Democracy in America | Political polarisation

Not persuaded

If it is nearly impossible to change people's minds on divisive issues, why is public support for gay marriage booming?

By M.S.

IT IS almost impossible to change people's opinions on divisive political issues by arguing with them. This is rather depressing for opinion journalists and others in the advocacy business, but the social-science research is fairly conclusive. There is even evidence of a "backfire effect" in which attempts to change people's minds only make them dig in deeper.

Last month, however, the radio programme This American Life reported on a technique that seemed to work. A gay-rights group in California had sent door-to-door canvassers to have open, non-confrontational conversations with opponents of gay marriage. A study by two political scientists found that after a single non-confrontational conversation with a gay canvasser, support for gay marriage among the 972 subjects rose several percentage points, and that this effect was lasting. The study was so striking that the journal Science immediately accepted it for publication last December.

Does that sound encouraging? Too bad! It seems the data were bunk. As the wonderful website Retraction Watch explains, the study by Donald Green, a professor at Columbia, and Michael LaCour, a graduate student at the University of California, Los Angeles, was so extraordinary that two other graduate students quickly tried to replicate it. Their first step was to re-do what Mr LaCour had supposedly done: carry out a paid online survey to determine respondents' "baseline" attitudes before sending canvassers to try to change their minds. But they couldn't get people online to respond at the rates Mr LaCour claimed to have garnered. When these students called the online survey company Mr LaCour said he had worked with, it had never heard of him. They performed statistical analysis on his pre- and post-intervention data and found it suspiciously uniform. When they asked Mr LaCour to provide them with his original pre-survey responses, he claimed he had accidentally deleted the file. They presented all of this to Mr Green, who is widely respected in his field and was not involved in the initial survey work. Mr Green gave Mr LaCour a chance to explain the discrepancies, and when he could not, asked Science to withdraw the paper. The magazine has published an "expression of concern".

This does not mean that exploratory two-way conversations are not a good way to persuade people to change their minds. Maybe the conversations worked. Without pre-survey data, we have no idea. The gay-rights group that provided the canvassers for the study, the Leadership Lab at the Los Angeles LGBT Centre, developed the technique and still believes in it. (The Leadership Lab says it had no idea the people at the addresses Mr LaCour was sending them had not in fact been pre-surveyed online.) And the theory accords with social-science research on the "illusion of understanding", which shows that when people are first asked to explain their own conception of a phenomenon, they often recognise how fragmentary it is and become more open to alternative positions.

But without this study, we are thrown back on the weight of previous research. That showed that when an issue has become polarised, persuasive argument will generally have no lasting effect on opinions, or will prompt people to rationalise their convictions. And gay marriage is certainly a polarised issue.

All of which raises the question: if persuading people to change their minds about gay marriage is so hard, why are people changing their minds about gay marriage?

Gallup reported this week that its latest Values and Beliefs poll shows public support for gay marriage at an all-time high of 60%, up five percentage points from last year. The rise of support for gay marriage since 1996 appears unstoppable, apart from brief reversals keyed to the presidential-election years of 2004, 2008 and 2012. In the Washington PostChris Cilizza contrasts this with the similarly partisan issues of abortion and the death penalty, where polls show back-and-forth shifts in support levels but no lasting trend towards resolution, and certainly not with the strength and speed of gay marriage.

Mr Cilizza cites the views of two pollsters, one Republican and one Democrat, who both point to the gradual process of gay people coming out of the closet, helping their relatives and friends who might once have opposed gay marriage to accept it. This is certainly part of the equation, but it seems inadequate to explain a swing from 37% support in 2005 to 60% support just ten years later. Not enough gay people came out of the closet in that period. Besides, for decades many people—including several of the gay-marriage opponents in the This American Life episode—had gay friends or family but did not believe they should be allowed to marry. Something else beyond familiarity with gay people is happening to drive this massive swing of public opinion. What is it?

In his book "The Righteous Mind: Why Good People Are Divided by Politics and Religion", Jonathan Haidt, a social psychologist at New York University, observes that most of the moral reasoning people do is oriented not towards discovering the truth, but towards justifying their beliefs to others in their social group. "Moral reasoning is more like a politician seeking votes than a scientist seeking truth," Mr Haidt writes. "We are obsessively concerned about what others think of us." For the most part, people select their moral beliefs the way they select their clothes, asking themselves whether this or that opinion is appropriate to their identity and how it will look to their friends. When they do engage in moral reasoning, they do it to justify taking the position necessary to fit in. If people's moral stances are shifting rapidly, it is because they are getting signals from others in their group that a different belief is now acceptable.

America's arguments on issues such as abortion, the death penalty, immigration and climate change have ground to a stalemate largely because of this sort of group-opinion dynamic. The two sides in these debates have been appropriated by liberal-Democratic and conservative-Republican factions, and group rivalry leaves members unable to budge. For a long time, it looked like the same thing might happen to gay marriage. From the mid-1990s through the mid-2000s Republican support was essentially static, going from 16% in 1996 to 17% in 2008, while Democratic support rose from 33% to 53%. But starting in 2009, something changed. Suddenly, partisans on both sides started moving in tandem: Republican support for gay marriage has since risen to 37% while Democratic support has reached 76%, according to Gallup. Pew polling results are less stark, but they show the same pattern: Republican support was nearly flat at 20% from 2001 to 2008, then started rising to 30% by 2014.

There are all sorts of reasons why Republicans might be sympathetic to gay-marriage arguments: libertarian beliefs that the state should not meddle in sexual affairs, the culturally conservative nature of marriage itself, the influence of mass media, the rise of a young gay-friendly generation. But these factors all existed from 1996 to 2008 too, to little effect. So what happened in 2009 that pushed Republicans to change their minds? The loss of the presidency in 2008 thrust the party into a period of intense ideological turmoil, which upended the moral consensus.

On some issues, including health care and climate change, the tea-party revolution drove Republicans further to the right and created a new orthodoxy. But on other issues, such as foreign policy and gay marriage, differences that had been papered over suddenly broke free. In 2009 Theodore Olson, the libertarian-leaning Republican lawyer who had served as solicitor general under George W. Bush, filed suit to overturn California's Proposition 8 outlawing gay marriage. Ken Mehlman, the GOP's party chairman until 2007, came out of the closet in 2010 and began campaigning for gay marriage a year later. Other senior Republicans began arguing that the party had to change tack or risk losing young voters, who overwhelmingly support same-sex marriage. A few years earlier these stances would have been seen as betrayals of the evangelical Christians who were at the centre of the party’s electoral strategy. But the GOP’s big loss in 2008 undermined this strategy, and allowed a faction of the Republican elite to switch sides.

To bring things back to our original dilemma: it is not actually hopeless to try to persuade people to change their opinions on divisive issues. Rather, it is usually hopeless for individuals to try to persuade other individuals to oppose the opinions of their political teams on divisive issues. One does not change the direction of a herd by lassoing horses one by one. Individuals, by and large, do not change their minds; groups do.

The Los Angeles gay-marriage persuasion experiment may or may not have had an effect on the 972 people it reached, but it had nothing to do with the huge gain in approval of gay marriage in America over the past five years. The political shift happened because a split opened up at the centre of a large ideological group, and its members started to sense that different attitudes were acceptable. That does not occur very often. But when it does, people can change their opinions very quickly indeed.

Discover more

The fifth Democratic primary debate showed that a cull is overdue

Thinning out the field of Democrats could focus minds on the way to Iowa’s caucuses

The election for Kentucky’s governor will be a referendum on Donald Trump

Matt Bevin, the unpopular incumbent, hopes to survive a formidable challenge by aligning himself with the president


A state court blocks North Carolina’s Republican-friendly map

The gerrymandering fix could help Democrats keep the House in 2020