Fall 2012

Beyond Biology? Gay Parenting and the Conjugal Ideal

Paul Sullins Download Article

Bos, Henry, Parenting in Planned Lesbian Families (Amsterdam University Press, 2004).

Brodzinsky, David M. and Pertman, Adam (eds), Adoption by Lesbians and Gay Men: A New Dimension in Family Diversity (New York: Oxford University Press, 2011).

Tasker, Fiona and Bigner, Jerry , Gay and Lesbian Parenting: New Directions (Binghamton, NY: Haworth Press, 2007).

Conjugality requires the joining of complementary sex organs by persons properly disposed to do so. Absent such con-joining (the root of con-jugal), according to most natural or religious accounts, men and women lack the particular character of relationship that is necessary for the mutual fulfillment of erotic love and – which is the point at issue in this review – optimal for the raising of children.

On this view, of course, homosexual partners, who lack conjugality, cannot raise children as competently as heterosexual partners who possess it. As a recent Catholic teaching document expresses: “the absence of sexual complementarity in [homosexual] unions creates obstacles in the normal development of children” such that “[a]llowing children to be adopted by persons living in such unions would actually mean doing violence to these children.” Moreover “the possibility of using recently discovered methods of artificial reproduction… does nothing to alter this inadequacy.” [1]

The three books reviewed here all take strong exception to this claim. In different ways each advances a rapidly-growing body of social science studies designed to show that there is no discernible difference in well-being between children raised by homosexual partners and those raised in heterosexual households. Together they offer about a dozen new studies and, in well-written review chapters, summarize the growing stream of several score empirical studies produced on this topic since the 1970s. With few exceptions, these studies follow a common form: each compares some set of homosexual families with a set of heterosexual ones on various measures of child adjustment or well-being, and offers the conclusion that there is no difference between the two sets of children.

Bos’s careful Dutch study (hereafter “B”) avoids overt policy claims; but not so the edited collections by Tasker and Bigner (hereafter “TB”) and Brodzinsky and Pertman (“BP”), which move quickly from research to advocacy. Both of these begin with a foreword by veteran gay researcher Susan Golombok which confidently asserts that “research on adoption and… on children born through assisted reproduction… shows that genetic relatedness to parents is less important for children’s psychological well-being than are positive family relationships” (BP, p. vi), and that “the presence of a father or heterosexual role models is not essential for children’s development,” nor does it matter “whether the mother is lesbian or heterosexual” (TB, p. xx).

For these researchers, any debate over biology and the family is over. BP declares: “We have moved from a view of the family in which structure is what matters – the presence of a mom, a dad, and their biological children – to a greater understanding of the importance of the quality of family relationships for children’s psychological well-being (p. vi). There is no attempt at even-handedness or objectivity in the face of this “unprecedented transformation” in Western family life.

In both TB and BP, as their subtitles suggest, there is a breathless sense of breaking new ground, overcoming prejudice and securing long-denied rights. TB begins with one of the editors describing his own experience coming out as a gay parent; BP concludes with several chapters advocating legal and policy reform, including best practices for gay-friendly adoption agencies.

Gay “marriage” and gay adoption are, of course, mutually reinforcing reforms (BP, p. 22), since if gays are raising children as well as straights, there is no conjugal advantage to heterosexual relationships to justify exclusively heterosexual marriage. Continued resistance to abandoning conjugal or natural biological relations in families is dismissed as lingering prejudice. Indeed, the research represented in these volumes has been widely accepted as definitive, and a basis for discontinuing social and legal restrictions on gay parenting, by most scholarly, legal and child welfare agencies.

Notwithstanding this wide cultural acceptance, from a perspective that takes conjugality seriously, and by the ordinary canons of scientific evidence, this research is deeply flawed. The remainder of this review will elaborate this claim by a close examination of some of the scientific deficiencies, then of the philosophical blind spots, evident in the literature presented in these volumes.

Science or Propaganda?
Despite the aforementioned endorsements, and its impressiveness to non-specialists, including many judges, this is not a serious body of scientific research. Its methodological flaws alone, detailed elsewhere,[2] preclude any claim of objectivity. Almost all included samples of gay parents are comprised of persons recruited from pro-gay parenting organizations or friendship networks (selection bias); are far too small for reliable statistical inference (generally fewer than 50 cases); and/or rely on parents’ self-assessment of their own parenting practices and children’s well-being (attribution bias). When one asks a small group of gay parenting advocates how their own children are faring, reasonable persons may be permitted to doubt the objectivity of the answer. Exclude these three sources of bias and ninety percent of this literature is eliminated. Exclude studies by researchers who are themselves gay, funded by a pro-gay-parenting organization, or have indicated a prior normative support for gay parenting, and there is nothing left.

The fact is that, despite dozens of studies announcing there is “no difference” in child outcomes, two serious limitations preclude credibility for such claims. First, a reliable inferential sample of gay fathers (as distinct from gay mothers) has yet to be produced. As the Henehan et al. chapter in TB notes: “Only one study since 1978 has recruited more than 40 gay fathers; all others used smaller samples.… the only empirical study to compare gay fathers with non-fathers sampled 30 gay fathers, all of whom were members of a Catholic parenting group.” Almost all the research on gay parenting, in other words, has examined lesbian parents. Studies that do include gay male parents, as well as a large body of research on heterosexual families, have found significant gender differences in parenting practices and experience. But study conclusions that do not examine gay males at all are routinely, and improperly, extended to gay male couples, stating “no differences” for children raised by “gay parents” or “GLBT parents.” Second, no research has followed child outcomes into adulthood. Almost all of the research examines children under age 12; very little looks at adolescents, and none at outcomes past age 25. Research on all types of families, for example divorced, single parent, or blended families, has found that the effects on children of parenting and family form persist, often powerfully, well into early adulthood and throughout the life cycle. Yet whatever anyone can claim to know about the well-being or adjustment of children raised by homosexuals relates only to transient conditions during childhood. Despite repeated broad claims, we simply do not know what will be the effects on children of having homosexual parents.

Finally, this literature lacks a key feature of genuine scientific discourse: the willingness to disclose and honestly address opposing views and difficult facts.[3] All studies in this literature ignore basic facts which a reasonable person might find relevant to parenting effects, particularly adoption policy. Space permits only a few examples. Never mentioned is the well-attested fact that gay relationships are much less stable than heterosexual ones, even where social stigma for gay unions is lacking. For example, data on Scandinavian gay marriages, which have existed since 1989, show that lesbian marriages are more than twice as likely to end in divorce as are heterosexual marriages; nearly a third of lesbian marriages (but only one in eight conjugal marriages) in Scandinavia divorce within five years.[4] The fact that the US Centers for Disease Control report that nearly one in 5 gay males, versus less than one in 300 heterosexual males, are currently HIV-positive,[5] or that repeated research finds that intimate-partner violence and emotional abuse is twice (or more) as prevalent among gays as among heterosexual partners,[6] is not deemed relevant to parenting or adoption prospects in this research. Ignoring obvious difficulties such as these is a characteristic of propaganda, not genuine scientific discourse.

It is, of course, possible that these methodological weaknesses could be overcome in future research. A deeper problem is presented by the anti-conjugal assumptions that animate the composition of the research. These are not likely to change, as they express a more profound philosophical rejection of the human body and of the conjugal family that is pervasive in the culture from which this body of research springs. Due to their complex character, these assumptions are best made visible by means of a couple of extended examples.

Better Children through Planning
B, methodologically among the best studies in this literature, nonetheless illustrates the problem of pervasive and debilitating non-conjugal assumptions, in this case about wantedness and fertility. The author Henny Bos, a Dutch researcher, matched 100 lesbian couples who conceived through assisted reproduction with 100 heterosexual couples in order to compare their parenting characteristics and children’s outcomes, a method that should magnify relevant differences between the two groups. The lesbian participants were chosen from population registers, not self-selected, and were matched with heterosexual couples virtually identical to them in socioeconomic characteristics and relationship duration or stability.

A central issue examined in this study is the motivation and desire to have children, “because the desire and motivation to have children are assumed to affect parenting and the parent-child relationship” (B, p. 9). Bos found that the lesbians planned and desired their children more intently than did heterosexual parents. On a 6-point scale measuring the intensity of desire for a child, lesbian biological mothers scored higher than heterosexual mothers by about a fifth (19.2%) and higher than fathers by over a quarter (28.6%).

The hypothesis here is that lesbians conceiving through assisted reproduction are likely to be even better parents than heterosexual couples because they are more invested in becoming parents in the first place. A related study co-authored by Bos explains that lesbian mothers using assisted reproduction have a

“commitment even before their offspring were born to be fully engaged in the process of parenting. During pregnancy, the prospective mothers took classes and formed support groups to learn about childrearing. They were actively involved in the education of their children and aspired to remain close to them, however unique their interests, orientations and preferences may be. To the extent that the [lesbian] mothers may have achieved this goal, numerous studies showed that having a satisfying relationship with one’s parents is associated with a more favorable adolescent adjustment.” [7]

Here the cultural norm of parental subjectivity is fully operationalized to the displacement of the body. In a culture where “wantedness” confers value on a child, lesbian children who, as the study’s title reflects, are all “planned,” are deemed likely to have better parenting than those of heterosexual partners, who are often not planned. The anti-conjugal implication is that assisted reproduction must be superior to sexual relations because in the former the child is more intently wanted. Humans often have sex, as we all know, for other reasons than a strong desire to produce a child.

Remarkably, the effect of assisted reproduction on intensity of desire is not taken into account for the heterosexual mothers, despite the fact that the study repeatedly acknowledges that, with respect to the very variables on which the author later finds differences, “lesbian women are in a situation comparable to that of infertile heterosexual couples” using assisted reproduction. Both lesbian couples and infertile heterosexual couples, the author notes, “have to go through a long and difficult process before they finally get pregnant, and the decision to have children is not taken easily,” hold “the expectation that parenthood would provide life-fulfillment [to be] considerably more important,” and “spend more time reflecting on their desire to have a child than fertile heterosexual couples do” (B, p. 45). Notwithstanding this, in assessing intensity of desire for a child, the study compares lesbians engaged in artificial reproduction, not with comparable heterosexual couples using assisted reproduction, but with heterosexual couples who conceived naturally. The result is that the central finding that there is a lesbian/heterosexual gap in desire for a child is rendered spurious, an artifact of this choice.

In a study so otherwise careful in its methods, this rather obvious weakness is never even acknowledged. The author seems blind to the problem. Like the emphasis on wantedness, this blind spot also reflects a larger cultural blindness about sex. By comparing lesbian and heterosexual mothers while ignoring the fact that the former are reproducing artificially while the latter are reproducing naturally, the methods employed in the study embody the assumption that will become the study’s conclusion, that sexual orientation is indifferent to fertility. On this view, the true distinction between persons is who they choose to have sex with, not whether they are able to beget children.

By contrast, a conjugal view affirms that fertility is not an added quality of human sex relations, but is intrinsic to them; and a lack of fertility is always to some extent tragic. Conjugal awareness thus subordinates human sexual relations to the possibility of offspring. Openness to a child unbidden – a prospect that no lesbian couple can face – is at the heart of conjugality. The entire relationship is shaped by that ideal possibility – is open to life – by dispositions to permanence, fidelity, and unreservedness, as expressed most clearly in the cultural norm of conjugal marriage. Such dispositions, as the coming of children themselves, do not restrict but rather enable sexual fulfillment.

Lesbianism and infertile heterosexual partners are similar, this study does not have the categories to see, because lesbianism is a form of voluntary infertility. From a conjugal point of view, it is not surprising that the intent to parent a child should be heightened among human mating partners who cannot conceive, or that lesbians should turn to artificial reproduction. Having sex that cannot produce children and begetting children without sex are twin behaviors, which both deny the intrinsic link between sex and fertility.

In several other ways, the complications involved in lesbian planned reproduction, far from overcoming biological limitations, actually underscore the link between sex and biological fertility. Among lesbian couples, the conceiving “biological mother” is typically much more attached and invested in the child than is her partner, the “social mother.” The only difference, of course, is that the former is genetically related to the child. While biological mothers desire a child more intensely than do naturally-conceiving heterosexual mothers, social mothers do not. Throughout her study the author compares lesbian social mothers to heterosexual fathers, not to heterosexual mothers, an implicit acknowledgement that the lesbian partner relationships mirror in some ways those of genetic parents. In all respects in which the social mothers differ from the biological mothers, they do so in the direction of heterosexual fathers. But, while fathers are also less intensely desirous of children than are their partners, on most measures the difference between heterosexual mothers and fathers – both genetically related to the child – is much smaller than that between the lesbian partners.

At the most fundamental level, this study falsifies any claim of “no differences.” The very results designed to demonstrate no parenting deficits for gay parents cannot avoid underscoring the dramatic differences between gay and conjugal parents in which biology is clearly implicated. While dutifully reporting that the study found “no differences between the psychological adjustment of children in lesbian and those in heterosexual families”, the study concludes, surprisingly, that there are nonetheless very real, structural differences between heterosexual and lesbian families. In the words of the author: “In sum, the findings show that lesbian families differ from heterosexual families in several respects.… The observed differences do not originate solely from gender, the position of lesbian social mothers in society, and/or the absence or presence of a biological link. It is revealed that these variables create a different kind of family structure that has consequences for the quality of the parent-child relationship.”

Interchangeable Findings
In addition to assisted reproduction, gay couples also acquire children from one of two sources: a former heterosexual relationship, or by adoption. Gay adoption research, the focus of BP and a major theme of TB, consistently concludes that children adopted by gays, including gay males, show “no difference” in well-being compared to those adopted by heterosexual couples and, as aforementioned, calls for an array of supporting reforms moving beyond the biological family.

Underlying these global pronouncements is a startling paucity of actual information. To date there have been only two studies, using subjective measures with small non-random samples, of children adopted by gay couples. Conclusions on gay adoptions, therefore, are based on studies of the adjustment and outcomes, not of children adopted into gay families, but of the natural offspring of gay parents conceived in prior heterosexual relationships. This procedure is justifiable, indeed conceivable, only on assumptions that already reject the importance of conjugality in family relations.

A central chapter in BP by Patterson and Wainwright, two veteran gay parenting scholars, illustrates the problem. To examine the question, “Should the sexual orientation of prospective adoptive parents be considered as a factor when making placements of minor children into adoptive homes?” they adduce data from the National Longitudinal Study of Adolescent Health, a large and respectable government-funded body of data on adolescents (BP, p. 85). It certainly seems that this dataset could provide valuable information on the outcomes of children adopted by same-sex parents; except that we learn, as the article proceeds, that there are no such children. The authors found, out of over twelve thousand cases, only fifty-two adolescents raised in same-sex families. Only two of these were adopted; just six were raised by male same-sex couples. Because two and six are far too few for statistical comparisons, these eight cases were excluded, and the authors proceeded to analyze, and draw conclusions for gay adoptions, from the remaining forty-four adolescents in gay households, none of whom were adopted and none of whom were being raised by gay males.

In any other area of social science, at this point the research effort would end for lack of evidence, with judgment on the underlying question being deferred until more information was available. But Patterson and Wainwright inform us that “data about the development of children who have been born to lesbian or gay parents and who are being reared by them” are “relevant” to the question of gay adoption (BP, p. 86). Quite how they are relevant is not made clear; what is clear is that the home experience of such children is not exactly comparable to adoption. At the time a child born to a gay parent was conceived, and in many cases for several formative years, the gay parent involved was in a heterosexual relationship, not a homosexual one; and all such children still have a noncustodial heterosexual parent who contributes to the childrearing situation and typically provides additional financial resources. These factors cannot be present in any adoption by two homosexual partners. To say that the experience of these children is relevant to those adopted by gay partners is like saying we can assess the outcomes of heterosexually-adopted children by looking at step-parent families instead of at adopted children.

The authors’ judgment that parenting outcomes are transferable between these two types of relationships which are structurally very different leads directly to their conclusion, which is that sexual orientation does not matter for child well-being, since the prior judgment precludes consideration of some important ways in which it might matter. Both judgment and conclusion reflect the assumption, universal in this research, that persons in family roles are fundamentally interchangeable by such biological matters as gender, sexual orientation and (in this case) genetic relatedness. This denial of the body is to some extent self-fulfilling in the celebrated diversity of family forms that devolve from the conjugal ideal.

Conjugality and Human Diversity
Sexual complementarity proposes, at minimum, a lower limit to the interchangeability of parents. If humans are not complementary by sex, then many types of family arrangement are possible; if they are complementary, then family types are much more limited; and to the extent that complementarity does not merely affirm the binary character of the sexes, but is a condition for expressing other levels of human uniqueness and irreplaceable worth, the range of acceptable family types becomes even more limited. If sexual difference “goes all the way down” in human ontology, then there can be only one proper family type. Many shapes can be constructed from the rearrangement of uniform blocks; but a puzzle with many unique pieces has only one solution.

To the extent humans are interchangeable in sexual and family roles, it matters little whether a child has two mommies or two daddies, or for that matter whether her parents are permanently committed, change partners or she changes parents. In past decades, in fact, all of these claims have been put forward at one time or another. Research has demonstrated definitively that impermanency (cohabitation), changing partners (divorce), and changing parents (remarriage or step-parenting) all inhibit the child’s well-being when compared to conjugal marriages. It is likely, though we do not yet know for sure, that the same will be true of single-sex partners.

The fundamental flaw, not just of the particular studies reviewed, but of the entire body of gay parenting research, is its blindness to conjugality. The cause of this blindness may be philosophical, but the result is expressed in a very practical effect. Remarkably, despite (or perhaps because of) the fact that in virtually every study ever performed on families a strong, stable marriage has been found to be the most powerful predictor of child well-being, not one study of gay and heterosexual parents has yet separated out married heterosexual couples from the rest. “Homosexual” families are simply compared with “heterosexual” ones – lumping together married, cohabiting, blended, and sometimes even single parents – to conclude that there is “no difference” in child outcomes. Gay parenting appears to make “no difference” for children only because the culture in which the practice is now increasingly approved has lost sight of what the difference that it makes may be.

This confused procedure, ironically, establishes beyond doubt that there are substantial differences in child outcomes between gay families and married ones, because we already know that there are significant differences in child outcomes between married families and other non-conjugal heterosexual family types. That gay parents raise children that are “no different” from most heterosexual partners doesn’t demonstrate the adequacy of gay parenting, but simply confirms the deficiency of most heterosexual pairings, which have long ago left behind the outdated and restrictive notion of conjugality. In a world where most children are raised in relationships that are already conjugally deficient, it may well be that having homosexual parents does no further or greater damage. The question in confronting gay parenting is not simply whether gay parents are acceptable, but whether such a world is acceptable.


[1] Joseph Ratzinger and Angelo Amato, Considerations Regarding Proposals to Give Legal Recognition to Unions Between Homosexual Persons(Congregation for the Doctrine of the Faith, 2003), sec. 7.

[2] Lynn D. Wardle, “A Response to the ‘Conservative Case’ for Same-Sex Marriage: Same-Sex Marriage and ‘the Tragedy of the Commons,’” BYU Journal of Public Law 22: 2 (January 2008), pp. 441–74; Wardle, “The Potential Impact of Homosexual Parenting on Children,” and cf. Carlos A. Ball and Janice Farrell Pea, “Warring with Wardle: Morality, Social Science, and Gay and Lesbian Parents,” University of Illinois Law Review (1998), p. 253.

[3] Wardle, “The Potential Impact of Homosexual Parenting on Children,” p. 838, similarly notes that “the willingness to honestly state opposing positions, to meet those arguments directly, and commitment to the fair-and-vigorous-exchange-of-informed-opinions ideal of… scholarship is generally absent from most of the current law review literature addressing same-sex marriage and homosexual parenting.”

[4] Gunnar Andersson et al., “The Demographics of Same-Sex Marriages in Norway and Sweden, ” Demography 43: 1 (February 2006), pp. 79–98.

[5] U.S. Centers for Disease Control and Prevention (CDC), “HIV Surveillance – United States, 1981-2008,” (n.d.), http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6021a2.htm.

[6] Government of Canada, “Measuring Violence Against Women: Statistical Trends 2006,” Statistics Canada Catalogue no. 85-570-XIE (2006), p. 39; Susan C. Turell, “A Descriptive Analysis of Same-Sex Relationship Violence for a Diverse Sample,” Journal of Family Violence 15: 3 (2000), pp. 281–93; “Silence Ending About Abuse in Gay Relationships – New York Times,” New York Times (November 6, 2000), http://www.nytimes.com/2000/11/06/us/silence-ending-about-abuse-in-gay-relationships.html?pagewanted=all&src=pm.

[7] N. Gartrell and H. Bos, “US National Longitudinal Lesbian Family Study: Psychological Adjustment of 17-Year-Old Adolescents,” Pediatrics 126:1 (June 7, 2010), pp. 33–4.