December 2012

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

Blogs & Sites We Read

Blog powered by Typepad

Search Site

  • Search Site



  • Counter

Become a Fan

Cat Quote

  • "He who dislikes the cat, was in his former life, a rat."

« Saving Pakistan... and India? Omar Ali | Main | International Workers' Day »

April 24, 2012


@ prasad:

Excellent piece of writing. This is why I like reading reviews of books. My shortcoming in being able to devour large numbers of books is helped by the fact that at least I know, generally, what all those books are about. The more important thing, for me, is learning a great deal more from a critical assessment by the reviewer than I could bring to, or take away from, a first reading on my own.

I'd like to comment about several parts of your review, but I'll do it in manageable chunks. Your criticism in "Things You Might DO with Foundations," uncovers an unforgivable sin in social science research. Going to press with the most high level and abstract 'findings,' without doing an in depth examination of more informative questions in inexcusable and suspect. A decent scientist might present findings and 'progress to date,' with a mention of the general direction of follow-on research, or where we go from here.

I will quote one piece of your writing here, to save the reader scrolling up and back.

"When you get a hundred thousand people to fill out detailed questionnaires cataloguing their attitudes, together with detailed demographic information, there’s a lot you can probe with this information, to see if particular attitudes to specific political questions covary with scores along various moral foundations....What would be valuable is this sort of information, but broken up along particular political and moral questions. Some possible examples:

"Natural food, or environmental pollution: purity plays a role in addition to care, but how much? As you ramp up interest in recycling, what happens to overall scores along the harm, fairness, or purity dimensions? Or in-group, for that matter - maybe recycling is about belonging to a good group and hating people in the bad one! Conservative wags often say ''Recycling'' is a liberal religion built around defilement of the land. Well, to what extent are they right?"

Haidt could keep his research assistants busy for a couple of years with IMPORTANT AND INFORMATIVE questions like these. Whether his RAs are really smart undergraduates, or run-of-the-mill cadres of graduate student slaves, he should be producing this research for journals and graduate student dissertations.

A good researcher knows that this kind of probing, the absence of which you brought to our attention, is part of an iterative process of refining and extending - even debunking - earlier ideas and research methods. He showed a modicum of refinement by increasing the number of foundational factors, as well as changing labels. This is fine, but the term 'modicum' is still operative.

Prasad, I'm going to read this several times. You need a 3QD writing prize for this -- at least! Back when I have something thought out to say...

Prasad, thanks for taking the trouble of writing a detailed and salient critique of Jonathan Haidt's methods and motives.

I have a somewhat better than sketchy idea of where Haidt is coming from, having read his views off and on since around 2004 or 2005. But I have never analyzed his data and claims in any formal way. My take on his views are more "instinctive" than "rational."

The foundations of morality that Haidt has picked for comparison between conservatives and liberals / libertarians are mostly adequate and I will grant him that perhaps our initial moral reactions do spring more from instinct than any rational thought. However, does the "rational rider" always steer toward the path which the "instinctive elephant" wanders towards, in order to justify the instinctive morality? Perhaps we do that a lot, making post hoc rational justifications for our instantaneous reaction, but not always. In fact most progress (civilized behavior, as we call it) in human as well as animal and environmental rights have taken place precisely because we bucked our instinctive morality. Also, how much of what Haidt calls "instinct" is actually cultural learning from a very early age? Take for example, the food habits of strict vegetarians from birth, the likes of which we find in India. For them the habit is a life-long cultural/ religious one. The choice is guided more by "purity" (the ick factor) than by any rational reasoning of "harm" (to the animal). But ask any Hindu / Jain vegetarian why they don't eat meat and they are most likely to explain it as a moral choice based on compassion and fairness towards other species. In this case, the rider does follow the elephant. On the other hand, the vegetarianism of one time meat eaters who make that choice later in life, even if they like eating meat, is usually based on thought out rational ethical considerations. Clearly, for them the rider charts the course of morality. Similarly for most human rights strides regarding race, gender, religious liberty and animal suffering and environmental preservation have occurred because people decided to act "against" their instinctive (some of it indistiguishable from early cultural training) moral reactions.

You mentioned the reaction of many heterosexuals towards homosexual acts. It is probably right to say that instinctual reactions of the former will contain a certain degree of the "ick" factor in imagining it. However, as we see in the recent mentality of most people towards gays, the evolving "moral judgments" are being made on a rational basis of fairness that are often contrary to their instinctive judgments of purity. Here is a very interesting review in the New Yorker of Dale Carpenter's new book, "Flagrant Conduct" that chronicles the story of the famous 2003 case of Lawrence vs Texas that succeeded in striking down the anti-sodomy laws in several (mostly southern) states in the US by a 6-3 decision. In his dissent, Antonin Scalia (sided with by William Rehnquist and Clarence Thomas) said that the majority was taking sides in a "cultur war" that opened the door to the "end of all morals legislation" and "judicial imposition of homosexual marriage." The majority opinion written by Anthony Kennedy and the concurring opinion by Sandra Day O'Connor invoked "personal liberty" and the "equal protection clause." Although Scalia is the one who brought up "morality," I would like to ask Haidt which of the two opinions in his opinion, is really the moral one - the seemingly "purity" based argument from the right or the "fairness" one of the majority. That the law was not entirely based on purity / sanctity, but may also have been guided by "authority/hierarchy" (with the heterosexual male at the top of the pyramid) is indicated by the fact that heterosexual sodomy and bestiality were not crimes in Texas at the time.

The Bhubaneshwar study is really a strange one coming from a self-identified liberal (at the time). I can quite see that a liberal might be comfortable or even charmed as a temporary guest in a conservative, strictly structured, seemingly happy and peaceful setting as the one Haidt found himself in Bhubaneshwar. (People go to gawk at the Amish in rural Pennsylvania). But to come away with the notion that a society based so deeply on hierarchy and purity may be happier "because" of those values, is a really shoddy piece of scholarship. Did he ask questions of the women and servants which went beyond, "Are you happy?" I would like to know. People can say that they are happy under amazingly surprising and even tragic conditions.

What exactly did Haidt ask the members of the Bhubaneshwar households? In a hierarchical, class-caste-gender inequality ridden (as also an attitude of resignation to fate) place like Bhubaneshwar, India, a question like, "Are you satisfied with your life?" would most probably elicit a response in the affirmative from almost all. Remember that the women and the servants also venerate authority - the employer / husband / father - more than we imagine. If on the other hand, Haidt had asked the servants something like, "Do you wish you owned your own land? Can you afford to send your children to school? Do they have easy access to a doctor?" or the women, "Can you make financial decisions on your own behalf or the household? Did you have any say in your marriage decisions? Do you wish you had a job outside the home?" etc., he may have got quite different answers.

I took Haidt's morality survey. I came out as ultra-liberal on all the questions. What does that mean, however? That I have certain criteria on which I base my moral judgment? Yes. But am I necessarily going to act on those instincts / rational arguments in the real world? Probably not. I too have Durkheimian tendencies of "structural functionality" based on stability and tradition. And that I think is a mortal flaw in Haidt's study. He interprets "self identification" on a moral scale as moral action. Doesn't work that way all the time which is why we surprise one another. That liberals are more predictable in the eye of the conservatives and not vice versa may mean a whole lot of things, not all of them very complimentary to conservatives nor a proof of the bone-headed lack of imagination on the part of liberals. The latter may actually have allowed for the benefit of the doubt towards the right, based on decency. BTW, why does Haidt lump "purity and decency" in the same category in his survey? Must both necessarily go hand in hand? That is another example of his own implicit bias.

As a research psychologist, this whole thing is very embarrassing. Here we have Prasad, a physicist, and Ruchira, a chemist, thinking the thoughts and asking the questions that Haidt should have thought and asked a long time ago. Haidt is an ersatz social scientist giving social science a bad name. More later.

I have always had this suspicion about Haidt who self identifies as a liberal but repeatedly, under the guise of toning down the shrill rhetoric across the ideological divide, comes out in favor of conservative values! It's almost as if he is saying, "If only the liberals chose to be a bit more like the conservatives, we would all get along." Very infuriating, especially to see liberals eagerly climb this illusory bandwagon of harmony by jettisoning their own values.

Got a bit curious about Haidt's morality quiz, and tried this shortened version on Bill Moyer's website. I thought some of my answers might be categorized as more conservative than liberal, and was surprised to find my score tended to the 'less traditional' and 'more progressive' than those who identify themselves as liberals. So is the test a bad one or just plain blind to all aspects that categorize a person's leanings, concentrating solely on the response to crime and punishment as a marker?

I haven't finished Haidt's book yet, and will comment later when I do so. Till now, I don't think he has made a very good case for why liberals should try to be more like conservatives. Nor has he started making a case for why liberals could effectively adopt tactics so frequently used with great effect by conservatives to 'rally the base'.

I came across some of Haidt survey questions online a while back and saw a basic problem with them. He seems to assume that moral reactions are flawed if the person who has them cannot justify them in a particular case. I would agree with his premise that moral responses are primarily emotional. Morality has to override very strong emotional drives towards selfishness, so it has to also be an emotional drive. But they also have to consist of relatively simple principles applicable to a variety of situations. One cannot respond to each novel situation with a novel emotion; there are many more situations than emotions. So moral feelings are rules of thumb; they can make sense generally, and it is silly to expect more. So that fact that Haidt can construct improbable situations where we cannot justify our instinctive moral reactions means nothing. Sex with dead animals is generally a health hazard, so evolution has provided us with an aversion to it. The fact that Haidt can contrive a situation, pretty much specific to modernity, where the danger is not present just means that the instinctive reaction did not develop in response to this kind of situation (a store-bought, and therefore sterile, dead chicken), which is pretty obvious. Likewise, with the incest example. Evolution does not account for condoms. It is no mystery why.

I also wonder whether health-based aversions are the same as moral insttincts. I know they are sometimes treated as such, and that this is part of what Haidt argues, but did he use the specific term "moral" in his surveys, or did he just ask, as above, whether such acts were "wrong". "Wrong" can have many meanings besides the moral, and, to the extent he treats the two as synonyms, he's committing a basic error. Take consuming feces. I do think there is a sense in which this is "wrong", but it is not a moral sense, it is a health instinct. There was a recent case where a woman had to medically consume the feces of another to replenish her own fauna. Strange and gross, but not a moral issue to me, and I think not to most, either liberal or conservative. Sex with dead chickens is about thhe same category.

This was a wonderful read. The skepticism toward evo-psych reduction and a dislike of those chasing the trendy patina of science in the dark waters of human motivation are what I most applaud (Prasad's terrific brain on display hardly needs mentioning as an additional source of pleasure). Not that I object to including evolutionary studies or cross comparisons between groups and other primates. Its just that premature shout of "Eureka!" that irks - when we all look around and realize that values still have to be created and argued. What we've adapted for we're not bound to necessarily. Job really well done!


Excellent review. I came across it actually a few minutes after I teared up listening to 40,000 Norwegians singing "My Rainbow Race". I don't think liberals are as bad at the group cohesion thing as Haidt claims. What does Haidt make of the fact that the left has always had better songs and done public protest better than the right?

I also agree with Ruchira in wishing that Haidt could have thought more critically about his experiences in the Indian village. It is possible to be empathetic without abandoning critique. Does he address why it's virtually a cultural universal that the female body is thought to be more in need of purification than the male body? This has serious implications in the "fairness" and "liberty" dimensions.

(OT: I love how you say "babby")

Unrelated to Jonathan Haidt or Prasad's review, here is a nice explanation of why it is important to ask the appropriate question in surveys to elicit the proper moral / political response. A letter to the editor in today's Houston Chronicle.

Vicki: Thanks for the leaving your comment here instead of at 3 QD or Facebook where you may have seen the link. Not everyone does that!

Lotsa traffic between here an 3QD! I have thoughts in both places. There are actions that cause a common harm -- harm to the self, and to the idea of the self as a fair dealer who is truthful. Maybe doing a thing you can get away with harms no one but you -- don't you count? Doesn't your idea of yourself as a person who can be relied on for doing the right thing count? You are the one you have to live with, after all, and that's not a foundation to whack away at or let erode.

Excellent piece Prasad, I'm a big fan of his moral foundations theory, but your review expands on some quibbles I've been having.

I do wish folks would hold off coming up with evolutionary explanations for social phenomena. I mean, I don't doubt the fundamental assumption of evolutionary psychology, that who we are is very much determined and constrained by millennia of selection, but there's no reason why explanations for things that we commonly understand as complex, such as religion, art, liberty etc. should have simple explanations. The scientist's instinct to look for the simplest possible explanation that fits the data doesn't work well here, especially when the data is knowledge that isn't very amenable to scientific analysis.

It looks like Haidt is doing this in his book, and if I understand Prasad correctly, he could have dropped the whole evolutionary thing without loss.

On the other hand, I am a little bit more concerned with the proliferation of foundations. Shouldn't this be more settled? Or are we seeing more and more examples that only kinda-sorta fit a foundation? For instance...

Supposing someone at a maternity ward swaps two newborns, who end up going home with the wrong families. No-one is the wiser.

That's wrong, isn't it? But from which moral foundation?

Two years later, the hospital discovers the swap. Should they tell the families? If they do, it will cause a great deal of pain, which is bad on the harm/care foundation. If they don't, which moral foundation has been violated?

I borrow your car without your permission while you're away, and return it without you ever knowing. Sure, I caused some wear on the car, but I left extra fuel in the tank, which is worth more. Did I do wrong?

What if I have sex with you pretending I'm your husband? You consent under the false pretence, enjoy the experience and never find out. In many jurisdictions I've committed the crime of rape. I did wrong, but on which moral foundation?

Elatia, my moral intuition is that I would hurt another person somehow, not just myself, in my examples, even if they never know it.

I suppose I want to add another moral foundation, perhaps "integrity/violation". It feels like such actions are a violation of something about someone: their truth, their integrity. People have needs, desires, purposes that are part of who they are, which it feels wrong to subvert. But there's also a connection with purity/sanctity: violations of sanctity seem to be violations of some concept of integrity.

Elatia and Sagredo,
It is interesting that you bring this up. I am not quite sure exactly where Haidt is on the core basis of morality. He claims that we all have instinctive moral compasses by which we judge ethical actions (according to him, even six month old human babies have a rudimentary sense of harm and meanness) which to me would mean that we therefore would use it not only to judge others but also our own 'purity' or as Sagredo puts it better, integrity.

Yet I am not very clear as to what Haidt sees as the value of this moral compass. He describes a debate between Socrates and Plato's brother Glaucon regarding exactly this scenario - of private vs public morality. Socrates came out on the side of personal integrity - that our morality acts as a self checking ethical mechanism whether or not we get caught and regardless of what others may think of us. Glaucon claimed that our ethical actions are meant to protect and enhance our social reputation and not our self worth. Haidt thinks Glaucon correctly recognized the social aspect of morality which ensures that bad actions will always bring bad consequences for the individual. The Socratic version, according Haidt, is a 'cerebral' decision irrespective of consequences, that values the self above society. He calls Socrates (and Plato) wrong and Glaucon a hero for recognizing that morality is based on emotion and not reason. Our moral compass therefore, says Haidt, is mostly a 'sociometer.' I am not sure if he is making a value judgment of the two views or merely an 'evo-psychological' observation. But then what about the "ick" factor, the one he says arises out of our innate sense of purity which I take would prevent us from certain acts even if no one was aware of it? He seems to want it both ways. If Socrates was being coldly rational in deciding that ethics are as much about remaining true to oneself as being just to others, doesn't the Glauconian view of ethics as only a deterrent to societal punishment sound even more calculating? I am not sure I understood the purity of this argument. Perhaps he means to say that as social animals, our instincts are more geared towards pleasing the plurality than ourselves. I am quite confused by the sophistry.

I think that very early in life, someone has to teach you that behaving badly to others precisely equals behaving ruinously to yourself. You need an "injury barometer" based on empathy and self-respect, and when you do go wrong -- as everyone does -- you need to be taught that coming clean and humbling yourself is the best action: you have not gotten away with anything, you are seeking to repair harm and be personally restored. Children can learn this better than adults, if only because an adult who has not learned this is probably going to be a "situation ethicist" for life, casting ethics as an exercise in enlightened self-interest, in keeping his nose clean, and in image management.

About 25 years ago, John Shad gave the Harvard Business School 10m to fund a study of ethics in business. One wag remarked, "You can't buy a lot of ethics for only 10m..." Exactly! You can't buy any, and if you did buy them, they are not ethics.

Thanks all for reading, and for the comments.

Ruchira, I didn't think carefully enough about his 'elephant/rider' chapters, and still haven't! Some musings -
- I remember Haidt mentioned there was one and only one study showing people have the ability to "act against" their intuitive responses. Funnily, it was also, if I recall, the only study described where people were given time to think about their response. For the most part the cognitive bias/implicit association type results he showed involved people making snap decisions. It's not exactly surprising to see intuition predominate under that circumstance!
- I think you're right about the possibility that the intuition/reason "weighting" differs for day-to-day human decisions and big society-wide progressive changes of the type you mention. But Haidt might still say in those cases that what's really changing minds is empathetic identification and consciousness raising, not rational argumentation, per se.
- In general there's the issue for Haidt that these sorts of results might turn out to be at the macro level what the behavioral economics literature has been - lots of flashy results in lab settings with small numbers of people, but when you integrate over large time periods and amounts and numbers of people nothing much remains. You're left with efficient markets plus animal spirits and not much "value" has been added. Here too, if you want to understand directed change over time, the "noise" might average out, and you're left with things that actually have an impulsive effect over time, which (whether intuition or reason or whatever else) needn't be the same as what matters when you quiz twenty people at time t.
- One specific thing we should all keep in mind is that a lot of the time the leading dynamic is 'demographic' - one side of an issue finds it hard to maintain itself in new generations. A generation after women get the vote no-one can understand why that was ever an issue. The question of what changes minds *in the under-18 set* gains especial importance then. This too can be seen as either mostly "intuitive" or "rational" (children lack the knowledge, experience, intellect needed to make good arguments and are swayed unduly by fashion and authority vs they're imperfectly inculcated in the bigotries, prejudices and preconceptions underwriting the current order.)

Sujatha, you've got me wondering whether Haidt tried to get people from different cultures and such to pose questions. Maybe if he got people in blue collar communities and Bhubaneshwar to pose survey questions he might see richer patterns (of course, this would tend to underscore his point about moral philosophy as the effort by professors in universities trying to understand their undergraduate students - you've got to spread out beyond the professor, not just beyond the student). I think re test design and how the questions Haidt poses affect what he sees, this is right up Norman's alley! (hint)

Martin, that's a very interesting point - what's a rule worth if every time you think it's not working you ignore it? Or you might even say the thing is bad because prohibited, that law deserves respect, etc. I think Haidt would reply that this is an option that was implicitly always available to his respondents. They *could* reply and say what you did, but instead they think the behavior in the case, in that situation, itself is wrong without being able to articulate why. Still, this is a somewhat complicated thought that may not occur immediately to someone, particularly under stressful conditions. I wonder how many people would take advantage of such a resource if made aware of it before the start of testing.
I'm less certain that people don't regard cases like the probiotic poop consumption one as morally interesting. I witnessed recently (in the context of a general discussion of 'lab meat') two very smart people arguing sincerely whether eating lab-made human meat would be cannibalism (hence wrong). Without access to a "purity/disgust/sanctity" type concern, it's hard to even understand what they're producing words about. At least with these concepts in mind you can diagnose what exactly is going on in their heads. Where Haidt and I would differ is that I think when you analyze this sort of situation, you can get people to change their minds, not necessarily by analysis alone, but in conjunction with a suitable social context (say if some fuzzy religion decided eating lab-made fake human meat had sacramental value) And such argument is better at dissolving intuitions based on some foundations than those relying on others (by which I mean particularly the new ones he introduces).

Jesse - I was actually surprised to find myself taking this tack. I don't as a rule object to evo-psych talk either (though neither am I swallowing every just-so story told), and am a pretty happy camper in the reductionistic, scientistic, materialistic group. It's just hard for me to see Haidt's evo ideas as being contextually useful even if true, which is especially bad when he's using somewhat marginal evo-arguments in addition. Plus as you say it's rather sloppy since he ignores the central ethical questions posed by this sort of analysis. I suppose Haidt would say my strong intuitive dislike for him outweighs my lukewarm response to evo-psych and positive intuitive response to reductionism.

Vicky - it was kinda silly but ever since the something awful animation I can't even write 'baby' without wanting to put a second b there :) More seriously, I think you make a very good point about how the left has done pretty well for itself in terms of solidarity movements, marches, protests, rallies and the like. Is the claim that the right is better at community organizing even true? Hmm, now that I think of it, it's not at all clear. I suppose they've got the advantage re weekly gathering in churches (I'm pretty sure this declines a moderate amount going from right to left). The point you and Ruchira make about women in traditional societies is all-important. It's well and good for Burke types to say traditions and institutions can have value that we can't immediately see with reason, but all too often when you look at unequal institutions like this, that "value" basically amounts to people in charge making things cushy for themselves. And all types of people haven't exactly had the same chance to be in charge.

Elatia - I suppose you could shoehorn into care or fairness, at least for the case where someone else filches your good name. Otherwise it might fit to a decent extent into two of his new, as yet unincluded foundations - honesty/integrity etc and maybe also property. I agree that it's not clear that these sorts of foundations are getting, at a fundamental level, issues pertaining to the ways of proper self-cultivation, or really any ethical issues connecting a person to himself. Haidt does say (if I recall) that he personally is attracted to virtue ethics, so maybe he's written about this somewhere.

Sagredo - I was thinking of the foundations-count question as being sort of like the features-count question when making a map. In addition to purely definitional issues (what's a city/what's a feature) you have the issue of needing to cut off your dots and squiggles somewhere, and it's probably not a deep science where you do it. He's trying to provide a basis for the ethical aspects of folk psychology, but presumably he doesn't intend or expect to capture everything. I do agree that if he turns out to need (say) twenty foundations to truly capture moral intuitions, it's a lot less interesting and useful than if it turns out he can get away with six or seven.
I think with your example scenarios, we're seeing again the issue of whether he's really posing varied enough questions to capture the range of human moral concerns, or whether the fact that he has six, not twenty-six, foundations has to do with the imaginative limits of his team. I think here again he would do well to talk more to colleagues in anthropology and literature departments instead of the trendy bio-crowd. To be sure, those guys sometimes don't like science-y stuff much, but they have the advantage of having spoken to lots of different people.

FULL DISCLOSURE: I have not read Jonathan Haidt's book. However, I have read Prasad's earlier, and more lengthy, draft, and a number of reviews. I watched a number of videos, including TED, of Haidt talking about "The Righteous Mind," a video interview with Robert Wright, and a Public Radio interview with Leonard Lopate. Also, I took one of his surveys and will take a few more. So here goes.

All social science research will have some flaws - some serious and some not. It is very easy for any experienced researcher, like myself, to spot almost all of them, eventually. The temptation is great to turn a critique of another person's research into an opportunity to tell the world how smart the critic is - especially compared to the author of the 'flawed' research. However, there are serious, fundamental flaws in his research and interpretations.

1. Correlation versus Causation

Interpreting Correlation as Causation is a fundamental problem, if not a deadly one, for any interpretation of Haidt's data. Haidt's interesting moral dimensions (also called personal styles in other contexts,) that are used to differentiate Liberals and Conservatives, are only - and exclusively - correlates of self-identification with a label. That label has no definition other than an operational one - the person says I own this label. There is nothing in the data to suggest that these personal styles, Haidt's moral dimensions, have any bearing on the political behavior of the respondents. By political behavior I mean voting behavior, contributing to a candidate or cause, or working against an opposing candidate or cause. It is the simplest and the worst mistake any social scientist can make. Correlation is interpreted as causation.

There is a way to examine causation, and to determine if these personal styles actually influence political behavior. You set up a study in which you change something in the personal style of the individual, and then look for a change in behavior. The reason why you don't see Haidt (nor many other social scientists) doing this is because it is too frakkin' hard, and takes too damn long. Striking while the iron is hot may be advantageous in publishing a NY Times best seller. I've been in the survey research field for more than 40 years. It takes a great deal of discipline, and 'moral' integrity to attenuate the impulse to run to publication with data that are, in the end, more apparent than real.

Here is an example of misinterpreting correlation as causation. It comes from another area of research of several decades ago. It was known as the Locus of Control (LOC) studies. Briefly, LOC describes the source of the reinforcements (this was in the day when Skinner and behaviorism were the IN THING) that determine the outcomes of our lives. The 'good' source of LOC was from within the person, an Internal LOC. If the source of life's reinforcements came from the outside, External LOC, that was 'bad.' With Internal LOC, a person felt that they had control over the sources of reinforcement in their lives. With External LOC control of your life was in the hands of other people and events in the world. Mentally healthy people had an Internal LOC, and mentally ill people had an External LOC. White people had an Internal LOC, and Black people felt that forces that controlled the reinforcements in their lives were outside themselves - External LOC. The same held for more educated people versus less educated people; People with more money, etc. etc. etc.

Everyone but everyone believed that LOC was a determiner (A CAUSE) of behavior (success versus failure) in the real world. Black people would do so much better, it was believed, if they understood that they should find their LOC inside themselves. This reinforced a belief that failure of Blacks in society was a direct result of a character flaw. Research in later years showed that a belief in an External LOC was not a character flaw. Rather, it was an intelligent and accurate assessment of their lives and the world in which they were living. Internal versus External LOC were correlates of race/ethic classification, not a determiner of behavior of peoples in different groups.

More managable chunks will be revealed, below.

Excellent review! I think that's a pretty fair assessment of Haidt's ideas concerning the liberal/conservative divide.

Haidt is a respected social psychologist with a great career. However, I think his grand theory is poppycock. In particular, to say that liberals don't value loyalty, authority, and sanctity is ridiculous. Just spend some time reading over on Democratic Underground and you'll see all of these moral dimensions represented in spades. They're just represented differently from conservatives. And in fact, I would say that, when it comes to conservative politicians at least, that those three moral dimensions are underrepresented.

I've come to be suspicious of any theory which tries to find some fundamental element that explains why most people are conservative and some are liberal. Primarily because their definitions of liberal and conservative are ahistorical and culturally hidebound. The conservatives of today are very different than the conservatives of the 1940s, as are the liberals. The fact of the matter is that both groups have been infected with individualist narratives -- the right with Randian pseudo-philosophy, and the left with hyper-rationalist techno-Utopianism. See Adam Curtis' brilliant 4-part documentary series The Century of the Self to see how the latter plays out (specifically part three I think). It's available free for watching online or download at

I just don't think you can make these sweeping generalizations about liberals and conservatives. The same goes for religious and areligious folk too. Humans aren't that simple.

Just for added context: Haidt was complaining about "liberal bias" in academia just last year.

@ Leo:

"However, I think his grand theory is poppycock." LIKE!

Thanks for mentioning "The Century of the Self." I second the motion.

OK Leo, I'm drifting a little off-topic here but here's my grand unified theory of why people are liberal and conservative (or do I mean left- and right-wing?): All malice is attributable to two root causes, greed and envy, often mixed together. Greed is the desire for more regardless of how that hurts others. Envy is the desire to hurt others who have something we don't. The balance of whether we have more one than the other leads to the two wings of the political scale: I hardly need to say which one is which. But then I tend to put "liberal" in the middle, "socialist" on the left and "capitalist" on the right.

Does Haidt mention restorative justice at all? It seems like restorative justice is based more on the harm/care foundation, whereas retributive justice is based more on fairness.

@ Sagredo:

Your question makes the point that Haidt should have thought out his ideas a whole lot better.

More on methodology:

2. Throw away your horoscopes. Personality tells all.

The potential explanatory power of the concept of personality, since the earliest days of Sigmund Freud, captivated the imaginations of people in medicine, psychology, literature, advertising, human potential, education, law, child rearing, and politics. The study of Haidt's moral dimensions is a study of personality, albeit a circumscribed view of personality, or personal styles. The more accurately you can describe personality, the more successful you will be in predicting behavioral, emotional, and mental outcomes in many aspects of life. At least, that was the hope and expectation.

Haidt's approach is akin to Adorno's study of "The Authoritarian Personality," from the 1940s through the 1960s. It was an attempt to explain the origins, rise, and behaviors of the Fascist State. Actually, it was even more specific than that. Adorno wanted to explain the Nazi State under Hitler, and the atrocities of the holocaust. Discovering a specific personality type that would explain everything was a reasonable hypothesis with which to start.

In the end, foundational personality traits, or personal styles, were found to have little utility in predicting specific outcomes in real life, or in generalizing to a wide array of situations and environments. Yet, Haidt's expectation is that he is going to do just that, and expose the inner workings of moral dimensions for Liberals, Libertarians, and Conservatives. Better still, he is going to do so without ever defining these three concepts. Factor analysis of self-reports is not a definition. It's a circular reasoning cop out.

Adorno failed because there was no underlying personality type that made Fascism, Hitler, Nazi Germany, and the Holocaust possible. Everybody and every personality type participated.

Haidt fails because his explanations are akin to roll-your-own Freudian interpretations of everyday life. A signature with loops and extensions that go, mostly, below the line, indicate a dominant Id. If they are, mostly, above the body of the signature, then you have a dominant superego. When you have no non-circular definitions, and no research showing a link to actual behavior, then you are stuck in trying to generalize. Otherwise, your explantions and interpretations are classed with the likes of deceptive illusions that can appear as either a hag or as an attractive young woman. His interpretations are no better than parlor tricks.

@ Ruchira:

The Houston Chronicle article that you linked, above, references studies done by the Survey Research Center, Institute for Social Research, University of Michigan. A number of my colleagues came from SRC-IRC. None of them would be caught dead producing the superficial explanations of survey data as Haidt does. The conclusions in the SRC-IRC research reflect far more careful thought, over longer periods of time, and from better developed survey instruments and administrations than Haidt could ever hope to accomplish.

I recommend the article in Ruchira's link to everyone.

It seems like we're no further along in understanding why the right embraces "traditional values" so tightly with one arm and the creative destruction of unregulated capitalism just as tightly with the other...

@ Vicki:

In Adorno's research on "The Authoritarian Personality," there was no single compelling explanation for Fascism, Nazism, Hitler, and the Holocaust. That's because these were not a unified, seamless phenomenon, cut from whole cloth. Many things came together to produce them, and any explanation will be at least as complex. It seemed obvious that we should find a Germanic personality that explained it all. He didn't find it.

Now, Haidt wants us to believe that six moral dimensions separate separate the left from the right, and the right leads by a score of 6 to 3; and the left will never be able to make up the difference.

Anyone else want to propose a single explanation that describes the political spectrum for most of the industrialized world? And yes, it's that ridiculous.

Vicki: Neither can we explain why they venerate "loyalty and authority" more than those on the left and yet numerous public and political icons on the right "cut and run" (as John Murtha famously put it) when they had the opportunity to serve their country by enlisting in the military. As I said in an earlier comment, self identification in a morality survey does not necessarily translate to moral actions.

Ruchira, your point leads to another values consideration I wish Haidt had thought to treat, in the left v. right context. The values we talk about are the values we wish we had. The values we live are the ones we actually have. Thus you see passionate defenders of a market economy accepting corporate welfare, and you see socialists embracing totalitarianism if it's the only way they see to save their corner of socialism.

I think maybe Haidt has been "hot" and newsy for a while, and, as that never lasts forever, it made him hurry the book along? If so, into which category would he put such a decision, even or especially if it were unconscious?

Haidt has a supporter in Thomas Edsall, see

Methinks that the Obama campaign have read Haidt's book and adapted some of what he suggests to appeal to the conservative voters. Vide the agonizing of the GOP over the 'politicization' of OBL's death. Must be major cognitive dissonance for the average conservative voter to realize that the 'liberals' have snatched that ice-cream from the GOP babby.

@Norm: Thank you for bringing your expertise on psychological surveys and their pitfalls to our attention.

@Sagredo: I am now about 50% into Haidt's book. So far, I have seen one reference to "retributive justice" and none to the "restorative" variety. But they are two sides of the coin so, it is possible to predict what he will say about the latter. As for retributive justice, Haidt says that conservatives are more in its favor because it fulfills the foundation of "fairness & cheating" based on proportionality - tit-for-tat - that according to him, is a fundamental human emotion dating back to the times of the ancient tribes. He also points out that tit-for-tat finds less favor with liberals when related to punishment because in their estimation, retributive justice violates the foundation of "harm & care" by increasingly the sum total of harm. He does't elaborate which one is a superior instinct but does go on to claim that one of the reasons Democrats lose elections in white, working class communities (a more sympathetic interpretation of What's the Matter with Kansas)
because they miss the emotional moral point of retributive justice.

Frankly, I am getting a little mentally mired in trying to keep up with Haidt's evolving narrative although the main thrust of his argument is pretty simple. As Prasad points out in the review, the book is a compilation of his research over the years and Haidt tweaks his methods as more and more angles open up in his mind and the "moral foundations" are revised. There seem to be numerous "on-the-one-hand and on-the-other-hand" types of explanations forming along the way. But then the human mind is a complex thing and Haidt (or any one else, for that matter) lacks the complete and necessary set of tools required to put its workings into neat little shelves.

Jonathan Haidt conquers the White House:

Who says House Republicans and President Obama never work together? From

In order to recognize the American spirit of loyalty and the sacrifices that so many have made for our Nation, the Congress, by Public Law 85-529 as amended, has designated May 1 of each year as "Loyalty Day." On this day, let us reaffirm our allegiance to the United States of America, our Constitution, and our founding values.

NOW, THEREFORE, I, BARACK OBAMA, President of the United States of America, do hereby proclaim May 1, 2012, as Loyalty Day. This Loyalty Day, I call upon all the people of the United States to join in support of this national observance, whether by displaying the flag of the United States or pledging allegiance to the Republic for which it stands.

IN WITNESS WHEREOF, I have hereunto set my hand this first day of May, in the year of our Lord two thousand twelve, and of the Independence of the United States of America the two hundred and thirty-sixth.

May Day! May Day! May Day!

President Barack Obama is being body-snatched by Jonathan (The Righteous Mind) Haidt. He declares May 1, 2012 as Loyalty Day.

May Day! May Day! May Day!

Link dump:

The comments to this entry are closed.