Why have some of the nation's most vehement anti-gay activists—Ted Haggard, Larry Craig—had gay sex scandals of their own? An op-ed in the New York Times' Sunday Review section tries to explain. The authors of the piece, two research psychologists, say they have "empirical evidence that homophobia can result, at least in part, from the suppression of same-sex desire." Their argument—summed up in the Times headline as "Homophobic? Maybe You're Gay"—promises to resolve a long-running debate in the field. For at least 15 years, scientists have been trying to use objective laboratory measures to prove the he-who-smelt-it-dealt-it theory of human sexuality. Has a research team based at the University of Rochester finally done it?
The new study works like an elaborate game of "homo say what?": Evidence of private, homosexual urges is elicited by subtle verbal cues. The researchers start by asking college freshmen, mostly women, to rate their sexual orientation on a scale from 1 to 10 (1 means completely straight; 5 means bisexual; 10 means totally gay) and then to say how much they agree with politically charged statements like, "Gay people make me nervous" and "I would feel uncomfortable having a gay roommate." Once the students have been characterized according to their relative degrees of gayness and homophobia, they're shown a series of icons or photos of wedding-cake figurines on a computer monitor—two women, two men, or a man with a woman—and told to label each one as being "gay" or "straight." In a final twist, some of the "gay" and "straight" images are preceded on the screen by a subliminal verbal cue—a word flashed quickly on the screen that reads either me or others. If seeing the word me shortens a student's reaction time for the gay-themed imagery, it's taken as a sign of her implicit homosexuality. On a subconscious level, at least, she's associating the word me with gayness.
Many researchers have used setups like this one—known as an "implicit association test"—to dig up evidence of covert inclinations or even racial prejudice. The idea is that it takes people less time to make connections between words or images when those connections conform to prior beliefs. A person might respond more quickly to the word blue if she'd been cued with the word sky, or—more disturbing—she might be faster on the word man after being cued with the word president. The Rochester team adapted this idea for their measure of sexuality: A student's secret gay identity could be revealed by testing whether she responded more quickly to me-cued gay pictures than to me-cued straight ones.
Applying this logic, the researchers found that among the college freshmen in their study, more than one-fifth of those who described themselves as very straight showed signs of covert homosexuality on the me-cued trials. And these "discrepant" (secretly gay) students happened to be the ones most likely to have expressed anti-gay sentiments on the pre-test survey.
Should we trust this interpretation of the data? In the Times op-ed, the authors claim that the reaction-time task "reliably distinguishes between self-identified straight individuals and those who self-identify as lesbian, gay or bisexual." Their formal write-up of the work for the Journal of Personality and Social Psychology is a bit less sanguine on the method, citing just one other study that has used this approach, and saying it "showed moderate correspondence with participants’ self-reported sexual orientation."
Whatever the precedents, their homo-say-what task leaves itself open to an easy, alternative interpretation. It could be that both gay people and homophobic straight people responded more quickly to the gay-themed imagery because they were all secretly gay. Or it could be that both gay people and homophobic straight people are more keyed up by gayness in general. A homosexual might be more attuned to a picture of two men because it aligns with his personal interests—no surprise there. But a homophobe would be more attuned to it for the opposite reason: It runs counter to his personal interests; it makes him nervous. The sociologist Michael Kimmel has argued that some men are less afraid of gay people than they are of being labeled as gay (and thus emasculated) themselves. By that logic, me-gay pairings would be particularly nerve-racking to true homophobes. And it's well-known that these two factors—salience and anxiety—tend to shrink reaction times. People get a little speedy when something upsets them, or turns them on.
The Rochester team might have tried to rule out this interpretation by comparing the me-cued trials with the others-cued trials. If the homosexuals and homophobes had changed their response times only when they were cued to think about themselves (and not when cued to think about other people), it would strengthen the case that personal identification with the imagery drove the effect. But even that would leave open some questions. After all, it makes sense that a me-cued, gay trial would cause a homophobe more anxiety than one innocuously cued for others. In any case, either the paper's authors neglected to make this me-others comparison, or they chose not to report its results in the paper. (I reached out to several of them by phone and email, but no one could provide more details.)
Another question arises from the use of icons and wedding-cake figures to signify gayness and straightness. In the Times op-ed, the authors promise evidence of suppressed "same-sex desire." But as sexuality stigma expert Gregory Herek points out, these chaste and abstracted images are several steps removed from real or even imagined erotic behavior. Other studies of the same phenomenon, he says, have used images or video clips of couples embracing, or half-dressed.
Without more compelling laboratory data or more explicitly sexual test stimuli, it's hard to argue that the Rochester study demonstrates anything about secret homosexual urges. There's a much simpler interpretation at hand: College freshmen who claim to be nervous around gay people are probably a little nervous around gay people. And college freshmen who call themselves gay are probably somewhat gay.
If the reaction-time tests can't provide a satisfying proof, is there any hope left for the he-who-smelt-it-dealt-it theory of sexuality? Could scientists ever demonstrate that, as Freud suggested, homophobes are reacting to subconscious, gay urges? Maybe if there were some more direct way to measure a man's private sexual desires, we'd be able to tell for sure whether he was a closet homo. Imagine if you put a bunch of homophobes and more tolerant straight people into a room and forced them to watch man-on-man sex films while measuring the size of their erections with some kind of circumferential strain gauge. Would the gay-haters be revealed by the size of their boners?