Friday, November 16, 2007

Ockham's Razor's Self-Excising

Ockham's Razor is the principle that, when devising theories to explain stuff, theoretical entities should not be multiplied unnecessarily. It is used by some to justify atheism, and theism by others; but in fact it's useless. Consider how we would actually explain some observations: You see a cat walking behind a sofa, then just the sofa, and then a very similar cat emerges from behind the other side of the sofa. A natural way to proceed is to perceive one cat walking behind a sofa. You could imagine there were two cattish things in succession, but it's obviously more realistic to postulate the one cat.
......Was that an application of Ockham's Razor, or a trivial application of common sense? If it was Ockham's Razor, then why not postulate, instead of all such cats (and dogs and sofas and so forth) in the world, just the one demon, who has hypnotised us... oh yes, because that's obviously unrealistic! Similarly we could regard all the electrons and positrons in the world as a single electron going forwards and backwards in time, at least mathematically, but should we think of that as what actually happens, and believe in just the one electron? Ockham's Razor says we should (if it says anything); but what do you think?
......Where Ockham's Razor does seem clearly applicable (e.g. I don't explain the appearance of the cat from behind the sofa by postulating a cat and an invisible and intangible splodge of stuff quite unlike anything else) there's no need to apply anything other than a more general principle, such as that one's beliefs should be reasonable (a belief in such a splodge would clearly have no reason for being; it's only philosophers who would think of such a thing). So insofar as it's true it's unnecessary, and so it ought not (by it's own application) to be one of our principles.

3 comments:

Cliff said...

I think we shouldn't think of it as just one electron, but I am worried the only thing I am employing to make that choice is, "common sense". I think Ockham's Razor seems to look for something other than truth, and truth is what I am interested in so that is why I reject the razor. I just don't know how to deal with instances where our evidence is consistent with a large range of possible states the world might actually be in. Ockham's razor might help us there but as you point out, employing it seems to have an equal chance at giving us a theory that is as far away from truth as not employing it.

Jeff said...

It seem's to me that a lot hinges on how we define the term "unnecessarily." I think the cat example is actually a pretty good example of why ockham's razor is helpful. By positing two cats you unnecessarily multiplied entities. (I'll grant you that cat's aren't theoretical but actual entities, but I'm not sure that changes anything.)

As for the case of the time traveling electron, it seems to me that we have two choices:
A) believe the common sense account that there are gazillions of seperate positrons and electrons.
B) believe in one time traveling positron or electron.

Unless somebody could provide me evidence for the (B) I would say that the time-traveling ability is an unnecessary bit of positing. Accepting (A) doesn't require that I believe positrons and electrons can time travel and is therefore the more parsimonious explanation.

In practice, I think that Ockham's Razor often looks on the ground a lot like trivial applications of common sense. I think it's most clearly useful when deciding between two alternative explanations of the same phenemona:
if two competing theories provide the same explanatory value then we should choose the simpler theory. (Here simpler is the theory which asks us to make fewer logical jumps; the one that has fewer theoretical entities.)

language said...

this