Moral Diversity vs. Evaluative Diversity

The Righteous MindIn his most recent book, The Righteous Mind: Why good people are divided by politics and religion, Jonathan Haidt (pronounced like “height”) reminds the reader at various points that he is telling his story in a roundabout way because typical readers would reject straight-up truth. The first four chapters are devoted to evidence that the average non-psychopath is irrational, able to learn truth only “in love” (as Ephesians 4:15 puts it). The Righteous Mind debuted at #6 on the New York Times best seller list for nonfiction hardcover, so, if you find it difficult to believe the claims in the summary below, you might want to try the roundabout version instead.

The Purpose of Division

Why are good people divided? Haidt devoted an entire chapter to defend the theory of group selection which entails that diversity will evolve if diversification is advantageous for groups. On page 365, Haidt summarized his conclusions about this advantage:

I suggested that liberals and conservatives are like yin and yang—both are “necessary elements of a healthy state of political life,” as John Stuart Mill put it.

In a similar way, bone cells and muscle cells are both necessary to the functioning of the human body, and it is for the good of the body that its cells divide and specialize.

To test the theory that diversification is advantageous for groups, one would want to compare the success of groups with different levels of diversity. Such evidence was collected by Douglas Wilde, a professor of design at Stanford University. His students divided into teams to develop designs submitted to intercollegiate competitions which were judged by blind-review. In some years, Wilde allowed students to form their own teams; in other years he forced them to team up with people who tended to think differently. Wilde, and the design professors who replicated this experiment at other colleges, found that forcing teams to be evaluatively diverse increased both internal conflict and win rates.

Instead of citing the research by the design professors, Haidt cited the research of Richard Sosis who found that the average religious commune founded in the nineteenth century United States was six times as likely as the average secular one to last over 20 years. Again, the research compared the success of different groups, but Sosis’ measure of success was longevity, while Wilde’s measure of success was win rate. Wilde’s measure would be irrelevant if we encountered a society that could survive well-enough with poor designs (i.e. had no competitors or environmental disasters pending to require rapid improvement of social designs).

The problem with Sosis’ research is that he did not manipulate or measure diversity. It is debatable whether the religious communes were more or less diverse than the secular ones. Communes are intrinsically anti-conservative—they are rebellions against the status-quo—yet religious communes have a commitment to norms. Thus, religious communes might be more likely to attract both liberals and conservatives, and it makes sense to expect them to be more diverse. Some of the greatest religious role-models created new norms while rebelling against the norms of their day (e.g. Muhammad, Jesus, Buddha, Gandhi, Confucius), yet Haidt offers an explanation which implies that religious communes would be less diverse (pg 342):

A commune that valued self-expression over conformity and that prized the virtue of tolerance over loyalty… would have lower moral capital than a commune that valued conformity and loyalty. The stricter commune would be better able to suppress or regulate selfishness, and would therefore be more likely to endure.

In Wilde’s research, the superior teams had heightened internal conflict, but Haidt’s explanation of Sosis’ research implies that we should expect the opposite. This may just be an example of Haidt trying to tell the story in a roundabout way. The bottom line is that Sosis’ research would need to be repeated with actual measures of diversity. Until then, we have Wilde’s results to support Haidt’s final conclusion that diversity is advantageous.

Proximate Causes of Division

From an evolutionary perspective, one could say that the cells of our bodies specialize into diverse types because this brings advantages to the body as a whole, but it is also correct to say that cells specialize because they are genetically programmed to do so. Genes are a proximate cause. In a similar way, while Haidt points to group-selection as the ultimate cause of division, he also points to research indicating that genetic and physiological differences (products of evolution) predispose us to disagree with one another.

After summarizing some of the research described in greater detail in John Hibbing and Kevin Smith’s Predisposed: Liberals, Conservatives, and the Biology of Political Differences, Haidt attempts to navigate the controversial issue of how our natures interact with nurture. This comes to a head in the recounting of Keith Richards’ testimony that he became a liberal when he was betrayed by the choir master of his school (pg 330):

Richards may have been predisposed by his personality [and genes] to become a liberal, but his politics were not predestined. Had his teachers treated him differently… he could have ended up in a more conventional job surrounded by conservative colleagues and sharing their moral matrix.

Of course a sufficiently controlled environment can manipulate the typical person into developing values contrary to his/her own genetic predisposition. Haidt also mentions that sufficiently controlled environments can flip a switch he calls the “hive switch” to shift a person’s values temporarily. He discusses oxytocin regulation, but dopamine regulation and ego depletion would be other such switches. However, Haidt stops short of discussing what the costs of manipulating people’s values might be.

Assuming one were to manipulate an environment to promote conservativism, it would see a decline in liberalism. If this sufficiently unbalances the society, then, according to the theory Haidt quoted from John Stuart Mill, it would collapse like an unbalanced ecosystem. That is one example of a cost. It is a cost to the group.

But we should also consider the consequences for an individual like Keith Richards. How would he like to have values contrary to his predispositions? Would he be frustrated like a short basketball player, a gay person in a heterosexual marriage, or someone with high IQ who cannot access the Intenet? Keith Richards is the lead guitarist of The Rolling Stones—it is difficult to imagine him being so successful in that role without genes predisposing him against conservativism—how would it have felt not to exercise those genes? Here’s one theory:

Theory #1: In more tolerant environments, people are more likely to hold values which align with their genetic predispositions and those who have such alignment experience better mental well-being (e.g. greater engagement in their career, family and community, and less depression, apathy, guilt, and desire to commit suicide).

To test this theory, psychologists would measure the values, predispositions and mental-well-being of people in environments with different levels of evaluativism. The benefits of this research could be huge: if it confirms the theory above, we could use it to improve mental well-being for our children and grandchildren. Most of the people with jobs today are not happy with their jobs, and our own lives might not be so bleak if our grandparents had conducted this research. So we have to ask, “Why have no psychologists tested this theory?”

Haidt’s subtitle “Why good people are divided by politics and religion” seems to ask about the causes of intolerance. If it turns out that intolerance has such significantly negative health consequences, that discovering them would motivate us to be more tolerant, then it is fair to say we are intolerant because psychologists have not measured those consequences. Psychologists have determined that suicide is the 10th leading cause of death in the U.S. and that gay youth facing anti-gay environments are more likely to attempt suicide, but this just a beginning to measuring the consequences of intolerance. Homophobia isn’t the only form of discrimination, and mental distress includes more than just suicide.

A 2014 study by Shanto Iyengar and Sean Westwood found that 80% of us, if asked to judge a scholarship competition, would discriminate against applicants with opposing values. That kind of discrimination is called “evaluativism” and the researchers offer every reason to believe it is pervasive, producing every manner of frustration. For the 13 years previous to that study, the only major study comparing kinds of discrimination was Haidt’s own study with Evan Rosenberg and Holly Hom. They found that people discriminate far more on the basis of values than on the basis of demographic differences, such as race, class and religion. His conclusion, in 2001, was that values diversity (which they called “moral diversity”) creates so much discrimination that it must be a bad kind of diversity.

In The Righteous Mind Haidt cited his 2001 study only in a footnote to his recommendation about how to make a team, company, school or other organization more “hivish, happy and productive” (pg 277):

Increase similarity, not diversity. To make a human hive, you want to make everyone feel like a family. So don’t call attention to the racial and ethnic differences; make them less relevant by ramping up similarity and celebrating the group’s shared values and common identity.

Again, Haidt implies that our aim should be to minimize internal conflict. As Haidt would predict, in years when Wilde didn’t draw attention to evaluative diversity, his students self-segregated and experienced less internal conflict. But the hivishness and happiness did not improve production; the consequence of self-segregation was inferior designs. Furthermore, if we do not raise awareness of evaluativism in awarding scholarships (and presumably jobs as well), Iyengar and Westwood’s research indicates the awards will be significantly and systematically biased. Aiming to minimize conflict is short-sighted.

Perhaps the worst tragedy to come from ignoring differences is implied by a 2009 twin study by Peter Hatemi, Carolyn Funk, Sarah Medland, Hermine Maes, Judy Silberg, Nicholas Martin, and Lindon Eaves which found that people’s values are less likely to align with their genetic predispositions while they remain in their parent’s homes. This does not indicate intentional discriminationparents are unaware of evaluative differencesyet even accidentally preventing one’s child from aligning with his/her genetic predispositions could diminish his/her mental well-being. What parent would want to remain ignorant of differences, if accepting those differences could save their child from wishing he/she were dead?

Again, the truth is so harsh that one can understand why Haidt might want to soften the blow. Would you believe a psychologist who told you that our failure to understand differences has made normal parenting is so oppressive that getting away from parents faster could save children from wanting to commit suicide?

Moral Diversity vs. Evaluative Diversity

Aside from his 2001 study, Haidt’s most important experiment may have been the development of the Moral Foundations Questionnaire (MFQ) which measures peoples beliefs that morality is about each of the following six values: Liberty/oppression, Fairness/cheating, Care/harm, Loyalty/betrayal, Authority/subversion, and Sanctity/degradation.

This research created a stir because moral psychology was previously dominated by the theory that there is one best moral type. As it turns out, people who rate themselves as politically conservative tend to consider all six values in their definition of “morality,” whereas people who rate themselves politically liberal tend to emphasize Care/harm and discount the last three values, and people who rate themselves as libertarians tend to emphasize Liberty/oppression and discount the last four values. Thus, the MFQ demonstrates that political types are moral types. Since it is unacceptable to conclude that one political type is better than the others, the dominant theory moral psychology was overturned.

In chapter 8, Haidt admits that his list of values might not be complete; in fact, one of the six values was not on the original list, so it has already been revised once. Given what we know about GRIN types, one might think the next revision should be to add “Originality/orthodoxy” and “Effectiveness/ inefficiency.” While some people do value original ideas and effective strategies, it is debatable whether the value qualify as “moral.” For example, the debate over whether the ends justify the means may be seen as a debate over whether Effectiveness is a moral value.

As part of his roundabout story-telling, Haidt saves his own definitions of morality and moral capital until the last two chapters:

Moral capital refers to the degree to which a community possesses interlocking sets of value, virtues, norms, practices, identities, institutions, and technologies that mesh well with evolved psychological mechanisms and thereby enable the community to suppress or regulate selfishness and make cooperation possible.

The values of Originality and Effectiveness do not necessarily suppress selfishness, so they would not qualify as “moral” values by this definition. They would probably qualify, however, under Ayn Rand’s definition of “moral.” Does Haidt have a scientific basis for dismissing Rand’s perspective? Haidt admits that Loyalty, Authority, and Sanctity do not qualify as “moral” by liberal definitionsdoes he have a scientific basis for dismissing the liberal perspective as well? To the contrary, Haidt concludes that the diverse perspectives are interdependent, so he is painted into a corner.

Haidt describes himself as a liberal who wants to understand conservatives on their own terms, so it makes sense that he would accept a conservative definition of “morality,” and it makes sense that this definition would produce a survey instrument that focuses on conservative values. Reaching across the isle is noble. However, a partisan definition is still a partisan definition, even if entertained by a psychologist from the opposing party.

The advantage of the term “evaluative diversity”  over “moral diversity” is to escape the non-scientific bias that will necessarily result from having to define “moral”. All values are evaluative, whether they are moral or not. Thus, evaluative diversity includes Liberty/oppression, Fairness/cheating, Care/harm, Loyalty/betrayal, Authority/subversion, Sanctity/degradation, Originality/orthodoxy, and Effectiveness/inefficiency (and perhaps more).

Unfortunately, there is no field of “evaluative psychology.” The field Haidt inherited and now leads is called “moral psychology”and that isn’t his faultso he finds himself asking people “Is it [morally] wrong for a brother and sister to have sex?” Depending on their own definitions of “morality” (or whether they even bother to have one), some people may find such questions nutty. I’m not God—why ask me? However, Haidt has already revolutionized his field. Asking him to strike the word “moral” from its name might be asking too much.