Arnold Kling  

Where Haidt Falls Short

PRINT
Confusion About Greenspan... Rodrik's Question...

Going back to Will Wilkinson's exposition.


Haidt’s research leads him to posit five psychological foundations of human moral sentiment, each with a distinct evolutionary history and function, which he labels harm, reciprocity, ingroup, hierarchy, and purity. While the five foundations are universal, cultures build upon each to varying degrees. Imagine five adjustable slides on a stereo equalizer that can be turned up or down to produce different balances of sound...If you’re a sharia devotee ready to stone adulterers and slaughter infidels, you have purity and ingroup pushed up to eleven. PETA members, who vibrate to the pain of other species, have turned ingroup way down and harm way up.

...Democrat-leaning liberals draw almost exclusively from harm and reciprocity, Republican-leaning conservatives draw more from the whole range of moral emotion.


On the last point, color me skeptical.

Liberals less concerned with in-group solidarity? True, they might not have an American flag on their lawn. It would leave less room for the "impeach them both" sign.

The liberals on the County Council that is considering a ban on trans fats would never be accused of appealing to the primitive emotions of purity/disgust.

And if research shows that liberals are relatively indifferent to hierarchy and authority, who am I to question that? After all, this is only a blog post, not a peer-reviewed journal. And I don't have tenure.


Comments and Sharing


CATEGORIES: Political Economy



COMMENTS (9 to date)
Rue Des Quatre Vents writes:

One thing that I'd like to ask Haidt about is how his survey's capture feelings about hierarchy, authority, and in-group loyalties.

Something in his conclusions worry me: I'm no so certain his questions accurately portray the loyalties people feel to certain institutions and secondly, I don't believe they accurately indicate what people would in fact do in certain situations. To see what I'm trying to get at, recall Milgram's shock experiments. First off, people's actions did not conform to what they expected on a survey. On the survey, people routintely claimed that they would not turn up the voltage past some low threshold. And yet, when the chips were down, they consistently did so.

Which brings me to my second point. Participants were willing to shock others because they were beholden to the authority of science.

I do not believe Milgram collected information about the participants's political views. But since the results were so far reaching, I find it hard to believe that so-called liberals, those who, as Haidt claims, profess to be mostly concerned about non-harm, would act any differently. Liberals have their hierarchies and their authorities. They just don't work in churches.

Matthew c writes:

Wonderful critique, absolutely spot on. The belief that liberals do not play the "in-group solidarity" and "disgust" cards is very biased and shows a strong lack of self-awareness.

Haight's essay is well worth reading, though, for his attention to how most supposedly rational thought about controversial topics is actually rationalizing positions already held.

This is the key to overcoming bias -- to see this kind of thought process going on in ourselves when we engage in it.

Again, a strong dose of humility is required to notice when our "rational" thought process is actually a "rationalizing" thought process. Introspection, while often unfairly derided, can play a key role here. . .

Buzzcut writes:

I took some of his surveys. They were very vague and very difficult to complete. I'm sure if I took them a second time, I would answer some of the questions differently (because of the vagueness).

I really question the vallidity of the whole enterprise. It is all based on deeply flawed surveys.

Your liberal/conservative analysis rings false to me. Like a lot of dialectical constructs where everyone's encourage to pick a side, the loser in my opinion is the concept of giving primacy to empiricism. Are you on the non-liberal team, Arnold Kling? Is Paul Krugman on the liberal team? And is that getting in the way of empirically derived solutions to the aging and existential challenges the three of us face?

I think these dialectics are more about representational privilege plays, often with tragedy of the commons effects if they're not centered around giving primacy to our persistence odds.

As such, I think they're performances more than good faith efforts to create the best models of reality and effective solutions to our existential challenges.

A good positive contrast is your post on Aubrey de Gray's book and Caplan's writings about changing enfranchisement rules for more economically effective decision-making.

Matt writes:

See, this guy is not a dualist, and the proofs go in completely different directions.

His claim is even deeper, that there are at least five instinctual basis for morality.

His proof starts to break down when he claims the sentiments can be adjustable.

Here is the dis-proof.

If the five basic instincts are adjustable, then one set of adjustments would predominate, for there is an evolutionary advantage if all primate organizations operate from a mutually understood set of common adjustments. Two humans, regardless of culture, must operate from the same set of adjustments to get the maximum human interaction efficiency.

Probabalistically, the adjustments can vary about the mean but must fall off as gaussian.

Hence, he is not talking about instinctual basis.

So the dualists trouble himself to put these categories under the proper instinct. Each of the five must be a characteristic of one instinct, or the other, or a descriptio of the interaction. None can apply equally to both instincts independently.

Most dualists postulate two very different instincts, for evolution would have its two instinctual basis occupy unique and different demands of evolutionary imperitives.

Troy Camplin writes:

It's not that liberals don't play those cards -- they're not stupid, they know what works -- it's that they themsevles are not affected by them as much as are conservatives. Conservatives want freedom to associate with people the way they want -- that is ingroup preference. Liberals try to eliminate all ingroups, atomize us, so that we are left with nothing but the government to organize us.

In general, I think Haidt is right. However, having taken his surveys myself, I would say that too much emphasis is put on government-as-an-indication-of-ingroup, whereas it is possible that one can be locally communitarian and nationally libertarian.

Matt writes:

Reading a little more, he has no premises about which to infer evolution's path.

Under most evolutinary dogma, instincts would have to arrive hierarchically and temporally i time. I don't see that.

Caliban Darklock writes:

While liberals may indeed understand these concepts, the assertion is that they don't use them in making decisions.

A liberal tends to view ingroups, heirarchies, and notions of purity as relativistic notions which are fundamentally arbitrary and hold no meaning. The conservative, however, understands that these arbitrary distinctions are the very thing that makes us human - and they hold tremendous meaning.

So while a liberal may indeed "clique up" with his ingroup, he will not side with the members of that ingroup simply because they are part of it. He will side with them when he agrees, and dissent when he does not. To do otherwise would be hypocrisy.

The conservative, on the other hand, frequently refuses to dissent with his ingroup because it is his ingroup. He will toe the party line on policy matters even if he does not agree, because that is what his ingroup believes. If his own behavior does not align with that belief, he does not see the problem.

I'm sure you can connect the dots with any number of recent events. The key point here is that neither the liberal nor the conservative is objectively wrong - they simply disagree on the importance of these matters.

MT57 writes:

As far as I can tell, all redistributive arguments, which I shall assume are liberal, stem from resentment over relative social position. Where that sentiment fits into Haidt's categories, I do not know. But the assertion that liberals are not driven, or are less driven, by ingroup sentiments, I find ludicrous, having visited numerous leftleaning blogs daily for the last few years (I am a Democrat). The whole drama of academic freedom vs. pc is a example of ingroup dynamics if not hierarchical behavior (when they control the dept) by the left, trying to define who is admissible to the group and who should be allowed to remain and in what role. The recent moveon.org controversy is another example - support moveon.org, don't break ranks, why did we elect these people if they're not doing what we elected them for, etc. - that's what you read on leftleaning blogs and the comments to them.

Comments for this entry have been closed
Return to top