the non-eliminativist must also argue for the ontological existence of relations - not an easy task — RussellA
Indeed, but let's have a go. First, let's distinguish
ontological existence from other merely everyday kinds of existence. Now 'ontological' means, roughly, 'pertaining to existence'. So we are looking for a category of existence that is related to existence. Well, perhaps that's not too hard, after all. I guess every category
of existence will qualify. If it is a category
of existence then it pertains to existence. In the same way, if we go looking for canine dogs, feline cats and primate humans we will not be disappointed. We can tuck them into our metaphorical hunting sack along with the ontological existence.
Next problem. We have to show that relations exist. We already know (from above) that,
if relations exist, then they have the special ontological kind of existence required - because everything that exists has that special kind of existence. But we don't yet know
whether relations exist.
Let's borrow an example from the opposing camp. Glasgow is west of Edinburgh - so we are told. We are further led to believe that 'being west of' is a 'relation'. Now, what would it mean for such a relation
to exist? It could mean that if we turn over the whole universe item by item we will find at least one thing that is to the west of another thing. On the other hand, if we get to the end of all the items in the universe and have not encountered a single thing that is to the west of any other thing then we can say that such a relation
does not exist. If 'a relation exists'
does not mean that, then I will need to confess I do not know what it
does mean. OK so far. That leaves us with the small problem of turning over every item in the universe. But hold on! We just picked an example of the
very thing we are looking for. There is a case - at least one case - of something being to the west of something else. From which it follows that there is at least one case of something being in relation to something else. From which if finally follows that relations exist. Putting this conclusion together with our earlier one, we can see that relations have ontological existence.
"But which of the two cities does the relation 'being west of' exist
in?" Well, relations do have
ontological existence. But they are not objects that we can lug around with us. We can turn over every object in Glasgow and Edinburgh and we will not find any such thing. That's not because it's very small or particularly elusive. It's because it's not a kind of thing. It's because - oh, gosh, see Ryle and anyone who has written about category mistakes for the last 70 years....
Need I go on? (No, Cuthbert. Not only need you not go on. You need not even have started.)