Skip to content

Category and Concept

January 22, 2012

I’m currently reading “Reclaiming Cognition,” a collection of essays arguing that thought is “fully embodied” – interactive and entangled in the particularities of our biology and environment. It’s one of the most stimulating and challenging books I’ve read in some time. Challenging less because of its difficulty and more because it confronts some of my most basic assumptions of how to make a science of psychology.

Specifically, the essay “Reclaiming Concepts” by Eleanor Rosch argues that we need to abandon trying to understanding meaning through substitutibility or a set formalism of any kind. So immina go through some of her points and suggest why they may not be detrimental after all, especially if we think in terms of substitution not between lexemes but between linguistic constructions. And not in terms of definitive substitutibility, but in terms of a normed measure of the acceptability of the substitution. So here are a couple of Rosch’s complaints and my responses:

1) “Judgment of similarity: Less good examples of categories are judged more similar to good example than vice versa. This violates the way similarity is treated in logic, where similarity measures are symmetrical and reversible.”

For example, because a chair is a more prototypical example of furniture than a shelf, a shelf might be judged more similar to a chair than a chair is to a shelf. But Rosch is confusing the proposed measure of substitutibility with how acceptable is the substitution between “A is similar to B” and “B is similar to A.” A bit more formally, if S is that measure, a function from pairs of constructions to a real number between zero and one, then S(“A is similar to B”, “B is similar to A”) has nothing to do with whether or not S is symmetric.

2) Novelty and non-composability.

Novelty: We can come up with and create examples for new concepts on the fly. If I asked you to name good places to hide from the mafia, you’d be able to do it even if you had never thought about it before.

Non-composability: People will judge a guppy to be a bad example of a fish and a bad example of a pet, but a good example of a pet fish.

However, these two issues resolve each other. If we’ve witnessed pet fish before, than we don’t need to compose our judgments of them from our judgments of pets and fish. And in so far as we can composite our understanding of concepts, we can construct new ones. So we only need composability in the case of novelty of a concept. Has anyone done experiments to see if our judgments are composed when novel? I hope so. It seems worthwhile.

Alright! That’s all I have to say on the topic for now.  Thanks for reading. Please don’t hesitate to question or comment!

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: