Allen and "Essentially Different"

Recently, Karen at Family Scholars Blog posted a link to an interview with Douglas Allen that she found interesting. Allen, according to his article's bio, is an "expert on the economics of social institutions."

Like Karen, I also found the interview, examining the "economics of same-sex marriage," interesting. I like the idea of measuring (or trying to measure) the effect of a social policy in terms of its costs and benefits. The devil, of course, is always in the details, however.

It is not an easy task to, for instance, measure the cost (if any) to existing married couples of allowing same-sex couples into the institution of marriage. And, in the interview, Allen does not clearly or adequately articulate his quantitative (or qualitative) methodology in a manner that I find convincing.

In fairness, that's probably because the piece is in the form of an interview, rather than an article, an academic work, or even a blog post. What I found problematic, even given the medium, is that Allen made several contentious claims about "how heterosexual, gay and lesbian [*sic] relationships are essentially different," and he did so without adequate supporting evidence.

"Essentially." That word has a specific meaning and, unfortunately, Allen and his interviewer were using it erroneously. To say that a group has an essential characteristic is to say that that characteristic is a necessary, indispensable characteristic, upon which a person's definition as being a part of that group depends. An essential characteristic is a characteristic that is, by definition, shared by all members of the group.

Thus, to say that heterosexual, gay, and lesbian relationships are "essentially different" is to say that all heterosexual relationships share essential characteristics that they do not, indeed cannot by definition, share with "gay or lesbian [*sic]" relationships. Likewise it is to say that "gay and lesbian [*sic]" relationships share essential characteristics (and differences) that they do not share with heterosexual relationships.

(*Note: Notice how Allen and the interviewer refer to same-sex relationships as "gay and lesbian relationships." Aren't bisexual and queer-identified people sometimes in "gay and lesbian" relationships too? Does sexual orientation matter? Do people "care" whether a person is gay or lesbian? Read on to find out!)

From there, answering the question as to how heterosexual and "gay and lesbian" couples are "essentially different," Allen goes on to attach further importance and distinction to the gay and lesbian identity. Apparently, gays and lesbians are essentially different from heterosexuals (in ways other than who they are sexually attracted to). He claims:

"For both gay men and lesbians, they are more likely to have multiple sex partners, both as singles and couples."

Right off, you notice that Allen says gay men and lesbians are "more likely" to do something, implying that all gay men and lesbians are more likely to do this thing. And well, that's an incredibly difficult claim to prove, but I'd love to see him try! I suspect that what he's done is categorize an average difference as an "essential" difference.

Relatedly, you notice that this claim isn't cited. In fact, none of his claims are cited. Within the interview, Allen generally refers to two papers he wrote (and which are fully cited at the end of the interview), but these are law journal articles, not works of original research directly supporting his contentions on relationship differences.

It would be appropriate here to note that the thing about law journals is that they're edited and staffed by law students. That's not a statement against those who publish in law journals, who often are legit academic types. It's just that errors and misrepresentations are going to happen when the primary cite checkers are students with only 1-2 years of law school experience, no experience as practicing attorneys, and who are often doing this work on top of a full load of coursework and internships/jobs.

So, when I looked up his law journal articles, I wasn't surprised to find that the evidence "supporting" his above claim was a pretty egregious misrepresentation of a study. In "Who Should Be Allowed Into the Marriage Franchise," published in the Drake Law Review, he claims:

"A number of studies have found gay couples to have explicit open-marriage agreements in about fifty percent of unions."

Gay male couples only, or male and female same-sex relationships?

It's important to be specific when using the word "gay," because the word is not consistently used. I regularly see it used as "gay male," "gay and bisexual male," "gay and lesbian," "gay, lesbian, and bisexual," and even sometimes "LGBT." If one is making claims about "gay couples," especially when discussing how these couples are different from other couples, it is impossible to write clearly and accurately if one doesn't understand that these varying usages create ambiguity.

Also, despite Allen's use of the plural "a number of studies," he cites only one study:

"Colleen C. Hoff et al., Serostatus Differences and Agreements About Sex with Outside Partners Among Gay Male Couples, 21 AIDS EDUC. & PREVENTION 25, 32 (2009)"

This study is a study of 191 gay male couples in the San Francisco Bay Area who were recruited specifically to provide a mix of HIV statuses. From this study, Allen then concludes that "this type of behavior contrasts significantly with heterosexual relationships in which open marriages are extremely rare."

Well, maybe. But let's not pretend that the monogamy practices of 191 gay male couples of varying HIV serostatuses in San Francisco are representative of the practices of all same-sex couples in the United States. And let's not pretend that non-monogamy is an "essential" difference between heterosexual and "gay and lesbian" couples.

My more general point here is that Allen's interview, in particular, is unlikely to be convincing to those who don't already agree with him about things. When people make provocative and controversial statements that are inaccurate, and do so in a flippant, unsupported manner, it is especially frustrating. It takes time and effort to cite check and then counter misrepresentations in a reasoned manner.

As family scholars, isn't one of our primary interests accuracy, even if it's not always politically correct to recognize that interest?