Re: Sortal Individual Thesis
Bill,
The psychological literature about similarity is what I was
thinking about when I wrote that note.
Two entities x and y are similar if there exists some predicate
P that is true of both. I agree that there are a lot of
psychological problems in determining whether P is true of x.
But identity does not depend on just one predicate; it requires
*every* predicate P that is true or false of x to be equally
true or false of y. The psychological problems of evaluating
every P are orders of magnitude more than evaluating one P.
Before Chris M. jumps on me for confusing psychology with
metaphysics, let me emphasize that I am not using psychology
to *define* metaphysical terms. I am just using psychology,
epistemology, or commonsense, as a guideline for helping me
decide which concepts are better candidates for primitives.
> For example, Nicola's sortals are said to carry (necessary or
>sufficient) identity criteria. Sortals are allowed to subsume other
>properties. Now, one of the things Nicola says is that if P is a
>sortal and subsumes Q, then Q inherits an identity criterion, namely
>that of P. So, if you have also defined Q to be such that it can
>have no identity criterion - then you've made a fundamental mistake,
>whatever your notion of identity. In fact - Nicola's approach doesn't
>even rely on the fact that you *know* what the identity criterion is
>for some sortal - only that you know there is one.
My major criticism was that sortal identity collapses to
ordinary identity when you have a type hierarchy. If x=y
according to the type Entity, then x=y according to any subtype.
That makes the notion of sortal identity vacuous.
John