The question may seem an odd one. If read as asking “Do there exist any objects that do not exist?” the answer is obviously No because the question presupposes a contradiction. But there is another way of reading the question that can be gotten at if we reflect on the relation between actuals, possibles that are not actual, and impossibles. Here are some uncontroversial examples of each:
Actuals: dogs, cats, grass, the Eiffel Tower, the Sun, etc.
Non-Actual Possibles: unicorns, the twin towers of the World Trade Center (they were actual, but are no longer), dinosaurs, etc.
Impossibles: Square circles, married bachelors, four-sided triangles, etc.
Now, in one sense of the word, to say something “exists” is to say that it is actual or real. But that can’t be the sense implied when we say that something “is” possible but non-actual or that something “is” impossible because both of those categories exclude actuality. So we have to recognize at least one additional sense of “exists” besides “is actual”. What could that sense be?
We can begin by noticing that the class of non-actual possibles and the class of impossibles are both non-empty. This fits nicely with Frege’s claim that “affirmation of existence is … denial of the number nought.” Following Peter Geach, I’ll call this the “there is” sense of “exists”. This is the sense being used in a sentence like “There is a possible world in which JFK dies a natural death.” Obviously, this is not saying that such a world is actual, but only that the class of possible worlds includes at least one in which JFK dies a natural death. In other words, the class of possible worlds in which JFK dies a natural death is non-empty.
With this distinction between the “actuality” and the “there is” senses of “exists” we can now read the opening question as follows: “Is the class of non-actual objects non-empty?” In light of the examples of impossibles and non-actual possibles above, the answer to this question would appear to be Yes.
This raises an interesting question: What is the ontological status of non-actual (or “mere”) possibles and of impossibles? On the one hand, we don’t want to reify them à la Meinong into quasi-actual entities that have a sort of independent existence [Addendum: Bill Vallicella informs me that this is not an accurate statement of Meinong’s view.]. On the other hand, we don’t want to dismiss them as irrelevant. Practically speaking, it makes a difference whether the things we attempt are really possible or not. My offhand thought is that to avoid both extremes we need to say that possibility and impossibility are parasitic on actuality–something is really possible iff what is actual does not preclude it and really impossible iff what is actual does preclude it.
Do you think it would be helpful to make a further distinction between logical possibilities and physical (or empirical) possibilities? Or is that just a rewording of what you had in mind?
You are arguing in effect for a double sense of the word ‘is’: the ‘there is’ sense and the ‘existence’ sense. I dispute this. If I say ‘no bridge crosses the river at Hammersmith’, I am contradicting someone who says any of the following, taken in their ordinary or everyday sense.
‘a bridge crosses the river at Hammersmith’
‘there is a bridge that crosses the river at Hammersmith’
‘a bridge crossing the river at Hammersmith exists’
‘there exists a bridge which crosses the river at Hammersmith’ &c
It may be there is an extraordinary sense of such expressions, but you give no compelling argument for this. You say ‘to say something “exists” is to say that it is actual or real’. But this takes us down a slippery slope. It suggests we are saying OF something that it is actual or real, of which we might also truly have said it was not real. But when I say there is no bridge crossing the river at Hammersmith, I am not saying of anything that it is not real. I am saying that there is no such thing. Once again, it is the Latin philosophers we have to blame for this, with their introduction of technical terms like ‘a parte rei’, ‘realiter’, ‘realis’ and so on. All of these are derived from ‘res’ or ‘thing’. But the ordinary sense of ‘thing’ has no metaphysical connotations. To say that such a bridge is real, is just to say that there is such a thing, that of all the things there are, Hammersmith bridge is one of them. To say it is not real, means there is no such thing.
You say ‘We can begin by noticing that the class of non-actual possibles and the class of impossibles are both non-empty.’ That seems obviously untrue & question-begging. The class of bridges crossing the river at Hammersmith is obviously empty, for there is no such bridge. The class of round squares is empty, for no square is round (= there is no round square, no round square exists, nothing is round and square &c &c).
You say that unicorns are ‘non-actual possibles’. I’m not sure what that means. Does it mean that it is possible that there are unicorns, but actually there aren’t? How does this imply that anything is a unicorn. You can read ‘non-actual possible’ as a sort of adjective, thus ‘unicorns are non-actual possibles’ implies that some things (unicorns) have the property of being non-actual but possible. But this begs the question. There aren’t any things that are non-actual, in the sense required. ‘F’s are non-actual’ just means that there aren’t any F’s, not that there are, but they are ‘non-actual’.
By the way, I was a pupil of C.J.F. Williams, who famously asked whether, if told that blue buttercups did not exist, he would ‘have felt obliged to examine several specimens of blue buttercup before concluding that none of them exist, that as a variety blue buttercup lacks existence’.
Alan, I must now apologise, for I haven’t addressed your main argument.
>>
In one sense of the word, to say something “exists” is to say that it is actual or real. But that can’t be the sense implied when we say that something “is” possible but non-actual or that something “is” impossible because both of those categories exclude actuality. So we have to recognize at least one additional sense of “exists” besides “is actual”. What could that sense be?
< < The argument is that if some things are possible but not actual (unicorns, say) then *there are* some things that are possible but not actual. And clearly we have to concede an ‘ordinary’ sense in which some things are actual, some things are possible but not actual, and some things just aren’t possible at all. But does this imply the two senses of ‘there are’? We have (A) Some things are possible, though there aren’t such things
(B) There are some things which are possible, though there aren’t such things
(B) seems paradoxical, because there is no way of reading ‘there are’ in the different senses required. But if (A) is true (& I concede it seems true), we have to explain how it is true. I don’t have an answer to that.
PS I mentioned this post in my own weblog here:
http://ocham.blogspot.com/
There is some way of linking posts but I couldn’t work out how to do it.
David,
Yeah, I think there is an important distinction between logical and physical possibilities, but I would propose that both are delimited by actualities. Physical possibilities are delimited by the laws of nature–no laws of nature, no physical possibilities or impossibilities. Logical possibilities are delimited by the fundamental laws of logic, esp. non-contradiction–no laws of logical, no logical possibilities or impossibilities.
Hi Ocham,
I see you’ve started a new blog as well. “Beyond Necessity”–I like the name. I’ve added it to my blogroll.
With regard to your comments, I’m not surprised that a student of Williams would take issue with what I’m saying. From what I’ve learned in the week or so that I’ve been reading on the topic of “existence”, he was a stauch proponent of the dominant Frege-Russell-Quine-etc. doctrine that the only existential use of “is” or “exists” is as a second-order predicate (a predicate of predicates). With all due respect to Williams (RIP), I think he (and Frege and Russell …) are wrong on this point.
Anyway, you suggest that my distinction between the “is actual” and the “there is” senses of “exists” creates problems. You say, “It suggests we are saying OF something that it is actual or real, of which we might also truly have said it was not real.” And this, so the objection goes, leads to absurdities like predicating ‘non-actuality’ of non-existent things. In response I would say that actuality being predicable of things does not mean that non-actuality is predicable of things. Instead of saying “Blue buttercups are non-actual” I would say “It is not the case that blue buttercups are actual”, and I would deny that the latter can be transformed into the former salve significatione.
You also write, “The argument is that if some things are possible but not actual (unicorns, say) then *there are* some things that are possible but not actual. And clearly we have to concede an ‘ordinary’ sense in which some things are actual, some things are possible but not actual, and some things just aren’t possible at all. But does this imply the two senses of ‘there are’? We have
(A) Some things are possible, though there aren’t such things
(B) There are some things which are possible, though there aren’t such things
(B) seems paradoxical, because there is no way of reading ‘there are’ in the different senses required. But if (A) is true (& I concede it seems true), we have to explain how it is true. I don’t have an answer to that.
In reply, (A) and (B) seem to me to say the same thing. I agree that (B) sounds paradoxical and that’s why I think we need two senses of “exists”. Given the “is actual”/”there is” distinction, we can express matters without paradox as follows:
(C) There are possible things that are not actual.
The “there is” sense of “exists” says of some describable class (say, the class of superheros) that it is non-empty in the relevant domain of quantification (which may vary from context to context–when talking about possible worlds we range over the domain of abstract possibilia; in other contexts we may want to restrict ourselves to the domain of actualia or expand to include abstract impossibilia as well). In the domain of the Marvel Comics universe that class is non-empty–it contains the X-men, Spiderman, etc. In the domain of actualia, it is empty.
But the “there is” sense can’t tell us where the domain of actuality lies. Given just the “there is” sense we have no way of picking out “the actual world” from other “possible worlds”. That’s where the “is actual” sense comes in. It tells us what ex-ists, what is real as opposed to what is unreal, fictitious, imaginary, etc.