We welcome a second post by Patrick Egan on the future of same-sex marriage. His first post is here.
My post here yesterday raised the question of why the five-justice majority in U.S. v. Windsor—which struck down a section of the Defense of Marriage Act by relying in part on the Constitution’s due process and equal protection principles—declined to issue a “fifty-state ruling” that would have made marriage equality the law of the land. I surmised that one explanation is that the justices sympathetic to gay marriage were reluctant to issue a sweeping ruling for fear of halting the steady momentum that same-sex marriage has enjoyed of late in public opinion and policy. I noted Justice Ruth Bader Ginsburg’s recent public statements asserting that Roe v. Wade had just this sort of deleterious effect on the movement for abortion rights and thus the ruling had gone too far.
The legal patchwork faced by married gay couples in the wake of Windsor now make the state-by-state battles that have been taking place over the right to marriage even more important. In their midst will be sympathetic politicians and judges who nevertheless hesitate to usher in sweeping changes due to the kinds of misgivings expressed by Justice Ginsburg. But a close look at public opinion trends regarding the issue of same-sex marriage suggests that the proper comparison is not the polarizing issue of abortion (and Roe v. Wade), but the once controversial and now settled issue of interracial marriage (and its landmark Supreme Court decision, Loving v. Virginia). Generational trends in public opinion on gay marriage look much more like those on interracial marriage than abortion, and by all indications, Americans are moving toward a national consensus.
We can of course never be absolutely sure about the future trajectory of public opinion. But without getting too technical, one indication that aggregate American attitudes are permanently changing in, say, a liberal direction is when (1) younger generations hold significantly more liberal attitudes than their older counterparts and (2) people’s attitudes do not become more conservative as they age. In the absence of any other stimuli, these two phenomena lead to a gradual change in overall opinion toward a level equal to the attitudes held by the youngest cohort. Furthermore, if people’s attitudes become more liberal as they age, than national opinion will change at an even faster rate.
We can see if this is the case with regard to abortion, interracial marriage and gay marriage with data from questions posed by the magisterial General Social Survey to nationally representative samples of American adults over impressively long spans of time. (For question wording, follow the links in the previous sentence.) The figure below plots opinion trends from the GSS on these three issues by cohorts of Americans born at twenty-year intervals.
The left-hand graph shows that abortion attitudes initially displayed some slight generational differences, with later cohorts holding more liberal opinions than older ones. But these generational differences never resulted in aggregate over-time opinion change, as newer birth cohorts—those born in the 1960s and later—emerged with more conservative attitudes than those of the generation born just beforehand. Furthermore, Americans’ opinions did not become appreciatively more liberal as they aged in any consistent way across generations. The result is the stasis in American opinion on abortion witnessed for the past four decades (as shown in the figure accompanying my previous post).
Compare these trends to those on interracial marriage, displayed in the middle graph. Here, dramatic inter-cohort differences existed in the early 1970s—on the order of 50 percentage points between the earliest and latest generations. These differences persisted over time, contributing to a gradual liberalization of aggregate national opinion this issue. This trend was bolstered by the fact that as Americans of every generation got older, they became more liberal on interracial marriage. By the beginning of the 21st century, a near-universal consensus had emerged that these marriages should be legal—a consensus so strong that the GSS dropped the survey question after 2002.
The final graph depicts the GSS generational trends for the legalization of same-sex marriage. (Although the GSS was one of the first surveys to ask a question about gay marriage in 1988, the question was not included again until 2004.) Here we see notable generational trends coming to the fore, and they look strikingly similar to those on interracial marriage. Over the past decade, opinion on gay marriage has been marked by large inter-cohort differences—once again, a gap reaching as large as about 50 percentage points between the youngest and oldest groups of Americans. Furthermore, if we ignore some bumpiness in the trends due to sampling error, Americans in each generation appear to have become more accepting of gay marriage as they have aged. For example, support among those born in the 1960s and 1970s rose by an astounding 20 percentage points between 2004 and 2012. Similar patterns have been found in other polls conducted over time, in particular those conducted by the Pew Research Center.
All told, our best bet is that attitudes on gay marriage are likely to continue to parallel those on interracial marriage. A point prediction of where the national consensus might eventually lie would be more than 7 out of 10 Americans favoring marriage equality, which is the level of support currently found among the youngest generation.
Over at the New Republic, Nate Cohn takes a less sanguine view. Noting the current opposition to gay marriage among young evangelicals, he foresees continuing resistance to marriage equality among this important segment of the American electorate—and the Republican Party’s coalition. I don’t disagree that religious resistance from some quarters will persist. But what Cohn views as a problematic hurdle to “universal support” for gay marriage in the long term I see as more akin to religious resistance to national consensuses over previously controversial issues such as contraception, the consumption of alcohol, and no-fault divorce. Many deeply religious people see these issues as matters of morality and licentiousness, but in all but a few places they do not have the numbers to affect public policy regarding them.
In U.S. v. Windsor, Justice Anthony Kennedy writes that the “injury and indignity” inflicted by DOMA on lawfully married same-sex couples “violates basic due process and equal protection principles.” He no doubt expresses the sentiment of many judges and politicians who will contend with same-sex marriage in the time to come. When they are deciding just how far they can go in addressing the injuries and indignities faced by gay and lesbian couples, the trends shown here should provide some comfort that their actions are likely to promote—rather than hinder—the slow and steady national momentum toward support for marriage equality.
 The irony is not lost on me that the quest for visual clarity necessitates using essentially rainbow-colored graphs for a blog post on gay marriage.