On Haidt & Relativism
Here are some issues it might be interesting to discuss either in the seminar or on the web site:
I. Should Haidt be a Moral Relativist (or
perhaps a Moral Eliminativist)?
In “The Emotional Dog & its Rational Tail,” Jon Haidt defends what he calls a social intuitionist model of moral judgment. He suggests that this view is of a piece with “Intuitionism in philosophy” which he characterizes as “the view that there are moral truths (it is a form of objectivism, not emotivism), and that when people grasp these truths they do so not by a process of ratiocination and reflection, but rather by a process more akin to perception, in which one ‘just sees without argument that they are and must be true.’” (ms, p. 2)
There are, however, some prima facie differences between perceptual intuitions & judgements on the one hand and moral intuitions & judgments on the other. These differences, I maintain, make it singularly implausible to view Haidt’s version of intuitionism as “a form of objectivism.” Rather, I think, Haidt is better viewed as advocating a version of relativism – though it is importantly different from Harman’s version. One might also make the case that Haidt’s view entails an “error theory” and thus supports moral eliminativism.
Some prima facie differences between moral and perceptual intuitions:
i) Moral intuitions seem to be culturally variable in ways that perceptual intuitions are not. Shweder offer lots of examples of Indian moral intuitions that are quite different from those common among Americans. And Haidt’s experiments show that high and low SES people have quite different intuitions about the wrongness of eating the family dog or masturbating using a chicken carcass that one then eats. There are no parallel differences between these groups in their perceptual intuitions.
A possible objection here turns on the alleged “theory-ladenness” of perception. I’m not sure how far the point could be pushed, however. It’s true that modern Americans make perceptual judgments that New Guinea tribesman cannot make (& vice versa): Eg. That’s a jet plane. or That’s a computer monitor. But it is less clear that there are many cases of disagreement in which people in one group judge that x is F and those in the other group judge that x is not F.
ii) According to Haidt, moral intuition (or crucial parts of the system subserving it) is largely acquired as the result of interactions with peers in late childhood and adolescence, and people who have very different experiences in this critical period end up with (dispositions to have) very different moral intuitions. Here’s the passage in which he summarizes this part of his view:
Putting together all of the developmental theories and findings presented above yields the following expansion of the social intuitionist model: Moral development is primarily a matter of the maturation and cultural shaping of intuitions. People can acquire explicit propositional knowledge about right and wrong in adulthood, but it is primarily through participation in custom complexes (Shweder, et al., 1998) involving sensory, motor, and other forms of implicit knowledge (Fiske, 1999; Lieberman, 2000; Shore, 1996), that are shared with one's peers during the sensitive period of late childhood and adolescence (Harris, 1995; Huttenlocher, 1994; Minoura, 1992) that one comes to feel, physically and emotionally (Damasio, 1994; Lakoff & Johnson, 1999), the self-evident truth of moral propositions. (ms, p. 35)
iii) Moral intuitions can be affected by personal relationships and by various sorts of “framing” in which the analogies between the case at hand and other sorts of cases is made more salient.
Now, of course, one might maintain that moral intuitions have little or nothing to do with the business of discovering moral truth. But that clearly is not Haidt’s view. One position that Haidt might find (and indeed, did find) attractive is the sort of relativism that Harman characterizes as follows:
For the purposes of assigning truth conditions, a judgment of the form, it would be morally wrong of P to D, has to be understood as elliptical for a judgment of the form, in relation to moral framework M, it would be morally wrong of P to D. Similarly for other moral judgments.
What Harman calls “a moral framework” might be identified with the cluster of moral-intuition-determining mental states and dispositions that, according to Haidt, one acquires in late childhood and adolescence.
B. Why Haidt might be an
eliminativist
One of the themes that is central in Haidt’s paper is that moral intuitions (or judgments) present themselves as being objectively and universally true. There are lots of passages like the following in his paper:
A moral world view must be so widely shared that it takes on the feeling of objective truth. It is for this reason that morally different minorities (e.g., Jews, communists, homosexuals, and even conservatives within liberal professions) have been purged, expelled, or even killed by cohesive moral communities. (ms, p. 14, italics added)
Rules prohibiting moral violations are judged, even by young children, to be universally applicable and unalterable. (ms, p. 6, italics added)
The stories were carefully constructed so that no plausible harm could be found, and most participants directly stated that nobody was hurt by the actions in question. Yet they still usually said the actions were wrong, and universally wrong. (ms, pp. 6-7. italics added)
Now, if it is really the case that presupposition about objectivity and universality are built into moral intuitions and judgments, one wonders whether one might be able to mount a Mackie-style “error-theory” argument whose conclusion is that all moral claims are false.
An analogy that both Harman & I have urged in the past is between moral intuitions and grammatical intuitions. Both, it is plausible to suppose, are subserved by a complex body of largely tacit information which is acquired during development and on which there may be strong innate constraints. The person who takes his moral intuitions to be universal and objective is like the person who takes his grammatical intuitions to be universal and objective – and who thus views people who speak different dialects or languages as mistaken about something.
II. Does Haidt’s work pose
problems for traditional views in moral philosophy?
A. Kant
At several points, Haidt suggested that his view about moral reasoning posed problems for Kantian and neo-Kantian views in ethics. Very roughly, the idea seems to be this: Kant requires that genuinely moral actions be motivated by judgements that result from a certain sort of reasoning process. However, if Haidt is right then perhaps no action is ever motivated in this way. Worse yet, perhaps humans are simply incapable of engaging in the sort of objective moral reasoning process that Kant requires.
I would be very interested in hearing some reactions to this from people who know more about Kant’s ethics than I do.
B. Ideal Observer, Qualified
Attitude and Secondary Quality theories
There is a family of theories in meta-ethics which share something like the following idea: Moral truths are the judgments that would be made by a suitably idealized judge or observer. (Or, on Brandt’s “qualified attitude” theory, moral truths describe situations about which judge would have a positive or negative “attitude”. When we say ‘x is good’ we are saying that I would have a positive attitude toward x if I were forming that attitude under appropriately idealized circumstances.) In telling the story about the attributes that the ideal observer must have, one can build in a variety of non-moral properties (the observer is fully informed about the relevant facts, in a calm state of mind, bears no special relation to any of the people involved, etc.) though it is cheating to require that the observer has any specific moral views.
Now I am inclined to think that the work Haidt reviews makes it singularly unlikely that all ideal observers would make the same judgments (or have the same attitudes). So, if moral truths are the judgments that all ideal observers would agree on, then there are no moral truths. (Brandt explicitly allows that his version of the theory may turn out to be relativistic in this way.) Though I am hardly an expert on the topic, it strikes me that recently trendy “secondary quality” accounts of moral properties are, at bottom, just variations on the ideal observer theme. As such, I think they are equally threatened by the sorts of evidence that Haidt reviews.
I trust there are people in the seminar who know much more about this family of theories than I do. It would be interesting to hear how they react to the claim that the literature Haidt reviews undermines the theories.