The Righteous Mind
Dec. 26th, 2012 06:15 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
selenite0Jonathan Haidt has spent his career trying to develop a scientific explantion of human moral drives. An early version of his "Moral Foundations Theory" annoyed me enough to get over 300 words into an unpublished rebuttal. Enough other people actually shared their reactions to force him into revising his theory to account for patterns of morality he hadn't encountered before (i.e., conservatives and libertarians).
The updated version is described in his new book The Righteous Mind and I have to confess he has thoroughly addressed my complaints. Haidt describes moral codes as made of combinations of six foundations the way a flavor is made of the five tastes. The moral "taste buds" are set on spectrums of Care/Harm, Authority/Submission, Sanctity/Degradation, Fairness/Cheating, Liberty/Oppression, and Loyalty/Betrayal. Humans use morality as a tool to bind themselves into groups for pursuit of a common goal: simple survival for a family or forager band, building something for a project team, or an ideology for a political party. Haidt discusses examples of liberal, conservative, and libertarian moral codes.
The best way to check a new scientific theory is to find a dataset not used during its creation and test the theory against that. If Haidt's theory is correct people have been responding to these foundations for thousands of years as they created codes for their cultures. Some of those sets of moral principles should closely match his models. I grew up with the "Seven Deadly Sins" as a guide to what behavior was to be most avoided (the 10 commandments didn't come up as much, not being a sculptor). So if Haidt's foundations match them I'd consider it useful evidence that he's come up with a good working model.
Some of the deadly sins match exactly to their corresponding foundations. "Care/Harm" is violated by someone feeling Wrath. The sin of Pride indicates someone is not accepting his proper place in the "Authority/Submission" hierarchy. The "Sanctity/Degradation" foundation concerns respect for sacred things (crucifixes, Korans, the words of MLKjr) and the human body, particularly sexual violations. So Lust was a much more common source of violations in medieval times than modern provocations such as flinging dung at an image of the Madonna. The "Fairness/Cheating" foundation dictates that rewards should be proportional to the effort people put in. Someone violating that is committing Sloth.
Identifying a sin for the "Liberty/Oppression" foundation seems hard until you realize Haidt isn't defining "oppression" in the libertarian sense of having your autonomy violated but in the liberal one of having a smaller share of resources than other members of society. That clearly fits with the sins of Greed and Gluttony.
This leaves one member in each set: "Loyalty/Betrayal" and Envy. They're not a close match. Loyalty/betrayal refers to behavior towards one's group while Envy is an emotion aimed at an individual. But envy is "antigroupish" in making individuals work against other members of their own group.
Overall I think Haidt's MFT is a good model of human moral instincts. Different moral codes place varying weights on the individual foundations. Traditional/conservative moralities place roughly equal weights on all six foundations. Liberals focus on care/harm, liberty/oppression, and fairness/cheating (some are actually opposed to the sanctity, loyalty, and authority foundations). Libertarians put an extremely high weight on the liberty/oppression foundation, focusing on personal autonomy rather than equal division, with a lesser emphasis on fairness/cheating and a low one on care/harm.
Another bit of folk wisdom Haidt confirms is "man is not a rational animal, he is a rationalizing one." Haidt discusses the history of analyzing whether conscious reasoning has more effect on our decisions than instinct and did considerable research showing instinct wins whenever there's a conflict. His preferred metaphor is of the rational mind as the rider on an elephant. If the elephant doesn't know where to go the rider can direct it to an option, but if the elephant leans even slightly the rider is busy coming up with explanations for why that must be the correct direction.
Group behavior in Haidt's model involves people subordinating their personal identities to a group one. It can appear in many forms—families, forager bands, church groups, work teams, and even fans of sporting events. Haidt describes a college football game in detail as something designed to produce the feeling of “groupishness” that people seek. Some recreational pharmaceuticals such as peyote and ecstasy have similar effects.
To test Haidt’s theory of groups I compared it to the analysis in the software development book Peopleware. Programmers are usually thought of as loners but Lister and DeMarco describe how cohesive teams produce better in both speed and quality. The authors confessed that they could not come up with any advice for how to help programmers form teams. Instead they wrote the chapter “Teamicide” describing corporate practices that inhibit or disrupt team formation. These all speak directly to Haidt’s theory. First is separating the team members either physically or by splitting their time among different projects. Physical proximity—usually as a dense group—is a key ingredient in forming Haidt’s groups. Other practices correspond to the moral foundations. “Bureaucracy,” “Quality reduction of the project,” “Phony deadlines,” and “Those Damn Posters and Plaques” all degrade the significance of the team’s central focus, the software project they’re working on. Team members need to respect the sanctity of the goal to believe in their team. Likewise “Defensive Management” and “Clique Control” subvert the group’s respect for their supervisors and remove the supporting moral authority of the team leadership. “The Side Effects of Overtime” becomes a straight fairness/cheating issue. Lister and DeMarco’s statement that humans naturally form close-knit teams if not prevented from doing so is another confirmation of Haidt’s model.
The foundations most violated by teamicide are also the ones most likely to be ignored by liberal morality. This matches with the observed preference of the software industry for liberal causes. It seems that people who practice liberal morality still have their behavior affected by foundations they don’t usually respect.
Haidt isn’t shy about his liberal beliefs. His research was partly driven by trying to understand how Republican politicians connected better with voters than so many Democratic ones did. Kerry frustrated Haidt by by his refusal to use all the moral foundations in his messages. Obama did a much better job of fulfilling Haidt’s hopes. The 2012 campaign particularly used the “loyalty/betrayal” foundation in appealing to various groups to get out and vote. Romney’s attempt to appeal to Americans as a whole was less successful.
I suppose some people would find The Righteous Mind to be totally obvious. As someone who’s never had a good grasp of how human minds work it was full of revelations for me. There’s certainly a lot of counter-productive activity in modern life that could be avoided if people had a better idea of how each other would react. The fundamental problem is disagreement over goals. Haidt’s analysis won’t help us come to a consensus on that. It does give me the hope that if we can acknowledge that the other sides in our debates have honest motives for their desired outcomes we’ll be more civil in trying to resolve them. There are compromises we can make if we’re willing to--and if the decision-makers are willing to give up the personal power involved.
The updated version is described in his new book The Righteous Mind and I have to confess he has thoroughly addressed my complaints. Haidt describes moral codes as made of combinations of six foundations the way a flavor is made of the five tastes. The moral "taste buds" are set on spectrums of Care/Harm, Authority/Submission, Sanctity/Degradation, Fairness/Cheating, Liberty/Oppression, and Loyalty/Betrayal. Humans use morality as a tool to bind themselves into groups for pursuit of a common goal: simple survival for a family or forager band, building something for a project team, or an ideology for a political party. Haidt discusses examples of liberal, conservative, and libertarian moral codes.
The best way to check a new scientific theory is to find a dataset not used during its creation and test the theory against that. If Haidt's theory is correct people have been responding to these foundations for thousands of years as they created codes for their cultures. Some of those sets of moral principles should closely match his models. I grew up with the "Seven Deadly Sins" as a guide to what behavior was to be most avoided (the 10 commandments didn't come up as much, not being a sculptor). So if Haidt's foundations match them I'd consider it useful evidence that he's come up with a good working model.
Some of the deadly sins match exactly to their corresponding foundations. "Care/Harm" is violated by someone feeling Wrath. The sin of Pride indicates someone is not accepting his proper place in the "Authority/Submission" hierarchy. The "Sanctity/Degradation" foundation concerns respect for sacred things (crucifixes, Korans, the words of MLKjr) and the human body, particularly sexual violations. So Lust was a much more common source of violations in medieval times than modern provocations such as flinging dung at an image of the Madonna. The "Fairness/Cheating" foundation dictates that rewards should be proportional to the effort people put in. Someone violating that is committing Sloth.
Identifying a sin for the "Liberty/Oppression" foundation seems hard until you realize Haidt isn't defining "oppression" in the libertarian sense of having your autonomy violated but in the liberal one of having a smaller share of resources than other members of society. That clearly fits with the sins of Greed and Gluttony.
This leaves one member in each set: "Loyalty/Betrayal" and Envy. They're not a close match. Loyalty/betrayal refers to behavior towards one's group while Envy is an emotion aimed at an individual. But envy is "antigroupish" in making individuals work against other members of their own group.
Overall I think Haidt's MFT is a good model of human moral instincts. Different moral codes place varying weights on the individual foundations. Traditional/conservative moralities place roughly equal weights on all six foundations. Liberals focus on care/harm, liberty/oppression, and fairness/cheating (some are actually opposed to the sanctity, loyalty, and authority foundations). Libertarians put an extremely high weight on the liberty/oppression foundation, focusing on personal autonomy rather than equal division, with a lesser emphasis on fairness/cheating and a low one on care/harm.
Another bit of folk wisdom Haidt confirms is "man is not a rational animal, he is a rationalizing one." Haidt discusses the history of analyzing whether conscious reasoning has more effect on our decisions than instinct and did considerable research showing instinct wins whenever there's a conflict. His preferred metaphor is of the rational mind as the rider on an elephant. If the elephant doesn't know where to go the rider can direct it to an option, but if the elephant leans even slightly the rider is busy coming up with explanations for why that must be the correct direction.
Group behavior in Haidt's model involves people subordinating their personal identities to a group one. It can appear in many forms—families, forager bands, church groups, work teams, and even fans of sporting events. Haidt describes a college football game in detail as something designed to produce the feeling of “groupishness” that people seek. Some recreational pharmaceuticals such as peyote and ecstasy have similar effects.
To test Haidt’s theory of groups I compared it to the analysis in the software development book Peopleware. Programmers are usually thought of as loners but Lister and DeMarco describe how cohesive teams produce better in both speed and quality. The authors confessed that they could not come up with any advice for how to help programmers form teams. Instead they wrote the chapter “Teamicide” describing corporate practices that inhibit or disrupt team formation. These all speak directly to Haidt’s theory. First is separating the team members either physically or by splitting their time among different projects. Physical proximity—usually as a dense group—is a key ingredient in forming Haidt’s groups. Other practices correspond to the moral foundations. “Bureaucracy,” “Quality reduction of the project,” “Phony deadlines,” and “Those Damn Posters and Plaques” all degrade the significance of the team’s central focus, the software project they’re working on. Team members need to respect the sanctity of the goal to believe in their team. Likewise “Defensive Management” and “Clique Control” subvert the group’s respect for their supervisors and remove the supporting moral authority of the team leadership. “The Side Effects of Overtime” becomes a straight fairness/cheating issue. Lister and DeMarco’s statement that humans naturally form close-knit teams if not prevented from doing so is another confirmation of Haidt’s model.
The foundations most violated by teamicide are also the ones most likely to be ignored by liberal morality. This matches with the observed preference of the software industry for liberal causes. It seems that people who practice liberal morality still have their behavior affected by foundations they don’t usually respect.
Haidt isn’t shy about his liberal beliefs. His research was partly driven by trying to understand how Republican politicians connected better with voters than so many Democratic ones did. Kerry frustrated Haidt by by his refusal to use all the moral foundations in his messages. Obama did a much better job of fulfilling Haidt’s hopes. The 2012 campaign particularly used the “loyalty/betrayal” foundation in appealing to various groups to get out and vote. Romney’s attempt to appeal to Americans as a whole was less successful.
I suppose some people would find The Righteous Mind to be totally obvious. As someone who’s never had a good grasp of how human minds work it was full of revelations for me. There’s certainly a lot of counter-productive activity in modern life that could be avoided if people had a better idea of how each other would react. The fundamental problem is disagreement over goals. Haidt’s analysis won’t help us come to a consensus on that. It does give me the hope that if we can acknowledge that the other sides in our debates have honest motives for their desired outcomes we’ll be more civil in trying to resolve them. There are compromises we can make if we’re willing to--and if the decision-makers are willing to give up the personal power involved.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2012-12-27 04:31 pm (UTC)This isn't far off from the list C. S. Lewis used in the Appendix to THE ABOLITION OF MAN (UK title MEN WITHOUT CHESTS).
no subject
Date: 2012-12-28 05:57 am (UTC)no subject
Date: 2012-12-28 06:47 am (UTC)no subject
Date: 2012-12-28 05:31 pm (UTC)no subject
Date: 2012-12-27 11:42 pm (UTC)My gut feeling from reading your summary and the Wikpedia entry on Haidt is that either "Care/Harm" is too narrow or he has a more expansive interpretation of that spectrum than a colloquial interpretation of those words would indicate. Since he is trying to describe a basis for group standards of morality (expressed in individuals), I'd go with "Altruism/Selfishness" instead of/in addition to "Care/Harm". "Care/Harm" would seem to be focused outside your group, where as all the other pairings are internal. Even then, "Care/Harm" doesn't seem to have a strong basis in human psychology.
Speaking of other groups, the other thing that isn't obvious from these pairings is if he is trying to deal with moral relations between different groups. I see nothing (other than "Care/Harm") that relates to the morality of treating with others/outsiders. "Trusting/Suspicious" or something akin might be needed to describe inter-group morality.
I guess I should go read it and see what he has to say.
Thanks!
no subject
Date: 2012-12-27 11:47 pm (UTC)no subject
Date: 2012-12-28 12:34 am (UTC)Offhand I'd consider "harmony" captured in the Authority and Loyalty foundations but I don't know the connotations well enough to know how well that actually fits.
The model certainly isn't perfect. The problem is capturing every subtle flavor (for example, separating out equality of results and resistance to privacy invasions) gives us 20-30 foundations which is too fuzzy for any useful analysis.
no subject
Date: 2012-12-28 05:11 am (UTC)I'll give him the benefit of the doubt, but I was thinking more along the line of Confusionism when I made the Oriental vs Occidental comment. Taoism and Buddism, too but Confusionism has the greatest emphasis on societal harmony. For me that culture doesn't feel like it derives from completely from Authority & Loyalty. They are indeed key values relating to Social Harmony, but for me they don't capture the emphasis on Order for its own sake in Confuscian moral thought. Your mileage may vary.
That's a lot stew from one carrot, so I really should read his book before I ladle on more opinion. Thanks again for pointing me at it.
no subject
Date: 2012-12-28 05:58 am (UTC)no subject
Date: 2012-12-28 05:22 am (UTC)Those are the general, over-arching standards: the other items are Do's and Don't's within them.
The Righteous Mind
Date: 2012-12-29 11:57 pm (UTC)no subject
Date: 2013-06-19 05:20 pm (UTC)You can horsewhip your Gascony archers, or torture your Picardy spears;
But don't try that game on the Saxon; you'll have the whole brood round your ears. (http://www.kipling.org.uk/poems_normansaxon.htm)
Clearly Kipling thought the Normans placed a low weight on the Care/Harm foundation.