Monday, August 12, 2019

shaming racists: behavioral speculations

A few weeks back, I whiled away an hour or two of an afternoon with my friend Oli, keeping her company in the parking garage office where she sometimes works. I forget what we were discussing that prompted it, but Oli told me she's a strong believer in the virtues of shaming as a means toward changing peoples' habits for the better. She wasn't talking about dogpiling, doxxing, and death threats, but rather the basic practice of establishing a taboo and relying on members of a group to socially punish those who violate it.

I concurred with her to the extent that it can work: the example that came to my mind had to do with smoking. Nobody automatically assumes they have your permission to light a cigarette inside your car or house these days, and part of the reason for this must have to do with enough hosts sternly asking enough guests to kindly take it outside. The social group erected a boundary and enforced it by calling out infractors, eventually establishing a new standard of behavior which we observe with little need for continuous policing or squabbling.

Our conversation flindered on, with neither of us dwelling on the point. We were already in the middle of disputing at least two other topics (I enjoy talking to Oli because she's intense, and because we can argue ideas in good faith without the mood growing acrimonious) and I didn't want to swerve too far away from the matter at hand. But I did continue thinking about it afterwards.

Shaming is more often than not a palliative: it targets the symptoms of undesirable habits instead of their causes. Any educator worth his or her pension will tell you that the stick works best when paired with the carrot. By itself, punishment is a short-term solution that must be applied and reapplied and reapplied for as long as the factors responsible for the problem continue to operate—and punishment usually leaves those factors untouched.

How, then, to account for the usefulness of shaming in stopping people from stinking up your car without your permission? It works because the smoker ultimately receives a reward—a cigarette—for deferring from lighting up in a closed, shared space. In this case, the host's ire functions as a guard rail. The carrot and the stick are wielded in tandem to achieve a satisfactory result.

To see why shaming, in and of itself, is an ineffective means of social control, we need only look to current events. The ghastly spectacle of a crowd chanting "send them back" at a Trump rally in North Carolina last month (to say nothing of last week's massacre in El Paso) was all the reminder anyone should need that the United States is still sick with racism.

What has been our strategy for combating this problem? Just what Superman recommended in the late 1940s: calling out people who express bigoted sentiments.

While there have been institutional efforts at combating racist attitudes through education, diversity training, etc., the most frequently used tool in the reformist's arsenal has been shaming. Linguist John McWhorter has often said that "racist" is only slightly less toxic a label as "pedophile." Even racists, he says, are embarrassed to be called racists. And yet, white racism continues to stain the United States' social fabric. If such a widely acknowledged social ailment has not been cured, part of the reason must be an ineffective mode of treatment.

In Science and Human Behavior (1951), behaviorist par excellence BF Skinner dedicates an entire chapter to punishment and its inefficacy. Punishment, by his (admittedly circular) definition, entails the introduction of a negative stimulus or the withdrawing of a positive stimulus during the performance of a given action. It needn't occur in a social context: the behavior of touching the glass of a kerosene lantern is "punished" through blistered fingertips. But no matter what agency delivers it, and whatever the particulars, the effects upon the punished organism are generally the same. Skinner enumerates three of these.

The first effect of punishment is to interrupt whatever activity the organism was engaged in. If you scream at a child loud enough for jumping on the bed, he or she will probably stop jumping. If you spray a cat with water for scratching the upholstery, the cat will recoil, and in doing so will stop scratching the sofa, at least for that moment. When I put my fingers on that damned kerosene lantern, I didn't keep them there for long.

Secondly, punishment has an abiding effect, which tends to emerge as emotional responses to occasions which resemble those at which punishment has previously occurred. I was going to link to a clip from A Clockwork Orange to illustrate this, but couldn't find what I had in mind. These scenes from a Simpsons parody will do, I think. It's an extreme (and fictional, and ridiculous) case, but in principle, it's pretty accurate. If you spot your ex on the street after an acrimonious breakup, your stomach goes tense. If your classmates make fun of you and your Monkees lunchbox, you might feel a sense of dread the next time you carry it toward the bus stop. I obviously still have misgivings about kerosene lanterns twenty-five years later, or I wouldn't so readily and vividly recall burning myself whenever I see one.

It should also be noted that someone who really likes cupcakes in the above scenario will also be affected by the thwarting of the previously reinforced behavior of reaching for them. The physiological events pursuant to the suppression of strong behavior after it has been punished are usually experienced as an aversive internal stimulus. Any neurotic can tell you that, as can any smoker who's tried to quit after a bout of nicotine sickness.

Skinner describes the third effect of punishment:
If a given response is followed by an aversive stimulus, any stimulation which accompanies the response, whether it arises from the behavior itself or from concurrent circumstances, will be conditioned. We have just appealed to this formula in accounting for conditioned emotional reflexes and predispositions, but the same process also leads to the conditioning of aversive stimuli which serve as negative reinforcers. Any behavior which reduces this conditioned aversive stimulation will be reinforced.
To recur to the cat: if you spray a hydrophobic feline enough times for raking its claws across the sofa, it will stop whatever it's doing, or disappear, at the sight of the bottle and nozzle. These avoidant behaviors have been negatively reinforced. If it is consistently sprayed every time it gets near the couch, it's less likely to wander toward the couch. "Not approaching the couch" carries no meaning in a behavioral analysis; "approaching x, y, or z while in the couch's vicinity" does. There are no such things as negatives in behavior.
The most important effect of punishment, then, is to establish aversive conditions which are avoided by the behavior of "doing something else." It is important——for both practical and theoretical reasons——to specify this behavior. Sometimes it is merely "doing nothing" in the form of standing still. Sometimes it is behavior appropriate to other current variables which are not, however, sufficient to explain the level of probability of the behavior without supposing that the individual is also acting "for the sake of keeping out of trouble."
The effect of punishment in setting up behavior which competes with, and may displace, the punished response is most commonly described by saying that the individual represses the behavior ... If there is any repressing force or agent, it is simply the incompatible response. The individual contributes to the process by executing the response ... No change in the strength of the punished response is implied.
Bolded for emphasis. By "incompatible response," Skinner refers to the "something else" that is reinforced as outlined above.
If punishment is repeatedly avoided, the conditioned negative reinforcer undergoes extinction. Incompatible behavior is then less and less strongly reinforced, and the punished behavior eventually emerges ... If punishment is discontinued, the behavior may emerge in full strength.
By itself, punishment only changes behavior as long as it is consistently applied, or continues to hang over the individual as a credible threat. If the spray bottle goes missing, our cat will likely start tearing up the sofa again after a while.

Let's try to imagine, from a behavioristic perspective, what happens when and after we shame a racist. This probably won't be easy, and not just because we're (understandably) reluctant to empathize with somebody who holds odious beliefs. Before we can start, we'll have to work out what we're talking about when we talk about racism, and without specifying terms like "hate," "bigotry," etc. as causes. We can use them as non-technical descriptors for the sake of vernacular clarity, but it's against the rules to treat ideas as motivators, or feelings as anything more than the epiphenomena of bodily states.

For the record, I acknowledge that this is a thought exercise, and as such, it is wholly speculative and prone to error. I can only hope that it contains at least a kernel of veracity.

Paul Klee, Struck aus der Liste (1933)

A behavioral analysis can't begin with a nebulous concept, even one so (apparently) well-defined as "racism." Let's start with the racist himself instead.

Like any human being, his locus of behavior is derived from his total history. A white American isn't congenitally wired to dislike or mistrust brown people: his environment teaches him to. (Going forward, let's assume that we're only talking about white racism in the United States.) We can outline a few probable characteristics of the individual who has been conditioned, one way or another, to behave in ways we would typically describe as being racist.

• At its most benign (to the extent that it can be), racism implies that the perceptible characteristics of a certain group or groups, or some indefinite set of stimuli associated with those groups, produce responses in the individual which consistently differ from responses elicited by stimuli associated with fair-skinned gentiles. These variables (skin color, speech patterns, verbal referents, etc.) take on a relevance insofar as their presence is apt to alter the probabilities of his responses to a given occasion. In the individual who is subtly "prejudiced," the presence of these relevant variables might be observed as a reluctance to engage with or trust people in whom these variables occur—and, conversely, as an entrenched preference for people who display a different set of variables (i.e. fair skin, typically "white" names, etc.).

The above characteristics are descriptive of the most widespread manifestation of racism. The individual described may deny being a racist, and believe himself when he says so. He may not be aware of his biases, or he may believe that his prejudices are matters of common sense: "why shouldn't I feel a little nervous when someone wearing a turban boards the plane I'm on?"

Note: events share aspects, and these areas of overlap can and do elicit similar responses, however the total occasion may differ. Skinnerian behaviorism contends that these "associations" aren't essentially, ontologically conceptual, but the products of innumerable conditioned reinforcers and the myriad simultaneous responses they may elicit. (Remember, no organism is ever doing only one thing, or encountering a single stimulus, at any given instant.) It's much easier to fudge a descriptive account of bias with recourse to mental activity—which we here treat as an epiphenomenon—though a thoroughgoing cognitive explication would be no less onerous.

• In more acute cases, the individual shows strong aversive responses to the relevant variables mentioned earlier. We could perhaps grade racism in terms of the severity of these responses, or by the range of stimuli which evoke them—such as in cases where an anti-Semite is virtually incapable of reading a news report without being reminded of something for which the Jews must be at fault.

• We can expect the "strong" racist to be reinforced by aggressive displays (even if these are subtle, such as a dirty look or a sotto voce slur) when confronted by the relevant variables and the averse stimulation they present.

Skinner: "Behavior which inflicts damage is reinforced in anger and is subsequently controlled by the conditions which control anger." This is probably a consequence of our species' evolution: among primates, a readiness to retaliate when attacked or challenged probably helps to ensure that individual's survival. We needn't assert that the individual must be enraged for Skinner's proposition to apply. We typically categorize emotions when the bodily states underlying their topography would probably be better measured through sets of quantitative data (which is obviously impractical in everyday situations). Whatever averse emotional state (indicative of a bodily state) is produced by the relevant stimuli in the racist's case provides the context in which these operations emerge and acquire the dimensions of a hidebound habit.

• The individual who is "merely" racially biased may be reinforced by avoidant responses to the relevant variables rather than by verbally or passively aggressive behavior. For example, when boarding a bus, the probability of his taking a given seat will be decreased if his sitting there would place him beside a dark-skinned person.

• People with racial prejudices—whether we're talking about radical white nationalists, fathers who don't like the idea of their daughters dating black men, or people who feel subtly unnerved when they're the only white person in the hospital waiting room—reinforce each other. A person having a few drinks with his friends says: "I'm no racist, but [insert statement here]."  Unless he is censured, or the remark is pointedly ignored, he is likely to receive the conditioned social reinforcers of approval and agreement. The diffuse array of behaviors, overt and cognitive, associated with whatever he opines, may also be reinforced. If his peers validate the wisdom of his supporting the police in a publicized case of alleged racial bias, voicing such opinions and allowing them to inform his actions will be reinforced. If hearing himself opine about the dangers of the "Hispanic invasion" is self-reinforcing to the individual, hearing a talk radio host vociferate about the same will also be reinforcing (and vice versa).

Arnold Schönberg, Der Rote Blick (1910)

To what extent does shaming a racist alter these tendencies?

Let's consider the occasion of punishment and its probable effects. We'll have to settle for imagining events in the abstract, independent of the particulars necessary for a truly rigorous examination. Let's say the individual tells a racial/racist joke to the wrong audience; or, during a discussion about current events and politics, he admits more about his prejudices than perhaps he intends to, and finds the rest of the group bearing down on him; or he makes an ill-advised Facebook post, or tweets something he really shouldn't, and gets more attention than he ever wanted. What happens then?

More bullets.

• Unless the punishment ultimately results in a the formulation of new habits which displace the above behaviors and recondition the individual to the relevant variables, we can fairly say that any cessation of racist expressions represents only a surficial change. Achieving such a drastic and deep-reaching result through punishment alone will likely carry ancillary consequences of the sort that a therapist is usually called upon to mitigate.

• In the long term, punishment may only influence the probability of outward behavior. While his speech and social media activity may be punished, it will be apparent to him that he remains safe if he keeps his (unchanged) opinions to himself. If he is habitually outspoken or loquacious in group settings, the incompatible behavior (keeping quiet) which obstructs the previously punished response (speaking his mind) may supply conditioned internal stimuli that produce an averse emotional state: he may feel frustration, resentment, etc. Under these circumstances, it is possible that occasions where he is able to speak freely may be more reinforcing than they might have been otherwise. This also implies that the cessation of objectionable behavior may be confined to the milieus in which punishment has occurred.

• The occasion of punishment may condition the individual with regard to any number of his responses to the stimuli coinciding with the punishing event. If the shaming is effective, any of these stimuli may arouse aversive emotional responses on future occasions. Say the punished behavior's context is a workplace conversation about immigration wherein the individual says: "I think we need strong border control because I don't want illegal immigrants breaking into and squatting in my summer home." (Yes, somebody actually said this.) If he is pilloried for saying so (not unjustifiably), future occasions involving referents to, say, Latin American immigrants, may evoke some measure of anxiety or stress, amplifying whatever subtle negative internal stimuli that already tended to accompany such occasions. In other words, punishment may result in the increased severity of his biases against the relevant group.

• Behavior that is reinforced by one peer group may be punished by the other, and vice versa. The individual may work in an office with staunch social progressives and share their convictions (to some extent) when he is in their presence, while his old drinking buddies may be "anti-PC." If, during a night out, one of his friends says something that would appall the members of the individual's workplace milieu (and maybe the individual too, if he were in that environment), his response may be more accommodating, and he may find himself sympathizing. He may admit his friend "has a point." If he engages with these friends and receives conditioned social reinforcers for expressing agreement, the private dimension of agreeing with what is being said may increase in strength. The individual's "true beliefs"—how he tends to respond to the relevant variables when he is at home by himself, or in the anonymity of a public street—may oscillate. Whether he more often behaves more in line with one milieu or the other when he is removed from both will be influenced by factors such as the strength of the reinforcement he receives from each, and on the generality of the conditioned stimuli (external and internal) associated with those reinforcements.

• Punishment invariably conditions the stimuli proximate to the agency which administers it. The individual's responses to this person or persons might include the arousal of an averse emotional state (again: anxiety, resentment, etc.). If the punishing agency are members of his peer group, and if his further exchanges with them are for the most part amicable, and if the adverse responses are superseded by positive ones, the negative effects of the conditioning may abate. He might even come around to their way of thinking by observing and correcting his behavior with regard to the relevant variables, particularly if his peer group rewards him for doing so in addition to withholding punishment. If no forthcoming positive reinforcement from the punishing agency attenuates the lasting impacts of the punishment (typical where interactions through the internet are concerned), it may transpire that stimuli associated with the punishing agency are conditioned such that they elicit an adverse emotional response.

This last one is of particular importance. A recent piece in The Atlantic quoted Trump supporters at his last rally in Cincinnati objecting to being called racists:
And that’s what the rally-goers I spoke to last night seemed most nonplussed by—not so much that Trump had been roundly condemned in recent days as a racist, or a bigot, but that they, by virtue of association, had been as well. But rather than distancing them from Trump, the accusations have only seemed to strengthen their support of this president. To back down, they suggested, would be to bow down to the scourge of political correctness. 
“We’re all tired of being called racists,” a 74-year-old bespectacled white man named Richard Haines told me. “You open your mouth, you’re a racist. My daughter is a liberal, and she’s [using the word] all the time. We don’t talk politics; we can’t—all the time she always accuses me of hate.” ...
Roseanna, who wore a red T-shirt, white shorts, and a MAGA hat ... defended Trump’s statements about Baltimore. “He didn’t say nothing about the color of somebody’s skin,” Roseanna said, yet it seemed like everyone was “wishing him toward ‘He’s a bigot.’ “I’m sick to death of it. I have 13 grandchildren—13,” she continued. “Four of them are biracial, black and white; another two of them are black and white; and another two of them are Singapore and white. You think I’m a racist? I go and I give them kids kisses like nobody’s business.” ... 
“...They got a little carried away there,” [another rallygoer] said of the Greenville crowd. Morris, too, said he was tired of being called a racist. Just yesterday, he said, he’d given a stranger $20 to help his foster child, who was black. And he sends money as often as he can to a school charity in the Dominican Republic. So if anybody started a chant like that, Morris said, “I’ll tell them, ‘Shut it down. You’re acting like them. We’re not them.’ The Democrats—they call names, they accuse, they’re always slandering, they always have a negative.”
Not only does shaming usually fail to change attitudes, it disposes its objects against the people who are actively committed to solving the problem of white racism. The rancor from the political center to the right directed against so-called social justice warriors is less a reaction against the SJWs' convictions per se than against the liberality with which they dispense punishment in the service of their aims.

The typical social-media savvy progressive, in this regard, resembles the "tough on crime" politicians of the 1980s and 1990s whom, ironically, he or she typically despises. The principle is the same: the social environment has generated an epidemic of deleterious behavior, and the preferred solution is to come down hard on violators instead of intervening such that the social environment produces fewer of them. A progressive who believes that the way to deal with a racist is to divest him of his income and essentially send him into exile with a brand on his face differs little in his or her operational acumen from a Reagan-era Republican who believes that the way to deal with a drug dealer from an impoverished urban neighborhood is to lock him up for forty years. Progressives have the acuity to recognize that the scourge of violent crime is best solved by actions and policy that emphasize positive reinforcement in the areas affected: youth programs, employment opportunities and training, higher wages, improved public schools, etc. Apparently their insights toward breaking the cycle of poverty and violence don't carry over to the matter of widespread racial prejudice.

We can appreciate this insofar as we recognize the satisfaction one derives from striking out against an irritant: directly asserting control over our environment is almost always reinforcing. And the palliative of punishment may be the only practicable (and certainly the most expedient) means of dealing with the problem of racism, given the difficulty of altering something so inaccessible as several million individual attitudes.

If we want to purge a cultural illness like racism (to whatever extent it can be purged), some strategy of encouragement and positive reinforcement must be employed in addition to shaming. Punishment, by itself, commonly suppresses the signals of the problem, and sends those who feel attacked by such methods of control into behavior-amplifying, sub rosa niches of likeminded peers—which can have deadly results, as we've recently (and repeatedly) witnessed. Moreover, when the national group has a leader such as the one we've got now, the invariability of receiving punishment for racist behavior is divested of its potency as a controlling variable—as we've also been seeing.

How do we dispose a white person to respond well to people of different ethnic backgrounds and skin tones? Place him in a setting where he can (or rather, must) form meaningful, positively-reinforcing relationships with them on a level socioeconomic plane. (The latter specification is important: whites must frequently encounter other-than-whites as neighbors and social equals, and vice versa, for the variables relating to race/ethnicity to lose their relevance with regard to the behavior of bias.) Absent sweeping strategic population- and economic redistribution programs, I don't see this happening anytime soon. But some sort of carrot must supplement the employment of the stick if the stick is to be used with any lasting efficacy. A called-out racist too often remains a racist.


  1. I've been struggling with this a lot lately, especially with my older white relatives, who so often use the Biden racist bone response - joining the Tea Party would say otherwise, as would the response to Mike Brown, as would having voted for Trump. So.... no desire to improve. And shaming and cutting out of lives only validates their feelings - it's not their beliefs that got them punished, but (to them) their beliefs being so true that the other people in their lives can't accept it.

    But if someone's willing to admit they've fucked up and want to be better, how does that begin? I'm willing to forgive but do we put the onus on non-white people in our lives to do that? That doesn't check out either.

    1. I just typed up a long response and then my browser ate it. Sigh.

      Summarizing: I don't think your relatives are going to change in this regard unless they pack up and move somewhere more integrated (or integration comes to them). Forming relationships with members of the "other" is the only way they cease to be "others." So, really, they're probably set in their ways for life at this point. Where you go from there is up to you, but there'll be even less possibility that they'll even for a moment consider a different viewpoint if their only friendly relations are the people who see things the same way they do.

  2. Lots of stuff to chew over here. One could argue you've outlined why social media echo-chambers have been so effective at dividing people, especially those with "subtle" racism: any resentment at being shamed gets reinforced by someone else, and if your daughter doesn't agree with you, well, Bill Davies, of Lake Charles, MI and your favourite Facebook group, does. There's a whole dimension to consider with a shameless US president to rally around. No wonder facts matter less than ever (though I doubt facts have mattered much to a lot of people to begin with).

  3. Trump supporters are not "racists", they worship black people and would gladly die for Israel. Your whole premise is wrong.