Wednesday, February 17, 2021

Diversity Training has a Rationality Problem

I have a theory that for a certain type of person, of which I include myself, resistance to things like racial quotas and mandatory diversity training has nothing to do with wanting to uphold my privilege and more to do with what I reluctantly call "coercion aversion."

To give an analogy, in writing about authoritarian Trump supporters, Jonathan Haidt used the metaphor of a forehead button.

"In times of low moral threat, when [authortarians] perceive that the country is relatively unified and the moral order is not being subverted, they are not particularly intolerant (Stenner finds). But, when they perceive that the moral order is falling apart ... its as though a button is pushed on their forehead that says “in case of moral threat, lock down the borders, kick out those who are different, and punish those who are morally deviant.” 

David Brooks made an interesting point about the failure of diversity trainings, writing:

"Our training model of “teaching people to be good” is based on the illusion that you can change people’s minds and behaviors by presenting them with new information and new thoughts."

The diversity training says "you are biased at the subconscious level. Now you know that and can stop being biased."

I think two things are going on. One is that coercive tactics push the coercion button on my forehead and make me resistant to policies that tell businesses how to do their hiring. Two is that diversity trainings have a rationality problem in that they think they can solve an intuitive problem at the rational level.

Fortunately, I think the work of Richard Thaler and Cass Susstein can provide solutions to both problems.

The Emotional Power of Habit

Has anyone ever successfully lost weight by reading about how it's bad to eat processed foods and enduringly committed themselves to a lifetime of salads? I've gained and lost over 30 pounds countless times and have found two things that have helped me lose weight: diet and exercise. More specifically, making exercise a habit and following Weight Watchers. 

The reason Weight Watchers works is mental. The process of tracking the points value of each meal reinforces the idea that everything you consume has consequences. It makes you accountable for everything that you eat. It also helps with exercise. Thirty minutes on the treadmill means I can have a beer tonight.

In other words, it inculcates a habit. Once you form that habit, the program becomes easier. You don't have to think about it as much. 

It's the same way with exercise. Once working out becomes a habit it no longer feels like a chore but a necessity, like eating or sleeping. 

At the rational level I'm punching calories into an app. At the subconscious level I'm training my mind to cultivate good eating habits. Understanding the logical benefits of weight loss doesn't cause you to lose weight. You have to train your subconscious mind through habituation.

Riding the Elephant

If you want to change someone's mind, Jonathan Haidt said you need to "talk to the elephant, not the rider." Telling someone that their support for Blue Lives Matter is not rational but a reflection of being high on the authority/subversion principle is not going to change their attitudes about ending qualified immunity. That argument is too rational, too meta.

A better method would be to frame your argument so it appeals to a person's intuitive respect for authority. Maybe focus on how police unions incentivize bad behavior by protecting cops who violate the civil liberties and personal freedoms of law-abiding Americans. Worse, it doesn't do enough to protect the good cops who care about justice.

In everyday rhetoric, this is just an appeal to emotion. But on a psychological level, Haidt understood that you don't change behavior by appealing to the rational mind. 

This is what the diversity training referenced in Brooks' article does not get. In fact, most people don't get it. At the structural level, you need policies that affect the subconscious, emotional, intuitive level of the mind if you want to see change.

The best examples of this are Richard Thaler and Cass Susstein's work in Nudge. Sticking with the weight loss example, they found the most effective strategy was to make a bet. Two people who want to lose weight put $100 into a pot. Whoever loses the most weight after six months wins the money. This works  not because people understand that obesity carries certain health risks. It works because of loss aversion; people don't like losing money. They took a universal emotional instinct and showed us how to manipulate it for our own benefit.

Avoiding the Coercion Button

I also think the work of Nudge and libertarian paternalism can help with the coercion aversion resistance to racial hiring quotas. 

An example from Nudge showed that you could increase the chances that a product in the school lunch line will be purchased depending on where you stock it (eye level, right by the register, etc.) They called this choice architecture. What if researchers found that search committees tend to favor the first resume they look at when given a stack of potential candidates?  Using this information, you could make sure that any black candidates appeared first. This would create a positive subconscious bias to combat any implicit bias. 

I do not have a racial essentialist button on my forehead that causes me to oppose any antiracist idea involving racial quotas. But I do have a coercion button on my forehead, which is why my idea of placing black resumes in the most favorable order doesn't bother me. If it works, I actually think it's a great idea.

Libertarian Antiracism

Whatever biases diversity training hopes to correct, it should look for solutions that instill healthy habits of the unconscious mind. In other words, instead of combating negative subconscious biases that are harmful to minorities with rational thinking, they should combat negative subconscious biases with positive subconscious biases that are helpful to minorities. 

This can be achieved by good choice architecture and making tweaks to the environment but we cannot figure out these tweaks without investing in the research (meaning actual science and not critical theory) and holding diversity trainings accountable by measuring whether they do what they say they're going to do.