London Ed recommended to me Patrick Hamilton's 1941 booze novel, Hangover Square. It gets off to a slow start, but quickly picks up speed and now has me in its grip. I'm on p. 60. The main character is one George Harvey Bone.
Ed gives this argument in an earlier thread:
(*) Bone, who is depicted by Hamilton as a sad alcoholic, is living in a flat in Earl’s Court.
The argument is that either the predicates ‘is depicted by Hamilton as a sad alcoholic’ and ‘is living in a flat in Earl’s Court’ have no subject, or they have the same subject. Either way, van Inwagen’s theory is wrong.
If they have no subject, then ‘is depicted by Hamilton as a sad alcoholic’ has no subject, but PvI argues that the subject is an abstract object. If they have the same subject, then if the subject of ‘is depicted by Hamilton as a sad alcoholic’ is an abstract object, then so is the subject of ‘is living in a flat in Earl’s Court’, which he also denies.
Either way, his theory cannot explain sentences like the one above.
The first thing I would point out (and this comports somewhat with a comment by David Brightly in the earlier thread) is that (*) can be reasonably parsed as a conjunction, the conjuncts of which belong to different categories of fiction (not fictional) discourse:
(*) Bone is depicted by Hamilton as a sad alcoholic & Bone lives in Earl's Court.
The two different categories are, first, the category of sentences we use when we engage in lit-crit discourse about fictional characters 'from the outside' while yet attending carefully to the 'internal' details of the fictional work. An example of such a sentence would be the following. "George Bone, like Don Birnham of Charles Jackson's 1944 Lost Weekend, have girlfriends, but Netta, the inamorata of the former, is a devil whereas Helen, the beloved of Birnham, is an angel."
Now that sentence I just wrote might be a second-rate bit of lit-crit, but it is a sentence that occurs in neither booze novel, nor is it entirely external to either novel. It is not entirely external because it reports details internal to the novels and it either gets them right or gets them wrong. 'George Bone is a purely fictional character,' by contrast, is an entirely external sentence. That sentence does not occur in the novel, and indeed it cannot occur within the novel (as opposed to within a bit of text preceding the novel proper, or as an authorial aside in a footnote) unless it were put into the mouth of a character. It cannot occur therein, because, within the world of Hangover Square, George Harvey Bone is precisely real, not fictional. As the same goes for Earl's Court, although it is also a real place in London. (One could, I suppose, argue that the Earl's Court of the novel is a fictional Earl's Court and thus distinct from the real-world Earl's Court. Holy moly, this is tricky stuff.)
The second category I mentioned comprises sentences that are either wholly internal to pieces of fiction or sentences that occur in synopses and summaries but could occur internally to pieces of fictions. For example, the second conjunct of (*):
C2. Bone lives in Earl's Court.
(C2) is probably too flat-footed a sentence to occur in a novel as good as Hangover Square, but it could have occurred therein and it could easily figure in a summary of the novel. (C1), however, namely,
C1. Bone is depicted by Hamilton as a sad alcoholic
could not have occurred in Hangover Square.
Now as I understand things, the grammatical subject of a sentence is a linguistic item, a word or a phrase. Thus (C1) and (C2) have the same grammatical subject, namely, the proper name 'Bone.' The grammatical subject is to be distinguished from its extralinguistic referent, if there is one. Call that the real subject. ('Logical subject' doesn't cut it since we do not typically refer to items on the logical plane such as propositions.)
So I take London Ed in his above-quoted animadversion to be referring to the real subjects of (C1) and (C2) when he uses 'subject.' He poses a dilemma for van Inwagen's view. Either the conjuncts have no subject or they have the same subject.
They cannot have no subject on van Inwagen's view because the subject of (C1) is an abstract object. And they cannot have the same subject, because then both conjuncts would have as real subject an abstract object. That cannot be, since on van Inwagen's view, and quite plausibly to boot, the subject of (C2) cannot be an abstract object. No abstract object lives or resides at any particular place. Abstract objects don't hang out or get hung over.
So, Ed concludes, van Inwagen's theory cannot explain (*).
Now my metaphilosophy teaches that no theory is any good on this topic or on any other. The problems of philosophy are most of them genuine, some of them humanly important, but none of them soluble. They are genuine intellectual knots that we cannot untie. That's about as good as it gets when it comes to "nailing my colours to the mast" as Ed demands that I do.
In other words, I am not advocating a particular theory as superior to Ed's, whatever exactly it is. (I am not being 'snarky' to use a Gen-X expression; I really don't know exactly what his theory is.) I don't think that van Inwagen's theory is unproblematic and I am not advocating it.
But I do think that Ed has failed to refute van Inwagen. The reason is because he conflates the two categories of fiction sentences lately distinguished, the category of lit-crit sentences like (C1), and the category of sentences that either do or could occur within pieces of fiction, an example being (C2).
Defending van Inwagen, I reject Ed's disjunction, namely: Either the conjuncts have no subject or they have the same subject. They have neither the same subject nor no subject. One has a subject and the other doesn't. (C1) has as its subject an abstract object and (C2) has as its subject nothing at all.
That's what van Inwagen could say to Ed so as to neutralize Ed's objection.
I am glad you are enjoying it. I just finished it. I was gripped, as I always am by money in old books, by the cost of things. A taxi from King’s Cross to Earl’s Court costs 6 shillings. Today more like £12, so a multiplier of about 40. But then a meal out at a swanky restaurant with the girl costs £2 16s, whereas the equivalent, at that type of restaurant, would cost close to £200, possibly more. So the relative values have changed.
But I digress. You say that PvI could easily resolve the problem for his theory by rejecting the driving premiss, which is that in this
(*) Bone, who is depicted by Hamilton as a sad alcoholic, is living in a flat in Earl’s Court
the predicate ‘is living in a flat in Earl’s Court’ has Bone as subject, and so does ‘is depicted by Hamilton as a sad alcoholic’. How can you say they are different? It’s like
() Socrates is standing and is arguing
Where quite obviously ‘is standing’ and ‘is arguing’ have Socrates as subject. Whatever satisfies one, satisfies the other. Biting on the legendary granite.
Also, the problem is easily resolved. We can rewrite (*) as
(**) Hamilton says (in the book) that Bone is living in a flat in Earl’s Court, and that Bone is a sad alcoholic
The only main change is to make explicit the story operator ‘Hamilton says in the book …’. Here it is also clear, and also uncontroversial, that the predicates have the same subject.
This is a very simple solution, and we should not be looking for complicated problems.
Posted by: Ed | Saturday, August 02, 2014 at 02:40 PM
Actually we have got deeper into this rabbit hole than I thought. You are apparently suggesting that the Inwagean can resolve any apparent inconsistency between A and B in
Something is A and is B
by replying that the predicate ‘is A’ is denoted by that sentence to satisfy a different object that ‘is B’ is denoted to satisfy. But then we could try to resolve
Something is a fictional character and is not a fictional character
in just that way. The predicate ‘is a fictional character’ is satisfied by the Inwagean abstract object, and ‘is not a fictional character’ is satisfied by something else. But wait, what is this ‘something else’? If anything, it’s Bone himself. But Inwageans don’t believe in Bone himself, since they don’t believe that intra-fictional sentences are satisfied by anything.
Meinongians can say that ‘is not a fictional character’ is satisfied by a Meinongian intentional thingy, but then they face the deeper problem of how “Something is a fictional character and is not a fictional character” expresses anything but a contradiction. Is it equivalent to “Something fictional character is not a fictional character”? Do they regard that is a contradiction or not?
Once you start messing with predication you have hit bedrock, solid granite, and you shouldn’t be messing with such large, immoveable and unbiteable masses.
Posted by: Ed | Sunday, August 03, 2014 at 01:01 AM
Ed, wouldn't the fact that 'Bone' is in an extensional context in (*), and an intentional context in (**) also count as a significant change?
Or, perhaps more interestingly, does it follow on your view that all singular terms in fiction discourse that appear to be in extensional contexts are in fact in intensional contexts as a result of the implicit (but, I take it, ubiquitous, at least when we're talking about fictional characters) story operator?
On the point of the OP, I think that van Inwagen would also say that in (c2), the predicate 'lives in Earl's Court' is ascribed (in the technical sense of ascription (A) that van Inwagen develops) to Bone in the following way:
A(lives in Earl Court, Bone, Hangover Square)
That is, van Inwagen would deny that Bone has the property, 'Lives in Earl Court,' though he would ascribe 'lives in Earl Court' to him in the sense that the property 'bears a certain intimate relation' to Bone (in the sense that (to use van Inwagen's example), for the Cartesian, 'being 6' tall' bears 'a certain intimate relation' to the (real) person, who as an immaterial substance cannot actually be possessed of that property).
However, with (c1) the property 'is depicted by Hamilton as a sad alcoholic' is predicated of Bone, which van Inwagen would take to be a 'character in a novel' (which are abstract objects that form a subset of 'creatures of fiction' which is itself a subset of 'theoretical entities of literary criticism'.)
So we could, on van Inwagen's theory, distinguish (c1) from (c2) by saying that the former involves predication, while the latter involves ascription, in addition to the point raised by Dr. Vallicella that the two can be distinguished, again on van Inwagen's theory, by the fact that strictly speaking, (c1) expresses a proposition, while (c2), since its subject has no referent, fails to express a proposition (though on other theories it may express a 'gappy' proposition).
What I'm not clear about is how, on van Inwagen's account, we can say that (c2) (if we suppose it occurs in the novel) both fails to express a proposition and succeeds in ascription, for it seems to me that a necessary condition of successful ascription is to successfully express a proposition. Further, we're ascribing (though not predicating) a property to Bone in (c2) -- how, then, can the subject term be empty? (Wouldn't that result in the ascription relation A (lives in Earl Court, _ , Hangover Square), which seems to me to resemble a gappy proposition, and which hence would imply that van Inwagen's account, at least of ascription, would inherit all the difficulties with gappy propositions?)
Posted by: Eric | Sunday, August 03, 2014 at 06:08 AM
Hold on Ed. Didn't we agree here that there can be circumstances when to say 'Bone is a fictional character' is not to predicate something of Bone but to deny his existence altogether? It seems we must 'mess with predication' at least to the extent of regarding 'is a fictional character', in some circumstances, as a pseudo-predicate, if indeed a predicate at all.
Posted by: David Brightly | Sunday, August 03, 2014 at 09:14 AM
I'm looking over Sainsbury's 'Fiction and Fictionalism', and he makes an interesting point (though he makes it almost as an aside): if there are fictional characters, are there also fictional events? That is, why are we more inclined to be realists about fictional characters than realists about fictional events?
So, if I read a book in which Germany wins WWII due to the skill of a fictional character General Zimmerman, why are we likely (at least pre-theoretically) to think that Zimmerman has a referent, but the event described does not? (Or, if the fact that we all know that Germany lost WWII is judged to be a good reason, then take an account of a fictional war, e.g. the war in LoTR.)
To put it another way: if we know the fictional character Frodo via stories about his exploits, and we take Frodo to be an abstract object, does it follow that the events in which he is said to have taken part in are also abstracta? If not, what becomes of much of the content we ascribe to Frodo, who is almost entirely known through the telling of these fictional events?
For example, we may say, 'Frodo is brave' is true only on the basis of the events he's described as having taken part in -- that is, suppose that nowhere in the story are we directly told that he's brave, but rather that we infer it from the story's account of his actions. Now if the events are not real in the same way Frodo is, then how can we say it's true that Frodo is brave? It seems as if the defender of realism about fictional characters is also committed to realism about fictional events (at least given the way natural language seems to work in these areas).
Posted by: Eric | Sunday, August 03, 2014 at 02:51 PM
David: "Hold on Ed. Didn't we agree here that there can be circumstances when to say 'Bone is a fictional character' is not to predicate something of Bone but to deny his existence altogether?"
Absolutely, and nothing has changed. "Bone is a fictional character" = "Some story says there is such a person as Bone, but there isn't such a person"
Eric:"Ed, wouldn't the fact that 'Bone' is in an extensional context in (*), and an intentional context in (**) also count as a significant change? "
I argue that they are intensional in both, but the operator in (*) is implicit, rather than explicit. Compare
Tolkien says that hobbits live in middle earth. They are shy creatures that live in holes.
The second sentence contains no 'Tolkien says' operator. But it is clearly there, implicitly. For example, in uttering the second sentence, I am not committing myself to the existence of such shy creatures.
>>Or, perhaps more interestingly, does it follow on your view that all singular terms in fiction discourse that appear to be in extensional contexts are in fact in intensional contexts as a result of the implicit (but, I take it, ubiquitous, at least when we're talking about fictional characters) story operator?
Yes, obviously.
>>That is, van Inwagen would deny that Bone has the property, 'Lives in Earl Court,' though he would ascribe 'lives in Earl Court' to him in the sense that the property 'bears a certain intimate relation' to Bone
Yes, but this absurd relation of 'ascription' is simply something he had to cobble together to avoid the apparent paradox.
>>What I'm not clear about is how, on van Inwagen's account, we can say that (c2) (if we suppose it occurs in the novel) both fails to express a proposition and succeeds in ascription
AFAIK, intra-fictional sentences do not succeed in ascription, according to PvI. He needs ascription to explain how extra-fictional sentences like 'Bone is a lonely alcoholic' are true in some sense. As they occur in the fiction itself, they express no proposition and have no truth value.
Note my last sentence above "As they occur in the fiction itself, they express no proposition and have no truth value. " I am not committing myself to the truth of what PvI says, on the contrary, I am denying it. But as the sentence follows one which is talking about what PvI says, I hope it is clear that I meant "According to my understanding of Van Inwagen, as they occur in the fiction itself, they express no proposition and have no truth value. "
>>Further, we're ascribing (though not predicating) a property to Bone in (c2) -- how, then, can the subject term be empty? (Wouldn't that result in the ascription relation A (lives in Earl Court, _ , Hangover Square), which seems to me to resemble a gappy proposition, and which hence would imply that van Inwagen's account, at least of ascription, would inherit all the difficulties with gappy propositions?)<<
On Van Inwagen’s view, the gappy proposition is satisfied by an abstract entity, the fictional character. Remember that on his view, fictional characters exist as abstract entities. To avoid the difficulty that abstract entities have no gender, can’t be fat, or tall etc, he develops a dual-sense copula type theory. Sometimes ‘is’ expresses predication (i.e. ‘Mrs Gamp is fictional character’), sometimes it expresses ascription or ‘holding’ (i.e. ‘Mrs Gamp is fat’).
It’s a perfect example of the ad-hocery that is rampant in standard philosophy, a sort of ad-hocery that would be straightforwardly rejected in the field of science. There is also a whiff of the ‘argument from mystery’, to my mind.
Posted by: london ed | Monday, August 04, 2014 at 02:25 AM