How To Fix DEI

The current system is broken. Here’s what truly inclusive institutions would look like.

(Photo by PeopleImages via Getty.)

This article is part of an ongoing Persuasion series on the future of universities.

Universities are in crisis—losing public support, shaken by internal divisions, facing angry donors and alumni, and increasingly straying from their core mission of intellectual curiosity and open inquiry. Our series, which is made possible by the generous support of the Arthur Vining Davis Foundations, will consist of a collection of longform essays and podcast interviews aimed at helping higher education tackle this crisis. 

In today’s installment, Rachel Kleinfeld explores the challenges facing any society as it grows more diverse; summarizes the literature on DEI and the backlash it faces; and offers a blueprint for a healthier way to teach diversity and inclusion in universities. To read the other installments in the series, from Eboo Patel’s blueprint for establishing dedicated pluralism programs to William Deresiewicz’s exploration of off-campus learning, click here!

– Yascha and the Persuasion team.

P.S. Do you know someone who would value Persuasion’s content? As the move back to campus begins, this month we're offering 25% off an annual subscription to students, faculty, and parents.

---

If universities, corporations, philanthropy, and other major U.S. institutions were engaged in a concerted effort to increase racism and misogyny, many Americans would be outraged. And yet, the current way many Diversity Equity and Inclusion (DEI) programs are being conducted may be doing precisely that.

Recent assaults on DEI from the right have caused people who care about a diverse, inclusive America to circle the wagons against attacks from people who, at best, want to pretend the country is already colorblind, and at worst, want a return to an era when White, Christian men presided unquestioned at the top of the status hierarchy. But proponents of diversity do need to alter these programs—not to please those who want to go backward, but to help America become the inclusive nation it needs to be moving forward.

Democracy and demographic change: the choices

If America is to remain a democracy, demographic reality means that its politics and culture must become ever more inclusive. According to U.S. census data, in 1860 Whites comprised 86% of the population. A century later, the number had gone up slightly to 89%. And then it began to fall, fast. By 2023, only 62 percent of Americans identified as solely White.¹

In the post-1960’s era the country also experienced a vast alteration in the role of women. Despite recent attempts to roll back the clock, one only has to watch an episode of Mad Men to realize how far women have come. There has been a simultaneous revolution in gay rights and a freefall in Christian religious belief. In other words, a nation in which White men were considered the natural and rightful holders of most power in public and private life, and where being Christian and straight was the overwhelming social norm, transitioned within a single generation to a vastly different reality.

A democracy in this position can move in three directions:

• It can create laws and a culture in which all of its diverse citizens have an equal voice, and enough standing, opportunity, and respect to make that voice heard.

• It can reduce its commitment to democracy by using laws and cultural vilification campaigns to undermine equality of citizenship for everyone who is not the dominant identity, as multiple states did with laws that denied the vote to Blacks during the Jim Crow era, and as some countries continue to do today by criminalizing the expression of minority identities.

• It can reduce its demographic diversity, a course of action that requires getting rid of millions of people who identify as gay or feminist, or as members of racial, ethnic, or gender minorities. Countries that have taken this path may cap racially diverse immigration or deport, jail, or murder enough people that it convinces others like them to leave.

These are stark choices, but they are not theoretical. Like a number of other democracies, the United States has chosen both of the latter two options in the past. From the 1920s through 1965 it put a tight ceiling on racially diverse immigration, leading to the aforementioned rising percentage of White Americans. In the 1890s through 1910s, an upwelling of lynchings combined with the advent of repressive laws helped catalyze the Great Migration of millions of Blacks out of the South. Today, the Trump campaign has plans to severely limit legal immigration and deport millions who have lived here, often for decades, while minorities are facing the highest sustained level of hate crimes this century.

Justin Gest has written about the handful of countries in which an erstwhile demographic majority transitioned to minority status. Singapore and Bahrain, already authoritarian states, responded with repression of the sort described in the latter two strategies. Two tiny island states, Mauritius and Trinidad and Tobago, maintained their democracies, but today have racialized political parties and ongoing tension. That’s it—the full list of sovereign countries that have dethroned their dominant majority.

To put it plainly: no large country has ever successfully transitioned to a majority minority demographic while maintaining a full democracy. Experience suggests that such a transition is much more likely to be blocked than to succeed.

So those of us who want the country to move towards inclusion must be serious in considering how to get there. How can the United States shift from a dominant culture that is White, male, straight, and Christian to one that has no demographic assumptions at its center?

Can DEI deliver an inclusive America?

Enter DEI. The multitude of programs and ideas implied by that acronym, which emerged in the 1970s but spread rapidly following the George Floyd murder of 2020, was supposed to achieve three things:

1. Foster widespread acceptance of a diverse society among the dominant culture.

2. Build greater equality of power and voice for members of previously marginalized groups.

3. Give all races, ethnicities, and sexual and gender groups a feeling of inclusion and belonging as equal members of a shared culture, not add-ons to a dominant norm.

These are the goals the United States needs to achieve in order to transition to a truly pluralistic democracy, and I will refer to them as the goals of pluralism. But do current DEI programs achieve them?

Stunningly for such an important effort to quickly change the basic power structure in a society, very little about the effectiveness of DEI in practice is known at all. There have, of course, been a multitude of positive studies where participants in DEI programs fill out a questionnaire before and after participating and self-report that they gained understanding and acceptance. But a serious 2009 review by Elizabeth Paluck and Donald Green that looked at 985 studies of antibias intervention found few that even met the methodological criteria for good evaluations. In 2021 they updated their review, adding two additional co-authors and looking at 418 more studies. They found unclear impact because of obvious methodological failures, along with serious concerns about publication bias and failure to publicly share data, which often hides researcher bias.

Even among these flawed studies, only two considered whether behavior—rather than just self-reported attitudes—changed. And where positive attitudinal change existed, it seemed short lived. Another study tested 17 interventions to reduce bias and found that half had no effect. They took the nine interventions that showed some impact and retested them on over six thousand North American university students—and found that the impact degraded within a few hours, and at most lasted two days. Other meta-studies have similar findings: a 2022 review of diversity training programs found that “enthusiasm for, and monetary investment in, diversity training has outpaced the available evidence that such programs are effective in achieving their goals.”

Perhaps most damning, most of these studies are lab experiments—almost no one is even attempting to seriously evaluate the actual programs being used to implement DEI in the real world.

Such null findings could prove that these programs are simply a waste of billions of dollars and millions of hours spent by universities, businesses, and philanthropy. But there are hints that some of the techniques as they are being implemented in real life are doing active harm to achieving a more inclusive and diverse America.

Effects on individuals: racial and gender backlash

A number of diversity trainings have been found to activate bias instead of decreasing it. For instance, one study found that asking White college students who identify with their Whiteness to consider “White privilege” increased their racial resentment. Another experiment showed that when White Americans who identified strongly with their Whiteness were shown messages related to multiculturalism, it increased their sense of threat—which in turn heightened their prejudice and made them want to be more dominant over other groups. In fact, just learning about a coming minority-majority America led Whites in one experiment to express more bias and spend more time with other Whites. Another study found that Anglo men self-reported less positive attitudes towards women after diversity education.

Studies demonstrating that DEI interventions may trigger retrograde attitudes are worrying. Even more concerning are evaluations that show behavioral changes that increase inequality. In lab experiments, men who supported the traditional gender hierarchy reacted to a perceived threat to their status by discriminating against a hypothetical woman’s resume. The same holds in real life: a look at mandatory corporate diversity initiatives found that they led to a nearly ten percent decrease in Black women in management, a five percent decrease in Asian American managers, and no offsetting increase for Black men or Hispanics.

Meanwhile, a host of studies shows that White men are particularly likely to avoid diversity trainings in the first place, which has led many universities to make their DEI programs mandatory. But that “solution” is even more problematic. Being forced to take any type of compulsory program is associated with backlash—people don’t like to be forced to do things. The backlash generated by mandatory DEI training has been found in many different studies and situations. The issue is not always participants who begin the program as more racist or misogynistic. In many cases participants simply rebel against compulsion—one study suggested that participants who felt socially compelled to reduce prejudice ended up showing greater bias, both explicit and implicit, than they had prior to the intervention.

Effects on democracy: polarization and voting backlash

DEI is not operating in a vacuum—it is competing in a marketplace of ideas. The particular type of content and DEI methodology that has been adopted at many universities is hardly the only option—but it is especially pernicious both for its illiberal content and the way in which it is catalyzing polarization and voter backlash.

Too often, the DEI curriculum chosen by universities centers on the supposedly “anti-racist” ideology popularized by writers such as Ibram X. Kendi and Robin DiAngelo. This set of ideas promotes an almost Calvinistic view of personhood. Instead of free will, connection, and hope that together we can make positive change, students are taught that they are born into an identity ascribed to them by others’ perceptions. Some are oppressed—regardless of their personal experiences and feelings. Others—Whites, men, Jews, colonists—are oppressors.

This description of reality is problematic—but what is even more dangerous is its prescriptive content. Oppressors cannot escape the identity of oppressor. At best they can do penance for their privilege by performing the role of a silent ally. While these allies should listen to the oppressed, they should not fool themselves into believing that they can actually understand, because true cross-cultural understanding is impossible. They should silently defer to the experience of others who deserve to speak because of the group they were born into.

Even for people who wish to be allies in the fight for equality of all races, genders, ethnicities and identities, that is a lot to accept. Claiming that the randomness of one’s birth makes one inherently oppressive or toxic and that there is no way out is, of course, morally wrong, whatever its directionality. It is not moral progress to group individuals by accidents of birth, set them in a hierarchy, and place them on a ladder of worth—whether that ladder puts White Christian men on top, or reverses the order by ascribing points to minority identity markers.

But it may be particularly hard to take for some men. While White men as a group remain at the top level of dominance in money and power in American society, many individual men are nowhere near that level. Only 36% of White men even make it to college, along with less than a third of Black men and just over a quarter of Hispanic men—and women out-graduate all of them by a rate approaching 1.5 to 1.

Unsurprisingly, a series of focus groups found that participants in university DEI programs felt that they were being told they were horrible people because they were White, male, and middle class. That is often purposeful: a thrust of many courses is the importance of discomfort. But creating discomfort and anger in those you are trying to persuade is not usually effective for changing minds.

And while many men hear university DEI programs telling them that they are oppressive and must shut up, listen, and do penance for their privilege, there are other communities that are welcoming them and providing a sense of belonging and answers. Unfortunately, that welcoming community is often the toxic cyberworld known as the “manosphere.” Nearly half of young men aged 18 to 25 told Equimundo that they trust one or more of the “men’s rights,” anti-feminist influencers such as Andrew Tate, a self-described misogynist and one of the most famous stars on TikTok, or Jordan Peterson, whose self-help manual 12 Rules for Life is a life guide for many men, but who has also claimed that feminists have “an unconscious wish for brutal male domination.” Other welcoming communities include White supremacists, who have been recruiting young men on gaming platforms for years, and whose efforts more than doubled after 2021.

These influencers legitimize the White, male, Christian status hierarchy—and by doing so, they increase the likelihood of backlash when these views are challenged. Perhaps unsurprisingly, given that this competition for hearts and minds is ignored in the implementation of DEI programs, studies show record rates of young men claiming that they are discriminated against, and believing that men make better political leaders—higher than any other male age group. Young men also show record levels of personal despair compared to other age and gender groups, which one study suggests may be leading them to a desire to control others and fantasies of dominance. Steve Bannon early on recognized the power of drifting young men, and Donald Trump is now speaking to the manosphere directly to turn a cultural shift into a political one.

Now, any political change like this is likely to have many causes. But it sure looks as if DEI as it is currently being practiced is adding bricks to the very wall its proponents claim to want to knock down. While leaders of the MAGA movement are spearheading the greatest rollback of women’s rights in fifty years and contributing to an atmosphere in which hate crimes and threats against minorities and women are at the highest point this century, DEI as it is popularly practiced appears to be motivating more people to support these noxious goals.

How do polarized democracies make major social change?

Anyone who wants to advance the goals of pluralism needs to begin from an understanding that they are working within a polarized country—and reckon with what that means for their strategy.

According to polarization scholars Jennifer McCoy and Murat Somer, no wealthy, consolidated democracy has been as perniciously polarized for as long as the United States. Of the less fulsome democracies that have experienced similar levels, the track record is not good: the majority faced democratic degradation, and nearly all of those descended into authoritarianism. The few that managed to depolarize all repolarized a few years later.

There are many people who claim that backlash and polarization is an inevitable byproduct of social progress. It should therefore be seen as a sign of success. That’s a convenient posture, but it ignores everything we know about how highly polarized democracies make major social change.

Polarized countries are stuck in a vicious cycle. Reforms desired by one side are immediately dismissed as wrong by the other side. Polarized societies cannot move a major social change forward—instead, they whipsaw from one extreme to another, with each side dismantling the others’ gains when they win control. As Shamil Idriss, Executive Director of an organization that works to diffuse conflict and build peace in the most divided societies on earth, writes: “adversarial advocacy yields victories that last just long enough for adversaries to [counter-]organize and generate backlash.”

As with many polarized countries, this cycle opens the door to authoritarianism. In the United States, the most affectively polarized Americans are more willing to bend the rules for their own side to maintain control. We see this in the refusal of polarized college students to tolerate disfavored forms of speech. We also see it politically among polarized Republicans, who indicate much less willingness than Democrats to accept normal democratic checks and balances. Social justice reformers are correct: a polarized country does generate inevitable backlash to change if it is advanced by only one side. Where they are wrong is thinking they can simply overwhelm the other side and win.

Any program to enhance pluralism must therefore make depolarization a first-order goal.

Start with social contact

Polarized democracies get out of cycles of conflict by creating spaces, arguments, and ideas that can get large cross-sections of people to come together across their polarized divides to work towards shared goals. Achieving pluralism involves building coalitions of unlikely allies by searching out common ground on which to begin pushing shared ideas forward.

Luckily, we have reams of research dating back to the 1960s about how to foster positive relationships and understanding across difference—and how to reduce prejudice. Since the 1960s, social contact theory (not to be confused with social contract theory from political science) has been found to be highly effective, especially for young people in the United States, on whom it has been most tested. Programs that foster contact across difference allow people the vulnerability to question and diffuse their own prejudices.

Particularly for people resistant to change, social contact, emphasizing what people have in common, and fostering positive feelings and relationships, is just what the doctor ordered. In fact, serious studies have shown that programs grounded in social contact theory can change behavior even when people don’t think they’ve changed—as when Christians and Muslims in Iraq who underwent a social contact-based soccer program reduced their prejudicial discrimination in real life—even while reporting unchanged attitudes. As researchers Dobbin and Kalev found in a study of diversity training in corporations, many of the most effective programs fostered contact across difference but were not designed to reduce prejudice; they managed to reduce prejudice as a side effect of bringing people together for extended time as equals to accomplish common goals. Nor is social contact only good for changing the attitudes of the dominant group (the first of the three pluralist goals)—cross-group friendships also are important for building a sense of belonging among marginalized identities.

---

How To Put a Country Back Together

---

Gordon Allport, the father of social contact theory, posits four conditions necessary for it to work: equality of participants, approval of leaders, working together rather than in competition, and a common goal. A massive meta-study has shown that social contact may work even without these conditions being met—but that their inclusion enhances effectiveness. Luckily, universities offer the perfect setting. They bring students together as equals. Dorms and dining halls minimize status differences, and the university membership itself creates a “third identity” that students all share, which does not negate, but layers atop their existing identities. Having students do something constructive together is the goal of nearly all clubs and activities.

I’ve written before that social contact does not, in and of itself, change societies. For that, people need to use these relationships to build coalitions that unite across polarizing divides to make change. But social contact is the essential first step towards ensuring social change isn’t swept away by backlash.

Doing DEI effectively

Many DEI leaders would insist that by requiring students to come together regardless of identity to learn in the same mandatory courses, they are doing social contact theory. The problem is that the characteristic techniques of DEI as it is currently practiced likely short-circuit these hypothetical gains. For instance, DEI programs make diversity a possible minefield in which a wrong step can lead to social cancellation. A 2004 study found that dominant group members often avoid intergroup contact because they fear being seen as racist. Meanwhile, encouraging ethnic affinity groups—as many universities do—does the opposite of building contact and changing minds.

So what must be done differently to make programs actually serve the goals of pluralism?

1. Insist on equality and inclusion for all.

Social contact theory works best when identity groups and individuals are treated as moral equals and are not pitted in competition against one another. A 5,000-person study looking at the perception of a variety of civic and democracy-related words found that “racial equity” was the term least seen by participants as “meant for them” and the term least associated with “bringing people together.”

By contrast, a framework of “inclusion for all” requires that we call everyone in, rather than positing diversity as a zero-sum competition for status. To create a more equal society, we must engage in diversity programming in which those from both minority and majority identities see the initiatives as potentially positive for them and the groups with which they identify. For instance, an intervention that treated gender bias not as a moral wound, not as a problem particular to men, but merely as a habit that both men and women might have inadvertently picked up, and that emphasized the ability of individuals to alter their own habits, is among the most rigorously researched and effective DEI program. Not only did academic departments that featured this intervention report more positive feelings—but two years later, they had hired more female faculty members than other departments.

2. Complexify identity.

Relationships of respect enable individuals to disagree while also remembering that each of us is more complex than our perspective on any single issue. That viewpoint comes more easily when DEI programs complexify identity.

Seeing other people through a lens of power, privilege, and oppression simplifies people into stereotypes. It also pushes group interactions into discomfort and lack of authenticity at best, and actual conflict at worst. Societies where having one identity is predictive of having another (i.e. being urban, liberal, and a minority) such that identities “stack,” are much more likely to fall into conflict. Pretending that how someone looks or identifies predicts how they feel increases the likelihood of violence and discrimination because people think that when they see how someone looks, it predicts how they think and vote.

Programs to embrace diversity should make room for all people to have complex identities in which all their parts do not line up with the majority of their group. In the words of Diana McLain Smith “closing the distance across groups … requires us to open up the space within groups.” This entails accepting the diversity within as well as between groups.

The same need for complexity extends to culture. Of course, there are noxious instances of cultural appropriation that are meant to convey disrespect or involve outright theft. But the concept of cultural appropriation is itself linked to a dangerous history; the very idea of a “pure” culture is one steeped in eugenics and blood. Nor is it historically accurate. Universities should, instead, acknowledge the constant interplay of cultures, how they have borrowed from each other since humanity began. Few cultures have been constant oppressors—more commonly, people who have been harmed themselves do harm to others, so that groups and individuals are often both oppressed and oppressor. Acknowledging moral and historical complexity lets no one off the hook, and requires all students to be aware of their own potential to both do good and harm to others.

3. Require rigor and curiosity.

DEI courses should help students understand and grapple with the behavioral science that affects how we interact across diversity: the way our brains are wired to create “us” and “them,” and how the creation of groups affects our behavior and causes motivated cognition, for instance. Such rigor should simply be a given in a university setting, even for professors who believe their main job is not to teach but to inculcate cultural change. Only by understanding why people divide into groups and how our brains and societies work can the social practices that embed dominant groups be overcome.

One form of rigor that is particularly essential to doing diversity differently is an end to trigger warnings and the fear of identity harm they entail. Diversity courses should help students get curious about one another and about themselves—not shut down questioning as too sensitive or hurtful. Toni Morrison said that speech is violence. It is not. Words can lead to violence—and as a scholar of political violence, I know that a very particular type of dehumanizing speech does precisely that. But conversations that explore ideas—even ideas that make some people uncomfortable—are fundamentally different, if they are pursued in a setting of respect for individuals and curiosity.

Moreover, decades of study into how to overcome anxiety shows that exposure actually reduces fear—while avoiding triggering content and even reading trigger warnings can increase anxiety, and might be particularly bad for the precise population they are intended to support: survivors of trauma, among whom studies have shown trigger warnings can worsen PTSD.

4. Deepen a sense of agency and possibility.

A pessimism pervades the DEI movement. While claiming to want massive structural change, many DEI tenets simultaneously imply that such change is impossible. It is true that American history—like that of most countries—is a trajectory of lurching, uneven, and often backwards motion. But it is also true, in the words of Transportation Secretary Pete Buttigieg, that his “kind of [gay family] life went from impossible to possible, from possible to real, from real to almost ordinary, in less than half a lifetime.” It is equally true that this process was not inevitable but required careful choices and strategies to make it happen.

Right now, as Steven Levitsky and Daniel Ziblatt write in Tyranny of the Minority, America has the largest majority of people it has ever had who want a multiracial and inclusive country. Since 1980, the number of African Americans in Congress has tripled, Latinos have nearly 8 times the representation they used to have, and Native Americans have gone from zero to five Members of Congress. While in 1973, 64% of Americans wouldn't support laws that banned discrimination in home sales, preferring to let homeowners decide for themselves, by 2014, those percentages had more than flipped, with 70% willing to vote for a law that made home sale discrimination illegal. Intermarriage has skyrocketed, while the number of people who oppose marriage outside of one’s race is now far lower than those who oppose marrying someone from the opposite political party!

If change is not possible, then taking a maximally polarizing posture harms nothing and enables one to signal one’s identity. But if change is possible, and one claims to want it, that forces students to make the effort to understand how to get there and to take responsibility for the outcomes of their actions.

The path to pluralism

Universities are bottleneck institutions—individuals who make it through have vastly greater life chances than those who do not attend or do not graduate. Their immense societal importance means that universities have a responsibility to enable their students to flourish in a diverse, pluralistic society in which people are entitled to disagree. We do not need universities to create a scorched earth, polarized battlefield.

The path to a pluralistic society is not inevitable. That makes it even more imperative that those who seek to broaden the dominant status hierarchy don’t fool around. This is no time for playing with interventions with no track record of success that might cause backlash. A goal as important and significant as pluralism needs to implement programs based on evidence of what works.

Excellent efforts that fit that bill are being promulgated by thinkers such as Kwame Anthony Appiah, Irshad Manji, Danielle Allen, and Eboo Patel—or to go back further, John Dewey, John Stuart Mill, and others who have thought deeply about how to educate for a pluralistic society of belonging and equality. Their ideas are ready to go. They have never been more urgent.

Dr. Rachel Kleinfeld is a senior fellow at the Carnegie Endowment for International Peace.

---

Follow Persuasion on Twitter, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.

And, to receive pieces like this in your inbox and support our work, subscribe below:

1. My capitalization of White and Black follows the logic of Kwame Anthony Appiah in The Atlantic, June 18, 2020—to remind readers that both of these and other racial categorizations are socially constructed and have no consistent basis in biological, cross-country, or even historical reality.