So, since I am auditing these courses and I do not have to do any work for them I figured I would instead keep track of what is going on here…any comments are welcome.
Russell’s Argument Against Existence as a Predicate
In the lecture’s on Logical Atomism Russell says, in response to a question, that the problem with treating existence as a property is that it then couldn’t fail to apply, and this is characteristic of a mistake. Kripke argued that we can eaisly define the existence predicate as ‘(Ey)(y=x)’ which can fail to apply. Russell must have been thinking of (x)Ey(y=x) which just says ‘everything exists’, but this isn’t a predicate…
Kripke on Fiction
Kripke takes fictional characters to exist as abstract objects. So ‘Sherlock Holmes’ is a name for an abstract fictional object. These abstract objects are not supposed to be thought of as some shadowy things (“no abstract object ever lived on Baker Street”) but, as he puts it, “exist in virtue of the story”. We can ask all kinds of empirical questions about these abstract objects, like ‘which fictional character is most written about by literary critics?’ etc.
In the story it is assumed that whatever conventions for naming are, they have been met. So fiction cannot be evidence for or against ANY semantic theory of names.
In the story existence claims are true. ‘Sherlock Holmes exists’ is true in the story ‘considered as actual’ in just the same way as ‘Kripke exists’ is true here in the actual world (because we assume that in the story we assume that whatever the right theory how names get their reference is met). But when evaluated outside of the story the sentence, though still true, is not true in the same sense. It does not pick out a person who was detective and who lived on Baker street, it picks out a certain fictional abstract object…is Kripke a two-dimensionalist with respect to fiction? Sounds like it to me….
These abstract objects can be vauge, depending on the story. So take the ghost that Hamlet sees. Suppose the story had been written so that it was unclear whether that ghost was real (in the story) or a hallucination. Then it would be the case that (metaphyscially, not epistemically) it would be undertermined what kind of abstract object the ghost was. Or to take an example more to my liking, take Pan’s Laberynth (SPOILER ALERT); we never really find out whether Pan and the other-worldly stuff is real or not. So the status of those fictional character’s is indeterminate…
The ‘fictional’ operator iterates. So there can be fictional fictional characters. An example of this is the Play that is put in in MacBeth…or one more to my tastes, Itchy and Scratchy from the Simpsons. They are fictional fictional characters.
When I asked if he thought that fictions were mini-worlds, he said no because some stories deliberately contain contradictions, whereas possible worlds don’t. When I asked if he thought possible worlds were fictions, he said no. But I don’t see why not. He claims that possible worlds are abstract objects, fictional worlds are abstract objects…when I pressed him on this he said ‘there is some connection’…I am interested to see how this will play out…It seems to me that the possible worlds should be a subset of the fictional worlds
Richard Brown 
I agree with Kripke on that. I’ve wrote this joke on my blog some time ago:
A person walks in a bar.
Does he exist?
The thing is that the person exists in the story. As far he is a person, he is existent. But as he is a fictional person, he is fictional existent, so he has fictional existence.
But not sure why Kripke would use ‘abstract’ to speak for this kind of imaginary existence. Why not simple talk about ‘imaginary’ or ‘fictional’?
Is ‘abstract’ supposed to be set vs. ‘concrete’?
I wonder if maybe ‘actual’ is better word to put vs. the fictional existence. The things that exist in imaginary stories, don’t *actually* exist. Wonder if this kind of linguistic point might also back-up your thoughts that possible worlds and fictional worlds should be treated in similar way – i.e. as non-actual.
Oh, also I’m skeptical about (Ey)(y=x)
I think what makes x not actually existent, is not that some such condition is false, but that we need to take in account the phenomenon of x – i.e. that people create works of fiction, and that certain author created such fictional story (e.g. book), and that the story included a fictional character x, which was named ‘x’.
In such case ‘x’ is a name for imagined intentional content, and I think that it is by virtue of that, it is not-actually-existent.
So, I think that to pin-down the fictional existence, we need to take into account the whole story of how the name got into use, and that it can’t be done by a (Ey)(x=y) criterion.
Hi y’all, possible worlds might be actual, whereas fictional worlds cannot be actual, or so I’d’ve thought, e.g. if someone wrote a fiction that happened to coincide exactly (so far as it went) with reality, it would still be a fiction; even if it contained a character just like you, it would not be about you (?)
Enigman, I would agree with you that fictional world can’t be actual.
But is the actual world a possible world? Richard from Et Cetera, had some time ago a couple of posts arguing how actual world is not a possible world.
Here is a link
oops, http://www.philosophyetc.net/2006/04/actual-world-is-not-possible-world.html is the address
Hey Tanasije, sorry for the delay!
You ask “why Kripke would use ‘abstract’ to speak for this kind of imaginary existence. Why not simple talk about ‘imaginary’ or ‘fictional’?
Is ‘abstract’ supposed to be set vs. ‘concrete’?”
I think (part of the) reason is that we can ask legitimate empirical questions about fictional characters, but we can’t about imaginary people (unless we mean by ‘imaginary’, ‘fictional’…of course the real question is what does Kripke think abstract objects are? I hope to find that out myself…
I don’t know about the contrast with actual because these abstract objects do actually exist. Also, in the story it is true that Holmes actually exists. Of course it is not true simpliciter, and in that sense then yes I do think it contrasts with ‘fictional’ and ‘possible’ in the same way. Good point.
I am not sure about the other stuff. It looks like you think that (Ey) (y=x) is supposed to be a critereon for identity, but it isn’t. Kripke was just showing that we can define an existence predicate. It is widely thought that it is inappropriate to treat existence as a property that an object has (or lacks). Kant was the first to point this out, but it was solidified by the logic developed by Frege and Russell. Existence, they heled, was not a property of objects, like being red or square, but was rather the quantifier ‘something’. Well, that is inelegantly expressed, but the idea is that when we say that x is red, we are attributing some property to it and in logical notiation that would then look like R(x), where ‘R’ is the predicate red. If existence were a predicate then to say that x exists is to say that E(x), which says that x has the property of existing. Russell taught us that this is wrong and that we should write x exists as (Ex) (Rx), it doesn’t attribute any property to any object, but quantifies over objects. Kripke’s point is that (Ey) (y=x) is a perfectly good predicate, it attributes a property, and it seems to capture what we mean when we say that a certain something exists, so it can ve used to make existence assertions in a proof (for instance)….
Oh, I forgot to reply to Enigman! Sorry!!
Tanasije is right to point out Richard C’s post on that stuff. It has been discussed a lot around here in some of the other posts, but the idea is Kripke’s idea about possible worlds. The real world, what Kripke calls ‘the massive scattered object that surrounds me’ is one thing, the various states it could be in another. So, consider two die. The real objects are one thing, the various states (l dice=3, right dice=1; LD=2,RD=4’etc) they can be in another. To say that a possible state of the die might be an actual state of the die is to say that one of the possible states will correctly describe the state that the real objects end up in.
But, I think your question can be reformultaed within this framework unproblematically. It becomes the question, “possible state might be actual states, in the above sense, but fictional states can’t, right?” When we are talking about the die above, I don’t see what the difference is at all. It seems to me I in effect told you (a really boring) story about these fictional objects named ‘left dice’ and ‘right dice’. I mean, there aren’t any particular die that I had in mind when I talked about one being 1 and the other being 4, or whatever. This particular story, f we are to believe Kripke, was taught to us in grade school.
Now, when we move to talking about somewthing much more unrulely, like talking about people and events on a larger scale, things get more complicated. So Kripke definately thinks that there are works of fiction that are about historical figures and that are metaphysically impossible. He gave an example of a story that was about Napoleon being born in a differnt time and still having all of his napoleon characteristics but not being a general because of his being born at the wrong time in history. This story is really about Napoleon, given that this is what is true in teh story. In the story it is supposed to actually be Napoleon who is born in another time. But, Kripke holds that this is metaphysically impossible because he holds that an objects origins are an essential property of that object. So it is not possible that Napoleon be born of any other parents. But that does not mean that this is true of every fictional work. Consider a story that takes as its premise the actual Hitler winning World War Two….
[…] 09/05/07 Kripke […]