I'm saying that the things we refer to by the word "iron" have 26 protons — Michael
We refer to all things with 26 protons as "iron", so in your proposition {the things we refer to by the word "iron"} is tautologous with {all things with 26 protons} so by substitution, your proposition is
the things we refer to by the word "iron" all things with 26 protons have 26 protons. True, but trivially true. It doesn't tell us anything non tautological. The non-trivial fact in this is the we
choose proton number to group elements where we could have chosen otherwise, and had we chosen otherwise "iron" would not even exist. As such iron (the class of metals, not the word "iron") is a human institutional fact, we declared that there shall be a class, determined by proton number, even before we named it "iron", the class itself (the thing being named) did not exist prior to our declaring it to.
When I use the word "iron" I might be referring to members of a class, but I nonetheless am referring to the members of the class, not the class. — Michael
I don't see how, you'd say "some iron", or "an example of iron", or "that iron". You'd have to indicate in some way the actual member other than by the word "iron", the word alone is insufficient to identify the object entirely
because it identifies only a class of objects. You could say "this object" (pointing) has 26 protons, but that's just a matter of belief based of trust in the institution that told you it was iron and told you all iron has 26 protons - it's not as if you looked. You're trusting (quite rightly) the human institutions behind that classification and naming process. If, on some weird science day the object, on investigation, turns out to have 26-and-a-half protons, something we'd previously thought impossible, we have to make an institutional decision as to whether it was iron, or not. up to that point, it's just 'stuff'. Once made, "all iron has 26 protons" is no longer true, not by virtue of us changing to what we apply the word "iron", but by us changing what iron is as a class (it now includes substances with 26-and-a-half protons, substances which didn't previously even have a class name).
It was for this reason that Ramsey was inclined to say (a little tongue in cheek) that there
are no facts, only events. A fairly extreme position on the status of universals, but one I've a lot of sympathy with.
I think you can both be right, — Xtrix
The realism discussion is one side saying it's word-to-world, while the other insists it is world-to-word; and they are both correct. — Banno
I think if we accept classes as human inventions, then the 'correctness' of a word-to-world fit depends largely on how existent the properties upon which the classes are founded actually are.
Say I could divide my book collection by author, by title or by publication year. I decide on author and create the class {all books by Wittgenstein}. The class is my invention, but its still a world-to-word fit because all the books are authored by someone and those in my class are authored by Wittgenstein.
But say I choose to divide my book collection by 'feel' and I create the class {books that make me feel happy}. Now we're perhaps a little more leery of saying this is a world-to-word fit, after all, how would we know that a book's being in the class {books that make me happy} isn't itself a reason why that book might make me happy (priming effect). We might be moving to a word-to-world fit.
So essentially, I see the problem as one of deciding to what extent our divvying up of the world affects the way we perceive potential members of sets - the extent to which the existence of the sets themselves has any determining influence over the decision about what objects are members of it. This problem, I think, goes all the way from active inference in perception (within my wheelhouse), to all the 'spooky' stuff in QM (none of which I understand, other than to know it's not a simple matter to say "this electron is...")