Last week, The Atlantic’s Vann R. Newkirk II laid out, in a thoughtful article, his views on an argument that has been taking place mostly on social media for a while, and which has ramped up considerably since Donald Trump’s shocking election victory last month. It’s an internecine fight among left-of-center folks, mostly, and it centers on the question of whether and to what extent progressive language on race and racism — specifically, the practice of explicitly “calling out” instances of racism when they crop up — is an effective tool for achieving various social-justice ends.
Newkirk’s article centers on the argument that, as his headline has it, “Sometimes There Are More Important Goals Than Civility.” He cites me as one of the liberals advancing the counterargument — that the most important goal in political activity is civility — and argues that this view in effect throws minorities under the bus, forcing them to engage in the exhausting task of “coddling” whites who engage in bigotry. He also argues that the research I and others have cited in defending more subtle approaches to persuasion “revolve[s] around episodic interpersonal interactions, and not necessarily around the complex sociological processes by which social mores are made, enforced, and internalized.”
I can understand why Newkirk hears, in liberal arguments against the efficacy of callouts, ugly echos of white people telling black people to “just calm down” over racism, but that isn’t my intent at all. Rather, my argument is simply that in situations where persuasion matters and is realistic, liberals tend to overestimate the effectiveness of calling out racism (and other forms of bigotry) by name, especially given that social scientists have come up with approaches that appear to work better and sidestep inevitably defensive responses (white people being one notoriously defensive group).
This doesn’t mean stigma and shaming never work, it doesn’t mean civility should always be the end goal, and it especially doesn’t mean that members of minority groups should ever be told the “right” way to react to racism or oppression. What it means is that sometimes, in the service of a specific political goal, liberals shouldn’t automatically stick to the same sorts of language and moral framing they use when talking among themselves. It also means that we need to be careful to not conflate two different things: asking a member of a minority group to engage with people who disrespect or dehumanize them for the sake of persuading that person to contribute to a particular, important political goal, versus policing or criticizing how members of minority groups react to oppression more generally, in situations where persuasion isn’t the point.
Newkirk begins his article by running down several articles expressing skepticism about the efficacy of callouts, including my piece on the political and persuasion limitations of Hillary Clinton’s “deplorables” remark and my colleague Drake Baer’s examination of the defensive way in which white people respond to accusations of racism. Then he lays out his thesis. There is a time and a place for civil, calm discourse, he writes, but it’s important to recognize that there are also many exceptions:
In the aggregate, though, these calls for civility threaten to impose a burden on people of color. If calling out racism is largely counterproductive, using a systemic definition like white supremacy is also unacceptable, and stigmatizing or shaming those who espouse racist beliefs is self-defeating, what tools remain? The only form of productive debate that people of color can engage in, it seems, is the gentle persuasion of white people who may or may not hold retrograde views.
That advice is of course probably most appealing to white Americans, for whom the social cost of being called racist may loom larger than the effects of racism itself, or for whom the ideal of a functioning marketplace of civil ideas is more important than the worry that they might be carved out of it. White Americans share a vested interest in not being called racist, straight people in not being called homophobic, and men in not being called misogynistic. Arguments in favor of civility cede valuable rhetorical ground by default and coddle people who may well know the score about their own views. Skepticism should be a default position here; instead, the bigoted views of individuals are privileged as artifacts of ignorance, and thus not considered as purposeful efforts to sabotage debate.
Minorities may suffer from racism and bigotry that goes well beyond incivility, but these arguments urge that it is their job in debate to remain civil, because that is the only productive way to reach across the aisle. As Singal notes, this is useful advice because racism is not always accepted as any single thing, and the parameters of the debate often depend on whether or not a belief or characteristic is actually racist.
Again, Newkirk is absolutely right that members of majority groups really have no right to dictate to members of minority groups the “correct” or “civil” way to react when they encounter bigotry. No one has any standing to dictate how another person should deal with any form of adversity, let alone one as painful as exposure to bigotry. This should be beyond debate.
But when I talk or write about the social science of persuasion, I’m simply referring to which tactics are most likely to work in those situations — and only those situations — where persuasion is a desired outcome. In many such situations, the target of the appeal may have significant ideological differences with the person making the appeal (or, to offer a less sugarcoated version, they may be racist and/or have racist views). But convincing people with different or less evolved views is one of the goals of just about any political movement — progressives need to be able to argue over tactics and messaging, since this is what political movements do. These arguments simply don’t apply to situations where there’s no specific “ask,” where you’re not trying to get someone to vote a certain way or sign a petition or whatever else.
Nor do persuasion arguments apply to the broad set of political acts and statements that aren’t about persuasion, anyway. Newkirk makes this point nicely:
For people who suffer the incivil burden of bigotry, [the claim that persuasion should always be the end goal of an argument] doesn’t quite hold up. Sometimes the goal of argument is to vent. Sometimes it is to simply tell the truth. Sometimes it’s just to loudly proclaim one’s own humanity. The general burden to always remain civil in arguments—even if it means coddling white egos and casting a blind eye to obvious bigotry—can even create that need to commit to truth-telling at any cost. Civil discussions with people who themselves may have already breached the bounds of civility are difficult. One way around that difficulty for marginalized folks is abandoning civility.
The labels of racism and bigotry can impose a social cost on bigoted actions, policy preferences, or speech, regardless of whether hearts or minds are changed. Stigma can be useful.
This crucial groundwork set — no one should be dictating “civility” toward perpetrators of bigotry, and persuasion isn’t always the goal — I do want to push back against Newkirk’s claim that recent persuasion research doesn’t account for the “complex sociological processes by which social mores are made, enforced, and internalized.” That isn’t true — the research that most informs my views on the limitations of certain types of callouts, at least, absolutely takes these factors into account, though to be fair I didn’t reference all of that research in the specific article of mine Newkirk cites.
Here it’s useful to look at two of the the most important behavior- and attitude-change studies of 2016, both of which highlight the potential limitations and complexities of simply “telling it like it is” as a tactic for getting people to act or think differently. Both studies show that more subtle, nuanced approaches are often a better bet.
In the first study, political scientists David Broockman and Joshua Kalla found that when residents of Miami-Dade County had brief conversations with canvassers who came to their doors to talk to them about transgender issues, and who followed a very specific technique in doing so, it induced surprisingly large and lasting upticks in the residents’ approval of trans people and their support for trans rights. The key thing, I wrote in my coverage, was to tie the conversation to the experiences of the voters themselves, which the canvassers did by “ask[ing] the subject whether they’d ever experienced stigma or negative judgment like a transgender person might”:
Broockman and Kalla think two mechanisms are at work here: “active processing” and “analogic perspective-taking.” Active processing simply means getting someone to actually engage with an issue in a thoughtful way — here Broockman referenced Daniel Kahneman’s famous divide between system 1, which is gut-level, quick-response thinking, and system 2, which involves deeper thought and rumination. Active processing, which the researchers think is triggered by asking respondents to explain their opinions on transgender people, nudges people into system 2 thinking. Analogic perspective-taking, meanwhile, is simply the act of putting yourself in someone else’s shoes and considering similarities between you and them.
In the mixing of the two is where the researchers think the magic lies: “Active processing signals that the perspective-taking you’re doing is not a shallow exercise,” said Broockman, “but one where you’re engaging with a lot of cognitive effort.” You’re not just saying, Yeah, yeah, transgender rights are great — got it. You’re in a place where you really can understand, in an intellectual way, that there are some similarities between a transgender person’s experience and your own.
The second study does offer support for certain types of stigma and shaming, but it also nicely emphasizes that wielding these tools requires a deft touch and an understanding of exactly how they operate. For the study, the psychologist Betsy Levy Paluck, the sociologist Hana Shepherd, and the political scientist Peter Aronow mapped the social networks of a bunch of middle schools in New Jersey, figured out which students were the most “connected,” and got to work. “Seed students were … encouraged to become the public face of opposition to these types of conflict,” the researchers write. Crucially, the research assistants who worked with the seed students gave them a ton of leeway to develop initiatives that they, the kids, thought would work: “This intervention model can be likened to a grassroots campaign in which the seed students took the lead and customized the intervention to address the problems they noted at their school,” wrote Paluck and her colleagues. “Notably, it lacked an educational or persuasive unit regarding adult-defined problems at their school.” In fact, one of the very reasons the study was even conducted was that there’s a severe paucity of evidence suggesting that traditional anti-bullying programs — which tend to include tactics like piling kids into an auditorium and having authority figures or outside speakers tell them that bullying is bad and harmful — are effective.
Paluck and her colleague’s different, more nuanced intervention was quite successful; schools in the pilot group experienced sizable reductions in conflict relative to a control group of schools which didn’t implement the program. Surely stigma and shaming contributed to the program’s success, and surely in the schools where the program took hold bullies were “called out” for their behavior. But that calling out didn’t occur in a top-down way; rather, it came from exactly those students who were seen as “social referents” within the school network. The underlying logic is simple: If I’m a Jersey teen and a stodgy principal tells me not to bully, I may laugh it off and go right on ahead abusing the nerds; if that message instead comes from some of the most influential kids in my class, and they have demonstrated a commitment to broadcasting and enforcing these norms, there will be a social cost to pay — potentially severe — if I continue to be a jerk.
So this bullying study contains a key insight: when stigma and shaming surrounding the violation of social norms works, it’s usually because they come from members of one’s in-group. To oversimplify a lot of research, shame and stigma coming instead from them — from a group the target of the shame or stigma doesn’t belong to or identify with — is likely to bounce right off people, and in some cases, especially when the subject in question has been politicized, may harden their behavior or views. In the Jersey schools, the anti-bullying messages had to come from kids, who were fluent in Jersey Adolescent, and not from adults, who weren’t respected in the same way as exemplars of acceptable norms.
This fits nicely into an argument which progressives often make: White people have an outsize role to play in talking to other white people — that is, other members of an in-group that matters a great deal in this country — about racial injustice. Given the dynamics of segregation and social networks in the U.S., there is a lot of wisdom to this view. A conservative-leaning white voter with reactionary, racist views is likely to going to ignore arguments about racial justice coming from liberal pundits or a Democratic campaign. But if their longtime neighbor pushes back on their belief that Black Lives Matter is a terrorist organization, or the church they have been a member of for years tacks in a racial-justice direction in response to the Flint water crisis, that could open a window a crack (though this is by no means guaranteed). Neighbors and churches and other institutions have a potentially important role to play in all this, and, yes, that role largely boils down to nudging social norms and enacting certain types of stigma.
Other progressive beliefs about stigma and shaming are less well-informed, though. Both the bullying and the canvassing study highlight the weaknesses of leaning too heavily on what could be called broad-based or top-down callout techniques to get people to act better or embrace social-justice ideals. Imagine an impassioned anti-bullying activist popping into one of Paluck’s schools, or a transgender-rights activist tagging along on one of the canvassing trips. If they weren’t familiar with the research motivating these studies, there’s a chance they would be shocked: In the school, they might ask, Why aren’t there any educational programs about bullying in this school? Why aren’t adults regularly telling kids not to bully, given how vital a message this is? On the canvassing trips, they might ask, Why are you spending so much time talking about the voters’ experiences instead of the alarming numbers of trans people murdered every year? Why are you sugarcoating this? In both cases, the researchers would answer: Because we think this works better, and because we think we have the evidence to prove it. In neither case would it be fair to accuse the researchers of “not caring” about bullying or trans rights, or of wanting to “coddle” bullies or transphobes by withholding certain sorts of information or callouts.
One paragraph from Newkirk’s article nicely encapsulates the dangers of progressive overconfidence about the effectiveness of stigma and shaming:
Motivated candidates and institutions can create social conditions and stigmas by which bigotry is diminished, and they also change the way in which media transmit information and people absorb it. Imagine if the same outrage manifest in media coverage about the ideas of microaggressions and safe spaces pioneered by marginalized people had been marshaled against stubborn implicit racial biases and resistance to multiculturalism among whites, or if the useless term “racially charged” in media descriptions of racist things had been replaced with something more potent, like “racist.”
Yes, publications of all stripes should be less mealymouthed about calling racism racism, but the question of whether candidates, institutions, and media “can create social conditions and stigmas by which bigotry is diminished” depends hugely on the specifics, and it’s often the case that this simply isn’t true. The New York Times has very limited power to shape the attitudes of people who don’t read the Times, or who view it as a mouthpiece for out-of-touch coastal elitism (setting aside whether you think this concept makes any sense). At a time when media is incredibly fragmented and everyone is constantly having their own worldview, well, coddled, there’s more reason than ever before to be skeptical of optimistic theories about the media’s ability to shape norms and behavior across partisan or ideological lines. That’s the point of all the social-science research on persuasion and norms and shaming — to figure out how and why shaming and stigma work, rather than to assume they always do.
One final, important point on the difference between asking members of minority groups to “coddle” their oppressors versus giving them the tools to be effective advocates and persuaders: Sometimes, political work or activism does entail asking members of marginalized groups to, in effect, face down some of the people perpetrating or contributing to their oppression. In the Broockman and Kalls study, many of the canvassers were themselves trans (one of the things the researchers wanted to test was whether trans canvassers were more or less effective than cisgender ones, but it turned out not to matter). But there was sound justification for them to temporarily suppress their desire to call out transphobia for what it is, or to react angrily to the slanders frequently leveled at trans people: They wanted to get these voters on their side, to potentially enlist them in the broader fight to secure the rights of trans people. If you’re asking people to do something for you, you do need to coddle them, in a sense. If you call them transphobes (or racists or misogynists), they are often going to respond defensively, which might squander any chance you had to convince them to do what you’re asking them to do.
But it doesn’t make sense to say trans people shouldn’t participate in this sort of canvassing, or that it’s unfair to ask them to — some trans activists are probably up for this sort of intense, persuasive work, while others might choose to get involved in other capacities. It’s a different sort of “ask” than making the indefensible demand that trans people respond “civilly” to transphobia they encounter in their everyday life, as opposed to on a canvassing trip.
Overall, this whole argument is more about degree than kind. Shaming and social stigma can work in certain situations, and no one should be arguing otherwise, or for any sort of universal, pro-civility rule in general. But stigma and shaming are not going to work if you don’t know how to use them, or if you assume that someone from a different group than your own is going to be receptive to moral appeals couched in language that’s familiar to you. I can’t speak for everyone making arguments about the shortcomings of certain types of callouts, but I have zero desire to “coddle racists.” Rather, I simply want progressives to adopt the most effective means of political persuasion possible.