2020 taught us how to fix this

This is the year that broke the truth. This is the year when millions of Americans — and not just your political opponents — seemed impervious to evidence, willing to believe the most outlandish things if it suited their biases, and eager to develop fervid animosities based on crude stereotypes.

Worse, this was the year that called into question the very processes by which our society supposedly makes progress.

So many of our hopes are based on the idea that the key to change is education. We can teach each other to be more informed and make better decisions. We can study social injustices and change our behaviour to fight them.

But this was the year that showed that our models for how we change minds or change behaviour are deeply flawed.

It turns out that if you tell someone their facts are wrong, you don't usually win them over; you just entrench false belief.

One of the most studied examples of this flawed model is racial diversity training. Over the last few decades, most large corporations and other institutions have begun racial diversity programs to combat the bias and racism pervasive in organisational life. The courses teach people about bias, they combat stereotypes and they encourage people to assume the perspectives of others in disadvantaged groups.

These programs are obviously well intended, and they often describe systemic racism accurately, but the bulk of the evidence, though not all of it, suggests they don't reduce discrimination. Firms that use such courses see no increase in managerial diversity. Sometimes they see an increase — not a decrease — in minority employee turnover.

Frank Dobbin and Alexandra Kalev offered a clear summary of the research in a 2018 essay in Anthropology Now. One meta-analysis of 985 studies of anti-bias interventions found little evidence that these programs reduced bias. Other studies sometimes do find a short-term change in attitudes, but very few find a widespread change in actual behaviour.

Dobbin and Kalev offer a few reasons for why these programs generally don't work as intended. First, "short-term educational interventions in general do not change people." This is as true for worker safety courses as it is for efforts to combat racism. Second, some researchers argue that the training activates stereotypes in people's minds rather than eliminating them. Third, training can make people complacent, thinking that because they went through the program they've solved the problem.

Fourth, the mandatory training makes many white participants feel left out, angry and resentful, actually decreasing their support for workplace diversity. Fifth, people don't like to be told what to think, and may rebel if they feel that they're being pressured to think a certain way.

These days a lot of the training is set up to combat implicit bias. This is based on research led by Anthony Greenwald, Mahzarin Banaji and Brian Nosek, showing that most Americans, and especially most white Americans, have hidden biases that influence who gets hired, who gets promoted and how people are seen.

Implicit bias is absolutely real. The problem is that courses to reduce its effects don't seem to work. As Greenwald told Knowable Magazine: "I see most implicit bias training as window dressing that looks good both internally to an organization and externally, as if you're concerned and trying to do something. But it can be deployed without actually achieving anything, which makes it in fact counterproductive."

Or, as Tiffany L. Green and Nao Hagiwara wrote in Scientific American this past August, "But to date, none of these interventions has been shown to result in permanent, long-term reductions of implicit bias scores or, more importantly, sustained and meaningful changes in behavior."

Part of the problem is that a lot of discrimination is structural; not in people's attitudes but in organisational practices and the way society is set up. Part of the problem, as Matt Martin writes in Fast Company, is: "There's surprisingly little correlation between most people's attitudes and behavior. And the correlation between bias and discrimination is weak."

Finally, our training model of "teaching people to be good" is based on the illusion that you can change people's minds and behaviours by presenting them with new information and new thoughts. If this were generally so, moral philosophers would behave better than the rest of us. They don't.

People change when they are put in new environments, in permanent relationship with diverse groups of people. Their embodied minds adapt to the environments in a million different ways we will never understand or be able to plan. Decades ago, the social psychologist Gordon Allport wrote about the contact hypothesis, that doing life together with people of other groups can reduce prejudice and change minds. It's how new emotional bonds are formed, how new conceptions of who is "us" and who is "them" come into being.

The superficial way to change minds and behaviour doesn't seem to work, to bridge either racial, partisan or class lines. Real change seems to involve putting bodies from different groups in the same room, on the same team and in the same neighbourhood. That's national service programs. That's residential integration programs across all lines of difference. That's workplace diversity, equity and inclusion — permanent physical integration, not training.

This points to a more fundamental vision of social change, but it is a hard won lesson from a bitterly divisive year.

©2020 The New York Times Company