Moral Minds
Richard Rorty’s review of Marc Hauser’s Moral Minds is pretty good. Hauser argues for a fairly strong moral nativism, involving a dedicated moral capacity, analogous to a Chomsky-style linguistic capacity. (Rawls floats the idea in Theory of Justice.) Rorty’s pretty widely known, but not a lot of non-philosophers know he was a top-flight philosopher of mind back when he was a philosopher (one of the first eliminative materialists), and he makes a pretty good case against the Chomsky analogy:
Hauser thinks that Noam Chomsky has shown that in at least one area — learning how to produce grammatical sentences — the latter sort of circuitry [i.e., general purpose] will not do the job. We need, Hauser says, a “radical rethinking of our ideas on morality, which is based on the analogy to language.” But the analogy seems fragile. Chomsky has argued, powerfully if not conclusively, that simple trial-and-error imitation of adult speakers cannot explain the speed and confidence with which children learn to talk: some special, dedicated mechanism must be at work. But is a parallel argument available to Hauser? For one thing, moral codes are not assimilated with any special rapidity. For another, the grammaticality of a sentence is rarely a matter of doubt or controversy, whereas moral dilemmas pull us in opposite directions and leave us uncertain. (Is it O.K. to kill a perfectly healthy but morally despicable person if her harvested organs would save the lives of five admirable people who need transplants? Ten people? Dozens?)
According to Chomsky, the parameters of the universal linguistic capacity can be set in different ways to produce the grammars of the various natural languages. But any setting of the parameters produces grammaticality, and is fully on par linguistically speaking. No language is better qua language, or more authentically languagey. Now, it may be that Yanomamo warriors, queer-stoning Islamists and gay Dutch vegans are all living out various dialects of morality, but if so, then it turns out that morality is a pretty useless category. The liberal morality of sympathy, reciprocity, and fairness, isn’t just an equivalent way of deploying moral judgment and emotion. It’s better than the alternatives. That’s basically the problem I’ve had with moral psychology based on Chomsky, such as John Mikhail’s and Sue Dwyer’s [pdf]. Rorty sums it up nicely.
Now, I’m a fan of Jonathan Haidt’s social intuitionist theory according to which specific moral emotion and moral judgment is a function of different settings on several general dimensions of moral emotion. This is also a kind of parameters approach, but, unlike Chomsky-based theories, it is grounded in emotion rather than a kind of innate knowledge (or “cognizance” to use Chomsky’s dodge word.) But the same critique applies. Certain ways of calibrating the dimensions of moral emotions are evidently, and seemingly paradoxically, immoral. Obviously, if you’re going to say that, you’re assuming the authority of one calibration as a secure basis for passing judgment on the others. Isn’t that arbitrary? Well, I think one thing to say is that it is possible to determine, in evolutionary terms, what moral capacities are for. As the environment of human interaction changes through history, certain ways of calibrating the moral sense fail to function in the appropriate way. So while we can say that a certain calibration is “a morality,” in the sense that it a way of deploying the moral capacity, it is not authoritatively moral, in the sense that it violates the principles of a calibration that does serve the proper function of morality given the present social/institutional setting.
Now, I don’t actually think that’s quite right. Because it’s not clear why the proper biological function of the moral capacity ought to have normative force. But I think it’s a place to start when trying to think through the bindingness of morality in a non-spooky natural world.
For a different view, John Mikhail defends Hauser’s book (which I haven’t read yet, by the way) on the Georgetown Law blog.
September 24th, 2006 20:07
Well, if morality is innate, like linguistic capacity, perhaps Chomsky’s own strong linguistic capacity is somehow related to his deficient moral capacity.
September 25th, 2006 00:37
Non-naturalistic truthmakers for moral claims are not “spooky”. They’re just non-natural. So what? For the life of me, I cannot figure out what the problem is.
Let me ask three questions to you then, Will:
1) What is naturalism, on your view?
2) Why are you one?
3) Why are the alternatives “spooky”?
September 25th, 2006 04:12
Will,
Just curious. Are you completely unsympathetic to an error theory about morality? Your comment about the problem of explaining the bindingness of morality in a natural world made me wonder…
September 25th, 2006 08:56
Rhadamanthus,
1) Everything is in space and time.
2) Because the best theories of the world quantify over nothint not in space and time. As the poet Quine said, “Nothing happens in the world, not the flutter of an eyelid, not the flicker of a thought without some redistribution of microphysical states.”
3) It is spooky to be outside space and time.
September 25th, 2006 09:05
Luka, You know, I’m confused about whether I have an error theory or not. My view is starting to become a bit like Gilbert Harman’s, which is relativistic. I guess I think that people’s moral conceptions, as they understand them, are usually false, but that false conceptions can roughly track real, non-queer rightness and wrongness…
September 25th, 2006 11:32
Quine believed in sets! And thanks for the reply.
I think naturalism has at least nine problems:
1) Universals broadly construed
The contingency of natural laws (why are there any?)
2) Modal properities and knowledge of them
3) Moral properties and knowledge of them
4) Moral responsibility
5) Epistemic norms and their grounding
6) Qualia
7) Intentionality, specifically real-world representation or ‘aboutness’
9) Selves
I guess you think:
1) Nominalists have the semantics
Natural laws have no explanation or you’re agnostic 9) Parfit’s right.
2) Quinean semantic reductions of modality work
3) Non-natural moral properties are queer
4) Compatibilism is true
5) Naturalizing epistemology won’t remove epistemic norms
6) Qualia can be eliminated or reduced
7) Intentional properties can be reduced to informational or computational properties
Those are my guesses, having read your blog for some time.
How’d I do?
September 25th, 2006 13:34
Rhad,
Well, that covers a lot of philosophy, so I’m hesitant to reply in detail. But, briefly.
About what I think, you did pretty well:
1) Not sure what you have in mind, but probably.
I have a weird naturalized trope theory of laws.
2) A kind of of non-essentialist Wittgensteinian/Kripkean combinatorialism will do.
3) There are no non-natural properties, so yes.
4) Yes.
5) No. Naturalizaing epistemology does remove norms, if you think of them as transcendental or immanent. Epistemic norms, like other norms, are either objective in an instrumental sense (if you’re aiming at truth, then think this way) or conventional.
6) What needs to be reduced?
7) Some kind of complicated story about causal chains of designation.
9) Or the buddhists are right.
September 26th, 2006 04:54
Will, I think you can be a little more careful about drawing the analogy between linguistics and morality.
It is true that every setting of the linguistic parameters produces a working language. And it is true that most linguists believe that no language is “better” than any other. But the second statement does not follow from the first!
One could easily imagine a case in which different parameter settings generate languages that are wildly different in terms of communicative efficiency, lyricism, their efficacy as a tool to help men get women into bed, or whatever scale you want to measure them with.
Linguists have found through OBSERVATION that all languages are about the same in terms of communicative efficiency and complexity. Chomsky’s “Principles and Parameters” approach does NOT explain this!
One could try to explain this observational fact via evolutionary pressure, but that is not a view that Chomsky would endorse. He is, in fact, quite opposed to evolutionary just-so stories.
What does this mean for morality? It means you could easily have a generative, parameter-setting approach to morality, AND have some of the generated moralities be awful and others be great. The purpose of the parameters model would not be to judge moralities, but to predict the set of moralities that are possible for humans to hold. That, I think, would be a very worthy goal.
September 26th, 2006 08:05
Rorty vs. the Moral Mind…
Heterodox analytic philosopher Richard Rorty reviewed psychologist Marc Hauser’s The Moral Mind last Friday in the ”Herald Tribune”. Rorty’s review is quite a disdainful piece of Hauser-bashing; yet Rorty does not fail to provide arguments agai…
September 26th, 2006 08:06
Rorty vs. the Moral Mind (bis)…
Rorty wrote : “Nazi parents found it easy to turn their children into conscientious little monsters.” He does not describe how nazi parents taught their children to be cruel with Jews. How did they accomplish such training ? By brainwashing as Pavlov…
September 26th, 2006 10:03
And another thing!
You quote Rorty as claiming that the grammaticality of a sentence is rarely a matter of doubt, while moral dilemmas leave us uncertain. But he may not be comparing equivalent sets.
There are huge numbers of sentences whose grammaticality is in doubt. A frequently-studied instance is the so-called “parasitic gap”, such as “Which paper did you read before you filed?” “What kind of person do you like whenever you meet?” and similar. Some people hate them, some people see nothing wrong with them, others are left in some kind of post-koan Zen state. These grammaticality judgements may be analogous to difficult moral quandaries.
On the other hand, depending on how you define “moral dilemma”, there may be huge numbers of them that are easy to decide and hence are not usually counted as dilemmas. For example, should I smack my mother for no reason? The answer is obvious to anyone, as obvious as the grammaticality of “The dog is happy”. These easy moral issues, which we solve quickly and unconsciously every day, may be analogous to standard grammatical sentences, which we produce quickly and unconsciously every day.
Note, too, that these easy moral questions are frequently mastered by the age of five to seven years, the same age that marks the mastery of the basics of one’s native tongue. My seven-year-old knows that one never says “dog the happy is”, and that stealing is morally wrong, regardless of whether you get caught. These laws of grammaticality and morality are unknown to most of the animal kingdom, regardless of how long they live. So Rorty’s assertion that moral codes are not assimilated especially quickly is also on shaky ground.
September 26th, 2006 12:55
Will, That’s an interesting position. I hadn’t quite thought of it as a possibility, I don’t think. I find the problem of the bindingness of morality to be fascinating, though. I tend to be sympathetic to the Mackie-style error theory: all moral judgments are false because there are no moral facts. - Anyway, good post.
October 17th, 2006 06:13
“But I think it’s a place to start when trying to think through the bindingness of morality in a non-spooky natural world.”
Analytical philosophy in a nutshell:
1) Deny the primacy of consciousness in the world; and
2) Now notice how hard it is to explain everything, and exert great efforts to come up with ingenious way to avoid admitting 1) was an error.