Friday, May 04, 2007

Occammed

During Karen Bennett's talk at the UT Austin conference, she said something that I think is worth repeating, again and again. It isn't even that big a point. In discussing presentism (or actualism, I forget which she was discussing at the time) she said something to defuse/dodge an objection from Occam's razor. I think it was in the context of using objects as proxies for the tensed-objects/possibilia that one loses if one is a presentist/actualist. She said, roughly, that Occam's razor mandates that one avoid multiplying entities beyond necessity, not that one must make do with less. Occam's razor is applicable once you've determined what things are necessary and what things you're attempting to smuggle in. Before those are settled, the possibility of doing without one or another entity, e.g. by using some sort of proxy, shouldn't force one to abandon the stuff that has proxies. Small point, but it seems like it is worth keeping in mind. Some of the discussions I've had about Grice's arguments against exclusive "or" in natural language seem to come down to the other person saying, look we can explain away exclusive "or" using the inclusive one, so we don't need the exclusive one by Occam's razor. But, this assumes that we've already determined what meanings are necessary and can begin shaving off the excess.

7 comments:

Justin said...

I'm a little bit confused by this entry, because I don't know quite how you're supposed to tell what meanings (or things in general) are necessary.

Justin said...

Half of that was accidentally posted before the other half was typed.

Anyway, what confuses me is that I would think that even the biggest fans of Occam's razor admit a need to determine which entities are necessary before appealing to considerations of parsimony. You look at the phenomena that are to be explained, see what you can explain given different sets of entities, and only then do you opt for the simpler account.

Maybe I'm wrong about the Occamists' approach. However, if I'm not, I don't see what alternate view of which entities are necessary we might use.

Shawn said...

You would think so, but I'm not sure that thought always makes it to the surface. I imagine there is a fair amount of premature Occamming going on in philosophy, although I can't point to much in the way of examples.

The question of how to decide what things (or what subsets of things) are necessary is a good one. I don't know in general. In the specific case, you'd probably have to look at what the theory calls for and what the use requires. When talking about semantics for languages as wholes, I'm not sure what sorts of constraints come into play.

Justin said...

I guess my residual complaint is that the dictum reads like it's offering some principle that exposes uses of Occam's razor as unfounded, but then the actual moral is just "use good judgment." But it's rare that one party to a dispute will say "Good judgment! Why, I'd been totally neglecting such a thing."

For a more concrete treatment of the point, try Onstad "Deciding to Have Optimism," 2004.

Lindsay said...

Yeah. Premature occamming is something to avoid. It happened to me the other night. My girlfriend says that it happens to every guy. Do you think that only male philosophers who have to worry about premature occamming? I've read a lot of Elizabeth Anscombe, and I'm pretty sure she's never prematurely occammed.

Shawn said...

Justin,
Point taken. I suppose part of the point is that one should be using good judgment. Neither side of a dispute of that sort is likely to give up due to lack of judgment.

It seems like one should explain what the criterion of necessity is before making a claim based on Occam's razor. Disputes about good judgment are likely to come down to that. While this won't convince anyone who is on the other side of the fence, it should shift the focus of the discussion.

I was thinking of Occam's razor as a form of optimization, and you must optimize relative to constraints. Depending on the constraints, one might want the fewest semantic entities possible. One might also not want that. My current favorite example is the biconditional. It is definable in propositional logic very easily, so few logicians are going to want to take it as primitive. (Occam that away.) If you are doing automated theorem proving, using a defined biconditional expands the formulas exponentially, which is bad. Under those constraints, one is likely to want a primitive biconditional. (Don't Occam that away.) That's the example that's been in the back of my mind.

Shawn said...

Lindsay,
If you prematurely Occam, you will have apologies to give.