Jordan's Blog - The problem of political polarization, Jonathan Haidt, and why I'm so goddamn righteous all the time (2)

Home | Archive | About | Subscribe

The problem of political polarization, Jonathan Haidt, and why I'm so goddamn righteous all the time (2)

Jordan 2018-02-04

In part 1 we learned that

Reasoning is terrible and morality is just emotions.

Ifyou understand that morality is about emotions we’ve evolved for protecting society, then you understand the world through Haidt’s eyes. Morality is less about preventing individual harm and more about preventing colony collapse. Societies are collections of groups bound together by moralistic feelings resting lightly atop rituals which have been honed and optimized by evolution to facilitate cooperation: tattoos, karaoke, the Haka.

First, if you pay attention, you start to notice the benefits of things like social justice — activists are bound together by shared feelings of liberation and oppression, care and harm — and religion. Then, if you think hard enough, you start to question their costs.

Haidt compares religion to college football. The naive individualistic perspective is to focus only on the game being played on the field, the cost, waste, loss of an ability to think rationally, and a long trail of victims. His elevated sociological view is that these games serve the function of creating a community, augmenting school spirit, attracting better students, and alumni donations. But is all of that cost, waste, loss of rational thought, and the long trail of victims necessary? Which toxic practices can you remove without destroying its functionality? Can you add more positive practices like blood drive competitions? How do you decrease the spite felt towards competitors Only top sports schools actually make money from their sports program — most schools lose money in the infinite arms race of funding — and even the money that is collected by the top schools is not newly generated capital, it’s alumni donations. Can you make sports rallies actually generate economic capital for a school, the way a programming hackathon or a research lab does?

If you don’t seethe Haidtian sociological view, you might start removing important components without realizing it (for example, it might be important to the ritual that players have a risk of physical injury), but if you don’t even see the naive individualistic view, you’ll never ask these questions about how we can do better.

And once you’re asking these questions you start to wonder why the people you disagree with aren’t doing the same from their side. Haidt repeatedly states that conservatives understand that tradition is good because it leads to increased moral capital. But do they truly “understand” this? How many would explain their beliefs this way? How many are following any intuition besides “it’s how I was brought up?” (Liberals do this too, let’s not act all innocent.) The actual thought process matters, even if the end judgement is the same. Haidt’s goal is to understand the other side, so if we’re conceding that authority over autonomy can promote social stability and moral behavior, and the conservatives are actually just thinking, “you can come up with whatever reason you like for yourself, but I know I’m right because I feel it in my gut,” then that might explain why so many people miss these questions.

I therefore present my alternative viewpoint: moral reasoning isn’t doomed because morality is emotional, it’s not actually that hard to reason well, and liberals aren’t just unable to connect with conservatives because they’re ignoring important moral emotions — the actual root cause of everything is:

People are uninterested and busy with other things.

With your new Haidtian sociological lens, ask yourself how liberals might come up with an argument for taxing the rich that appeals to everyone. If Haidt is right that liberals could convince conservatives if they knew how to tap into some of the other moral foundations, then we should be surprised when we hear the modern day narrative surrounding the tax rates of the 1%: that these people are born with trust funds, and their huge leg up means they hardly have to work at all and they end up way richer. This is a strong appeal to the foundation of fairness/cheating, one which Haidt says is associated with the conservative take on equality, yet when we look at this stance and it’s attached solution of taxing the rich, we can clearly see that real life does not reflect what our Haidtian theory predicts. Conservatives are not in favor of raising the tax rate on the rich.

This is because Haidt’s view is incomplete. Yes, some of our social disagreements come from an over-reliance on bad reasoning, an inability to reason well, and a failure to craft emotionally compelling narratives that help us relate to others who disagree. But most disagreements happen because people tend to be epistemically lazy. When you tell them that they can reason better, they’re not interested. They do not like to consider other options. They think that being proven wrong means losing. They want to keep doing the easy thing because anything else takes effort. Not inhuman effort, just effort. Don’t get me wrong, people are also amazing. Every innovative invention in all of history was invented by a human. Some people actually do put in effort. Heck, most people put effort into things they care about. It’s just the specific intersection of people who care about understanding opposing viewpoints, who want to overcome their own bias, and hope to find truth through reason and experimentation, and also have enough time and energy to put in effort, is not very big.

I think of The Righteous Mind as a manual explaining why this intersection is small. We start the book in this world where people who disagree with our political viewpoints are impossible to understand, and our own beliefs make perfect coherent logical sense. And Haidt slowly cracks this world open assumption by assumption until we are forced to admit that human nature is very different from what we think. We come out the other side wondering what the heck we should do if we can’t trust our own brains.

Well, what the heck should we do?

Haidt’s recommendations tend to be about ways we can take advantage of our nature rather than fighting it. We can build systems that make very little depend on truth-seeking reasoning, and very much depend on a systematic balance of poor reasoning, ideally between disagreeing parties with a desire to get along. As far as policy goes I think this is clever and creative, and we would probably do well as a society by listening to Haidt here. But consider the problem of addiction. Like bias in reasoning, as Haidt puts it, addiction is a built in feature of our brains, and not a bug that can be removed. That means that we could similarly get societal gains by creating systematic efforts to make addiction less likely and less dangerous. And we do. But does that mean that individual addicts shouldn’t also try to do something about it? I imagine Haidt critiquing our natural willpower, saying that the average human has very little, and addiction evolved in our brains because it served some evolutionary purpose (think sugar addiction) and we’re actually terrible as a species at using willpower to overcome addiction. And I want him to tell recovering addicts that we need to be wary of any individual’s ability to overcome addiction through sheer power of will. And I have to wonder if it some point he will stop and say “but you should definitely still try even though it’s hard.”

Correcting the balance of beliefs — an aside

“Millennials need to learn to be more assertive. Or maybe they need to be more polite. Actually, maybe they just need to be more in the middle, like my generation was. What are we talking about again? Oh yeah, can I get the pastrami?”

There’s a term for the special class of argument wherein two sides have opposite assumptions about the majority. There are people who are too aggressive and could be more polite, and there are also people who are too passive and could be more assertive, so the entire argument becomes about which side is being braver by opposing that majority. These are called bravery debates. If you wrote a think piece about how Millennials need to learn to be assertive, you’re entering the bravery debate with the assumption that the majority of Millennials are too polite, and you’re bravely trying to correct the balance.

Pictured above: millennial or something i dont care.

Similarly, maybe the point Haidt is making is that our society is over-reliant on individual reasoning, and he is trying to correct the balance by pointing out the flaws in this view. And if the majority started to understand that individual reasoning actually sucks and totally stops trying, he would switch to writing books about why we can’t rely only on sociological improvements.

But I don’t think Haidt sees this as a bravery debate.

If he did, he would want to cite studies showing that we overestimate our reasoning ability and underestimate the power of policy. Instead he spends a lot of time talking about how we are often convinced that we are being unbiased even when we’re not. Haidt thinks we can’t trust someone who claims to be unbiased because the evidence says that statistically they’re probably wrong. This would be like assuming that someone who tells you they quit drinking without AA is lying, and it sort of makes sense, but only if you think it takes superhuman willpower to overcome addiction by yourself. That’s one reason why we might see statistical trends of people being bad at overcoming bias, maybe it’s truly inhumanly hard to do. But following my alternative theory, maybe people just aren’t generally interested in it? When incentives align with interest we actually do okay at overcoming bias. For example, engineers routinely encounter the Planning Fallacy, in which people underestimate the amount of time it will take to do something. In this case it’s profitable to overcome this bias, so engineers keep track of their historical estimation error and use it to calibrate their future estimations. And when we need to fairly assess evidence, we can enforce Tetlock’s conditions and even create a community based around aligning reputation with honest truth seeking.

Chimp communities and bee hives

Individual improvement is not a priority for Haidt because he generally trusts our irrational human nature. We are 90% chimp and 10% bee, goes the motto of the third book: 90% self-interested and obsessed with protecting our reputations, 10% selfless and altruistic within our ingroups. Robert Putnam showed through a national survey that high levels of immigration and ethnic diversity are correlated with a reduction in social capital both within groups and between groups. When we emphasize our individuality, our differences, that self-interest really starts to show, but when we emphasize our groupishness, our similarities, only then does our society truly thrive.

Haidt writes that “religion helps people achieve together what they cannot achieve on their own.” He then adds “… but so does the Mafia. Is religion something that preys on, or at least turns its back on everyone else?” Religion brings out our groupishness, so does it bring out our selflessness also? Recall that religious people are more charitable than secular people on average. Many of these donations go to churches, but to Haidt’s point, this effect is still seen even when discounting those and looking only at donations to organizations that benefit all Americans, like the American Cancer Society.

So what’s the deal here? Well first of all charity is great, and I think we should all strive to join communities that make us more charitable even if they’re not religious. But I also think that Haidt slightly misses the point, or to put it another way, he’s looking at the wrong outgroup. The outgroup of religious Americans is not people with cancer, so it shouldn’t surprise us that they are happy to donate to the ACS. To use a somewhat circular definition:

Our true outgroup is the group we are least happy to donate to.

And not just because they are faraway strangers, but more often because they are nearby and we think we know them. Religious people donate more to charity, but don’t expect most of them to donate to Planned Parenthood. Haidt says religion is a prosocial construct because even though it doesn’t make you altruistic towards strangers, it makes you a better member of your community. It makes you kinder to your ingroup. Well I say, in most of the U.S, it’s easy to be kind to your ingroup. What’s hard is being kind to the people who don’t share your morals, the people that Haidt spent the first two thirds of the book trying to help us understand.

Haidt says that increasing love of the ingroup usually doesn’t lead to hatred of outgroups “under normal circumstances.” He cites two studies: Brewer and Campbell 1976, and de Dreu et al. 2011. The latter refers to a study on oxytocin as a drug that enhances ingroup cooperation and outgroup non-cooperation, but also points to de Dreu et al. 2010, which shows that oxytocin drives defensive, not offensive, aggression toward competing outgroups. Even if we say defensive aggression is not a sign of hatred, and I’m not sure we should, this study is focused on the effects of oxytocin, and there are plenty of other tribal mechanisms and brain chemicals which could lead to hatred of outgroups.

Also consider Molloy 1998, which cites Brewer and Campbell 1976 and concludes that “only when two groups perceive equal resource availability per effort spent do they naturally respect and tolerate each other… otherwise there will be mutual dislike, low trust, negative stereotypes, and groups will tend to take action against each other.” (Luckily this is fine because under normal circumstances people never feel like disadvantaged underdogs, right?) And if we still deny a relationship between increased ingrouping and increased outgrouping, there is at the very least a binary effect — you can’t have outgrouping at all if you don’t have ingrouping.

Take action, just not individually

The second half of Putnam’s point, which Haidt leaves out, is that over the long run, as we get to know each other, the effect of diversity decreasing trust tends to go away. He recommends that we “look for opportunities… in which for some period of time some shared value or shared identity trumps our racial or ethnic identity.” This is an appeal to individual improvement. Instead, Haidt argues that “the increasing Manichaeism [polarization] of American political life is not something we can address by signing pledges and resolving to be nicer. Our politics will become more civil when we find ways to change the procedures for electing politicians and the intuitions and environments within which they interact.” He makes an appeal for sociological improvements.

This tradeoff between individual and sociological appears over and over throughout the entire book. From religion to homo duplex to Durkheim, Haidt’s goal is to widen our view of societies from collections of individuals to a whole that is greater than the sum of its parts. Religion is a maypole dance — it makes no sense if you focus on an individual and their actions, and only when you see the dance with all participants in its entirety do you understand its purpose. Humans sometimes act with economic rationality, purchasing the cheapest of available goods, but are also prone to having their hive switch flipped on, acting out of allegiance to their tribe, and helping others at no benefit to themselves. His focus on why our nature makes individual solutions hard can seem defeatist at times, and he acknowledges the way he glosses over such issues: “because of the insurmountable power of the confirmation bias, counterarguments will have to be produced by those who disagree with me.”

The moral

Overall I agree with Haidt’s interpretation of the experimental evidence which shows that most study participants don’t reason very well, and that this causes a lot of conflict in conversations about politics. I disagree that this means reasoning is doomed, I prefer my own theory that reasoning is not hard, it’s just not a priority. Haidt was able to convince me that morality is probably not reducible to the principle of least harm, at least in the naive individualistic sense, even though I started out with a strong aversion to that idea. Sometimes helping the bee really does hurt the hive. And lastly I disagree with Haidt’s argument that we can best utilize human nature by coming up with systemic sociological changes which allow for more groupishness. I instead think that ingrouping tends to cause us to “other” those who are different than us and this is harmful to humanity overall; helping the hive can sometimes hurt the species. •