What the AIDS Crisis Did to Sex Research

In 1981, the Centers for Disease Control published a brief report noting that five young men in Los Angeles had been diagnosed with Pneumocystis carinii pneumonia, a lung infection so rare that the existence of five cases in one city was itself a medical event. All five men were gay. All five were previously healthy. Two were already dead by the time the report was published.

Within three years, the thing that would be named AIDS was killing thousands of people, and the federal government of the United States needed, with sudden and desperate urgency, to know exactly what gay men did in bed. This was a problem, because it had spent the previous thirty years making absolutely certain that nobody studied the subject.

The Government Needed Data It Had Spent Decades Refusing to Collect

The epidemiological math was simple and brutal. To stop a sexually transmitted epidemic, you need to know who is having sex with whom, how, how often, and with how many partners. You need to know about specific acts; whether anal intercourse is more efficient at transmitting the virus than oral sex; whether condoms work in practice and not just in theory; whether the number of sexual partners is a better predictor of risk than the type of sexual contact. You need granular, detailed, honest data about the actual sexual behavior of the population most affected.

The United States did not have this data. It had never wanted it. The last serious attempt to map American sexual behavior had been Kinsey’s surveys in the 1940s and 1950s, and the political establishment had spent the intervening decades making sure nobody repeated the exercise. There were no large-scale behavioral surveys. There were no reliable estimates of the size of the gay male population. There were no data on the frequency or variety of sexual practices among men who had sex with men. The government was trying to fight an epidemic with a map drawn from prejudice and assumption.

The assumptions were wrong. Public health officials in the early 1980s operated on a model of gay male sexuality that was essentially a caricature; bathhouses and anonymous encounters, an undifferentiated mass of promiscuity. Some gay men did have large numbers of sexual partners. Some straight people did too. What the data would eventually show, once anyone was allowed to collect it, was that the distribution of partner counts was similar across orientations; a small percentage of any population has a large number of partners, and the vast majority does not. But the policy response was built on the caricature, because the caricature was all they had.

This is what happens when you defund an entire field of research for thirty years and then need its findings overnight.

The Funding Arrived with Strings Attached

The National Institutes of Health began funding sex research in the mid-1980s at levels that would have been unimaginable a decade earlier. Researchers who had spent their careers scraping together grants for small-scale studies of sexual behavior suddenly found that the federal government was willing to pay for exactly the kind of large-scale behavioral surveys it had been blocking since Kinsey. The money was there. The urgency was real. And the strings were everywhere.

The first major attempt at a national sex survey, proposed in 1987 by a team led by sociologist Edward Laumann at the University of Chicago, was designed to be the Kinsey study done right; a probability-sampled survey of 20,000 Americans about their sexual behavior, funded by NIH at an estimated cost of $15 million. The study got through peer review. It got NIH approval. It was scientifically uncontroversial. And it was killed by Congress.

Senator Jesse Helms of North Carolina led the opposition. His argument was not scientific. It was that asking Americans about their sexual behavior was an invasion of privacy and that the government had no business knowing what people did in their bedrooms. This was the same Jesse Helms who had spent his career arguing that the government had every business regulating what people did in their bedrooms; sodomy laws were still on the books in multiple states, and Helms was their most vocal defender. The contradiction was not a contradiction to Helms. The state had the right to punish sexual behavior. It did not have the right to understand it. Knowledge was more dangerous than prohibition.

The Helms amendments, attached to NIH appropriations bills throughout the late 1980s and 1990s, specifically prohibited federal funding for surveys of sexual behavior. Laumann’s team eventually conducted a scaled-down version of their study using private foundation money; the National Health and Social Life Survey, published in 1994, surveyed 3,432 Americans instead of the planned 20,000. It remains one of the most cited sex surveys in history. It found, among other things, that the average American had far fewer sexual partners than popular culture suggested, that the vast majority of sex happened within committed relationships, and that sexual dysfunction was widespread and largely untreated. None of these findings were politically useful to anyone, which is probably why the study survived.

What Researchers Found When They Were Finally Allowed to Ask

The AIDS-driven research produced findings that contradicted nearly every assumption that public health policy had been built on. The research showed that sexual behavior existed on continuous distributions, not in discrete categories. Men who had sex with men were not a separate population; they were a behavior group that overlapped with men who also had sex with women, men who identified as straight, men who were married, men who would never use the word “gay” to describe themselves. The epidemiological category of “men who have sex with men,” or MSM, was invented specifically because the identity categories were useless for tracking a virus. The virus didn’t care what you called yourself. It cared what you did.

The research showed that condom use was not a simple binary. People didn’t either use condoms or not use them. They used them sometimes, with some partners, for some acts. The factors that predicted condom use were not the ones the public health campaigns assumed; not knowledge of HIV risk, which was near-universal by the late 1980s, but relationship dynamics, power differentials, substance use, and whether the person had access to condoms at the moment of the decision. Education campaigns built on the assumption that information changed behavior were failing because information was never the bottleneck. The bottleneck was everything else.

The research showed that the bathhouse model of gay male sexuality, the model that had driven the initial public health response, described a specific subculture in a specific set of cities during a specific historical period, not a universal truth about how gay men had sex. When researchers were finally able to survey large and representative samples of men who had sex with men, they found the same variation in sexual behavior that Kinsey had found in the general population forty years earlier; a wide distribution, with most people clustered in the middle and a long tail of high-activity individuals who were not representative of anything except themselves.

The data didn’t fit the narrative. The narrative was that AIDS was a disease of deviance, of excess, of a population whose sexual behavior was fundamentally different from the mainstream. The data showed that the behavior wasn’t fundamentally different. The virus was just more efficiently transmitted by certain acts, and the population that practiced those acts had been left without public health infrastructure because the country had decided that their sexuality didn’t deserve medical attention.

The Abstinence Contamination Poisoned the Well

The political interference didn’t stop with funding restrictions. It reached into the research itself. Throughout the 1990s and accelerating under the George W. Bush administration, abstinence-only sex education became the federal government’s primary approach to HIV prevention among young people. This wasn’t just an education policy. It was a research policy. Federal grants for HIV prevention research increasingly required that the interventions being studied include abstinence messaging. Researchers who wanted to study harm reduction, condom distribution, or comprehensive sex education found that their grant applications were scored lower or rejected outright if they didn’t include an abstinence component.

The science on abstinence-only education was unambiguous and damning. Multiple large-scale evaluations, including a congressionally mandated study published in 2007 by Mathematica Policy Research, found that abstinence-only programs had no effect on sexual behavior. Students who received abstinence-only education were no more likely to abstain from sex, no more likely to delay sexual initiation, and no less likely to contract STIs than students who received no sex education at all. The programs did, however, produce one measurable outcome: students who took virginity pledges were less likely to use condoms when they eventually had sex, because the pledge had framed condom use as a concession to failure rather than a health behavior.

This data was available. It was replicated. It was ignored. Between 1996 and 2009, the federal government spent over $1.5 billion on abstinence-only education programs that the government’s own research showed did not work. The money came directly out of the pool that could have funded evidence-based prevention research. Every dollar spent on proving that telling teenagers to wait was ineffective was a dollar not spent on figuring out what actually reduced transmission.

The researchers who produced these findings paid for them. Joycelyn Elders was fired as Surgeon General in 1994 for suggesting, at a United Nations conference on AIDS, that masturbation “perhaps should be taught” as an alternative to riskier sexual behavior. Her statement was medically uncontroversial and politically lethal. The message to every public health official who followed her was permanent: there are true things about sex that you cannot say in public if you want to keep your job.

The Data We Still Don’t Have

The AIDS crisis forced open a door that had been locked since Kinsey, but the door didn’t open all the way, and it didn’t stay open. The research that got funded was the research that served the epidemic; studies of transmission, risk behavior, condom efficacy, partner notification. The research that didn’t serve the epidemic directly stayed unfunded. This meant that the AIDS era produced an extraordinary body of knowledge about the mechanics of disease transmission and almost nothing about the broader landscape of human sexual behavior, health, and wellbeing.

The data on women’s sexual behavior barely expanded at all. AIDS was framed as a gay male disease for the first critical years of the epidemic, and by the time the heterosexual transmission risk was acknowledged, the research infrastructure was already built around MSM populations. Lesbian sexual health, which posed essentially zero HIV risk, became invisible. There are more peer-reviewed studies on the sexual behavior of male-to-female transsexuals than there are on the sexual health of lesbians, not because the population is larger but because the HIV connection made one group epidemiologically interesting and the absence of HIV risk made the other group epidemiologically invisible.

The data on sexual pleasure, satisfaction, and function outside a disease framework remained sparse. Studying what made sex good was not the same as studying what made sex dangerous, and only the dangerous version got funded. The orgasm gap, the consistent finding that women in heterosexual encounters have fewer orgasms than their male partners, was documented but not investigated with anything like the rigor that went into studying condom failure rates. The assumption embedded in the funding structure was that sex was a public health problem to be managed, not a human experience to be understood.

The data on sexual minorities beyond gay men remained thin. Bisexual populations were particularly underserved; researchers tended to collapse bisexual men into the MSM category for epidemiological purposes and ignore bisexual women entirely. Asexuality as a research subject barely existed. The full range of human sexual variation, the thing Kinsey had been trying to map when the government shut him down, remained unmapped.

The Window Closed Before the Work Was Done

By the mid-2000s, the acute phase of the AIDS crisis in the developed world was over. Antiretroviral therapy had transformed HIV from a death sentence into a manageable chronic condition for people with access to treatment. The political urgency that had forced the government to fund sex research evaporated. The funding didn’t disappear entirely, but it contracted, and the contraction followed the same pattern as always: research that could be framed as disease prevention survived, and research that was about understanding human sexuality as a thing in itself did not.

The AIDS crisis proved that the government would fund sex research when the alternative was mass death. It also proved that the government would stop funding sex research the moment the crisis became manageable. The window of tolerance was not about a new respect for scientific inquiry. It was about panic. When the panic subsided, the tolerance did too.

What remains is a body of knowledge shaped entirely by crisis. The most-studied sexual behavior in human history is anal intercourse between men, not because it is the most common or the most important, but because it was the most relevant to stopping an epidemic. The second most-studied is condom use, for the same reason. The behaviors that define most people’s actual sexual lives; the negotiation of desire in long-term relationships, the development of sexual identity across the lifespan, the mechanics of pleasure and satisfaction; remain understudied, because no epidemic has ever made them politically necessary to understand.

The next crisis will open the window again. It always does. The question is whether the country will have spent the intervening years building the research infrastructure to respond, or whether it will find itself, once more, trying to fight an epidemic with data it refused to collect.