In 2000, the Canadian Parliament enacted the Modernization of Benefits and Obligations Act, which extended benefits and obligations to common-law couples, be they of opposite sexes or the same sex. Furthermore, the definition of "spouse" was changed to include any two persons who have lived together in a "marriage-like" relationship for at least two years. In 2001, the Netherlands became the first country to legalize same-sex marriage and was followed by Belgium in 2003. The provinces of British Columbia and Ontario did so in 2003, and five other provinces and territories followed suit in 2004. While the U.S. government is officially opposed to same-sex marriages, the Canadian government is currently willing to allow it.

Even among those who are favourable to the change, question marks remain. Thus, issues pertaining to same-sex marriage loom large in the public consciousness and many questions recur in people's minds and in their inquiries in this respect. Some of the enquiries concern the quality of homosexual unions. A majority are concerned about lesbigays' (abbreviation for lesbians and gays) ability to parent and the effect on children of living with same-sex parents. Others contend that allowing same-sex couples to marry would devalue the institution of marriage.

We do not yet have research on same-sex marriage itself; thus it makes sense to examine the relationship of cohabiting homosexuals in order to have an idea of their functioning, division of labor, psychological satisfaction, and sexual relations, as is done for married and cohabiting heterosexual couples. However, except for large-scale studies on sexuality with representative samples of the adult population, there are no large-scale studies of the life of lesbigay couples. Especially lacking are follow-up studies of such couples living in long-term relationships. There are several reasons for this gap in our knowledge:

1. It is only recently that homosexual unions have been socially recognized. Hence, in the past, researchers had little incentive to study them.
2. It is still not possible to obtain large and representative samples of homosexual couples. It is only with such samples that we could draw conclusions applicable to the entirety of the homosexual couple population.
3. However, even here, another set of problems emerges. First, gays and lesbians have suffered from stigmatization and may not wish to share the intimate details of their lives with non-gay researchers for fear of being further victimized or categorized (Nelson, 1996:10). Second, these concerns may affect the way they respond to researchers' questions.
4. Quite a few good studies on couples and children in same-sex parent families have been carried out in the late 1990s and early 2000s. However, these suffer from the fact that the couples are often selected by the researchers or they are self-selected. Self-selection occurs when ads are placed in newspapers, newsletters, bulletin boards of gay community and health centers, or distributed on university campuses.

Obviously, such an approach draws relatively educated couples who maintain links to the homosexual community, may be activists, and are fairly young (Christopher and Sprecher, 2000). Less educated same-sex couples and those who have little to do with the homosexual subcultures or are older will not be represented in these samples (but, see Yip, 1997).

Furthermore, there may be specific characteristics of individuals/couples who accept the call of these ads. For instance, couples who want to make a pro-homosexuality statement or only couples who get along particularly well may elect to be interviewed or fill out a questionnaire--thus biasing the research results and making it difficult to interpret them.

There are many additional methodological problems in the research on same-sex-parent families, including small samples and lack of control for important variables (whether sociological or psychological). In other words, much of the research is still at the exploratory level so that it is possible that some of the findings herein reported will not hold under the scrutiny of better-designed analyses and from a wider spectrum of researchers who have expertise in various domains of family life.

It cannot be emphasized enough that the research upon which we can draw at this time remains embryonic and is often incomplete or inadequate methodologically. As such, any conclusions that can be drawn are tentative and serve to sharpen research questions yet to be pursued. Nevertheless, it is important to bring the evidence that is available to bear upon the discussions in which Canadians and parliamentarians are now involved. In this article, research about the following topics is summarized and critically assessed:

• demographic estimates of gay and lesbian couples
• conjugal relationships among lesbians and gays
• conflict within lesbigay relations
• fidelity within same-sex unions
• same-sex couples raising children
• children growing up with same-sex parents.

The summary review of empirical evidence about these themes provides a necessary foundation for a reasoned consideration of the question: would same-sex marriage devalue the institution of marriage? The paper concludes with a list of thus far unanswered questions that can serve as an agenda for future research.

As already pointed out, the research presented and issues discussed in this article have to be placed within their proper demographic context: homosexuals constitute a small minority. Furthermore, in terms of definition, there is not always a clear-cut demarcation between homosexual and heterosexual self-identity so that, at some point in their life cycles, but not necessarily throughout their entire adult years, a number of persons identify as homosexuals.

Many may have had a heterosexual past: based on three large American data sets, Black et al. (2000) estimate that possibly as many as 30% of gay men and 46% of lesbian women have been or are married heterosexually. In fact, until now, a good proportion of lesbians, especially those above 30, have discovered their identity later in life (Morris et al., 2001); they may identify as bisexual. Other persons do not identify as gay but nevertheless engage in homosexuals acts, such as in prisons.

The carefully designed studies carried out by Laumann et al. (1994) established that 2.8% of men identify as gay and l.4% of women identify as lesbian. These figures are similar to the rates at which men and women have exclusively same-sex relationships (3% for men and 1.6% for women). In the Black et al. (2000) sophisticated statistical analyses, 4.7% of men and 3.5% of women had had at least one same-sex experience since age 18. But only 2.5% of men and 1.4% of women had engaged in exclusively same-sex activities over the year preceding the survey. Overall, Gates and Sanders (2002) estimate that from 2% to an upper limit of 5% of men are gay and from 1% to 3.5% of women are lesbian. These estimates closely match the 4% in the Laumann et al. study.

The much heralded estimate of 10% of the population being gay originated from the hastily interpreted results of the poorly designed 1948 Kinsey Report: their sample was not representative and was largely self-selected. But even then Kinsey and his colleagues estimated that only 4% of males were exclusively homosexual (Kinsey et al., 1948). Yet, this figure was overlooked and it is the 10% estimate that caught the media's attention, became entrenched in the public's perception, and accepted by gays themselves. (Kinsey's subsequent 1953 report on female homosexuality did not present exact estimates.)

Lesbigays appear to be more numerous than they are because they live disproportionately in large metropolitan areas such as San Francisco and Toronto or in smaller cities that contain a major university: this concentration gives them higher social visibility (Black et al., 2000), allows them to advocate for equality, gives them greater media exposure, and at many universities, considerable presence in the curriculum of the social sciences.

Furthermore, there is unfortunately a tendency among homophobic persons and groups to exaggerate the size of the homosexual population in order to sustain their "worst" fear scenarios and prejudices. And, among some homosexuals, there is a tendency to inflate the numbers for political and social purposes. Hence, many misleading "facts" abound.

What is more relevant to our discussion below is the 2000 U.S. Census estimate that 1% of all couples sharing a household (married + cohabiting) are same-sex ones and the 2001 Canadian census estimate of a prevalence of 0.5% (Statistics Canada, 2002). It is likely that these proportions are somewhat higher in reality. Thus, it is among this 1% of the couple population that we will find those who will be seeking marriage and parenthood.

Furthermore, the 1990 U.S. census revealed that about 22% of households headed by lesbian couples had a child under 18 compared to 6% of those headed by gay partners (Black et al., 2000). In Canada, the 2001 census revealed that 15% of households headed by lesbian couples had children versus 3% among male same-sex households (Statistics Canada, 2002). Again, these numbers probably underestimate the situation to some extent. Moreover, these data omit lesbigays who are single (not partnered) and have a child living with them and those who are married heterosexually and have a child or children living with them. Black et al. estimate that over 28% of all lesbians and 14% of all gays have children living with them. Still, these numbers omit lesbigays whose ex-spouse has custody of their child or children following a divorce.

In general, lesbigay couples report a more egalitarian division of labour than married couples (Patterson, 2000). Kurdek (1998, 2001) compared gay and lesbian couples to married couples with and without children in order to study how satisfied partners were with their relationship. Gay couples reported more autonomy from each other in terms of activities, friendships, and decision-making than married dyads. Lesbian couples reported greater relationship satisfaction and more intimacy, autonomy, and equality than married couples as well as higher levels of positive problem solving.

A small study reported lesbian couples with children to be happier with their relationship than those without children (Koepke et al., 1992). This result is the opposite of what is often found among married couples: those without children report a higher average degree of happiness than those with children (Ambert, 2001b; Kurdek, 2001).

As is the case among heterosexuals, saying "I love you," showing physical affection, gift-giving, and cooking for the other are frequently-mentioned expressions of love among same-sex partners (Stiers, 1996). Gift-giving is more often mentioned among males than females, perhaps as a substitute for verbal expressions of affection. However, gay couples report lower levels of commitment than childless heterosexual couples (Kurdek, 2001). Both gay and lesbian couples have higher rates of dissolution than married ones (Kurdek, 1998).

One reason for the last difference is that, just as is the case among heterosexual couples who choose to cohabit rather than marry, many lesbigays choose to live together without being committed to a lasting and exclusive relationship (Ambert, 2001b). Without such a commitment, cohabitations dissolve more easily, be they same-sex or opposite-sex. For his part, Kurdek explained the higher rates of dissolution among lesbigay couples compared to married ones by the fact that nothing in the social world encourages same-sex couples to stay together, while the contrary occurs for married couples. The latter are supported by cultural values preventing them from leaving the relationship. Gay unions do not constitute a valued social institution as does marriage nor, for that matter, as do heterosexual cohabitations.

All of this means that there are no legal and social barriers preventing the dissolution of their relationships, and the same largely occurs for cohabiting heterosexual couples who do not have children. Socially, a same-sex relationship is already labelled deviant and few object if it breaks up.

In fact, Kurdek (2001) found that both gay and lesbian couples experienced lower levels of approval of their relationship from their respective families than was the case for heterosexual couples. Despite this lack of social acceptance and support as well as lack of disincentives to conjugal dissollution, around 85% of the same-sex couples remained together for the duration of Kurdek's five-year longitudinal study. It should be pointed out, however, that they had already been together for an average of 10 and 7 years, respectively, for gay and lesbian couples: they had thus survived the most unstable years, which are the first ones--both for marriage and cohabitation (Ambert, 2002b).

Overall, there are both similarities and differences in the way homosexual and heterosexual couples live their relationship. In many ways, the requirements of daily life place similar demands on them. What differs is that two men together, because of gender roles and biology, may function somewhat differently than two women together and, in turn, than a heterosexual couple. The following pages point to such areas of differences and similarities.

When Kurdek (1994) compared gay, lesbian, and heterosexual couples, he found little difference in the rank order of frequency of conflict concerning power, social issues, personal flaws, distrust, intimacy, and personal distance. Issues pertaining to intimacy and power were the most frequently-mentioned sources of conflict by all three types of couples. This is interesting in itself because it indicates that problems related to power are not entirely explainable in terms of gender roles or the patriarchal order because they exist in same-sex couples as well.

For all three types of couples, frequent conflict was negatively related to each partner's satisfaction with the relationship. But conflict over intimacy and power were more likely to result in lower levels of relationship satisfaction than were other sources of conflict. In fact, during the first year of this study, conflict over power was the most powerful predictor for deterioration of the relationship over time. These findings are in agreement with those of Gottman (1994) as well as Vangelisti and Huston (1994) for married couples.

All three types of couples engage in similar conflict resolution mechanisms and report a relatively low level of conflict (Kurdek, 1994). However, lesbian women resort to more constructive conflict-resolution styles. They also report making greater effort to resolve conflict, perhaps because as women they have been socialized to be attuned to a partner's distress more than is the case among other couples that include at least one man (Metz et al., 1994).

For its part, the purpose of verbal abuse is to dominate, exercise power, show who is "the boss." It is also used to rationalize or excuse one's bad behavior by demeaning the other. Verbal abuse consists of repeatedly calling one's partner epithets, including much foul language, berating and demeaning--put-downs, in other words--threatening, criticizing, even in public. It also sets a precedent upon which physical abuse can be added once civility has been eroded in the relationship. In fact, physical abuse is generally accompanied or preceded by verbal insults and attempts by one partner or both at intimidating the other (Sugarman et al., 1996).

Verbal abuse can also take the form of threats tantamount to psychological blackmail. One type of abuse that is unique to same-sex relationships is the threat of "outing." This consists in revealing that the other is gay or lesbian at his or her work, for instance, or to relatives who are unaware of the situation and might turn against the hapless person were they to be so informed (D'Augelli et al., 1998).

As Renzetti (1997b:74) has discovered among same-sex couples, psychological abuse is often tailored to fit a partner's vulnerabilities. She writes of two women with physical disabilities: "Their partners would abandon them in dangerous settings (i.e., an isolated wooded area) without their wheelchairs. Another woman who was diabetic stated that her partner would punish her by forcing her to eat sugar."

Apparently, then, physical violence in same-sex couples is as much a matter of power and control as it is in heterosexual couple violence, except that gender or patriarchal ideologies are not an issue (Waldner-Haugrud, 1999). The rate of violence among same-sex couples approximates that of heterosexuals, between 12 and 33% depending on the sample and measures (Straus and Gelles, 1990). The research carried out strictly on lesbigay samples shows slightly higher rates (Elliott, 1996). Depending on the studies, as many as 50% or as few as 17% of lesbians admit having been abused by a female partner. Of those who have been victimized, the same proportion report having also practiced abuse. But 30% of those who have not been abused report having been abusive.

These statistics are tentative because, as Renzetti (1997a:289) points out, no reliable study of the prevalence of same-sex partner abuse exists yet. Renzetti found that partner abuse often went hand in hand with child abuse, particularly toward the child of the victimized partner. These cases, however, have been documented only among lesbian couples whose children were from a previous marriage.

Statistics on domestic abuse may also be unreliable because this is a problem that the homosexual community often neglects or denies (Merrill and Wolfe, 2000) in part because of the fear of reinforcing negative stereotypes and also because it is erroneously believed that partner abuse occurs only in male-female couples (Renzetti and Miley, 1996). In fact, many lesbians "believe in the inherent goodness" of lesbian relationships (Riedmann et al., 2003). Furthermore, as Riedmann et al. point out, "both men and women in same-sex relationships are freer to be either dominant or submissive...and they may fight back more often than do" their heterosexual counterparts, "a situation that leads to confusion about who is the battered and who is the batterer" (p. 306).

In addition, domestic violence within same-sex unions receives less attention from the police. It is less reported and, when reported, is often interpreted as violence of one man against another rather than as domestic violence. However, many police departments in the U.S. are now well trained and deal with these cases efficiently (Island and Letellier, 1991). The therapeutic establishment and other concerned groups are also taking steps to address these problems (Leventhal and Lundy, 1999).

Bars are often important in the social life of homosexuals, which can lead to alcohol-related problems, including partner abuse and a high level of substance use among gay youths (Orenstein, 2001). As well, although lesbians do not have a higher rate of alcohol use, there is research to suggest that they may have a higher rate of problem drinking than women in general (Cochran et al., 2001). However, as is the case among heterosexual couples, alcohol is merely a facilitator of abuse. It is often used as an excuse by the abuser and even by those who have been victimized (Renzetti, 1997a).

Generally, no matter the type of union, males engage in extra-couple sex more than women, as indicated in the percentages provided by Blumstein and Schwartz below (1990). The respondents in the survey were men and women who reported at least one instance of sexuality outside their marriage or cohabitation in the past year:

The exact number of children living with two homosexual parents is not known. What we know from the 2001 census is that at least 3,000 same-sex couples are raising children in Canada today. A majority of these children were born to a mother-father unit that ended in divorce and the lesbian or gay parent obtained custody. Thus, a majority of these children have another parent who is heterosexual and, presumably, a good proportion of the children remain in contact with this parent. In addition, other homosexuals have children from a previous marriage and these children are in the custody of the non-gay parent.

But more and more lesbian couples give birth via donor insemination or adopt. Some gay couples also adopt, and in the U.S., particularly California, a few have had recourse to surrogate mothers. In other cases, a gay couple cooperates with a lesbian couple in producing children (via donor insemination) and raising them jointly, an arrangement that allows children to have parents of both sexes (Patterson and Chan, 1997).

Lesbian women prefer committed relationships far more than do gay men and are more monogamous sexually, as we saw in a preceding section (Fowlkes, 1994). As women, they are socialized to be kin keepers. These characteristics lead to the formation of a greater number of lesbian-based than gay-based families. In Canada, approximately 15% of lesbian couples are raising children in comparison to only 3% of gay couples. And, of course, lesbian mothers can produce children more easily.

As Slater (1995:94) points out, the lesbian parental role and cooperation is somewhat different depending on the family situation. When partners decide to have children together, the child "is the new member being incorporated into an existing unit." Both partners can be free to parent equally and the child may bolster their intimacy.

But the family challenges are different when one partner comes to the couple as a parent; the other partner, then, is the new family member. There are instances when both partners bring children from previous relationships. These situations are similar to those of stepparenting among heterosexuals. But these instances may also differ depending on whether the children came from a heterosexual marriage or from another lesbian union.

Lesbigays who have children often create a network of fictive kin or "chosen" family (friends, former partners, and willing relatives) for social and emotional support as well as to offer their children suitable adult role models of the other sex. This support network may be entirely gay but generally represents a mixture (Oswald, 2002). Lesbigay parents also have recourse to mainstream community organizations that are supportive of lesbigay families.

While there are still few gay-headed families, one can also presume that, for lesbigay parents, the daily routine is largely similar to that of heterosexual families (Nelson, 1996). Whoever their parents are, all children need love and supervision. They all need to be sheltered, fed, taken to school, and so on. As well, the family life cycle is similar for both same-sex and opposite-sex-parent families: children arrive, grow up, go through stages, and parenting flows accordingly.

In the population at large, there are three main concerns regarding the children of same-sex parents. These are the fears that the offspring will grow up to be psychologically maladjusted because of social stigma, that they will be molested by their parents or parent's partners, and that they will become homosexual themselves. None of these concerns have been supported by the research so far (Ross, 1994), except perhaps the last one, but only to a very small extent.

Again, it bears repeating that same-sex-parent families have been excluded from large, representative surveys on family life, and, for instance, none exists on child abuse in these families. The studies that exist have not included a wide range of developmental outcomes and, of course, few have followed the children raised by same-sex couples into adulthood (see Tasker and Golombok, for an exception, 1997).

Nevertheless, all the small studies put together arrive at fairly similar, positive results. In interpreting these results, it is important to remember that the young adults and most adolescents in these studies were originally born in heterosexual families that ended in divorce. It is therefore possible that some deficits/resiliencies caused by divorce may have been overlooked in the research designs.

For instance, we do not know whether, after a divorce, children of a lesbigay parent have fewer or more problems than children who have heterosexual parents. For these children, it may well be the divorce of their parents rather than the one parent's homosexuality which is the more important factor affecting their development--especially so since they have, after all, lived with at least one heterosexual parent until the divorce. Unfortunately, these studies have failed to separate the effects of the two variables: divorce and parents' sexual identity.

Tasker and Golombok (1995) initiated a longitudinal study to compare children whose single mother was lesbian with those whose mother was heterosexual. At the time, the children were on average 9.5 years old and were reinterviewed 14 years later The young adults with a lesbian mother reported a better relationship with her and with her partner than young adults whose mother was heterosexual and brought in a stepfather. The children of lesbians even had a better relationship with their father, probably because a mother's female partner does not appear to be competing with their father.

Hence, there is a possibility that children with a lesbian mother experience a less difficult time before and after their parents' divorce compared to children with two heterosexual parents--perhaps because of a lower level of parental conflict, less jealousy and fewer feelings of personal rejection. This possibility deserves to be researched.

Children living in same-sex-parent families are often teased by peers and shunned by their peers' parents and thus their lives may be more stressful (Morris et al., 2001:151). Yet, they do not seem to grow up disadvantaged emotionally and may even possess certain strengths of character such as tolerance, empathy, and contentment (Laird, 1993; Patterson, 2000). However, other anecdotal and clinical cases point to difficulties experienced by adolescents who "start to feel embarrassed by their parents' homosexuality" (DeAngelis, 2002), a situation that may be attenuated in future cohorts (Nelson, 1996).

A study comparing lesbian and heterosexual mothers, some single, some in couples, found no difference in children's adaptation and development around age 7, even though all had been reproduced by donor insemination, which is certainly an additional complication in a child's self-definition (Chan et al., 1998). The sample was largely upper-middle-class so that we do not know if the results would be replicated at a working-class level.

Overall, as parents, lesbian mothers are similar to heterosexual mothers; it is not their sexual orientation that emerges as an important variable but their identity as mothers (Lewin, 1993). Their children show few differences from other children (Parks, 1998), and whatever differences exist stem largely from the social stigma attached to homosexuality and consequent social rejection outside the home.

Children of homosexual partners usually adopt heterosexual identities (Bailey and Dawood, 1998). Bailey et al. (1995) compared adult sons who had spent many years with their gay father to sons who had lived only briefly with their gay father. There was no rise in the rate of homosexuality among the offspring with longer contact with their father. A small research by Costello (1997) found that many homosexual partners consciously avoid pressuring their children to conform to their sexual preferences.

Nevertheless, young people raised in same-sex-parent families are more tolerant of same-sex experimentation (Stacey, 1998), and they develop a homosexual identity slightly more often than do children in other families--whether heredity and learning are interrelated causal factors is impossible to evaluate at this point (Stacey and Biblarz, 2001).

But Tasker and Golombok (1995) found that, when mothers had had more lesbian partners and were more open about their sexuality when the children were young, there was a somewhat greater likelihood that, as young adults, these children would have a homosexual identity. However, we do not know what the proportions will be when these youths are older adults. Nor do we know to what extent a mother's serial same-sex relationships pose developmental problems similar to those found among children whose heterosexual mothers have had serial relationships, particularly multiple cohabitations (Dunifon and Kowaleski-Jones, 2002).

As far as worries over child sexual abuse, gay men are no more likely than heterosexuals to abuse children, and the same applies to lesbian women (Jenny et al., 1994). Homosexuality is not synonymous with pedophilia.

This question is not only one that is often asked but it is also at the core of the issue of same-sex marriage. It is a question with legal and moral substance. The answer hinges on the definition of marriage and on the type of marriage that lesbians and gays are seeking. The answer also hinges on the distinction between marriage and cohabitation.

Marriage is an institution which has so far been legally and socially defined as a union between a man and a woman, thus a heterosexual covenant. From my perspective, at least, this portion of the definition can be changed without devaluing marriage as an institution. More importantly, marriage is defined as a sexual, economic, social and emotional partnership involving obligations as well as rights. Furthermore, commitment is a key aspect of the institution of marriage and so is fidelity. When children are involved, the marriage also entails a parental relationship and the rights and obligations assumed jointly by parents to provide and care for their children. Until recently, in all societies of the world, marriage served as the basis for family formation.

However, high rates of divorce, of never-married parenthood, and, as we see below, of cohabitation, may have altered the meaning of marriage for many (Walker and McGraw, 2000): for a segment of the population, marriage has become a private relationship based on personal satisfaction rather than a covenant based on mutual fulfillment and responsibilities (Whitehead and Popenoe, 2001). Some scholars are concerned about the "stripping away of marriage from parenting" (Doherty et al., 2000). In most Western societies, an increasing proportion of male/female couples eschew marriage, at least for a certain time, and opt for cohabitation (Ambert, 2002a, b; 2005).

Cohabitation is seen as entailing fewer responsibilities at the legal, emotional, and economic levels, and is thus perceived to be a freer lifestyle (Wu, 2000). For instance, males who cohabit are less committed to their relationship than are married men (Bumpass et al., 1991). It may be one of the reasons why cohabitants have higher levels of depression than their married counterparts, even after controlling for sociodemographic factors (Brown, 2000). Fewer cohabitants than married persons feel that it is very important to have a lasting relationship in order to achieve a happy life (Milan and Peters, 2003). Cohabitants do not pool their financial resources as often as do married couples. They are also significantly more likely than daters and married couples to be abusive (Brownridge and Halli, 2001; Magdol et al., 1998). In fact, entry into cohabitation heightens a woman's risk of physical abuse (GSS, 2000; GSS; General Social Survey, Sev'er, 2002).

So far, lesbians and gays who wish to live together have been permitted only to cohabit: the door to the institution of marriage has been closed to them. Furthermore, even their cohabiting relationships were not socially sanctioned nor recognized. In the 1970s and the 1980s, the gay and lesbian cultures were most often characterized as countercultures and still are to a great extent--witness the contents of some of the gay pride parades which are a concern for many lesbigays who point out that these tourist-attraction events do not represent the reality of their daily lives.

In the past, lesbigays rejected motherhood and natality as well as traditional family values. Some of the banners read "Smash the Family" and "Smash Monogamy" (Stacey, 1998:17). This rejection of tradition was particularly strident among radical, feminist lesbians. In the midst of this rejection, a movement toward integration within the mainstream family culture emerged, with a focus on stable couple formation and family life--although many homosexuals are still against marriage which they see as a heterosexuel institution (Eleanor Brown, Globe and Mail, Aug. 8, 2002, p. A17).

The movement among gays and lesbians to form families and be recognized legally by both civil and religious authorities as married couples invites consideration of a number of ideological contradictions. Four are mentioned here.

The first contradiction stems from the fact that, although many gays and particularly lesbians are becoming more pro-family and pro-marriage--values that average citizens cherish--this cultural conversion has been largely rejected by the rest of society. Lesbigay families are often stigmatized as abnormal and immoral and parents are denied the right to marry, ironically, at a time when concerns are raised about the decline of marriage itself as a binding force and fundamental institution. One could argue here that allowing those lesbigays who intend to enter into a stable, committed, and sexually exclusive marriage would reinforce the value of the institution of marriage itself (Woodill in Mandell, 2004;200).

The second contradiction resides in that, among family scholars, it is often those who oppose or at least critique the institution of marriage for heterosexuals, because of its perceived consequences for women, who are in favor of homosexuals choosing to marry. This is a somewhat peculiar ideological double standard stemming from leftist values, based on the rights and freedoms of individuals rather than the obligations and responsibilities entailed by marriage, as well as on the rejection of tradition in general.

Often, such arguments see marriage as a patriarchal institution which oppresses women, yet reason that such an oppression does not occur among same-sex partners. We have, however, seen that studies on partner abuse disprove this line of reasoning. Furthermore, the research indicates that both partners harvest benefits in good to average marriages in terms of health, happiness, security, and wealth (Ambert, 2005; Simon, 2002; Waite and Gallagher, 2000).

The third contradiction resides in those who pursue the legalization of same-sex marriage simply as a matter of a civil rights victory. Although it can be argued that, as a group, lesbigays should have the right to marry legally, one does not get married to have rights. It is interesting that one rarely hears of "civil obligations" in any of these discussions.

Thus, one could also be concerned about the possibility that many lesbians and gays might marry, at least at the outset, as a political statement and as a celebration of newly-acquired rights. Such unions would result in extremely high rates of divorce soon after; this situation would lend credence to those who worry about the devaluation of marriage and marital stability were same-sex marriage legalized. Currently, the institution of marriage needs more, not less, stability.

The fourth ideological contradiction leads us back to the initial question about what is essential and crucial to the definition of marriage. It arises when some homosexuals wish to pursue an "open contract," that is, a relationship that includes extra-marital outlets (Sullivan, 1995). This is a contradiction in terms because marriage is meant to be a committed and exclusive relationship. It could be argued that gay men who espouse this definition of marriage should cohabit rather than marry. Demands for marriage on their part lends credence to those who are against homosexual marriages because such unions would weaken many aspects of the institution, particularly commitment and fidelity.

However, same-sex partners who wish to form an exclusive relationship based on love and cooperation do not present this contradiction. In fact, barring the "civil rights victory" possibility mentioned earlier, one could expect that, generally, lesbigay partners who will seek marriage will be a select group who have been together for many years and wish to legalize their commitment to each other and those who are "engaged" and wish to pursue an exclusive and committed relationships. Indeed, the couples whose marriages were celebrated at City Hall in Toronto in 2003 and 2004 generally fit these profiles.

In point of fact, questions of morality are raised about both heterosexuals and homosexuals who cohabit and want the same rights, but not the same responsibilities, as married couples. Thus, it would be important that homosexuals who intend to marry share a commitment to monogamy and fidelity. If not, cohabitation might be more appropriate for them.

It would also be important that, where lesbigay marriages are permitted by law, a similar cultural consensus concerning the unacceptability of same-sex extra-marital infidelities emerge. Otherwise there would be a double standard: one for heterosexual marriage and one for gay marriage. This would mean that homosexual marriages are "second best," (Woodill in Mandell, 2004;200). Furthermore, as implied in the Kurdek studies, lesbigays need social barriers against marital dissolution just as do heterosexuals.

Homosexuality itself presents a moral dilemma for many heterosexual people all over the world and most particularly for the religiously inclined. For some, the problem lies in a belief that homosexuality is abnormal and/or immoral. For others, what is morally reprehensible is the lifestyle that is often presented to the public by a segment of the homosexual population. On the one hand, this moral dilemma has to be recognized and respected. The Catholic Church and various other Churches and religion will not bless homosexual marriages.

However, some North American and European Churches have become more tolerant than the same Churches from Africa, Asia, and Latin America--blocks of countries which will soon form the majority of Christian Churches (Jenkins, 2002). Hence, the matter of homosexual marriage at the religious level is creating a rift within many religions as is currently the case among Anglicans. Furthermore, the matter of homosexual marriage is not even considered in other faiths, such as the Muslim.

On the other hand, at the civil level, in individualistic, democratic societies such as Canada and the U.S., everyone is entitled to his or her religion and moral beliefs about family life. Except for abusing one's children or one's partner, or practicing polygamy, liberal individualistic democracies ensure individuals' rights to live their lives and establish their relationships according to their personal beliefs. But this democratic right of individuals as family members also presumes an obligation: the tolerance of the different ways and values of other families, and this includes same-sex-parent families and couples.

As is obvious throughout this paper, far more research--and far more sophisticated research--is required on all aspects of the lives of same-sex parents and their children (Patterson, 2000). To begin with, we do not have systematic studies on parenting comparing lesbian with gay parents and those who have male versus female children. While some studies exist on the children born and raised in same-sex families, we do not know enough about adults who grew up in such families--as comparatively few such adults exist.

On another level, there is practically no literature including the important variables of class, religion, and ethnicity (but see Cantu', 2001; Yip, 1997). While there is an extensive literature on the negative impact of poverty on children, well into their adulthood (Ambert, 1998), there is no comparable information on poverty and homosexuals' children and young homosexuals themselves. Are gay adolescents who live in poverty affected differently? Is their life course more difficult than those who originate from the middle class? And how does this compare to class differences among heterosexual adolescents?

Furthermore, certain religions and ethnicities are particularly adverse to homosexuality. Is it more difficult to be a young black lesbigay than a white one, for instance? Are relationships with parents/siblings easier or more difficult than in white groups? Are black or Native homosexuals more or less likely than whites to live in exclusive and stable couples? Are gay couples more or less likely to be racially mixed? Above all, are nonwhite gay couples more likely to become parents or less so?

Generally, only the biological parent has any legal right to the child born to or brought into a same-sex union. After a break-up, the other partner who may have been very attached to and involved with the child may not even have visitation rights. The child is thus suddenly cut off from half of his or her parental unit, unless the two former partners are able to find an amicable solution. However, there are no indications that homosexual break-ups are friendlier and more harmonious than heterosexual ones; therefore, the lack of a legal status for the non-biological parent is a concern in the homosexual community. "When there are no clear guidelines, what often substitutes are conflict within the ex-couple" (DeAngelis, 2002).

This concern leads me to underline how little we know about separation among homosexual couples who have children living with them. How often does it occur? How does the process differ from that of heterosexuals? We already know that male homosexuals are less sexually exclusive, which, among heterosexuals is a risk factor in terms of relationship stability. But this does not tell us whether this applies to gays who go through the trouble of fathering. Still, it could be argued here that male couples might be at a greater risk of breaking up than male/female couples and perhaps than lesbian couples. It could also be argued that lesbians who bring one child into a relationship may have a high rate of union dissolution, as is the case among remarriages (Ambert, 2002b).

We have no research data on the effect of partner separation on children raised with same-sex parents. This is crucial in view of the large literature indicating some negative long-term effect of parental divorce on children (Ambert, 2002b), including higher rates of divorce later on (Amato and DeBoer, 2001).

Furthermore, how will a lack of biological relatedness to one parent impact on that child's relationship with this parent after divorce? We have a substantial research literature on the impact of parental absence after divorce (e.g., Sun and Li, 2002) but not for children of same-sex-parent households. For instance, we know that divorced fathers, on average, do not see their children often, and more so in the U. S. than Canada (Ambert, 2001b). Paternal disinvestment or the disinvestment of one parent is a costly problem and there is so far no reason to think that gays would be more conscientious in this respect.

Little attention has been paid to reconstituted lesbian-headed families and the stepparent role that then occurs (Parks, 1998). In these stepfamilies, there are unique obstacles to overcome because social support is lacking (Lynch, 2000). The lesbian mother's partner is less involved with the children than is the partner of a donor-inseminated woman; in the latter case, the partner often becomes the "other mother," although her legal status is not recognized (Nelson, 1996).

Another gap in our research is the question of repartnering. In point of fact, we also know little about this for heterosexuals who break-up after a cohabitational relationship. Another important question concerns future relationships of lesbigay parents' heterosexual children: will they be as commmitted to their partners and spouses later on or will they experience a higher rate of conjugal dissolution?

Then, as the life course of lesbians who already have adult children unfolds, they will become grandmothers. One would expect here that their role will differ very little from that of heterosexual grandmothers. At the same time, it would also be important to know more about grandparental relationships. Do lesbigay families receive less help from their children's grandparents?

One major drawback of the literature on same-sex domestic violence is that it does not study its effects on children. On the one hand, it seems reasonable to assume that same-sex couples who are involved in spousal abuse are less likely to have children than similar heterosexual couples. Cohabitation is more fragile than marriage and same-sex couples who do not get along and quarrel violently probably separate even before the thought of having children occurs to them. Moreover, reproducing children is a far more complicated project to carry out among homosexuals than among heterosexuals. It is not something that happens accidentally.

Therefore it is possible that a processus of selection is at work: those who decide to have children and are then able to achieve this goal may be a select group of more stable and devoted couples than are average heterosexual couples. Consequently, they would not fall prey to deviances, especially spousal abuse, that could endanger their relationship and their children's well-being. This is a hypothesis which deserves scrutiny. On the other hand, as seen earlier, a great proportion of children with one homosexual parent were born to that parent's previous union. That parent's new partner becomes a stepparent and such relationships may present a greater risk of domestic violence and child abuse, as mentioned by Renzetti (1997a).

And, finally, research should begin now to observe and monitor the consequences of the legislative changes that have already been introduced in Canada, the U.S., and elsewhere to provide legal acknowledgement of same-sex couples and same-sex parents raising children. In particular, if research on the implementation and consequences of domestic partnership registries, civil unions, and same-sex marriages introduced recently in various jurisdictions is not initiated now, a significant opportunity will have been missed.

The research questions outlined above are of intellectual and human interest in the fields of the sociology and psychology of the family. But equally to the point, research data would be very useful in terms of welfare and social policies as there is currently too little information that can guide policy makers, social workers, clinicians, as well as teachers and parents to understand the family strengths and weaknesses of same-sex couples and their children. These families are in many respects pioneering a "new" way of life and may need more social support than others. This social support should be forthcoming immediately, disregarding the outcome of the issue of same-sex marriage: married or not, same-sex-parent families exist now.