Women, Fire, and Dangerous Things

The cognitive view of the mind, and language, but no mention of anthropologists like Whorf.

Also see http://www.annualreviews.org/doi/pdf/10.1146/annurev.ps.32.020181.000513 ' CATEGORIZATION OF NATURAL OBJECTS by Carolyn B. Mervis and Eleanor Rosch ( )

read for Cog Sci at UCB

I have to hand it to George Lakoff. Despite his Chomsky training at MIT, he was able to break out and consider how people really construct metaphors and how much of language is really metaphor. All his works and those of his cohorts are insightful and evdn exciting, although this particular work has a sexist bias as the title reveals. I guess that was George's little joke ( )

Lakoff shows a knack in this book for combining dry, unimaginative, academic writing with an almost obscene lack of clarity.

Here's an excerpt of the sort of prose to which you'll be subjected while braving this tome: "But, most important, the EFFECTS = STRUCTURE and PROTOTYPE = REPRESENTATION INTERPRETATIONS are wildly inaccurate ways of understanding prototype and basic-level effects." (p. 142). You might think that in context this sentence would be decipherable. It isn't. Part of what Lakoff is getting at here is that semantic vagueness is not co-terminous with prototype effects. It would be nice if he would just say that.

And for having written a book so enthusiastic about the cognitive status of metaphor, its author sure has a hard time deploying the technique to any great literary or intellectual effect. To choose a particularly egregious example, in discussing dualism in Chapter 19, Lakoff refers to the dualist position as the "mind-as-machine paradigm." You read that right, and it is exactly as ass-backward as it sounds. People who are attracted to dualism are attracted to it precisely because it promises a means of escape from conceptualizing the mind as a machine! And in what sense can a machine be "disembodied," as Lakoff so fervently insists that the "mind-as-machine"-people believe the mind is? If you think the mind is disembodied, then the mind can't be a machine, because machines are embodied! Is this difficult?

Women, Fire, and Dangerous Things is emphatically not a rewarding read, but some of the ideas contained in it are worth being made aware of. The idea of "motivated" connections as distinguished from generative/predictable connections on the one hand, and arbitrary connections on the other seems to offer a valuable perspective on the categories demarcated by human language. Likewise, the idea that "basic-level" concepts arise out of human kinesthetics and neurophysiology, as opposed to divisions in nature itself, seems spot-on. The goodness of the ideas just makes you wish that they had been more artfully and clearly expressed.

As a sort of bonus, the book contains an excellent chapter covering the origins and history of mathematical formalism, and a relatively cogent explanation of Hilary Putnam's proof that "meaning" can't arise out of a mere isomorphism between inherently meaningless symbols and things in the world. I was unfamiliar with Putnam's result, and actually found it rather bracing.

On the other hand, the book is littered with typographical errors, and contains the ugliest typesetting ever to have escaped from a university press.