simont |
Mon 2005-07-25 12:51 |
That doesn't make the question inadequate, though. It just makes it a question which is only meaningful given some assumptions.
To take another analogy from maths: it's quite common to define something with a woolly definition, and prove a theorem that states that the woolly definition actually works out to be rigorous. For example (I'm not sure how strong your maths is), quotient spaces. You divide your source space (say, the integers) into equivalence classes (say, residue classes modulo p), and then you define the (say) addition operation on those classes by saying "the result of adding class A to class B is the equivalence class containing a+b, where a is an element of A and b of B". What's to stop this giving multiple different answers, if the class you get as your result depends on which elements you choose from A and B? Well, it turns out that this doesn't occur, because you can prove a theorem that says the answer is the same no matter which elements you choose - in other words, the apparently woolly definition is fine provided this theorem holds, which it does. It's true that this definition would be a bit useless if you picked a space, or a set of equivalence classes, for which the theorem failed to hold; So Don't Do That, Then.
Similarly, our definition of "identity" is conditional on some properties of the real world such as consciousnesses being indivisible and unforkable; but as long as those properties hold the definition is still valid and still useful. I agree that as soon as one of them stops holding, the definition of identity will begin to stretch or break; at best it will simply become non-transitive (if you and I are teleport-copies of the same original person, then perhaps we can both meaningfully say that we "are" that person, but we would have to agree that we are not one another), and at worst it will fragment into multiple concepts, and questions like "are you Dave Gorman" will tend to have a correct answer of "depends why you want to know". |
|