-sectional data.(6)

In order to place the United States and Britain in a broader cross-national perspective, I draw on data from the 1994 International Social Survey Programme module on 'Family and Changing Sex Roles' that was appended to the national social surveys in the United States, Britain, Ireland, Germany, Poland and Sweden. For general and technical descriptions of the ISSP, see Davis and Jowell (1989) and Jowell et al. (1993). Although these additional four countries have been chosen, in part, for pragmatic reasons concerning data accessibility, they also constitute a useful set on which to test the hypothesis that religion and, in particular the authority of the Catholic Church plays a crucial role in maintaining traditional beliefs. Comparable measures of religious affiliation and church attendance are used to examine the importance of cross-national variation in religion for explaining both within country and across country differences in attitudes towards sexual morality. For comparative multivariate analysis, in addition to age and sex, the respondent's experience of cohabitation and divorce, as well as educational attainment, are introduced as control variables. Both education and partnership experience vary considerably between the different countries and each, for reasons outlined in the previous section, may affect attitudes towards sexual morality. For the six-country survey of attitudes it is not possible to infer anything about social change, as there are only cross-sectional data collected at one point in time.

Changes in Sexual Morality in the United States and Britain

The GSS has included the same three items about attitudes towards premarital sex, extra-marital sex and sexual relations between adults of the same sex, since 1972. Similar questions on pre-marital and extra-marital sex were first posed in a 1965 National Opinion Research Survey. This series, spanning some thirty years, provides a unique opportunity to examine changes over time (see Appendix for exact question wording). The public's level of disapproval has varied considerably depending on the sexual behaviour in question, with levels of condemnation consistently high for extra-marital sex, somewhat lower for sexual relations between adults of the same sex, and very substantially less for attitudes towards pre-marital sex.(7) By the 1980s, disapproval of pre-marital sex had steadied at just over 40 per cent for women and about 30 per cent for men. This gender disparity has been evident throughout the last thirty years, with women, on average, always more conservative than men. However, the general pattern of change is quite similar for both sexes, with a strong trend towards less restrictive attitudes to pre-marital sex in the late 1960s and early 1970s, with the trend levelling off by the end of the decade, and staying relatively constant through the 1980s and 1990s [ILLUSTRATION FOR FIGURES 1a and 1b OMITTED]).

For much of the last thirty years, women have also tended to be more disapproving than men of extra-marital sex.(8) Interestingly, however, condemnation of extra-marital sex among men has increased, since the mid-1980s, leading to a convergence of attitudes among men and women. The heightened concern among men, coincides with the emergence of concern about the risk of the spread of AIDS among the heterosexual population. In marked contrast, attitudes condemning sexual relations between adults of the same sex have fallen sharply, since 1990, both among men and women.

These repeated cross-sectional survey data can be used to determine to what extent cohorts possess different attitudes and whether cohorts retain [TABULAR DATA FOR TABLE 1 OMITTED] those attitudes as they grow older. The analysis is done only for attitudes to pre-marital sex and homosexuality as attitudes to extra-marital sex have changed so little. The analysis presented in Table 1 was carried out as follows. Using four time points which are ten years apart - 1965, 1974/5, 1984/5 and 1994 - we constructed six ten-year groupings: 18-27, 28-37, 38-47, 48-57, 58-67 and 68 and over.(9) The figures for each age-group at the different points in time can be read across the rows of the table. Thus, for example, among the youngest age-group of men who were 18-27 in the 1965 survey 68 per cent disapproved of pre-marital sex. The level of disapproval fell sharply for this age-group to 15.5 per cent in the 1970s and then rose slightly in the 1980s to 19.8 per cent before dropping back to 15.4 per cent in 1994. The columns ([[Delta].sup.1]) show the results of comparing the same age-groups across the periods. For the youngest age-group there was a drop of 52.5 per cent between 1965 and 1975. Looking down the column it can be seen that, in this earliest time period, changes were far the most pronounced for the younger age-groups. The same is true for women although all but the oldest age-group show considerable attitudinal change. While in 1965 there was no age difference among men, there was a pronounced age difference among women, with older women being far more conservative. However, from 1972 onwards there are highly significant age effects for both men and women, although the generational gap is smaller in the 1980s and 1990s than it was in the earlier years.

In addition, it is possible to compare a given cohort category with itself across the four surveys. Thus the time-2 figure (1974-75) figure for the 28-37 age group, minus the time-1 figure (1965) for the 18-27 age-group, represents the intra-cohort change. In Table 1 the percentage in this group of men who disapprove of pre-marital sex has moved from 68 per cent to 29.8 per cent across time and thus the intra-cohort shift ([[Delta].sup.2]) is -38.2 per cent. In the 1960s to 1970s all the more recent cohorts became more tolerant of premarital sex.

The very substantial intra-cohort shifts of the earliest period towards a more liberal stance can be interpreted as mainly due to period factors rather than life-cycle or ageing effects for at least two reasons. First, as I have already argued, there is clear evidence that younger people stand to gain more than older people from the loosening of traditional constraints. Second, as people grow older, they have little incentive to change the basic conceptions with which they have learned to assess the propriety of situations (Ryder 1965). After the mid-1970s, attitudes among earlier cohorts of men continued to become less restrictive (there is no such clear-cut pattern among women), whereas more recent cohorts of men, if anything, moved in a more conservative direction. However, the attitudes of both men and women, who were in the 58-67 age-group in 1985, became substantially more conservative a decade later. Perhaps, this indicates that, in the eyes of the older generation, the changes in permissiveness were thought to be going too far.

[TABULAR DATA FOR TABLE 2 OMITTED]

These data can also be used to infer whether aggregate social change with respect to attitudes to pre-marital sex is due more to changes within cohorts or changes across cohorts. If there was no change in attitudes within cohorts, that is if any overall attitudinal change occurred entirely because later cohorts differed in attitudes from earlier ones, then the mean change shown in A2 would be zero. We can see from Table 1 that within cohort change plays a large role in changing attitudes towards pre-marital sex, in the earliest period. In that sense the change was 'revolutionary', in that it is indicative of a strong period effect. However, since that time, nearly all the aggregate level change is due to the much slower process of cohort replacement, although the reversal of stance that can be seen in the most recent period may mean that the acceptability of pre-marital sex has reached its limit in the United States.

The picture is rather different for attitudes concerning sexual relations between adults of the same sex. This time it is women, particularly the youngest women, who take a far more liberal stance than men. The youngest age-group of both men and women are more conservative in the mid-1980s than they were in the mid-1970s, presumably reflecting the AIDS scare, but this age-group has subsequently returned to a more liberal position. The stability of attitudes at the aggregate level between 1970 and 1980 conceals fluctuation within cohorts, especially among women. Since the 1980s a quite substantial proportion of the net increase in tolerance among women is due to change going on within the cohorts, as well as the change due to the process of cohort succession. At all years and for both men and women, earlier cohorts are less tolerant than more recent cohorts, although this generational difference is especially pronounced among women.

The BSA has posed a very similar set of items to those of the GSS since 1983, although in Britain the response categories changed in 1991 and 1994, becoming identical to those of the GSS. The exact wording of questions can be seen in the Appendix. As Figures 2a and 2b show, as in the United States, there is a clear differentiation of disapproval for the different sexual behaviours. Because of the different context and slight differences in wording, it is not possible to make any direct comparison of the absolute level of support of sexual morality in Britain and the United States. What can be compared, however, is the relative change among men and women within the two countries over the last ten years. Attitudes towards extra-marital sex are, like in the United States, by far the most censorious and disapproval has remained relatively constant over the decade. The same gender differences can be observed, with women less likely to condemn homosexual relations but more likely to be opposed to pre-marital sex than men. However, in Britain, there is a far larger gap between attitudes towards extra-marital sex and homosexuality than in the United States, especially among women. In Britain, disapproval of sexual relations between adults of the same sex, increased steadily, among both men and women, through to 1987 and then declined. For men, the change has been somewhat uneven, but, for women, intolerance fell sharply [TABULAR DATA FOR TABLE 3 OMITTED] resulting in a marked gender difference in the mid-1990s. In contrast, there has been a gender convergence in attitudes towards pre-marital sex, with men fluctuating around a 20 per cent disapproval mark, while women's disapproval fell off steadily, to reach a similar level to that of men, by 1994.

The analysis in Table 3 shows British attitudinal change broken down by cohort and gender, for both attitudes towards pre-marital sex and homosexual relations (extra-marital sex is again excluded as change has been negligible). As in Table 1, figures for each age-group at the two points in time can be read from the rows of the table. The columns ([[Delta].sup.1]) show the results of comparing the same age group in 1984 and 1994. For pre-marital sex, earlier cohorts of women have shown the greatest changes towards a more liberal position, whereas, for homosexuality, the reverse is the case and the youngest age-groups of women have changed most. By looking at the mean change associated with column ([[Delta].sup.1]) we can see that women's attitudes towards both pre-marital sex and homosexuality have changed twice as much as men. The mean change is not weighted to take account of the different numbers within each age-group. Otherwise, it would be the exact equivalent of the percentage shift across the relevant years shown in Figures 2a and 2b.

As in the earlier tables, the change in disapproval over time for a particular cohort can be read going down the diagonal of the columns of the table. As in the net change for age-groups over time, it is the earlier cohorts of women who have changed most in terms of becoming more liberal towards premarital sex; whereas for homosexuality, it is the more recent cohorts who have changed most. Nevertheless, there is little evidence of any revolutionary change over the decade, and most of the change in attitudes towards sexual morality is due to the slow process of cohort replacement. Thus, in both Britain and the United States there has been a marked reduction in disapproval of homosexuality over the past decade. However, in marked contrast to the United States, the British trend towards greater tolerance of pre-marital sex is continuing, especially among women.

British and American Attitudes: A Cross-national Perspective

In order to place British and American attitudes in a broader cross-national perspective we compare attitudes across six nations that have very different religious backgrounds. It has been suggested that in recent years, individuals increasingly have interpreted their religious commitments and beliefs in individualistic terms and less in terms of institutional loyalty and obligation (Thornton 1985). However, the extent to which this is true is likely to differ between countries. In predominantly Catholic countries, with high levels of church attendance, institutional resistance to the belief that sexual morality is a matter for individual decision is likely to be more successful than in Protestant societies where doctrine and culture combine to favour a more individualistic and permissive morality. However, the divide may be far less evident within national contexts. In American society, for example, Catholics and Protestants have demonstrated considerable convergence over the past decades. This convergence may reflect both the assimilative processes of educational expansion and the tendency of more recent Catholics to reject Papal rulings on sex and reproduction (Alba 1981; Greeley 1990). Indeed, in the American context, differences between fundamentalist and other Protestant denominations are likely to be more pertinent than divisions between Protestants and Catholics. However, the question that motivates this broader cross-national comparison is a more general one: 'To what extent do religious differences account for the cross-national variations in attitudes?'

To answer this question we use data from the 1994 International Social Survey programme from the United States, Britain, Ireland, Sweden, Germany and Poland. Details of the sample size, mode of survey and response rates are shown in Table 4.(10) These six nations differ not only in their religious backgrounds, but also in the related area of family practice, as the comparison of cohabitation and divorce experience shows.

In the Republic of Ireland the culture as well as individual religious practice reflects the dominance of the Catholic Church. Given the dominant influence that the Church exerts over family policy in Ireland, it would be surprising if Irish attitudes were not among the most traditional. Poland makes an interesting case for comparison because, like Ireland, the Catholic Church's influence is very strong. Yet, unlike Ireland, the Polish Catholic Church was in the recent past reduced to political opposition by the secular communist regime. In the 1990s, religion has enjoyed a resurgence of influence and, especially with a Polish Pope, the Catholic church wields considerable political power. Not surprisingly, as Table 4 shows, Poland is a relatively traditional country with regards to family practice.

In Germany, national differences in family policy have proved among the more contentious issues that have had to be resolved as part of reunification. In East Germany, state socialism has promoted gender egalitarianism in the workforce, while leaving the traditional gender roles in the private sphere of the family relatively unchanged. In West Germany, post-war family policy tended to be very patriarchal, although recent repeals have left the way open for women to embrace alternative lifestyles, with an increasing number opting out of marriage and the traditional wife/mother role.(11) Yet the German divide is not as straightforward as policy differences might indicate. The two situations have been characterised as public progressivism and private traditionalism in the East; and public traditionalism and private radicalism in the West (Chamberlayne 1994). Evidence for this can be seen in the fact that, despite the very different religious profiles of the two parts of Germany, the incidence of cohabitation and divorce is quite similar.

Sweden is clearly one of the most liberal of sexual regimes in Europe. The [TABULAR DATA FOR TABLE 4 OMITTED] practice and legal sanction of cohabitation far exceeds that of any non-Scandinavian country. Legislation on heterosexual cohabitees has also been extended to homosexual couples to ensure they have the same legal protection with regard to property (Hantrais and Letablier 1996). Sweden, despite the [TABULAR DATA FOR TABLE 5 OMITTED] widespread affiliation to the Lutheran Church, is an essentially secular society and the church has relatively little scope for exercising restraint on liberal sexual attitudes.

In the following cross-national comparison of attitudes we use the same items on pre-marital sex, homosexuality and extra-marital sex that are used in the first part of this paper (see Appendix for exact question wording). While in the first part of the paper our main focus was on gender and generation, we now include religion and religiosity (as measured by church attendance) as major discriminating variables. The percentages disapproving of the different sexual behaviours by country can be seen in the top row of Table 5. Regarding pre-marital sex, East Germany, Sweden and West Germany are by far the most liberal, with United States and Ireland at the other extreme. The range of disapproval across countries is very substantial with only 3 per cent disapproving of pre-marital sex in East Germany compared with 43 per cent in Ireland. It is interesting to note that Poland, despite its high Catholicism, adopts a quite moderate stance in comparison to the other countries. The pattern is, however, very different when it comes to attitudes concerning homosexuality and here the two Catholic countries are the most disapproving (Poland 82 per cent followed by Ireland 77 per cent). The United States is also quite traditional (74 per cent) while West Germany is the most liberal (52 per cent). East Germany, Sweden and Britain are, on this issue, in the middle. For extra-marital sex, attitudes are universally disapproving and the cross-country variation is small. However, Germany (both West and East) are significantly more liberal than other countries although, even then, over 80 per cent disapprove.

Using Multiple Classification Analysis (MCA), it is possible to compute the percentage adjustments from the grand mean for belonging to the different categories of the independent variables, while controlling for the effects of the other variables (see Andrews et al. 1973). Thus, the next two rows show the relative position that men and women take on these different attitudes (controlling for age, religion and church attendance). In Britain, for example, there is no change from the 15 per cent disapproval rate for either men or women; whereas for homosexuality 6 percentage points would need to be added for men (increasing disapproval to 70 per cent) while 7 percentage points would be subtracted for women (reducing disapproval to 57 per cent). (These figures differ slightly from those shown in Table 3 because of a small reduction in sample size due to the inclusion of the control variables.) In all countries there is substantial agreement between the sexes, regarding premarital sex, with a significant but only slight difference emerging in the United States and Poland, with women more traditional than men. For homosexuality, however, much stronger gender differences are evident with women always more liberal than men. For extra-marital sex, men and women tend to be equally condemning, with slightly more liberal views of men apparent in Sweden and East Germany.

The relationship of age and sexual morality tends to be very similar across Britain, the United States, Ireland and Poland with age-relationships most marked for pre-marital sex, somewhat less marked for homosexuality and least evident (if at all) for extra-marital sex. However, in both parts of Germany and Sweden, the age difference is most clear for homosexual relations, with the oldest age groups very substantially more likely to condemn sexual relations between same sex adults than the younger generations. What is particularly interesting here is the way the liberalisation of attitudes is associated with a particular age-group. In most countries across most issues, it is the age group that is 48-57 (who were in their early adulthood during the 1960s) that are likely to take a more liberal stance. The exceptions are in the two Catholic countries, Ireland and Poland, where, if age is an indicator of when attitudinal shifts occurred, the change in moral attitudes seems to have come a decade or so later.

Given the stance of the Catholic Church on procreative sex, it might seem that Catholics would be more hostile to homosexuality than Protestants, but this is only true of East Germany, where the Catholics are a very small minority group in a largely secular society. Having no religious affiliation, however, tends to reduce homophobia markedly, in both the United States and West Germany. The relatively liberal stance of the West Germans on extra-marital sex is strongly influenced by religion with the secular being most willing to tolerate adultery, but in general there is little variation within or across countries in the condemning of extra-marital sex. However, the impact of attending church weekly is very substantial for attitudes to both pre-marital sex and homosexuality, with those who attend church weekly being far more likely to condemn such behaviours, in all countries except East Germany where regular church attendance is rare.

In Table 6 we examine the effect of religion on country difference for premarital and homosexual relations. Extra-marital relations are not included as there is relatively little difference between countries. In Model 1 the percentage adjustments for each country are shown from the grand mean across all six countries (18 per cent for pre-marital sex). Thus we can see the country differences observed in Table 5 above. For example, Britain has an adjustment of -3 resulting in 15 per cent of (18-3) disapproving of pre-marital sex. In Model 2, relevant demographic variables are introduced (sex, age, educational attainment, and cohabitation and divorce experience). Controlling for these differences in demographic and partnership status hardly changes the differences associated with each country in attitudes to pre-marital sex. However, controlling for religion and religiosity (in Model 3) does make a substantial difference. In particular, in the two Catholic countries, Ireland and Poland, the strong positive adjustments for country (indicating a more traditional stance) are substantially reduced. Thus, much of their traditional stance can be explained by the religious influence of Catholicism. Similarly, in East Germany, the strong negative adjustment (indicating a more liberal stance) is [TABULAR DATA FOR TABLE 6 OMITTED] also reduced once religion is controlled. Thus, much of the East Germans' liberal stance towards pre-marital sex is explained by their relative secularisation.

Similar patterns can be observed when Model 1 and Model 3 are compared regarding attitudes to homosexual relations. Ireland and Poland become less traditional once religion is controlled and East Germany becomes less liberal. However, in contrast to the models for attitudes towards pre-marital sex, the demographic factors included in Model 2 do make a substantial difference for attitudes towards homosexuality. As predicted, education has a very significantly liberalising influence on attitudes concerning homosexual relations (the adjustments decrease monotonically from +9 for the lowest educational attainment to - 18 for the highest).

These data would be consistent with the interpretation that, in the United States, the relatively high proportion in higher education helps moderate the rather homophonic stance that Americans adopt, relative to their counterparts in other countries. However, the liberalising effect of education in America seems to be cancelled out by the greater degree of religiosity. Thus, the US adjustment for homosexual attitudes increases from 8 to 13 in Models 1 and 2 when education is controlled, but then decreases to 9 in Model 3, when religion is also controlled. It is plausible that there is a similar education effect in Sweden (once education is controlled, the country adjustment increases from -5 to +3). However, in Sweden, unlike in the United States, the liberalising effect of education and the greater acceptance of non-traditional partnerships combine with secularisation to help explain the Swedes' relatively tolerant stance regarding sexual relations between adults of the same sex.

Discussion and Conclusions

In this paper I have examined changes in men and women's attitudes to sexual morality across nations and time. The paper has two main objectives: first to examine to what extent there has been a revolution in sexual attitudes in the United States and Britain and whether the change in attitudes has continued through the 1990s; and, second, to place British and American attitudes in a broader cross-national perspective and investigate how far differences in religion and religiosity help explain within and between country variation in attitudes.

The double standard in sexual morality is not evident in our data, which go back to the mid-1960s in the United States and to the early 1980s in Britain. The fact that people express similar attitudes towards the sexual behaviour of men and women when it comes to pre-marital or extra-marital sex is important, but is no guarantee that double standards do not still exist in practice.(12)

There is a clear differentiation in people's attitudes to pre-marital, extramarital and homosexuality and it is only with respect to pre-marital sex that there has been a very dramatic shift in attitudes over the past decades. In the United States, there was a marked drop in disapproval between the mid-1960s and 1970s and by the 1980s only two-fifths of women and one-third of men felt that sex before marriage was wrong. This level of disapproval has held steady through the 1990s. In Britain, although women were more disapproving than men in the early 1980s, women have become increasingly liberal over the last decade, so that by the mid-1990s there is no gender difference in attitudes. Attitudes to extra-marital sex have held remarkably constant over the whole period, in both countries, with a very high level of condemnation and with women rather more disapproving than men. Interestingly, however, in the latter part of the 1980s, when the scare about the AIDS epidemic spreading to the heterosexual community was very high, men's disapproval increased, making men more like women in their level of censure.

There is a clear contrast between Britain and the United States in attitudes towards homosexuality. In the United States, homosexuality - while never regarded with quite the same degree of disapproval as extra-marital sex - has been subject to quite similar levels of condemnation, especially among men. In Britain disapproval of homosexuality is far below the levels of censure of adultery, especially among women. In the 1990s, however, there has been a marked fall in American levels of disapproval of sexual relations between adults of the same sex. The greater tolerance is likely to reflect the more balanced assessment of AIDS risk, the success of the gay rights movements, and the influence of the more liberal political regime under President Clinton. In Britain, the effects of the AIDS scare is clearly seen in the increase of disapproval from 1983 to 1987, but since then levels of condemnation have fallen slightly among men and quite substantially among women, leading to a marked gender difference in approval rates. This gender difference in homophobia may reflect differences in the way male and female identities relate to sexual orientation. Some men may feel a need to condemn homosexuality in order to assert their own masculinity which, traditionally, has been partly rooted in heterosexual prowess.

In order to explore the extent to which there has been a revolutionary change in attitudes we examined how far attitudinal change has been the result of period effects and how far it reflects cohort succession. It is clear that, in the United States, there was a revolutionary shift in attitudes to premarital sex between the 1960s and 1970s that affected all age groups, but especially the more recent cohorts. The marked change in attitudes continued among earlier cohorts through into the 1980s, but has subsequently reversed, suggesting that in the eyes of the older generation at least, changes in permissiveness were thought to have gone too far. In the United States, attitudinal change seems to have slowed or even stopped and the acceptability of pre-marital sex may have reached its limit. In Britain, attitudinal change is continuing, especially among the earliest cohorts of women. The generational divide is lessening over time and even among the over 60s, only a minority regard pre-marital sex as wrong.

In the United States, the stability that was seen in aggregate attitudes towards homosexuality conceals marked cohort fluctuations. Interestingly, women are considerably more prone to attitudinal change than men, both in the more traditional move in the 1980s in response to the AIDS crisis and conservative backlash, and again in the more liberal move in the 1990s. In Britain the pattern is similar and the more liberal shift in attitudes in the 1990s is by far the most evident among young women. This gender difference could be taken as further evidence that anti-homosexual attitudes are less deeply rooted in women's self-identity than they are for men. Thus, women are more likely to alter their attitudes in response to external events such as the AIDS crisis and socio-political change.

Our second major objective was to place British and American attitudes in a broader cross-national perspective by comparing attitudes to sexual morality in the United States, Britain, Ireland, Germany, Sweden and Poland. In particular, the concern was to investigate how far differences in religion accounted for within country and across country variation in attitudes. In this regard, one pertinent issue is whether, in those countries where the authority of the Catholic church is predominant, traditional attitudes concerning the marital and procreative imperative of sex prevail. Poland and Ireland make interesting contrast cases, in that in Poland the Church was reduced to political opposition by the communist regime; whereas in Ireland the Catholic church's religious monopoly has gone unchallenged. Ireland tends to be very traditional with respect to all aspects of sexual morality, whereas in Poland, although homosexual relations are condemned, attitudes to pre-marital and extra-marital sex are relatively moderate. The United States is very comparable to Ireland in its traditional stance. Yet within the United States, Catholics tend to be more liberal than Protestants which, in part, reflects the influence of fundamentalism. In both the former West Germany and in Britain there is little difference between Catholic and Protestant positions. The British, who have quite low rates of church attendance, are remarkable for their relative moderation in attitudes towards sexual morality. In Sweden although the few regular church attendees do take a more traditional stance with respect to pre-marital sex, the overwhelming majority adopt attitudes that are in line with liberal sexual practice and policy. It is not surprising that the former East Germany is one of the least traditional countries, but the former West Germans also adopt a relatively liberal stance that is at odds with post-war patriarchal family policy.

Church attendance has a very marked influence on sexual morality in all countries. Religiosity is a powerful counterbalance to the permissive trends regarding pre-marital sex and tolerance of homosexual relations. However, church influence is being exercised against the liberalising effect of time, as cohort effects and secularization trends are likely to combine in making attitudes more tolerant. Despite the widespread tendency of women to be more regular church attendees than men, women are more tolerant than men of same sex relations in all six nations.

Religion also helps explain some of the variation between nations in attitudes to sexual morality. Interestingly, the United States remains the most traditional country with respect to sexual morality even when the influence of its relatively high levels of church attendance are factored out. The puritan mentality runs deep. This might seem at odds with the astonishingly high rates of divorce in America, but perhaps the two are more complementary than they seem at first sight. It may be that although the consumer culture has extended sexual freedom, the puritan ethic serves to harness sexual activity into higher rates of divorce and remarriage. Ireland, however, becomes much closer to the cross-national norm, once religion is controlled. Given the huge cohort differences that can be seen in attitudes to sexual morality in Ireland, public opinion seems likely to diverge at an increasing rate from the traditional Catholic doctrines that limit sexual freedom. A similar pattern can be seen in Poland, although cohort effects are somewhat less pronounced than in Ireland.

This analysis highlights two important aspects of attitudes to sexual morality. First, the changes have not been as revolutionary as is often claimed. The language of sexual revolution has a momentum of its own and catapults us into the era of risk society, when if not quite 'anything goes', there is at least an almost limitless range of acceptable choice. People, however, do seem to show a remarkable degree of agreement about the relative culpability of different sexual behaviours. Moreover, with the exception of attitudes to premarital sex, attitudes have not changed very dramatically over the past few decades. Attitudes towards homosexuality are becoming more tolerant in Britain and the United States, especially among women, but the pace of change is fairly slow. Attitudes towards extra-marital sex have stayed remarkably constant and sexual fidelity is still very much an ideal. Doubtless ideals do not always translate into practice, but this is nothing new.

The second important thing to note is that it is wrong to discount the influence that the organised churches still exert. It is a very curious omission that in much of the theoretical discussion of the sexual revolution, the influence of religion is largely ignored. For example, in the analysis of the structuring of the sexual revolution that we quoted from in the introduction (Martin 1996), much is made of the various economic factors, but religious influence is rarely mentioned. Similarly, in the analysis of the Risk Society by Beck (1992), the demonopolisation of science figures very prominently, but it is as if, in post modern or post-traditional society, the role of established religion has long since past. Certainly, it is true that in this age of uncertainty, new fundamentalisms, whether secular or religious, Christian, Hindu, or Islamic, are trying to mould new moralistic proscriptions and prescriptions (Weeks 1995). However, the demise of traditional values has been over-stated and our analysis suggests that the old proscriptions and prescriptions are still influential in determining the sexual attitudes of the different nations.

Appendix: Exact Wording of Sexual Morality Items

British Social Attitudes

* If a man or a woman have sexual relations before marriage what would your general opinion be? Would it be always wrong, mostly wrong, sometimes wrong, rarely wrong or not wrong at all?

* What about a married man having sexual relations with a woman other than his wife? (1983 only)

* What about a married woman having sexual relations with a man other than her husband? (1983 only)

* What about a married person having sexual relations with someone other than his or her partner? (1984 onwards)

* What about sexual relations between two adults of the same sex?

NORC Survey (1965)

Please tell me whether you consider the following actions always wrong, almost always wrong, wrong only sometimes, or probably all right:

* A man has intimate relations with a woman he is engaged to and intends to marry.

* A married man has an extra-marital love affair.

General Social Survey (1972 onwards)

There's been a lot of discussion about the way morals and attitudes about sex are changing in this country.

* If a man and a woman have sex relations before marriage do you think it is always wrong, almost always wrong, wrong only sometimes, or not wrong at all?

* ... A married person having sexual relations with someone other than the marriage partner?

* ... Sexual relations between two adults of the same sex?

International Social Survey programme (1994)

* Do you think it is wrong or not wrong if a man and woman have sexual relations before marriage? (always wrong, almost always wrong, wrong only sometimes, or not wrong at all)

* What about a married person having sexual relations with someone other than his or her husband or wife?

* And what about sexual relations between two adults of the same sex?

Acknowledgements

The support of the Economic and Social Research Council is gratefully acknowledged. This work forms part of on-going research funded by the ESRC's Population and Household Change Initiative (#L315253024). I am grateful to Duane Alwin and Michael Braun for helpful comments and for earlier collaborative research on which this paper draws and to Kim Perren for her careful assistance with data handling and analysis. I am also grateful to anonymous reviewers of Sociology for helpful and constructive suggestions.

Notes

1. According to Martin (1996), the term dates back at least to the 1921 book, Die Sexuelle Revolution, published in Leipzig, written by Wilheim Heinrich Dreuw, which predates by nearly three decades the book by Wilheim Reich (1949), whose title translated into English is 'The Sexual Revolution'.

2. For example in England, the established church experienced its biggest annual drop in Sunday church attendance for twenty years. The decline is paralleled in the fall in attendance at Catholic mass. Together they indicate that the trend against collective worship, far from bottoming out, is continuing. Efforts to attract young people to church to ensure its vitality into the next century do not seem to be successful (Guardian, 7 February 1997).

3. As Martin (1996) points out, in the 1960s, the rise in women's employment in the United States was mainly among mothers. Thus assertions of any direct causal link between women's entry into paid employment and the sexual revolution is problematic. However, the underlying causes of the sexual revolution goes beyond the scope of this paper. Moreover, the objection that the timing is wrong is perhaps less serious than appears, because there is often a temporal lag between changing social conditions and changing social attitudes.

4. One manifestation of the new political conservatism was the funding difficulties experienced both by the British National Survey of Sexual Attitudes and Life Styles and the American National Health and Social Life Survey. Government sponsorship was withdrawn and private support had to be found.

5. Analysing attitudinal trends is not a straightforward matter and great care must be exercised to ensure comparability of question wording, context, response categories etc. and to take account of non-comparability when it occurs (see Schuman and Presser 1981; Alwin and Scott 1996). Sampling error must also be taken into consideration and, in the attitudinal items used in this paper, a conservative guide for interpreting confidence intervals for a sample size of 1,000, is to add or subtract 3 per cent. Thus, we can be 95 per cent certain that the true proportion is within 3 per cent (in either direction) of the proportions reported here.

6. Statistical techniques to arrive at the resolution of this identification problem have been proposed and debated elsewhere (e.g. Mason and Fienberg 1985; Firebaugh 1989; Alwin and Scott 1996). The tables presented in this paper show the standard cohort decomposition for the population level (see Mason and Lu 1988 and Scott et al. 1996).

7. As Thornton noted (1989) the marked decline in disapproval of pre-marital sex between 1965 and 1972 is likely to be an underestimate, because the 1965 question indicates a high degree of commitment to the marriage, whereas the question used from 1972 onwards makes no such assumption.

8. The reference for the 1965 question is to a man having an extra-marital affair rather than a woman. This is to ensure that the estimate of change is, if anything, conservative, although there was little evidence of a double standard.

9. Combined years were used to increase sample size and provide more stable estimates.

10. The ISSP provide functionally equivalent attitudinal measures from representative cross-national samples and are, therefore, the best survey data available for this comparison. Differences in survey organisational procedures means that it would be unwise to place great reliance on the absolute differences between countries in attitudes to sexual morality; although the relative disapproval of different sexual behaviours and their relationships with covariates are unlikely to be affected.

11. It was only in 1977, for example, that the law that forbade a wife to take a job without the consent of her husband was repealed.

12. In Britain, for example, there is evidence of persisting double standards in grounds for divorce. The majority of divorces are granted to women because of 'intolerable behaviour', but when divorce is granted to the man, the most common grounds is the wife's adultery (Social Trends 1996).

References

ALBA, R. D. 1981. 'The Twilight of Ethnicity Among American Catholics of European Ancestry'. Annals of the American Academy of Political and Social Science 454:86-97.

ALWIN, D. and SCOTT, J. 1996. 'Attitude Change- Its Measurement and Interpretation Using Longitudinal Surveys', pp. 75-106 in B. Taylor and K. Thomson (eds.) A Decade of Change in Social Values. Aldershot: Dartmouth.

ANDREWS, F., MORGAN, J., SONQUIST, J. and KLEM, L. 1973. Multiple Classification Analysis. Ann Arbor, Michigan: Institute of Social Research.

BECK, U. 1992. Risk Society: Towards A New Modernity. London: Sage.

BRAUN, M., SCOTT, J. and ALWIN, D. 1995. 'Nichtkonventionelle Lebensformen: Verhalten und Einstellungen in Ost und West' (Unconventional Family Forms in East and West Germany). German Sociological Association, Halle Germany, unpublished paper.

BUCK, N. and SCOTT, J. 1994. 'Household and Family Change'. In N. Buck, J. Gershuny, D. Rose and J. Scott (eds.) Changing Households: The British Household Panel Study 1990-1992. Colchester: ESRC Research Centre.

BUMPASS, L. and SWEET, J. 1989. 'National Estimates of Cohabitation'. Demography 26:615-625.

CARRIGAN, T., CONNELL, B. and LEE, J. 1987. 'Towards a New Sociology of Masculinity', in H. Brod (ed.) The Making of Masculinities. London: Allen & Unwin.

CASTELLS, M. 1997. The Power of Identity. Oxford: Blackwell.

CHAMBERLAYNE, P. 1994. 'Women and Social Policy', in J. Clasen and R. Freeman (eds.) Social Policy in Germany. Hemel Hempstead: Harvester.

DAVIS, J. and JOWELL, R. 1989. 'Measuring National Differences: An Introduction to the International Social Survey Programme ISSP', in R. Jowell, S. Witherspoon and L. Brook (eds.) British Social Attitudes: Special International Report, Aldershot: Gower.

DAVIS, J. and SMITH, T. 1994. General Social Surveys: Cumulative Codebook, 1972-1994. Chicago: National Opinion Research Centre.

DUNCOMBE, J. and MARSDEN, D. 1993. 'Love and Intimacy: The Gender Division of Emotion and 'Emotion Work', Sociology 27:221-42.

EHRENREICH, B. 1983. The Hearts of Men: American Dreams and the Flight from Commitment. New York: Anchor Books.

FIREBAUGH, G. 1989. 'Methods of Estimating Cohort Replacement Effects', in Clogg, C. C. (ed.) Sociological Methodology, Cambridge, Mass.: Basil Blackwell.

GIDDENS, A. 1996. In Defence of Sociology. Cambridge: Polity Press.

GLENN, N. and WEAVER, C. 1979. 'Attitudes Towards Premarital, Extramarital and Homosexual Relations in the US in the 1970s'. The Journal of Sex Research 15:108-18.

GREELEY, A. 1989. Religious Change in America. Cambridge, Mass.: Harvard University Press.

GREELEY, A. 1990. The Catholic Myth: The Behaviour and Beliefs of American Catholics. New York: Scribners.

HANTRAIS, L. and LETABLIER, M. 1996. Families and Family Policies in Europe. London: Longman.

HARDING, S. 1988. 'Trends in Permissiveness', in R. Jowell; S. Witherspoon and L. Brook (eds.) British Social Attitudes, the 5th Report, Aldershot: Gower.

HAYES, B. 1995. 'Religious Identification and Moral Attitudes: The British Case'. British Journal of Sociology 46:457-474.

INGELHART, R. 1990. Culture Shift In Advanced Industrial Societies. Princeton: Princeton University Press.

JOWELL, R., BROOK, L. and DOWDS, L. 1993. Technical Details of Surveys, Appendix 1, in Jowell, Brook and Dowds (eds.) International Social Attitudes: The loth BSA Report. Aldershot: Gower.

JOWELL, R. and PARK, A. 1996. 'Questioning British Attitudes', pp. 75-106 in B. Taylor and K. Thomson (eds.) A Decade of Change in Social Values. Aldershot: Dartmouth.

LAUMANN, E., GAGNON, J., MICHAEL, R. and MICHAEL, S. 1994. The Social Organization of Sexuality. Chicago: University of Chicago Press.

LUCKMANN, T. 1967. The Invisible Religion. New York: Macmillan.

MARTIN, J. 1996. 'Structuring the Sexual Revolution'. Theory and Society 25:105-51.

MASON, W. and FIENBERG, S. 1985. Cohort Analysis in Social Research: Beyond the Identification Problem. New York: Springer-Verlag.

MASON, K. and LU, Y-H. 1988. 'Attitudes Towards Women's Familial Roles: Changes in the United States: 1977-1985'. Gender and Society 2:39-57.

REISS, I. 1990. An End to Shame: Shaping our Next Sexual Revolution (with H. Reiss). Buffalo, NY: Prometheus Books.

RUSSELL, B. 1929. Marriage and Morals. London: Allen & Unwin.

RYDER, N. 1965. 'The Cohort as a Concept in the Study of Social Change'. American Sociological Review 30:843-61.

SAUNDERS, J. and EDWARDS, J. 1984. 'Extra-Marital Sexuality. A Predictive Model of Permissive Attitudes'. Journal of Marriage and Family 46:825-35.

SCHUMAN, H. and PRESSER, S. 1981. Questions and Answers in Attitude Surveys. New York: Academic Press.

SCOTT, J. 1998. 'Generational Changes in Attitudes to Abortion'. European Sociological Review 14:177-90.

SCOTT, J., ALWIN, D. and BRAUN, M. 1996. 'Generations and Changing Sex-Role Attitudes: Britain in a Cross-national Perspective'. Sociology 30:471-92.

SCOTT, J., BRAUN, M. and ALWIN, D. 1995 'Religion and Family Values: A Cross-national Comparison'. European Sociological Conference, Budapest, Hungary, unpublished paper.

SEIDMAN, S. 1992. Embattled Eros: Sexual Politics and Ethics in Contemporary America. New York: Routledge.

SINGH, B. 1980, 'Trends in Attitudes to Premarital Sexual Relations'. Journal of Marriage and Family 42:387-95.

SMITH, T. 1990 'Report: The Sexual Revolution?' Public Opinion Quarterly 54:415-35.

THORNTON, A. 1985. 'Family and Religion in a Changing World'. Journal of Marriage and Family 47:381-94.

THORNTON, A. 1989. 'Changing Attitudes Towards Family Issues in the United States'. Journal of Marriage and the Family 51:873-93.

WEEKS, J. 1989. Sex, Politics and Society: The Regulation of Sexuality since 1800. 2nd edn. Harlow: Longman.

WEEKS, J. 1995. Invented Moralities: Sexual Values in an Age of Uncertainty. Cambridge: Polity Press.

WELLINGS, K. and WADSWORTH, J. 1990. 'AIDS and the Moral Climate', in R. Jowell, S. Witherspoon and L. Brook (eds.) British Social Attitudes: The 7th Report. Aldershot: Gower.

WELLINGS, K., FIELD, J., JOHNSON, A. and WADSWORTH, J. (eds.) 1994. Sexual Behaviour in Britain. London: Penguin.

WILSON, B. 1982. Religion in a Sociological Perspective. Oxford: Oxford University Press.

Biographical Note: JACQUELINE SCOTT is an Assistant Director of Research in the Faculty of Social and Political Sciences and Fellow of Queens' College, University of Cambridge. Her current research interests include family change and lifecourse perspectives.

Email: