Sunday, May 01, 2005

What the Science Says

My post on marriage and language sparked a stellar debate with "Sally" in comments over a wide gamut of issues concerning gay rights. It concluded when I agreed to read some of the critical evaluations of the scientific evidence I had been heavily relying on. Specifically, I decided to read Brigham Young Law Professor Lynn Wardle's article "The Potential Impact of Homosexual Parenting on Children" (1997 U. Ill. L. Rev. 833). While I think he scored some points, ultimately I found his critique unpersuasive. Allow me to explain why.

First of all, I think Professor Wardle made a valuable addition to the literature in this article. We shouldn't accept scientific "findings" unquestioned, and Sally was absolutely right when she reminded me to be particularly cautious when said findings uphold our own deeply held beliefs. The other important thing to note is what Professor Wardle is not arguing: he does not believe that homosexuals should be excluded as a class from adopting or gaining custody over children. In addressing two arguments for allowing homosexual adoption of children, that "public policy should encourage formation of families, even nontraditional same-sex parenting families, because two parents (even of the same gender) are better than one," and "parenting by an adult who is engaged in a homosexual relationship may be the best option for a particular child," he admits that these
two policy arguments are variations of a common theme - that in the world we live in, persons who engage in extramarital sexual relations, including homosexual relations, may, in some cases, provide the best available parenting alternative for particular children. This argument recognizes that all parents are imperfect and acknowledges the tremendous failures of the alternatives to parenting (e.g., the abuse of institutional care, the limbo of foster care, and the tragedies of other governmental substitutes for in-home parenting, the disadvantages of single-parenting). It is hard to disagree that, despite the possible harms children may suffer from the extramarital sexual behavior, including homosexual behavior, of their parents, if the best interests of children is followed in all cases, there undeniably will be, and have been, some cases in which a parent engaged in extramarital sexual behavior will offer, and has provided, the least detrimental practical parenting alternative.

Thus, this article does not quarrel with the last two principles above or that there are and will be in this imperfect world exceptional cases in which less-than-ideal parents and less-than-perfect custody and adoption arrangements are the best options for a particular child or children. (842-43)

I do disagree with Wardle's rhetoric on the matter. Given the state of our adoption system, I doubt such "imperfect" cases are all that "extraordinary," nor do I feel that, even if there homosexuality is an "imperfection," it would constitute a negative factor more grievous than many other more typical negative factors that afflict heterosexual families. I also think it is facile to group a monogamous homosexual relationship in the same category as extramarital sexual behavior. Though technically this is accurate, it is a product of legal rules beyond the control of the parties (many whom would surely marry if given the chance). I hardly think the same harms apply to children in families with extramarital affairs as with those living in families with committed homosexual affairs (which yes, are technically "extramarital"). This conflation is a critical flaw throughout Wardle's article, but we'll get there later. Suffice to say, Wardle does not advocate categorical exclusion for homosexuals gaining custody of children (as is being advocated in Texas, was the subject of the Howard decision in Arkansas, and remains the law in Florida), but rather a rebuttable presumption against them, and a relatively weak one at that:
States should adopt a rebuttable presumption that ongoing homosexual relations by an adult seeking or exercising parental rights is not in the best interest of a child. The presumption should be rebuttable by a mere preponderance of the evidence. Unless rebutted, this presumption should be considered as one of the many nonexclusive "best interests of the child" factors that are weighed by courts deciding parenting issues. This presumption should apply in any case involving significant parental relationship with the child, including custody, adoption, and visitation.


It was also distressing to see Professor Wardle quote the profoundly discredited Dr. Paul Cameron. I do not know the credentials of most scientific experts opining on this issue; my default setting is to assume they are credible unless shown otherwise (for both sides). Dr. Cameron, however, is someone I do recognize as basically an ideological hack, kicked out of the APA and condemned by the ASA for consistently misrepresenting sociological research. Although Wardle only cited Cameron six times, it still reflected poorly on his judgment as a researcher and scholar.

Wardle does score some points when criticizing methodological flaws in the pro-equality research. Part of this, I feel, is due to the small availability of subjects--there just aren't that many homosexual parents available to study. More generally, however,
"the types of methodological flaws that Wardle discusses in his article (such as small sample sizes and lack of diversity in the samples) are by no means limited to the study of gay and lesbian families. [...]producing a representative probability sample can be expensive and difficult. Respondents chosen randomly will typically be diffused across the entire area inhabited by the population, rather than concentrated in areas easy for the researcher to access...
[...]
[furthermore,] there are ongoing efforts to address some of the methodological problems noted by Professor Wardle." [Carlos A. Ball, Janice Farrell Pea, "Warring With Wardle: Morality, Social Science, and Gay and Lesbian Parents," 1998 U. Ill. L. Rev. 253, 274-76]

As Wardle's article is from 1997, I'd assume there have been some advances in the literature which may shed light on these questions (any links to them would be greatly appreciated).

However, while some of his accusations are legitimate, some of the objections are rather silly. Wardle says that many of the homosexual families studied were "model" families, well-educated, upper middle class, with stable incomes, while "poor...uneducated, or troubled" homosexuals are left out. However, the point of these studies is to determine whether or not homosexuals, in absence of other disabilities which might affect their ability to be good parents, are possessed of inherent problems which might disable them from adopting or gaining custody over children. In this respect, it makes more sense to try and isolate the variable and eliminate other factors which are not unique to homosexual families at all (IE, "troubled" persons are likely to have issues with children regardless of their sexual orientation).

One consistent and problematic aspect of Wardle's critique is his constant conflation of homosexuality with another, unrelated relational issue. For example, he writes that:
Finally, it is reasonable to be concerned that ongoing parental homosexual sexual behavior is harmful to children because that seems to be the lesson of the most relevant and analogous human experience - the experience of extramarital sexual relations generally. (855)

But as I wrote above, this analogy isn't relevant at all--and it's Wardle who inadvertently indicates why:
Extramarital sexual behavior is associated with such harm to children as the breakup of their parents' marriage and the destabilizing, child-harming consequences of divorce. Parental extramarital relationships wound children, shaking, sometimes even destroying, their faith in marriage and in personal commitments of fidelity and intimacy. It hurts a child to learn that one parent has been unfaithful to the other. That pain is very real and very wrenching. Parental extramarital relationships also provide a dangerous model for children, serving to pass intergenerational self-destructive behavior on to children. The message of intergender and intergenerational carelessness, and family-sacrificing selfishness not only hurts, but also may have a programming effect on children. The lesson of sexual self-gratification at the expense of familial fidelity conveys a tragic message about both family commitments and responsible sexual behavior in our society. In these days of so many harmful, even deadly, sexually transmitted diseases, the risks may be physical as well as emotional. (855-56)

Not a single one of these problems apply to committed and monogamous homosexual relationships which a) do not break up a marriage, b) do not undermine the principles of loyalty and commitment c) do not provide a dangerous model for children (except, perhaps, insofar as they legitimize homosexuality, which would be a circular argument) d) are not any more "selfish" than any other committed relationship and e) do not have any raised risks of STDs. Of course it is true that for unfaithful homosexual couples, these issues do come into play--but then, they also come into play for unfaithful heterosexual couples. I would argue that the legal system already has the tools to evaluate and dispose of these cases.

He makes a similar slight of hand when talking about the need for dual-gender parenting. Virtually all of his research is a comparison to single-parent homes, not dual-parent homosexual homes. This is critical for several reasons. First of all, much of his argument focuses on issues are more acutely felt by single-parent homes than two-parent homosexual homes, but are theoretically overcomable, such as time availability and economic wherewithal. Second, there are inherent problems within single-parent homes that simply don't exist in homosexual houses--not that he doesn't try and connect them.
But the significance of the disadvantage of growing up without both father and mother in the home should not be trivialized, either, especially in this difficult time of our social history. Individually, the child must cope with the loss of example, counsel, and experience that living with the missing-gender parent would have provided, and must overcome the sense of loss, abandonment, and deprivation (often feelings of devaluation, inadequacy, and anger) caused by loss of the second-gender parent. Socially, the community may be burdened, not just to provide some quasi-parental experience with missing-gender adults for the growing child to compensate for what the missing-gender parent would have provided, but also to bear the adverse behavioral consequences often associated with single-gender parenting. (863-64)

Note how he shifts seamlessly from the experience of most single-parent homes (where the "loss" is literal, in the form of death or abandonment), to the experience of most homosexual homes, where the loss is more abstract and never ingrained in the experience of the child.

Never is this false equivalence made more clear than in this passage:
Children make the transition through developmental stages better, have stronger gender identity, are more confident of themselves, do better in school, have fewer emotional crises, and become functioning adults best when they are reared in two-parent, dual-gender families. As family sociologist David Popenoe puts it:
social science research is almost never conclusive... Yet in three decades of work as a social scientist, I know of few other bodies of data in which the weight of evidence is so decisively on one side of the issue: on the whole, for children, two-parent families are preferable to single-parent and stepfamilies.

As even a cursory reread of Popenoe's remarks indicates, he says nary a word about dual-gender homes, merely dual-parent homes compared to single-parent and step-parent families. For the reasons stated above, neither family formulation is comparable to a committed, monogamous homosexual relationship.

There are quite a few other, minor arguments in Wardle's article, and I also have my own minor objections to them. But I think I've hit upon the meat of the argument presented. As I said, Wardle does score a few points, regarding some methodological flaws in the scientific literature presented. Ultimately, however, I find his own position to be marred by its own flaws which severely undermine his advocacy.

2 comments:

alex said...

Do you know if there is a copy of Wardle's article online somewhere?

I'm a-priori skeptical about the claim that many of studies on homosexual parents have methodological problems. Roughly speaking, every study has methodological problems: its never as clean in the real world as it is in a textbook thought experiment. There are, of course, methodological problems that ought to disqualify a study completely; but these articles have undergone peer review designed to make sure there are no major issues. I'd be curious to learn what kind of criticisms Wardle makes.

David Schraub said...

I got Wardle's article on LexisNexis (I love being in college), so I don't know if it's available online in another form. Sorry.