Fear

Martha Nussbaum (2018) argues that fear lies at the core of all of the emotions that undermine democracy and promote support for authoritarian politics. I doubt that all of these emotions are reducible to fear. Resentment, for example, is often backward-looking, whereas fear is anticipatory. Nevertheless, I agree that fear is a powerful driver of our toxic discourse, particularly as it applies to fear of diverse others. Karen Stenner (2005) argues that fear of diversity—of opinions, lifestyles, and identities—powers authoritarian politics. Toxic discourse helps people manage their fear by subordinating, silencing, and marginalizing diverse others. So let us begin by considering how fear of diverse others can be reduced.

Dr. Ayaz Virji is a Muslim doctor with an interest in rural medicine who built a practice in Dawson, Minnesota, an overwhelmingly white Christian town. At first, the residents of Dawson welcomed him and his family. However, he and his family began to feel hostility during and after the 2016 election, in which Trump campaigned on a pledge to ban Muslims from entering the United States. Virji at first responded with anger to this hostility. He was taking care of the very people who were treating him and his family like suspected terrorists. Yet, instead of reciprocating hostility, Virji decided to ‘transform anger into understanding’ by considering why his fellow community members had become hostile. A Lutheran pastor in town, who was concerned about terrible things members of her prayer group were saying about Muslims, asked him to give a talk about Islam to the community. She thought that Virji could allay the fears underlying their hostility. Virji agreed to speak in Dawson and nearby towns.

In his talks, Virji explained ‘what 99.99%’ of Muslims believe, in contrast with stereotypes about Islam and terrorism. He invited his audience to consider how they would respond to similar demonization: ‘If you say, “[ISIS] is Islam”, then that’s like me saying, “Well, Christianity is David Koresh [the polygamous leader of a Christian cult, whose members died in a conflict with federal agents in 1993]”‘. Virji explained that he left a large urban practice for one in rural Minnesota because he wanted to take care of underserved communities. He showed family photos: ‘Look! We’re normal! . . . That’s our cat!’ He smiled and joked. And he spoke from the heart, describing his pain in being subjected to negative stereotyping. Not everyone was transformed by his talks. But many showed their appreciation with applause. As news about his talks spread, other communities invited him to speak (McCrummen 2017).

Virji’s talks exercised a kind of emotional intelligence that offers lessons for us all. Fear is a response to a felt perception of threat. Several factors contribute to such perceptions, including the sense that the target is powerful, hidden, and radically other—not belonging to the same social groups to which the perceiver belongs or sharing the same values. Without power, the target is in no position to harm. Visibility enables one to better assess whether the target really is threatening and to take precautions against threats. Shared identities and concerns disarm fear because they take advantage of ubiquitous ethnocentric or in-group preferences. In testifying from the heart about his personal experiences, day-to-day life, faith, and values, Virji disarmed the feelings that made him a target of fear-based hostility. He showed his vulnerability. His voice and body language conveyed sincerity and openness in confessing his faith and values—his commitments to caring for people just like those in his audiences and, drawing from his Islamic faith, to living in peace with others. In showing pictures of his life and everyday concerns, he showed how he shared salient identities and concerns with his audience—he is a family man, a member of the community, a pet owner, a middle-class American. In confessing his own distress and anger at being demonized, he showed that he had nothing to hide. No wonder one member of the audience who spoke during a question-and-answer session following his talk said, ‘I hear a lot of pain from you this evening . . . Um, I’m sorry’.

Of course, it would be unjust to expect every member of a marginalized group to assume the huge burdens Virji took on in his speaking tour. The work of reducing fear needs to be fairly distributed. It is also more effective and less costly if it occurs in the normal course of ordinary interactions than in special settings set aside for such burdensome emotional labor.

My point is that even many of those who express their fears aggressively can be reached. Once we see how much identity defense underlies these fearful responses, we can understand how self-defeating it is to shame people as bigots for being afraid. That only provokes more identity-protective aggression. The vast majority of people are also capable of sympathy. Sympathy counteracts fear by responding to sincere, heartfelt communication that conveys good will and common interests. So, communication that activates sympathy is liable to be more effective than shaming in disarming fear and aggression.

Resentment

Overcoming resentment raises different communicative challenges. Our difficulty is heightened by the fact that in many cases, our interlocutors are often causally implicated in the problem that needs to be solved. In this respect, for example, pandemics are not like earthquakes. For purely natural disasters such as earthquakes, others may be called to contribute to disaster relief even though no one was responsible for causing the disaster. No blame is implied when people are called to pitch in. By contrast, pandemics get worse when people behave in ways that increase the spread of disease—for instance, by attending large parties and refusing to wear masks or refusing to get vaccinated. Pointing this out to such people is difficult to separate from blaming and shaming. And blaming and shaming tend to spark resentment and resistance, as people dig in their heels to defend their status and social identities against an interlocutor perceived to be arrogantly asserting their moral superiority. Blaming and shaming seem to presume that the individual doesn’t care about harming others, when what may actually motivate them is distrust of those who claim that the measures they demand are necessary to prevent harm (Larson 2020).

The problem of disarming resentment is even more challenging when the issue involves an injustice in the relative status accorded to different social groups. By ‘status’ I refer not just to esteem but also to power, standing, and considerability. To have standing is to possess an acknowledged right to make claims on others and hold them to account for failure to heed those claims. To be accorded considerability is for others to weigh one’s interests in their deliberations about what to do, especially in public and corporate policy-making; for others to be responsive to one’s interests in their habits, social norms, and organizational policies; and for such consideration to be built into the social infrastructure—for example, in urban design and the siting of public goods and bads. These are the stakes in all challenges to racism, sexism, homophobia, and other types of group inequality.

Let’s focus on the case of racism, where the issues are clearest precisely because, in the post-civil rights, postcolonialist era, the principle of racial equality under the law is almost universally avowed. Few people openly call for a return to explicit, legally instituted racism among citizens. Beyond purely formal equality, it is also almost universally conceded that it is wrong to deliberately harm others on account of their race. Yet dismantling explicit legal racism and condemning deliberate harm on the basis of race are only the first two steps to ending racism in society. As Alexis de Tocqueville (2004) shrewdly observed,

There is a natural prejudice that leads a man to scorn a person who had been his inferior long after that person has become his equal. The real inequality resulting from fortune or law is always replaced by an imaginary inequality rooted in mores. (393)

That is, informal social norms of unequal treatment and consideration, deeply rooted in sometimes unconscious habits and attitudes of disdain, aversion, and neglect, survive the dismantling of formal racial hierarchy. In Tocqueville’s view, such attitudes and associated norms even tend to harden, to shore up a group’s accustomed superior status when legal supports for hierarchy are taken away.

How, then, can members of a group formerly recognized by law as a superior race reconcile their avowed repudiation of racism with persistent habits and attitudes—which, by their automatic nature, evade conscious control even if one wants to extinguish them—and with their desire to see themselves and be recognized by others as good people? One method is to define racism down. In 2018, Keith Burris, the editor of the Pittsburgh Post-Gazette, published a controversial unsigned editorial complaining about the excessively wide scope of the term ‘racist’:

We need to confine the word ‘racist’ to people like Bull Connor and Dylann Roof. For if every person who speaks inelegantly, or from a position of privilege, or ignorance, or expresses an idea we dislike, or happens to be a white male, is a racist, the term is devoid of meaning.

In Burris’s view, only people who do things such as setting bulldogs against antiracist demonstrators, murdering people out of hatred of their race, or spouting inflammatory racist ideology, should be called ‘racist’. Applied to anyone else, he claims that it is just a malicious insult and an outrageous slander, not an attempt to communicate a legitimate complaint about racial injustice. Hence, outside of this narrow range of cases, complaints about racism should not only be dismissed out of hand but met with a vigorous defense of one’s own standing.

Burris’s editorial conflates two issues: (1) the proper scope of the word ‘racist’, and (2) whether blacks or other historically subordinated groups have any legitimate complaints about contemporary systematic racial injustice, racial stigmatization, or other objectionable treatment that can be traced to the fact that they are racialized as not white. Let us postpone consideration of the word ‘racist’, which, in Burris’s eyes as in the perceptions of many white people, is mostly an expression of moral contempt. The deeper question is the second. Resort to second-order identity defense and esteem competition does not arise only because of the choice of words to communicate moral concerns about racial injustice. And resentment grounds much resistance to the claims for justice made by subordinated and marginalized groups, even when those claims are not couched in terms like ‘racist’.

Consider how a resentment-based framing played out in Marathon County, Wisconsin. In 2021, the county board convened a meeting to consider a resolution to declare the county ‘a community for all’, as called for by black and Hmong residents who comprised a small minority of the population and who experienced discrimination there. Some had to get a white person to vouch for them before they could rent an apartment. Others faced hostility from whites who assumed they must be on welfare. White opponents of the resolution denied that racism existed: they hadn’t heard of any discrimination. They claimed that race relations were fine until self-interested people started complaining about discrimination. Even raising the issue of discrimination was divisive. Accusations of discrimination amount to racism against white people. They complained that they were the ones being misunderstood and unfairly stigmatized: ‘You can’t come around and tell people that work their tails off from daylight to dark and tell them that they got white privilege and they’re racist and they’ve got to treat the Hmongs and the coloreds and the gays better because they’re racist. People are sick of it’. Even if there is discrimination, it’s not the county’s business to remedy it. And remedies would themselves simply ‘elevate one group of people above another’. One opponent of the resolution put the point succinctly: ‘Government cannot give someone something without taking it away from someone else’ (Epstein 2021).

This episode illustrates a standard playbook for resentful resistance that includes a battery of strategies besides overtly changing the subject to second-order concerns. It includes denying that the complainants suffer from discrimination or disadvantage, claiming that if the speaker is unaware of discrimination it must not exist, insisting without evidence on alternative explanations of complainants’ disadvantage, blaming the victims, dismissing the complainants as hypersensitive ‘snowflakes’ or self-interested agitators, disclaiming responsibility for addressing the complaint because the wrong was committed by others, rejecting complaints as inherently divisive, and insisting that bygones be bygones—that past wrongs should be forgotten, notwithstanding their continuing harmful effects. Resentment also underlies resistance to proposed remedies, expressed in claims that complainants are slandering their addressees, demanding ‘special rights’ rather than equal rights, or proposing to turn the tables against the addressees by engaging in ‘reverse discrimination’. In short, resistance to complaints is founded on denial of injustice, while resistance to remedies is founded on the idea that they must result in an unfair sacrifice imposed on one’s own group.

The critical function of the playbook is to deny the legitimacy of even giving complaints and remedies a serious hearing. Complaints are dismissed prior to a serious examination of evidence; evidence dismissed out of hand; this isn’t the right time or forum in which to hear them. The playbook is ready for deployment on every occasion. Those who deploy the playbook enact a central feature of a racist society: the insistence by one racial group that it is unaccountable to those it considers inferior, that the latter have no standing to bring complaints that the former must hear. As Justice Taney claimed, blacks are ‘so far inferior that they had no rights which the white man was bound to respect’.¹

But for whites consumed by racial resentment, that’s not how they see themselves. Arlie Hochschild (2016: ch. 9) argues that white racial resentment stems from a positional framing of claims for racial justice: here white people are, working hard, following the rules, minding their own business, patiently waiting in line for fulfillment of the American Dream when blacks and other minorities, who they assume have not been working so hard or following the rules, show up out of the blue and demand to cut in line ahead of them. As Heather McGhee has argued (2021: ch. 6), such zero-sum thinking is not just false—it is fatal for democracy. Whites in Marathon County didn’t object to claims that racism is disadvantaging black and Hmong people only because they felt that this threatened a redistribution of material resources away from themselves. They also objected because they felt that such claims amounted to an attack on their characters. They felt that white people as a group were being attacked as racists, and that this attack amounted to antiwhite racism. They rejected a first-order moral concern about discrimination and exclusion by interpreting it as a second-order claim about the relative moral worth of different racial groups: black and Hmong people were trying to raise themselves in status by denigrating white people as evil.

Can we communicate with others in ways that disarm such resentful framing of claims against discrimination? David Broockman and Joshua Kalla (2016) have demonstrated one method, which they call ‘analogic perspective-taking’. In a controlled door-to-door canvassing experiment in Florida, they tried this out with a proposed law designed to protect transgender people from discrimination in housing, employment, and public accommodations. Canvassers usually present one-sided arguments for the policy they advocate. This may succeed in preaching to the choir. But it has little positive influence on those who resentfully feel that the way the arguments are framed runs roughshod over their own perspectives. So Broockman and Kalla directed their canvassers not to argue for their own position but to listen attentively to how voters explained their own thinking about the issue. Canvassers invited voters to reflect on whether they had ever felt that they were judged negatively because they were viewed as different from the others. Practically everyone has felt this at some point, even if such experiences have not pervaded their lives. Voters were forthcoming in relating their experiences. The canvassers disclosed that they or a close friend were transgender, calmly described their or their friend’s analogous experiences, and encouraged them to consider how their own experiences might help them understand how transgender people feel. Voters were then asked to describe their reflections and whether this changed their minds about the issue. Canvassers were nonjudgmental at all stages in responding to voters’ perspectives. Broockman and Kalla found that voters who undertook this exercise were persistently more likely to support the antidiscrimination law than voters in a control group. They found similar results in an experiment concerning exclusionary immigration policy (Kalla and Broockman 2020).

I suggest that such nondirective, nonjudgmental conversations helped to allay resentful, zero-sum framing by several means. Attentive listening to others’ perspectives is an important way of expressing respect for others. Voters were invited to reflect without being pressured to respond in a particular way and without being judged negatively for their answers. By letting voters come to their own conclusions, the canvassers showed respect for voters’ autonomy and independent judgment. These discussions scrupulously avoided activating feelings that addressees were being unfairly attacked or demoted for who they are. The analogic perspective-taking exercise enlarges voters’ empathy for people different from them by encouraging them to see similarity between themselves and others. It invites them to feel not just sympathy but perhaps indignation on behalf of others similarly judged—a feeling that might also support their own sense of themselves as good people who aren’t bigots. An invitation to analogic perspective-taking thus offers a way to help people focus on the first-order moral concern without feeling that their identities and status are threatened.

Moral Pride and Contempt

Communication is a two-way street. We need to critically examine not only the problematic responses of those who are addressed, who may respond with fear and resentment to calls for remedy, but also how those who raise first-order concerns may express moral pride in themselves and contempt for those they address. Expressions of pride and contempt both enact and trigger status competition and thereby contribute to the problem.

The difficulty is that calling attention to unjust and harmful conduct may seem inherently to blame and shame the addressee and to place the speaker in a morally superior position to the addressee. Yet it is a moral error to believe that speakers assert their moral superiority in calling upon others to do their part to solve a problem. If a speaker had to be morally blameless or perfect in all matters, or on the matter in question, in order to be entitled to complain or call another to action on a common concern, then no one could have standing to complain, call attention to a problem, or call for joint action to address it. To impose such a prerequisite on making moral claims is a formula for universal impunity and moral failure. All moral discourse would consist in arguing over who is better than who rather than in what ought to be done.

Those who wish to focus others’ minds on the first-order moral concern should consider whether their communication is likely to induce others to make this moral mistake. They should therefore critically examine whether raising the concern would be constructive in context. The marginal moral good of being the ten thousandth person to publicly denounce a racially offensive joke by an unknown individual is certainly negative. Consider all the harms of piling on against the latest social media outrage. Doing so publicizes the joke more widely, spreading the harm. It inflicts harm on the joker grossly disproportionate to the offense. If the joker was trolling, joining in mass denunciation gratifies the joker’s wishes to upset many others and grab attention. Piling on is emotionally exhausting. It alienates people who could be called to constructive action with gentler means, by making them worry that the slightest misstep will lead to their social ostracism. It pushes others who don’t get the joke or see what is racist about it into sullen, resentful silence when gentler modes of communication might offer them moral insight.

Because communication is a two-way street, those who wish to raise first-order moral concerns to others would do well to consider how those others understand their words. Recall Buriss’s complaint about the word ‘racist’. In academic and progressive circles, ‘racism’ has acquired a very wide scope. Besides covering malicious acts of discrimination and violence, it covers all of the myriad ways in which society is organized so as to afford greater status, benefits, and opportunities to whites and to concentrate stigma, burdens, and deprivation on people of color. Many of these ways are built into the social infrastructure of de facto racial segregation, in such matters as the layout of highways; the boundaries between municipalities, school districts, and electoral constituencies; the siting of polluting industries; even the distribution of publicly provided goods such as parks and trees (Anderson 2010). Some of these patterns are due to deliberate discrimination decades earlier. Yet, like the concentration of polluting industries in areas where people of color live, they perpetuate racial inequalities for generations. Some are due to negligence or unconscious biases that, as a matter of fact, fail to accord equal consideration to people of color. Hence, in academic and progressive circles, the term ‘racist’ applies far beyond individual acts of deliberate malice.

Two difficulties arise when ‘racist’ is used for political communication across partisan and racial lines, however. First, outside of academic and progressive circles, almost no one talks like this. Especially among whites without a recent college degree, this term is reserved to express moral opprobrium directed against the most hateful and vile individuals. Hence, when people use the term ‘racist’ to communicate first-order moral concerns to people who use that word exclusively as a second-order moral judgment, this will predictably trigger a response focused on identity defense and esteem competition. In our polarized state, addressees are liable to express their second-order concerns with toxic discourse. They may interpret the expression of the first-order concern as just a matter of smug and contemptuous liberal posturing and respond with bullying. Alternatively, they may respond with what Robin DiAngelo (2018) calls ‘white fragility’: acting hurt, crying, and changing the subject to how they have been wounded simply by the fact that the concern was raised.

Second, because the term ‘racist’ is used with such wide scope, it often leaves addressees in the dark about what is wrong. The term is not precise enough to delineate a complaint in a way that can facilitate discussion of appropriate remedies. This is particularly problematic in the United States, where moral discourse tends to be highly individualistic: it presumes that if something is wrong, some specific individuals must be to blame. American moral imaginations rarely extend to an articulate critique of social systems, habits, or norms.

Until the broad-scope meaning of ‘racism’ becomes a matter of common usage, it would be wise, in communicating concerns about problematic racial phenomena across lines of party, ideology, and social identity, to use other terms to describe moral problems concerning race that are not also used as insults. It would also be wise to take care to precisely specify the concern at issue. Instead of calling some problematic conduct ‘racist’, one could describe how it hurts others in ways that implicate racial identities or race relations. For example, a statement might reinforce a stigmatizing racial stereotype. The use and sometimes even mention of a certain racial slur or symbol might be disrespectful to others. A policy may impose disproportionate costs on vulnerable members of the community who have been racialized as not white. Decisions might be made in ways that exclude the perspectives of those most likely to suffer from them, who have long been politically marginalized. The same racial group might be regularly asked to sacrifice for the sake of other groups, without reciprocation. It might be that no consideration has been given to how to mitigate the costs of a practice on some group or whether alternatives could secure better or more fair outcomes for everyone. The same level of virtue, merit, or desert may have been praised when exhibited by members of one group, disparaged when exhibited by members of another group. A standard of merit might put members of one group in a double bind, damned if they meet it, because it is viewed as unseemly for them to try, and damned if they don’t, because then they don’t merit whatever rewards are attached to meeting the standard. An institution may punish disproportionately the misdeeds of one group while offering lenience toward the equivalent misdeeds of another. A speech may sow distrust between racial groups, disrupting the formation of multiracial coalitions that could join forces to make the world better for all. Some technology might have been designed taking white skin as the norm, neglecting how it doesn’t work well for people with darker skin tones. This is just a sample of the myriad ways that racial injustice manifests in society.

When people articulate complaints more precisely, they are better able to make obvious what is harmful or unfair about what they are complaining about. They are also better able to tie their complaint to specific evidence. Most importantly, spelling things out may clarify that in many cases, the complaint involves negligence or the enduring unjust consequences of decisions made long ago, not malice on the part of the addressees. They involve failures of due care or consideration for others rather than deliberate attempts to harm. Much of this negligence is built into the design of institutions and public infrastructure rather than in individual minds. Pointing this out helps avoid any feelings that specific individuals are being attacked for being horrible people. It is perfectly proper to call upon our fellow citizens to remedy problems in how the institutions and infrastructure we all use are not working well or with an unfair allocation of burdens and benefits. That is what democracy is for.

More still needs to be done to avoid smugness, superiority, and contempt in communicating moral concerns to others. No one likes to be hectored, talked down to, or browbeaten into submission. We need especially to avoid two toxic inferential dynamics. One regards disagreements about morally consequential facts. It is tempting in such cases for people to make inferences of the following form: ‘You don’t believe what I do. So you must be stupid’. This inference supposes that you have the same evidence that I have, and that you evaluate it in the same way. In particular, it supposes that you trust the same sources. The second toxic pattern of inference regards disagreements about values: ‘You are behaving harmfully or in other ways that violate my values. So you must be evil’. This inference supposes that my values are obviously right, and that you are intentionally violating them.

Philosopher Robin Dembroff of Yale University offers wise advice, grounded in personal experience, against these temptations (Dembroff and Sosis 2019). Dembroff, who identifies as genderqueer and uses the pronoun ‘they’, was gender nonconforming as long as they can remember. Others are unable to consistently or confidently categorize Dembroff as male or female. They grew up in a Christian evangelical family and majored in philosophy at the Bible Institute of Los Angeles, California. Their upbringing featured intense enforcement of conventional gender norms, which Dembroff experienced as traumatic. Their community excluded any ideas by which Dembroff could attain self-understanding. Hence, they were closeted even to themself until they were twenty-one. Leaving the evangelical worldview was itself a wrenching experience.

Dembroff draws two vital lessons from their experience of epistemic ‘whiplash’ in quickly moving from a conservative evangelical to a progressive academic worldview. The first is to practice the virtue of epistemic humility. Dembroff once held evangelical beliefs with certainty, and then came to reject them. Recognizing that they could come to dramatically change their mind again, Dembroff advises caution in making sweeping claims and openness to alternative perspectives. Second, Dembroff urges ‘ideological mercy’. When confronting someone with beliefs one finds harmful or disrespectful, one should first try to understand that the other’s life experience led them to a worldview that appears obvious and right to them. From their point of view, their beliefs make sense and are what any decent person would affirm. Particularly if they were brought up in an insular community, they may never have encountered evidence or arguments that might call their views into question. Everyone they love and trust, everyone who has cared or looked out for them, may have avowed the same views. It would be wrong to leap to the conclusion that they have a bad character for holding harmful views without knowing how they came to hold those views.

But aren’t they being epistemically irresponsible in holding bigoted views that cause such harm to others? To be sure, some people have enough access to the evidence, and the ability to discern reliable from unreliable sources of information and methods of reasoning, to know that their beliefs are likely to be mistaken. But most people have little instruction in logic, statistics, detecting cognitive biases, or other critical reasoning skills, nor even in basic methods for telling the difference between serious journalism and propaganda. And even those who have had such instruction to high degrees often err. Instead of blaming people for failing to properly judge the trustworthiness of the sources of information they rely on, we need to understand why they distrust certain sources and ideas.

Ideological mercy demands that we meet people where they are. This requires that we first listen to their perspectives, to learn how they are thinking about the issue at stake. Listening carefully and respectfully to others’ stories may inform us of possible openings for constructive discussion. Such stories may reveal common ground that affords a basis for reasoning together about the issue in question. Or they may reveal questions or doubts that open avenues for exploration. Expressions of distrust based on misapprehension may offer insight into people’s deeper worries, which might be reduced by sincerely affirming certain values that they don’t realize are shared. In general, we should not attack others as stupid, evil, or irresponsible in holding harmful beliefs. And the more educated should not dismiss those who lack comparable access to knowledge and analytical methods.

Paul K. Chappell, a veteran who has dedicated his postmilitary life to promoting peace, has addressed many hostile and fearful audiences and learned from deep reflection and patient practice how to communicate concerns to those who find them threatening to their worldviews and identities. Like Dembroff, he argues that it is better to assume that someone who acts on harmful beliefs is ignorant of the harm they are doing rather than assume that they are evil or stupid. Given that the vast majority of harmful conduct is negligent rather than malicious, this is a fairly safe assumption. Acting on it opens space for communicating moral concerns in ways that minimize threats to people’s identities. Instead of accusing people of bad motives, one can calmly testify to the personal experience of being treated in harmful or disrespectful ways, indicating that one understands that they, or the people with whom they identify, may have acted unawares. We should appeal to the better angels of their natures, expressing confidence in their goodwill and willingness to change course once certain consequences of their conduct are brought to their attention (Chappell 2015: ch. 3).

In practicing a democratic ethos of communication, no one can claim omniscience about the solutions to problems. Public policies have different impacts on different members of society. Any given individual is differently affected and has limited knowledge of impacts on people who are differently situated from themselves. Indeed, given the complexities involved in tracing social causes, even the individuals most affected by a given policy often have partial and defective understandings of how they are affected and are prone to misattribute causes. Democratic discussion is a means we use to gather information about the diverse experiences of institutions, policies, and habits, so that we can explore reforms that are more responsive to the interests of all and so that their costs and benefits are more fairly distributed (Anderson 2006).

The fact that each of us speaks only from partial knowledge gives us further reason to practice epistemic humility. This entails that we cannot speak with confidence about solutions to problems without listening carefully to differently situated others. We can’t simply dictate solutions to them. Democratic communication of such concerns is a call to attention, not to specific, predefined actions. It is an invitation to dialogue about how to understand the concerns and how to address them. This is a joint activity, in which each gives reciprocal attention to the other’s concerns and values (Springer 2013: chs. 4–5). Such activity of course does not preclude, and indeed may require, preliminary caucusing with groups to which one belongs, to forge some ideas about common concerns in the smaller groups. But in joining dialogue across identity lines, openness to change, compromise, and cooperation is part of the democratic ethos.

Precisely because discussion that follows the democratic ethos involves serious listening to and active engagement with diverse points of view, from people who have different experiences, it enables the creative articulation of new possibilities that would not have been imagined from more parochial points of view. As John Dewey argued (1988), this is the promise of democracy. Yet, the realization of this promise requires that one be prepared to recognize one’s own point of view as only partial, contingent, and subject to revision. It requires that one listen to and address others with respect and concern. And it requires that we focus our attention on first-order concerns, avoiding modes of communication that either express or activate identity-based esteem competition.

Does this mean that we have to passively accept being the target of harassing discourse? No. It means that we should avoid responding in kind. This was the real message Michelle Obama was trying to convey when she said, ‘When they go low, we go high’. This wasn’t a boast or a contemptuous comparison between her side and opponents, although many thought it was. She was reporting how she instructed her daughters to respond to hate speech and how adults should model good behavior to their children. Two years later, reporter Philip Galanes (2018) asked her to explain how she would ‘go high’ in response to someone going ‘low’. He chose a relatively mild example from cable news: ‘I’m so sick of these #MeToo women playing the victim card’. Obama replied:

If you said that, I know I’m not going to change your mind in the moment. You’d just feel attacked. I’d have to understand why you feel that way. I’d have to be your friend and get into your pain and hurt, your fears. And that takes time. That’s the work that needs to happen around kitchen tables and in our communities. When I say ‘go high’, I’m not trying to win the argument. I’m trying to figure out how to understand you and how I can help you understand me.

This sounds like a lot of work. And it is. But Obama’s advice does work. She reported how things went on the campaign trail:

When you travel around this country, like Barack and I have, we’ve been fortunate to see the country in its fullness: sitting at people’s kitchen tables, going to people’s churches and veterans’ community halls. You learn two things: First, people are open to having strangers come into their homes and talk. And you know what? We would talk, and we would listen. And people would start going: ‘Oh! That’s who you are. I’ve heard all this stuff about you on Fox News, but you’re actually kind of reasonable’.

Of course, not everyone can be persuaded to engage in discussion guided by a democratic ethos. It may not be worth one’s (or anyone’s) time and trouble to persuade some people who might be persuaded. Each individual needs to decide where to invest their time. In many cases, this will involve communicating with people we already know and have some relationship with that is independent of politics—for example, as kin, neighbors, coworkers, parents of children on the same soccer team, and so forth. And saving democracy does not require that everyone be brought around. It only requires that enough are brought around to sustain a dominant climate of discussion guided by a democratic ethos.

Alas, one of the sources of affective partisan polarization is that we are more and more segregated by party identity, because most of our other identities, including those tied to where we live, align with our party identities (Klein 2020). So, many of us need to reach out beyond our regular social circles. It easier than one might think to do this, in settings designed to promote constructive discussion. I’m not talking about social media, much of which is driven by algorithms that massively amplify toxic discourse—both harassment and disinformation—at the expense of decent and honest communication (Vosoughi, Roy, and Aral 2018; Horwitz and Seetharaman 2020). Organizations such as Braver Angels (https://braverangels.org/), More in Common (https://www.moreincommon.com/), and the Local Voices Network (https://lvn.org/) bring together ordinary people from different walks of life for conversations designed to enhance mutual understanding, sometimes using technology to connect people from distant locations.

Political scientist Katherine Cramer, who works with the Local Voices Network (LVN), has spent her career listening to Americans talk about politics, and with one another. She observes that Democrats and Republicans have similar stereotypes about each other: each thinks the other side is dogmatic, hypocritical, fooled by the media they consume, driven by their identities to vote for their side no matter how unworthy their candidate is. Ordinary voters on both sides feel ignored and disrespected by the political system but see the other side as the enemy rather than focusing on the ways the system is failing them. LVN uses technology to connect diverse people from different communities for facilitated conversations to promote understanding. It strives to include people who have been left out of policy decisions and the media. Participants start by talking about the values that are important in their lives. They tell stories about their personal experiences, and share their hopes for how their local communities might improve. They listen to what distant participants say about their values, lives, and communities and reflect on their testimony. Conversations stress active listening, and personal experience over arguments. They engage issues, such as race and policing, that make people feel uncomfortable. But by focusing on personal experience rather than abstract principles or stereotypes about the other side, these conversations enable constructive dialogue about difficult issues and highlight neglected perspectives (Cramer 2019).

When diverse people from different walks of life come together to discuss politically fraught issues in contexts that facilitate discussion that discourages toxic discourse and follows a democratic ethos, ordinary people do engage in constructive discussion. James Fishkin is a political scientist who has conducted ‘deliberative polls’ following a democratic ethos for decades. In a 2019 experiment, his research team brought together a random sample of 523 US voters to discuss controversial topics such as immigration and the minimum wage, which voters indicated were high priority. Participants received policy briefings from experts from both political parties but without party labels attached. They also didn’t label themselves. Pre- and post-meeting polls showed that, after four days of deliberation, participants tended to move away from the most polarizing policies on their own side and move closer together. Perhaps more importantly, their confidence in democracy increased, and almost everyone agreed that they learned a lot about people different from themselves and about what their lives are like. Mutual understanding reduces distrust and division (Fishkin and Diamond 2019).

Similar exercises among real citizens that have real stakes have also yielded encouraging results. Ireland convened citizens’ assemblies to deliberate about constitutional reforms on abortion and marriage equality. Their proceedings, which can be viewed online, helped the broader voting public think about these issues before voting on referendums on these topics and led to voting that was better aligned with voters’ core values. It appears that the broader public trusted the thoughtful deliberations of a representative sample of ordinary citizens more than the squabbling of elite politicians and media figures (Suiter and Reidy 2020). In Miami-Dade County, Florida, politically diverse citizens participated in dozens of meetings with officials, business people, and civic organizations to discuss measures to address climate change. Despite the fact that ‘human-caused climate change’ is heard by many Republicans as akin to fighting words, moderated discussions that screened out toxic discourse enabled the county to come to consensus on 110 climate-change adaptation and mitigation actions. Importantly, leaders at these meetings rejected discourse disparaging Republicans as science deniers and pointed to the active participation of Republicans in promoting strategies to address the problems, such as sea-level rise, that everyone agreed were serious (Kahan 2015).

On January 20, 2021, president Joseph Biden delivered his inaugural address (Biden 2021). Exactly two weeks after a mob attacked the Capitol in an attempt to overturn the results of the presidential election, Biden spoke to the disinformation, fears, and divisions that threaten democracy in the United States. It is worth pondering his words:

I understand that many Americans view the future with some fear and trepidation. . . But the answer is not to turn inward, to retreat into competing factions, distrusting those who don’t look like you do, or worship the way you do, or don’t get their news from the same sources you do. We must end this uncivil war that pits red against blue, rural versus urban, conservative versus liberal. We can do this if we open our souls instead of hardening our hearts. If we show a little tolerance and humility. If we’re willing to stand in the other person’s shoes just for a moment.

Biden could have been channeling John Dewey, America’s greatest democratic theorist. In 1939, Dewey wrote an op-ed that addressed the crisis of democracy in the face of rising fascist and communist movements:

Intolerance, abuse, calling of names because of differences of opinion about religion or politics or business, as well as because of differences of race, color, wealth or degree of culture, are treason to the democratic way of life. For everything which bars freedom and fullness of communication sets up barriers that divide human beings into sets and cliques, into antagonistic sects and factions, and thereby undermines the democratic way of life. Merely legal guarantees of the civil liberties of free belief, free expression, free assembly are of little avail if in daily life freedom of communication, the give and take of ideas, facts, experiences, is choked by mutual suspicion, by abuse, by fear and hatred. These things destroy the essential condition of the democratic way of living even more effectually than open coercion which—as the example of totalitarian states proves—is effective only when it succeeds in breeding hate, suspicion, intolerance in the minds of individual human beings. (Dewey 1988: 228–29)

Citizens across many democracies today face a crisis of democratic backsliding due to politicians and media sources that promote distrust, division, resentment, and fear among the people. To respond to this toxic discourse with contempt and disdain only fans the flames of division by turning politics into a war over who is superior to whom. Toxic discourse suppresses free communication of first-order concerns and thereby derails democracy. We can all do our part to overcome this problem by practicing empathy, tolerance, humility, and mercy when we discuss politics. Assiduously practicing these virtues can defuse fear, resentment, and distrust and inspire reciprocal responses in others. Only then can we strengthen the foundations of a democratic way of life.