Celsus

Has anyone read John Dupré's review in American Scientist here? It's quite harsh on Pinker. Thoughts?

Joel

Doubting Didymus

Hmm. It seems preety convincing to me, but I have not read pinker, so it is possible he is being mischaracterised. If, however, dupre is being accurate, then the criticisms he levels seem to be quite damaging.

In fact, the more I hear about pinker, the less I like what I see. Perhaps I should set aside Jared Diamond for a second and read the language instinct...

pz

It's a very good review, and not quite harsh enough as far as I'm concerned. I was bothered by this remark, though:
That sounds too much like one reason he dislikes the book is that he dislikes the conclusion.

Doubting Didymus

Not really. Having started with an opinion, he simply states that the book has little to change said opinion. Its not that he dislikes the conclusion, he simply observes that pinker does nothing to convince his opponents.

pz

That wasn't the bit I object to. It's the idea that "biological determinism is more of a threat to reason". I think biological determinism is complete nonsense, but if the evidence backed it up, the only threat to reason would be rejecting it.

Doubting Didymus

Oh, I see what you're saying. I read it differently, you see. I thought he was saying biological determinism was a threat to reason because the evidence is against it, rather than there being a 'danger' of it being found true.

copernicus

I once took a seminar from CMU's Brian MacWhinney on this subject--somewhat before the days when Pinker became a popular proponent of Chomskyan innateness theory. MacWhinney, a psychologist, pointed out that no one really believes the extremes--the tabula rasa theory or the theory that everything in language is innate. The debate isn't really over innateness but over what is innate.

Clearly, nobody inherits English, Chinese, or Turkish. On the other hand, there are incredibly specific linguistic universals that are difficult to explain in the absence of some kind of innateness theory. For example, if a language tends to have sentences in which the verb comes at the end of the clause (e.g. Japanese, Hindi, Turkish), then it is extremely likely to have postpositions, suffixes, noun modifiers and relative clauses that precede nouns, and so on. If a language tends to have sentences in with the verb comes at the beginning of the clause (Breton, Tagalog, classical Arabic), then it is extremely likely to have prepositions, prefixes, noun modifiers and relative clauses that follow nouns, and so on. That is, verb-first and verb-last languages tend to be mirror images of each other in terms of grammatical structure. Why is that? Is there some kind of biological programming that causes humans to expect languages to have these properties?

I agree with Pinker that there must be a great deal of linguistic architecture that is biologically governed. On the other hand, I agree with MacWhinney that the interesting question is not whether language is innate but what in language is innate.

Doubting Didymus

Copernicus;

Dupre seems to agree with you in his critque:

Vorkosigan

The evolutionary psychology position, and Pinker's, IS interactionist. So we have strawmen constructions of ev psych being review by people strongly opposed to what they think is some simple-minded form of "biological determinism."

Dupre on Pinker:
"That book offers an egregious example of the development of the argument that if we decide that a given behavior would have been good for our ancestors in the Stone Age, then we must conclude that we almost certainly have evolved a tendency to produce it."

That is not the argument that ev psych people are promoting. Can no one write about this topic without dancing with straw men?

For example, Pinker attributes opposition to genetically modified foods to innate and intuitive essentialism. This provides an excuse¡Xof which Pinker avails himself¡Xfor dismissing without any analysis, or enumeration even, the criticisms that have been made of these technologies.

Here Dupre makes the mirror image error, dismissing Pinker's scientific conclusions -- which are in all probability correct -- because he doesn't like Pinker's political conclusions. The two are separate and should be kept separate by both the reviewer and by Pinker. Both have failed here.

The review is, well, not really very good. He seems a little pissed at the Americans for driving a truck through the postmodern critiques of science, and accuses them of the horrible crime of being middle-brow (constructivism is much more alive in England than in the States), since there is no space for a more detailed refutation -- which would, in any case, consist of an extremely long-winded statement of why the Americans are middle-brow. Clearly he has his own ax to grind -- see the title of his book down below -- and sees Pinker as merely another stone to sharpen his ax on.

Vorkosigan

Doubting Didymus

Dupre acknowledged that from the start:

He's saying that both sides are more reasonable and 'centrist' if you will, than either of them admits.