What Problem Does Literary Fiction Pose?

More than one.  Here is one.  And as old Chisholm used to say, you are not philosophizing unless you have a puzzle.  So try on this aporetic triad for size:

1. Purely fictional objects do not exist.

2. There are true  sentences about purely fictional objects, e.g., 'Sherlock Holmes is a detective' and 'Sherlock Holmes is purely fictional.'

3. If a sentence of the form Fa is true, then there exists an x such that 'a' refers to x.

The triad is logically inconsistent: any two limbs entail the negation of the remaining one. So the limbs cannot all be true despite the considerable plausibility of each.  So one of the propositions must be rejected.  But the first is nonnegotiable since it is true by definition.  The leaves two options: reject (2) or reject (3).

I want to avoid truck with Meinong if at all possible.  So I should like to adhere to (3).  There are no true singular sentences about what does not exist.

Suppose we reject (2).  One way to do this is by supplying a paraphrase in which the apparent reference to the nonexistent is replaced by real reference to the existent.  For example, the apparent reference to Sherlock, who does not exist, is replaced by real reference to a story in which he figures, a story that, of course, exists.  The elliptical approach is one way of implementing this paraphrastic strategy.  Accordingly,

4. Sherlock Holmes is a detective

and

5. Sherlock Holmes is fictional

are elliptical for, respectively,

6. In the Conan Doyle stories, Sherlock Holmes is a detective

and

7. In the Conan Doyle stories, Sherlock Holmes is fictional.

But note that while (5) is plainly true, (7) is plainly false.  The stroies represent the detective as a real individual, not a fictional individual!   So (7) cannot be taken as elliptical for (5) This is a serious problem for the 'story operator' approach.  Or consider the true

8. Sherlock Holmes does not exist.

(8) is surely not short for the false

9. In the Conan Doyle stories, Sherlock Holmes does not exist.

The point can be made with other 'extranuclear' predicates such as 'merely possible' and 'mythological.'  If I say that Pegasus is mythological, I don't mean that, according to legend, Pegasus is mythological. 

I'll end with a different challenge to the story operator approach.  Consider

10. Pinocchio was less of a liar than Barack Obama.

Whether you consider (1) true or false, it is certainly not elliptical for

11. In Carlo Collodi's The Adventures of Pinocchio (1883), Pinocchio was less of a liar than Barack Obama.

To put it vaguely, one problem with the story operator approach is that it traps fictional characters within particular stories, songs, legends, tales, etc. so that (i) it becomes difficult to understand how they can show up in different different stories, songs, etc. as they obviously do in the cases of Faust and Pinocchio, and (ii) it becomes difficult to understand how they can show up in comparisons with nonfictional individuals.

Is there a tenable solution to my triad or is it a genuine aporia? 

Related articles

A Paraphrastic Approach to Fictional Sentences

Obama the Liar Update

Comments

[Lukas Novak]

Hello, Bill,

what exactly keeps you from dropping (3), besides the post-Russellian standards of philosophical PC? I can't see any reason supporting (3), whereas (1) is evident conceptual truth and (2) is evident empirical truth (or if you doubt that, consider this reductio: what about the truth of the sentence "Sherlock Holmes is such that no true sentences can be formed about him"?) - so either (3) must be false, or logic itself. I'd say it is rational to choose (3), pace all the alleged "intuitions".

[Ed from London]

>>3. If a sentence of the form Fa is true, then there exists an x such that 'a' refers to x.

You say you want to adhere to this, but then you consider the sentence “In the Conan Doyle stories, Sherlock Holmes is a detective”. This is also of the form Fa, where ‘a’ is ‘Sherlock Holmes’, and F is “In the Conan Doyle stories, --- is a detective”. So you could have blocked the move at a much earlier stage.

Your (3) is false, in my view.

>>I want to avoid truck with Meinong if at all possible.

Is this a rhetorical ‘want to avoid truck’, or have you undergone some Damascene experience?

[Bill Vallicella]

Happy New Year, Lukas.

You solve the problem by rejecting (3). That is respectable. There are quite a few very intelligent Meinongians, and, as I have argued, van Inwagen hasn't refuted them. But just to be clear: are you maintaining that there are items that actually have properties but have no Being whatsoever and are therefore ausserseiend in Meinong's sense?

I'll need to write a separate post to explain why I cannot follow Meinong.

[Bill Vallicella]

Ed,

I take it you reject (3) because you don't accept (extralinguistic) reference at all! And not because you are a Meinongian. Unless you have had a drastic conversion experience!

>>You say you want to adhere to this, but then you consider the sentence “In the Conan Doyle stories, Sherlock Holmes is a detective”. This is also of the form Fa, where ‘a’ is ‘Sherlock Holmes’, and F is “In the Conan Doyle stories, --- is a detective”.<<

This smacks of sophistry, London Ed! You are entangling yourself in some sort of exportation fallacy.

Consider:

Ed believes that Holmes lives in Baker Street.

That is not equivalent to

Holmes is believed by Ed to live in Baker Street.

An illicit de dicto --> de re move.

The story operator cannot be transformed into a predicate.

[David Brightly]

Morning, Bill, and a Happy New Year to you.

For me, alarm bells were ringing before I reached (3). If we conjoin (1) and (2) we get

There are no PFOs yet there are true sentences about them,

and we should ask what the them refers to. Compare with

There are no leprechauns but there are true sentences about them.

So I think we can reject (1) and (2) in combination without worrying too deeply about what it means to be a PFO.

[Ed from London]

>>This smacks of sophistry, London Ed! You are entangling yourself in some sort of exportation fallacy.

I don't believe so, but I will check. I remember we had a similar dispute about placeholders and bound variables last year, which was purely about textbook stuff.

Otherwise we wouldn't get the substitution puzzles.

[Bill Vallicella]

Ed,

Is this a valid inference:

Tom believes that flying horses do not exist (de dicto)
ergo
Flying horses are believed by Tom to be nonexistent. (de re)

?

I say No. What say you? The premise is true. The conclusion is either false or lacks a truth-value.

[Ed from London]

>>Is this a valid inference: [Tom believes that flying horses do not exist (de dicto)ergo Flying horses are believed by Tom to be nonexistent. (de re) ? <<

Most decidedly not. 'T believes that Ex Fx' does not imply 'Ex T believes that Fx'

But in classical logic, 'T believes that Fa' does imply 'Ex T believes that Fx'. That is because 'a' is a constant.

Ask Peter Lupu.

[Bill Vallicella]

Happy New Year, David. Good to hear from you.

I take it you are saying that (1) and (2) are inconsistent such that there is no need to bring in (3).

I disagree. >>There are no leprechauns but there are true sentences about them.<<

According to Meinong, some items lack being, but one can still refer to them. So while there are no leprechauns, there are true sentences about them. 'Them' refers to nonexistent leprechauns!

We need (3) to derive a contradiction.

A confusion Ed may be falling into is to fail to distinguish between extralinguistic reference to the nonexistent and intralinguistic reference.

[Bill Vallicella]

Tom believes that Pegasus is winged
ergo
Pegasus is believed by Tom to be winged.

Invalid, say I. Premise true, conclusion false.

[Dominic Bolin]

I would add a qualification to (1). Purely fictional objects do not exist in the same way that animals, mountains, or other physical objects do, but I would not say that they do not exist at all. In this way, I think all three are true.

[Bill Vallicella]

That's a sol'n too. Distinguish between modes of existence.

[Lukas Novak]

Dear Bill,

happy new year to you, too! Yes, if "there is", "property" and "actually" are sufficiently broadly interpreted, then I say that there are items lacking any ontological status whatsoever but actually having some properties (though I am not sure whether that makes me a Meinongian).

Your attempt to block the arguments to prove this in terms of de re - de dicto fallacy does not convince me. Tom may well believe de dicto that Pegasus is winged, bud he may also truly believe de re about Pegassus that he is winged - which voids your response (if you say he cannot because there is no Pegasus, you're begging the question). And, even if Tom only believes de dicto, it can still be inferred that Pegasus is such that Tom believes de dicto that Pegasus is winged. For even the de dicto belief of Tom concerns, somehow, Pegasus.

Suppose Paul is Jack the Ripper but Tom does not know that. Tom believes de dicto that Jack the Ripper is a murderer, but does not believe de re about Paul that he is a murderer. (I would also say that he is in a position to believe de re about Jack the Ripper that he is a murderer, but let us not be distracted by this now.) Now I say that although Paul does not have, in virtue of Tom's belief, the property of being believed by Tom to be a murderer, he has, in virtue of Tom's belief, the property of being that very person which satisfies the description that Tom employs to fix the reference to the one about whom he believes to be a murderer. So, even de dicto beliefs endow their incidental objects with certain actual properties, even though these properties are not simple "passive beliefs".

Moreover, it is questionable whether the de re - de dicto distinction here actually makes any difference in this particular case about Pegasus (it depends on what exaclty is meant by this distinction).

[Ed from London]

>>Tom believes that Pegasus is winged
ergo
Pegasus is believed by Tom to be winged.
Invalid, say I. Premise true, conclusion false.<<

In classical logic, the premiss is false.

[Bill Vallicella]

Makes no sense, Ed. The premise is obviously true.

[Bill Vallicella]

Lukas,

You are indeed a Meinongian, though if I remember correctly you find this view in Scotus.

The de dicto/de re business was not directed at you or at disproving the Meinongian view. It was directed at Ed.

[David Brightly]

Hi Bill,
Yes, that's exactly my thought. May I make a methodological suggestion? Holding open the possibility that the Meinongians are right requires us to perform intellectual gymnastics. Rather than seeing a sentence like

(L) There are no leprechauns but there are true sentences about them,

as necessarily false by virtue of the ordinary meanings of its terms, we are asked to see it as contingently false and dependent on the truth of a possibly more questionable assertion such as your (3). This is too hard. Why not proceed on the assumption that the Meinongians are wrong and see if an acceptable theory of fiction can be found under that constraint? If this project fails, then we might return cap-in-hand to Meinong. But it's impossible to defend continuously against the Meinongian objections---they are so thoroughly undermining of ordinary thought. Previous attempts on this topic have run into the sands for just this reason.

[Bill Vallicella]

David Brightly writes:

>>Why not proceed on the assumption that the Meinongians are wrong and see if an acceptable theory of fiction can be found under that constraint? If this project fails, then we might return cap-in-hand to Meinong.<<

A very sensible approach, indeed the most sensible approach. Unfortunately, there is no acceptable theory of fiction that satisfies the constraint, although some fool themselves on this score. This is why Meinongians are a flourishing minority.

I am not saying that Meinongian theories are any good. I am saying that there is no good theory, and beyond that, that it is a good bet that the problem is insoluble by us.

Do you have a knock-down argument against the Meinongian view?

It is not obviously contradictory. Lukas Novak, above, writes, >>there are items lacking any ontological status whatsoever . . . <<

That smacks of contradiction: if there ARE such items, then they ARE, in which case they have Being, hence some ontological status!

But Lukas could have said: Some items have no ontological status whatsoever.

So I challenge you, David: Show that my formulation is self-contradictory, but without begging the question, i.e., without assuming that 'some' is an existential and not merely a particular quantifier.

[Bill Vallicella]

>>they are so thoroughly undermining of ordinary thought.<<

Interestingly, Meinongians would protest vociferously and claim that the contortions of the anti-Meinongian paraphasers undermine ordinary thought.

[London Ed]

I am still not with you Bill. We have a sentence in ordinary language

Tom believes that Pegasus is winged

You then have your statement (3)

3. If a sentence of the form Fa is true, then there exists an x such that 'a' refers to x.

Which implies you are translating into classical logic. I.e. Fa implies Ex Fx. How are you translating the first ‘Pegasus’ sentence into classical logic? If you do, you have to define ‘Pegasus’, and it is axiomatic that Ex x = Pegasus, and so ‘Ex x = Pegasus & Tom believes that Pegasus is winged’ is true. So if you define ‘Pegasus’, your inference is valid. If you don’t, then ‘Tom believes that Pegasus is winged’ is meaningless, i.e. is not true, and your inference is still valid.

That assumes you want some kind of interpretation into classical logic. If we are asking about the logic of ordinary language, then I still question your assumption (3).

Regarding (2) “2. There are true sentences about purely fictional objects, e.g., 'Sherlock Holmes is a detective' and 'Sherlock Holmes is purely fictional.'”

Why can’t the first sentence have the fiction operator and the second sentence not have it? Thus ‘In the SH stories Sherlock Holmes is a detective, but in reality there is no such person as SH’.

Obviously ‘In the SH stories there is no such person as SH’ is false.

[Bill Vallicella]

I don't agree.

The sense of 'Pegasus' is given by the 'the winged horse of Greek mythology.'

This is clearly invalid:

Tom believes that the winged horse of Greek mythology is winged. (True)
ergo
The winged horse of Greek mythology is believed by Tom to be winged. (False)

For nothing satisfies the def descr.

[David Brightly]

Bill,

>>Meinongians would protest vociferously...

Well, adapting a line from Mandy Rice-Davies, they would, wouldn't they?

I shan't attempt to refute the Meinongians, merely ignore them. We can have one philosophical theory presupposing another, surely?

Let me try to take the sting from your objection to the 'story operator' approach. You want to treat your (4) and (5) symmetrically wrt the story operator. I claim that there is a relevant difference in kind between the predicates 'is a detective' and 'is fictional' that justifies an asymmetric treatment. Here.

Also, you say that the story operator approach traps characters within stories. I think there are linguistic devices for 'lifting them out' of their containing stories. For example, the Pinocchio/Obama sentence can be rendered as

Pinocchio, as described in the story by Carlo Collodi, is less of a liar than Obama, in real life, is.

The effect here is that the scope of the first story operator is restricted to ‘Pinocchio’, and the second to 'Obama'.

[Ed from London]

I agree with David. You say (Prop. 2 in your main post) that there are true sentences about purely fictional objects, e.g., 'Sherlock Holmes is a detective' and 'Sherlock Holmes is purely fictional.'

Why can’t the first sentence have the fiction operator and the second sentence not have it? Thus ‘(a)In the SH stories Sherlock Holmes is a detective, but (b) in reality there is no such person as SH’.

[Joshua]

Sorry to distract from this erudite discussion with a lay logician's question (i.e., I teach secondary students to use sound categorical syllogism when making an argument), but I'm not clear why your repeated example is a fallacy, Bill. Why is "Pegasus is believed by Tom to be winged" false?

[Bill Vallicella]

Because there is nothing to which 'Pegasus' refers.

'Tom believes that Pegasus is winged,' by contrast, does not require the existence of Pegasus for its truth. This is because the proposition which is the accusative of Tom's belief state is Fregean: its subject-constituent is a sense.