The internet was abuzz this August with speculation about why Millennials have less sex than our elders did at our age. Perhaps economic precarity makes dating difficult, particularly for those living with their parents. Mediating our relationships through the internet could reduce physical contact. Or maybe ubiquitous pornography is replacing partnered sex, antidepressants are depressing libidos, higher standards of consent reduce the incidence of “pressured sex,” asexual people are more comfortable in their identities, or Millennials are simply too busy for sexual relationships.
Although this summer’s vibrant online discussions were sparked by a study published by three psychologists, as well as by Washington Post journalist Tara Bahrampour’s reporting on it in articles on August 1 and 2 (in which speculation about most of the potential causes above can already be found), they are essentially conversations about the history and sociology of sexuality. In a scramble to explain changes in sexual practices, though, many have overlooked two critical questions: whether the study actually demonstrates that Millennials are having less sex, and what behaviors we’re counting when we claim to measure sexual activity.
In “Sexual Inactivity During Young Adulthood Is More Common Among U.S. Millennials and iGen,” published on August 1 in the Archives of Sexual Behavior, Jean Twenge, Ryne Sherman, and Brooke E. Wells analyzed data from the General Social Survey, a regularly conducted study of American adults. In 1989, the GSS began asking participants, “Now thinking about the time since your 18th birthday (including the past 12 months), how many female partners have you had sex with?” It also asked the same question with regard to male partners.
Twenge and her coauthors found that when surveyed in their early 20s, 15% of subjects born in the early 1990s had no sexual partners since they turned 18, compared with only 6% of those born in the late 1960s and 12% of those born in the 1970s and 1980s. “Contrary to popular media conceptions of a ‘hookup generation’ more likely to engage in frequent casual sex,” they wrote, “a higher percentage of Millennials and iGen’ers reported no sexual partners as adults.”
For Twenge, Sherman, and Wells, this is the latest bout in an ongoing battle against the idea that Millennials have more sex than previous generations. Their previous article on the subject, “Changes in American Adults’ Sexual Behavior and Attitudes, 1972–2012,” also appeared in the Archives of Sexual Behavior and drew on the GSS. Millennials, it argued, actually had fewer sexual partners than their immediate elders in Generation X.
That line of research was prominently dismissed in a 2015 Vanity Fair article, “Tinder and the Dawn of the ‘Dating Apocalypse,’” by journalist Nancy Jo Sales, who found it contradicted her own reporting on the app-assisted promiscuity of young men and women. “In some ways,” said Sherman, the new study on celibacy was “a response” to Sales’ article.
This framing as debunking the myth of a “hookup generation” may be part of why media coverage has often treated this study as if it concerned the average frequency with which Millennials have sex, rather than the percentage who are celibate. “Today’s 20-year-olds are having less sex than the previous generation,” begins one representative story from CNN.
Even when journalists recognize that this is fundamentally a study about how many young people don’t have sex rather than how often the rest do, they miss something even more important: The General Social Survey data on which it bases its claim is incredibly ambiguous.
An answer to the question “How many … partners have you had sex with?” is as much about how the subject defines sex as it is about their behavior. Those who respond that they haven’t had any partners since turning 18, and who researchers interpret as celibate, might have engaged in sexual behavior that they don’t count as “having sex,” or might even have had sex with someone they don’t really consider a “partner.” Twenge and her coauthors acknowledge this ambiguity, and even that “interpretations of this question may have changed over time,” but conclude that since “alternatives to vaginal intercourse such as oral sex were less common behaviors in previous eras,” it’s unlikely that the increased rate of apparent celibacy can be explained by earlier generations counting these practices as sex while Millennials do not.
They are right that sexual practices become more and less common over time, which is only another way of saying that they have histories — even if these histories are more multivalent than can be represented by a simple frequency. In their 1987 article “The Sexual Scripting of Oral Genital Contacts,” for example, sociologists John Gagnon and William Simon observed that among college-educated young adults, “the incidence of oral sex in nonpaid relationships during the period before marriage appears to have nearly doubled” between Alfred Kinsey’s research around 1940 and their own interviews of students in 1967. In an addendum, they added that a 1973 study suggested that the percentage of college students who had engaged in oral sex then doubled again, to around 60%, over only six years. (The statistic in 2016, according to the National College Health Assessment, is around 70%.)
What the celibacy study’s authors overlook is not that sexual behavior changes over time, but that quantitative studies like the GSS are ambiguous enough to leave open a range of interpretations of how it changes. What if other sexual practices are replacing some penis-in-vagina (PIV) sex, for example, but survey respondents count them as “having sex” at a lower rate? If this is the case — and it seems likely that it is — then Twenge and her coauthors may well be right that more Millennials are abstaining from PIV sex than youth of earlier generations, but wrong that this constitutes a “rise in sexual inactivity.” While journalists and my friends-of-friends on Facebook write anxiously about the end of sex, then, the GSS data might instead be suggesting that fewer Millennials have PIV sex because we have more alternative ways of achieving intimacy and pleasure.
Or perhaps it isn’t. Even setting aside the challenges of relying on self-reporting, the GSS simply doesn’t ask the right questions to measure frequencies of sexual behavior. Perhaps more explicitly anatomical questions — such as “Has another human’s penis been in your vagina?” — could produce a more rigorous accounting.
Even better, though, would be an account of how sexual practices and the ways we think about them have actually changed together. “Sexual conduct is a historical phenomenon changing in time as well as in cultural place,” wrote Gagnon and Simon. “The bookkeeping measures that we have used capture only a portion of the picture.” Historians, trained in thinking interpretively about change and in bringing together different kinds of evidence, know that categories of behavior like “having sex” are not stable entities, but rather vary in meaning over time and space. Perhaps psychologists, historians, and sociologists can collaborate to produce more nuanced, and yet rigorous, studies of the history of human sexual expression.