What I'm suggesting is that a thing's identity is not (directly) determined by its pre-linguistic properties but by its linguistic categorisation. — Michael
I don't know what you mean by, 'a thing's identity.' Do you mean, whether it is a planet or not? But this is just not so; things were planets before anyone called them anything, and in those cases their physical properties were relevant and not to the extent we talked about them at all. In fact, we can counterfactually say, even if there were no language at all, there would still be planets.
This again seems to commit you, also, to saying that we can make things planets by calling them by a certain word, which is manifestly false (and again, your objection does not work, as I've outlined above).
Consider: suppose that to be water, something only needs to be called 'water,' and must be. Now, suppose I'm an alchemist, trying to transmute wood into water. According to your account, merely changing its physical properties is not enough: I cannot change wood into water, for example, by rearranging its molecular structure. Rather, the fact that I've done this will only be efficacious to the extent I've made something that people then go on to call 'water.' But this is absurd; to complete the transmutation, I only change the wood's physical qualities, to make it water. No one needs to then
call it water to
make it water. It was
already water; we call it water because that's what it is, it is not water because we call it that. Likewise, there are plenty of planets out there that no one ever has, or ever will, call anything. Yet planets they remain, in virtue of their physical qualities, and
not because of the tendency of these qualities to make us call them by a certain word. To put it simply, nouns denote properties; they do not denote the property of
being called by the very word. A little thought will show you that this is circular and impossible to institute in practice (but we can go over that too if you want).
This is obvious in the case of proper nouns. I was once Yahadreas and now I'm Michael (so not Yahadreas). Muhammad Ali was once Cassius Clay. I'm simply extending this principle to common nouns (a real-life example of this is the pre-op transexual who newly identifies as a different gender; "I was a woman but now I'm a man"). — Michael
Proper nouns are semantically distinct from common nouns. The former are referential expressions that denote individuals; the latter are predicative expressions that denote properties.
There is a way of talking about name-bearing properties and using proper names as common nouns; we can say there are two Michaels, for example, by which we mean there are two people named 'Michael.' And then, yes, I agree, there is nothing to being a Michael other than being called 'Michael.' That is because name-bearing properties are metalinguistic, but other properties, like being a planet, are not. And there are principled reasons for this, I think, but I don't want to get into it (I work on name-bearing and proper names, which is a fascinating subject in its own right). In many languages, proper names used as common nouns are marked differently from ordinary common nouns, showing quotative marking that highlights their metalinguistic character. This is because, I hold, they are derivative of the corresponding referential expression, much as, say, the common noun 'she,' meaning 'feminine individual,' is derivative of the pronoun.