« They All Had Jewish Names | Main | The War Goes On »

April 25, 2006


P. Froward

"He says that the meanings of words are ... wired into the brain..."

Is that *really* what he actually says? I agree, it's stark raving nonsense, but that's obvious to anybody over the age of six -- unless what Chompers really said is something other than that "cat" is hardwired to point to the thing sleeping on my couch. If that's the case, why is Knight saying something else?

And the following:

"Chomsky keeps changing his theories, and in absolutely fundamental ways. ...it is never in response to a new empirical observation or experimental finding. ...he explicitly states that he is against any such concept of science."

I'm no fan of Chomsky, personally or politically, but this stuff would be more compelling with some pointers to examples. Without examples, it's just one Marxist throwing his poop through the bars at another, and frankly I'd rather watch two drag queens pulling each other's hair. Be honest: Would you rather be stuck in an elevator with a drag queen, or a Marxist? Barring Norm Geras, the drag queen's going to be totally more fun to hang out with.

Mick H

Yes I'd normally agree, but I must admit to a weakness for Chris Knight, despite his Marxism - "Blood Relations" is such a marvellous book. And I've always had a feeling that despite Chomsky's well-deserved fame for overthrowing the old behaviourist line in linguistics, much of his current reputation is based on his name rather than his theories - ie, as Knight says, "more pope than Galileo".


The trouble is, Knight sounds a lot "more Pope than Galileo" than Chomsky does. Chomsky wants to keep science and ideology separate. Wasn't that exactly Galileo's point? We've been here before. Knight sounds like some old Soviet hack attacking Western "bourgeois liberal fascist" science while extolling the merits of Trofim Lysenko. I'd rather have my Chomsky-bashing done by an expert, such as linguistics professor Marc Miyake at his blog "Amritas" (he would agree the basic problem with the Chomster is: "He doesn’t make observations. He doesn’t test hypotheses. He doesn’t start with empirical facts."). But Marxism vs. Chomskyanism is just a Celebrity Death Match between two pseudo-sciences (and Chomsky has the moral advantage here, since AFAIK nobody's been murdered as a result of Universal Grammar).


It's reasonable for anyone interested in Chomsky's politics to also be interested in his linguistics, but it's very easy to get things wrong in various ways and I think this is what Knight does (although he is right that Chomsky has a rather pope-like position among some linguists).

He says that 'He doesn’t make observations. He doesn’t test hypotheses'. This is simply not true. Some of his work is a little light on data, but there are numerous places where he takes various observations (e.g. the fact 'he' and 'him' must be two different people in 'Does he expect to like him?' but may be the same person in 'Who does he expect to like him') and attempts to account for them, and there are plenty of examples of him revising his proposals because they don't handle some data. Knight also says ‘He doesn’t develop a theory and then stick with it, as Galileo did with his moving earth. Instead, Chomsky keeps changing his theories, and in absolutely fundamental ways. When he changes his mind, it is never in response to a new empirical observation or experimental finding’. Again this really isn’t right. Chomsky has assumed some form of transformational grammar for fifty years. It has changed but it remains a form of transformational grammar. (Some linguists would say that he has hung on to ideas that he should have rejected.) Sometimes he has changed his approach in response to conceptual considerations but there are plenty of examples of him changing his analyses in response to empirical observations. It is probably true that he hasn’t changed his views in response to experimental findings, but he would certainly agree that such findings could in principle require a change.

Knight also says that Chomsky ‘doesn’t work with scientists in neighbouring fields’. It is true that he hasn’t done a lot of this, but there are examples e.g.:

Hauser, M. D., Chomsky, N., and Fitch, W. T. (2002) The Faculty of Language: What Is It, Who Has It, and How Did It Evolve? Science, 298:1569--1579.

It is not really true that Chomsky thinks that ‘No form of political action can be justified by science’. He has talked in various places about the possibility of developing a scientific understanding of human nature that would have implications for political action although I suppose he thinks we don’t yet have such a scientific understanding.

It is very odd to suggest (if this is Knight’s suggestion) that Chomsky’s linguistics is ‘political through and through’ because he ‘defines language as not social’. There are good reasons, spelled out in many of Chomsky’s writings, for seeing language as a cognitive system. Of course, language use is largely a social phenomenon, but that is a different matter. I don’t believe there is anywhere in Chomsky’s writings where he suggests that ‘the meanings of words are not socially negotiated but wired into the brain in advance as features of the human genome’. Jerry Fodor has been known to say things like this, but not Chomsky. It would be quite extraordinary for Chomsky to say that the meanings of words are not socially negotiated given that a lot of political writings are concerned with how various words (‘imperialism’, ‘aggression’, etc) should be understood and hence are part of a social negotiation.


The last sentence of my comment should say 'lot of *his* political writings'.


" I've always had a feeling that despite Chomsky's well-deserved fame for overthrowing the old behaviourist line in linguistics ..."

But have you actually read Skinner on Verbal Behavior? I think it is time to resurrect it. Chomsky did a great diservice to psychology by his very political attack on Skinner.

Mick H

Well no, I'll admit I haven't read Skinner. I always assumed that was a battle not worth revisiting, but maybe I'm wrong.

Juan Golblado

Thanks very much for those two pointers to Chris Knight. Blood Relations has been on my back-burner list for a few years now, ever since I read The Evolution of Culture, edited by Dunbar, Knight and Camilla Power. The interview with Knight has moved me closer to actually reading it.

The comments to this entry are closed.