What kind of entity is a fictional character?

[DBT]

How can existence not be a property?

 

[ficino]

As I understand it, in predicate logic that goes back to Russell, it's a quantifier. In an existence claim, you are attributing certain properties to members of a set, and the existential quantifier marks that set as having at least one member. So taking "the present king of France is bald," you can analyze that as

(Ə x) (xF • xG) where F = present king of France and G = bald

I.e. there exists an x such that x is the present king of France and x is bald.

Since there is no x over which this conjunction holds true, the whole proposition is negated. So the original statement is not meaningless but simply false. On this account.

The properties are being king of France now and being bald. Existing isn't a third property.

I may have some of this wrong.

My understanding is that Russell's breakthrough, which helped lead to analytic philosophy, gained support in so many quarters because it resolved many puzzles of language and metaphysics.

Kant, though, had already attacked the Ontological Argument on the grounds that it wrongly treats existence as a predicate. The Stanford Encyclopedia article on Objects, which steve_bnk linked, goes into that.

I understand that Meinong, for one, held that existence is a predicate, and he has continued to have followers.

 

[ficino]

These are all just pretend referents, and statements about them technically are false. The story is not incomprehensible; it's just that, like all fictions, the utterances that compose it are literally false.

It would have been better if, instead of the above, I had written:

"These are all just pretend referents, and if we treat the author's utterances about them as statements, they are false. But since we all agree that fiction suspends real-world reference, the author's utterances about the characters don't count as statements/assertions. Fictional utterance is not assertion in the strict sense. The story is not incomprehensible; it's just that, like all fictions, the utterances that compose it are, if viewed as assertions, false, and if viewed as fictional discourse, not assertions."

 

[Togo]

Great post, Togo! I appreciate your thoughts. You are ahead of me, having written a paper on this. I'm just trying to catch up.

Meh, an essay, over a decade ago. Don't read too much into it.

Do you agree with those who posit Meinongian objects?

I'll be honest - I don't know enough about their opinions to be sure. I had a brief scan through Stanford's on-line philosophy articles. I'm broadly sympathetic with Many-worlds, and with Dual Copula, neo-Meinongian explanations, but I feel at a tangent to them.

The problem is that not that there are literally many seperate worlds, only some of which exist but all of which can be talked about, but that the real actual world consists of many worlds of discourse. That is, that things can be described in many different ways, all of which are potentially valid.

This becomes clear when we consider ambigious states. For example let us suppose that two people get married in a civil ceremony first of all, and then married in a church wedding thereafter, or married in a church, first, and then a temple. This kind of thing happens when the two families are of different religions, so each gets a service they approve of. But at what point are they married? What is their status between services? The answer depends very much on what you mean by married, and whether you deem it to be a legal, religious, or social state.

The problem for the logician here is that there are several different definitions, used in different contexts. This is a common problem, to the extent that some statements, such as 'he is my superior' are almost meaningless without the context, in this case of what hierarchy is being talked about. And it's not limited to social situations. The 'letter has been delivered' can mean several different things, 'he is red' can refer to communist sympathies or sunburn, and '11' means very different things in base 10 compared to base 2.

What's missing in the consideration of 'existance' is very much this kind of context, hence my sympathy with Many Worlds, which introduces the 'world' as context, and with Dual Copula, which focuses on the ambiguity of what 'is' really means.

So let's try adding it and see what happens. Superman exists within the context of the DC comic universe, a fictional experience that millions have partaken of. He does not exist in the real world. Similarly Mickey Mouse does not exist within the context of that universe - he is referenced in superman universe as a fictional character. Thus within the context of that universe, one exists and one does not, but within the context of the real world, neither exist.

As a second example consider our much married couple from earlier. In the context of social expectations, either of the religious ceremonies are sufficient to declare them married. In the context of religious belief, one of the ceremonies is important and the other irrelvent. In the eyes of the law, neither ceremony particularly matters, since the legal part is the signing of the register in front of witnesses. And so on.

In short (too late), attributes need context to render them meaningful, and the difficulties inherent in declaring that something exists or that it doesn't is that such a declaration is meaningless without some kind of context. The difficulty faced by existance as an attribute is not unique, but rather a classic example of a wider problem, which is that there are a great many attributes that require context to make them meaningful.

As to your three points, I need to think further, but can't resist trying a quickie response now. Here's my attempt. Against:

1): we can talk about the class of bachelors and the class of married people. These classes do not intersect; there is no member of both. hence, we cannot quantify over both classes. In the case you mention, it is false that the thing you are talking about is a married bachelor. You are just making an illicit conjunction of the predicates married and bachelor - illicit by the rules of language. Russell is still correct that it is not the case that there is some x that is married and is a bachelor. So you're really just talking about married x's and bachelors and making a mistake in conjoining them.

While I can see why this approach is tempting, I don't find it convincing. What does 'illicit by the rules of language' really mean? The problem is not with language, which has few if any problems in expressing logical contradictions, but that the meaning is contradictory. But expressing contradicting meanings is not 'illicit' or 'untrue', nor is it the case that when someone expresses a contradiction they are in fact talking about some other thing that is not contradictory. The point is that people can and do express logically contradictory ideas. They aren't secretly actually doing something else.

Going further, I'd say that expressing logical contradictions is not only commonplace, but necessary to understanding logic. With the idea of logic consistency comes logical inconsistency, and an inability to understand one leads at the very least to a flawed understanding of the other. We must know and understand logical contradictions to discuss them at all, or at least to avoid them.

You also mentioned the idea of logically distinct classes. There are of course, many way of defining interactions between sets, but one of the most common is that two sets are seperate if and only if there are no items that are a member of both sets, or more formally, that the set of the intersection of both sets is an empty set. To embrace Russel's claim that when we talk of logically inconsistent attributes we are in fact talking of something else is to render this definition meaningless, and to contradict some fairly unambigious mathematical assertions.

2): When an author writes a fiction, s/he makes pretend assertions about objects. The author of the Superman stories makes pretend assertions. They are pretend and not real assertions because they have no reference; Superman does not exist. When we talk about a fiction, though, our assertions do have a reference, because the fiction itself is a created artifact that is a real-world referent. So we and children can make genuine assertions about Superman, the created artifact. So no problem with our or critics' discourse about the properties of the text. The problem is with explaining the nature of the referent of the utterances put by the author into the fiction, but that's not the problem you bring up in 2).

I'm not generally a fan of definitions that only activate post-hoc. I'm talking about I guy I met at a party, called Tony. My wife claims that I didn't meet any such person, and I'm getting confused between two other people Timmy and Tyler. We discuss the matter. Am I making pretend assertions or real ones? According to what you've said, I'm making real assertions if Tony does exist, and pretend ones if he doesn't, but it should be clear from the example that I personally am not doing anything different in either case. Either I was mistaken or I wasn't, but I'm not actually pretending either way.

I'd prefer a definition set that put the ambiguity where it should be, in the actual concepts being discussed, rather than trying to hive it off into the actors.

3) the story isn't a counterexample against Russell because all the characters are part of the fiction, the boy as much as the princess. It doesn't matter how many back-levels, shall we say, exist in a narrative, i.e. a character who interacts with characters fictional to him, who may in turn interact with characters fictional to them. These are all just pretend referents, and statements about them technically are false. The story is not incomprehensible; it's just that, like all fictions, the utterances that compose it are literally false. But if they were true, it wouldn't be fiction.

Hm.. But it's not enough to declare them false, because their falsity or truth doesn't have any bearing on the case. What matters is whether they are coherent. Russel claims (from memory) are extremely strong - he claims that the idea of an existant unicorn doesn't make sense, even to the person talking about it - that it is literally non-sensical. But in the case of the story, the existance or otherwise of the characters is an essential plot point. Something coherent is being consistently communicated. We can establish, as an observed fact, that existance can coherently be treated as an attribute - within the context of a fiction. Hence my claim that Russel's position is counterfactual - people do in fact make sense of a relationship that he claims doesn't contain any.

OK, show me where I went wrong! And if you have any views about Meinong, I'd like to hear them. So far I'm not prepared to grant his "objects."

What's your objection to them?

 

[Togo]

It would have been better if, instead of the above, I had written:

"These are all just pretend referents, and if we treat the author's utterances about them as statements, they are false. But since we all agree that fiction suspends real-world reference, the author's utterances about the characters don't count as statements/assertions. Fictional utterance is not assertion in the strict sense. The story is not incomprehensible; it's just that, like all fictions, the utterances that compose it are, if viewed as assertions, false, and if viewed as fictional discourse, not assertions."

Does that really work?

Consider the statement. "Henry believes that the Loch Ness monster exists." Or simpler, still, "Henry believes that X." From what you've said, this is an assertion if Henry is a real person, and not an assertion if Henry is a fictional character. But in that case it is not possible to tell from the statement itself whether or not it is an assertion, since it depends on the properties of the subject. Which makes it impossible to produce a generalised logic. That leads to the uncomfortable situation where the statement "x=3" may or may be an assertion, depending on whether or not x is considered to be real.

 

[Perspicuo]

A fictional character may be a body of information, various concepts, imagined characters, creators, invisible beings, etc, cobbled together by a mind into a coherant but fictional entity.

How coherent? I doubt it. In every head they're imagined differently. For instance, you can have canonical Spock from Star Trek, versus fanfic Spock, but that is just one of the conceptualizations, and certainly, not even in the screenplay writers head is Spock a "logical" concept devoid of the interplay of fantasy, conscious and unconscious... pun intended.

 

[Perspicuo]

To believe a character in the mind is as existent as one in the physical world can hide a naive concept of existence. If Papa Smurf is blue, his blueness is not a real property of blueness as in the actual world.

An example: I can imagine a famous blue painting, is the concept blue? No, it is not, it is perhaps a binary instruction in my neural net, which by virtue of the process of fantasy which is neverending, it is never the same to begin with. The real painting's blueness is actually a "property" (actually a reproduceable phenomenon) due to calcium copper silicate in the pigments which make it look blue to my visual cortex by afference from my photoreceptors. There is no such physico-chemical phenomenon involving calcium copper silicate in my brain, therefore blueness is not a property of the painting as I imagine it. What happens in minds when the stimulus "Gainsborough's The Blue Boy" is given to a human being is not the same as the real thing--the reaction of photons on bouncing off the physical Blue Boy painting.

 

[ficino]

Does that really work?

Consider the statement. "Henry believes that the Loch Ness monster exists." Or simpler, still, "Henry believes that X." From what you've said, this is an assertion if Henry is a real person, and not an assertion if Henry is a fictional character. But in that case it is not possible to tell from the statement itself whether or not it is an assertion, since it depends on the properties of the subject. Which makes it impossible to produce a generalised logic. That leads to the uncomfortable situation where the statement "x=3" may or may be an assertion, depending on whether or not x is considered to be real.

Hi Togo, I'm excited about getting back to your longer post. Just for now, in reply to this:

My revised response to your 3. relies on speech-act theory about fictional discourse, a la John Searle. I haven't worked out yet whether I subscribe to it, but there is a lot that I find attractive.

On the Searlian account, any utterance about Henry will be taken as an assertion about a real-world Henry until we find out that we are dealing with fiction. Then, by the shared codes that govern writing and reception of fiction, we assume that Henry by definition doesn't exist. So we read utterances about him in the fiction w/ the same language rules we use for assertions but at the same time read them as pretend assertions, since Henry is pretend. So your "Henry believes that x" is an utterance either way but only an assertion if it has reference -- which it can't have if Henry does not exist. It's only a pretend assertion in fiction. So you're right, you can't tell from the utterance whether it is an assertion or a pretend assertion. You need to know the pragmatic context to ferret that out.

Searle also gets into one of the fascinating questions for me: what about real-world referents in fiction? If Henry says "3+3=6" in the fiction, I'm not sure whether Searle would say that THAT is an assertion because it has real-world reference. Searle does say that the theory of history stuck into War and Peace is Tolstoy's assertions and that the theory is not part of the story. He also says that there are real-world referents in that novel, like Napoleon. He doesn't really work this into his theory in a comprehensive way.

Others, like Jon-K. Adams, make EVERYTHING in the fiction a double of what is in the real world. I think that move preserves the boundary between fiction and non-fiction but forces us to drop many intuitions we have as readers. Just an example: a Greek historian told me that the description of the Battle of Salamis in Aeschylus' Persians is more reliable than what we find in Herodotus. But the drama, being fiction, would for Adams present, not the real-world battle, but a fictional one of similar description, so that Aeschylus asserts NOTHING about the real-world battle... gets very messy, and I think, unnecessarily so.

The real-world reference in fiction is really my top interest right now for what I'm trying to do with the discourse we get in Plato's dialogues!

OK, all I can type now. Till later, cheers, f

 

[DBT]

A fictional character may be a body of information, various concepts, imagined characters, creators, invisible beings, etc, cobbled together by a mind into a coherant but fictional entity.

How coherent? I doubt it. In every head they're imagined differently. For instance, you can have canonical Spock from Star Trek, versus fanfic Spock, but that is just one of the conceptualizations, and certainly, not even in the screenplay writers head is Spock a "logical" concept devoid of the interplay of fantasy, conscious and unconscious... pun intended.

I agree with that, but there must be a certain degree of coherence in order for the characters and their adventures to make sense, firstly to the author and that the reader can relate to the characters and events as they are described by the author.

Of course everyone does interpret the narrative in their own way, but that is also true of actual events. Ask 10 witnesses to an accident and you'll probably get 10 different versions of the event, albeit reasonably accurate overall.

 

[DBT]

As I understand it, in predicate logic that goes back to Russell, it's a quantifier. In an existence claim, you are attributing certain properties to members of a set, and the existential quantifier marks that set as having at least one member. So taking "the present king of France is bald," you can analyze that as

(Ə x) (xF • xG) where F = present king of France and G = bald

I.e. there exists an x such that x is the present king of France and x is bald.

Since there is no x over which this conjunction holds true, the whole proposition is negated. So the original statement is not meaningless but simply false. On this account.

The properties are being king of France now and being bald. Existing isn't a third property.

I may have some of this wrong.

My understanding is that Russell's breakthrough, which helped lead to analytic philosophy, gained support in so many quarters because it resolved many puzzles of language and metaphysics.

Kant, though, had already attacked the Ontological Argument on the grounds that it wrongly treats existence as a predicate. The Stanford Encyclopedia article on Objects, which steve_bnk linked, goes into that.

I understand that Meinong, for one, held that existence is a predicate, and he has continued to have followers.

As I understand it, existing is not a third property, or a second property, but a necessary quality of the objects and events of the world. So the present King of France either exists, or he does not exist, he is either bald or he is not bald, of which any of these claims may be falsified by the available evidence.

 

[ficino]

As I understand it, existing is not a third property, or a second property, but a necessary quality of the objects and events of the world. So the present King of France either exists, or he does not exist, he is either bald or he is not bald, of which any of these claims may be falsified by the available evidence.

I think we may be basically saying the same thing, but do you want to stick with "not a ...property, but a ...quality"? I would think property and quality are synonyms or virtually so.

 

[DBT]

As I understand it, existing is not a third property, or a second property, but a necessary quality of the objects and events of the world. So the present King of France either exists, or he does not exist, he is either bald or he is not bald, of which any of these claims may be falsified by the available evidence.

I think we may be basically saying the same thing, but do you want to stick with "not a ...property, but a ...quality"? I would think property and quality are synonyms or virtually so.

Yeah, virtually synonymous as far as I can see.

 

[Kharakov]

http://plato.stanford.edu/entries/nonexistent-objects/

IMO sounds like convoluted philosophical nonsense. You may be heading towards the 'are thoughts real discussion?

it is all biochemical states in your brain.. I believe the inability to distinguish between reality and fiction is considered a disorder.

OK, I thought this was a philosophy forum. But I'm down with it's all biochemical states in my brain. Reductionism rocks.

Are reductionism rocks a pun or something? Because sometimes I think physicalist's thoughts are formed out of them. My thoughts are formed out of thoughts thought. Ooops. Tyop. Not going to correct it though.

 

[ficino]

As I understand it, existing is not a third property, or a second property, but a necessary quality of the objects and events of the world. So the present King of France either exists, or he does not exist, he is either bald or he is not bald, of which any of these claims may be falsified by the available evidence.

I think we may be basically saying the same thing, but do you want to stick with "not a ...property, but a ...quality"? I would think property and quality are synonyms or virtually so.

Yeah, virtually synonymous as far as I can see.

I think I get it - you mean existence is a necessary predicate of things in the world, not a contingent predicate? If so, I take it that the scope of "necessary" is the if-then relation: if it's an object or event in the world, then necessarily it exists.

I think this account will mess up standard predicate logic, but it may be that I'm wrong about that consequence or that you're fine with it, given that you oppose some of what Russell advocated.

 

[DBT]

As I understand it, existing is not a third property, or a second property, but a necessary quality of the objects and events of the world. So the present King of France either exists, or he does not exist, he is either bald or he is not bald, of which any of these claims may be falsified by the available evidence.

I think we may be basically saying the same thing, but do you want to stick with "not a ...property, but a ...quality"? I would think property and quality are synonyms or virtually so.

Yeah, virtually synonymous as far as I can see.

I think I get it - you mean existence is a necessary predicate of things in the world, not a contingent predicate? If so, I take it that the scope of "necessary" is the if-then relation: if it's an object or event in the world, then necessarily it exists.

I think this account will mess up standard predicate logic, but it may be that I'm wrong about that consequence or that you're fine with it, given that you oppose some of what Russell advocated.

What exactly do you believe I said that opposes what Russell said?

 

[ficino]

I'm sorry, my error. I was confusing your posts with those of Togo. I'm losing track already of what's been said in this thread - must reread!

 

[Perspicuo]

A fictional character may be a body of information, various concepts, imagined characters, creators, invisible beings, etc, cobbled together by a mind into a coherant but fictional entity.

How coherent? I doubt it. In every head they're imagined differently. For instance, you can have canonical Spock from Star Trek, versus fanfic Spock, but that is just one of the conceptualizations, and certainly, not even in the screenplay writers head is Spock a "logical" concept devoid of the interplay of fantasy, conscious and unconscious... pun intended.

I agree with that, but there must be a certain degree of coherence in order for the characters and their adventures to make sense, firstly to the author and that the reader can relate to the characters and events as they are described by the author.

Of course everyone does interpret the narrative in their own way, but that is also true of actual events. Ask 10 witnesses to an accident and you'll probably get 10 different versions of the event, albeit reasonably accurate overall.

I'm comparing subjective experiences versus physical experiences, not subjective experiences versus interpretations of physical experiences (which is the same as comparing subjective experiences with subjective experiences). "Versions of the event" is not the same as "actual events".

 

[ficino]

Hm.. But it's not enough to declare them false, because their falsity or truth doesn't have any bearing on the case. What matters is whether they are coherent. Russel claims (from memory) are extremely strong - he claims that the idea of an existant unicorn doesn't make sense, even to the person talking about it - that it is literally non-sensical. But in the case of the story, the existance or otherwise of the characters is an essential plot point. Something coherent is being consistently communicated. We can establish, as an observed fact, that existance can coherently be treated as an attribute - within the context of a fiction. Hence my claim that Russel's position is counterfactual - people do in fact make sense of a relationship that he claims doesn't contain any.

Hello Togo, I think we're agreed that context matters in evaluating what authors do with sentences.

Here are two tries at refuting the notion that existence is a predicate of individuals. See what you think.

I will use the example of the fictional character, Anna Karenina. I assume there was no historical AK. I think it's legit to argue using an example, because if existence isn't a predicate in her case, the case would falsify the general thesis.

1 (a la Russell). The contradiction argument.
AK is a fictional character who is presented in the novel as existing and as having a unique identity.
a. Assume ex hypothesi that a set of predicates applies to her: Russian, woman, unhappily married, exists, etc.
b. But fictional characters by definition do not exist.
Therefore a. and b. entail a contradiction.
We jettison a. because b. is established by the meaning of "fictional."

It won't help to try to undo this refutation by accusing it of equivocating on "exist." If "existence" in fiction really bears a sense different from "existence" in the real world, so that it's spelled the same but means something else like "fictionality," then the refutation still works because "Existence" becomes no longer a predicate of AK. Something else, "fictionality" or whatever, becomes the predicate, so the original thesis turns out not to satisfy truth conditions after all.

2. The "does no work" argument (a la Kant).
AK is a fictional object that must be distinguished from other fictional objects in the novel. AK is described by a unique set of properties, which ex hypothesi includes existence. Her husband, Karenin, is described by a different, unique set of properties, which ex hypothesi includes existence. Same for all the other objects presented in the novel. Since "existence" is a property of each of them, it does no work in describing any object; "existence" is otiose in the description. But a predicate does work of describing a thing. Therefore existence is not a predicate of an individual.

It won't help to retort that the above argument confuses identification and description. If existence is a property of all the objects in the novel, it is tautological to repeat it in an assertion about any one object. By the definition of description, the information that is given about the described object is not true of at least one other object in the relevant domain.

It may be that existence is a predicate of concepts. I am also aware that some philosophers maintain against Russell, Frege et al that existence is a predicate (or property?) of individuals. I'm waiting to be convinced!

 

[DBT]

A fictional character may be a body of information, various concepts, imagined characters, creators, invisible beings, etc, cobbled together by a mind into a coherant but fictional entity.

How coherent? I doubt it. In every head they're imagined differently. For instance, you can have canonical Spock from Star Trek, versus fanfic Spock, but that is just one of the conceptualizations, and certainly, not even in the screenplay writers head is Spock a "logical" concept devoid of the interplay of fantasy, conscious and unconscious... pun intended.

I agree with that, but there must be a certain degree of coherence in order for the characters and their adventures to make sense, firstly to the author and that the reader can relate to the characters and events as they are described by the author.

Of course everyone does interpret the narrative in their own way, but that is also true of actual events. Ask 10 witnesses to an accident and you'll probably get 10 different versions of the event, albeit reasonably accurate overall.

I'm comparing subjective experiences versus physical experiences, not subjective experiences versus interpretations of physical experiences (which is the same as comparing subjective experiences with subjective experiences). "Versions of the event" is not the same as "actual events".

Of course, but all physical experiences are also subjective experiences. The actual events of the external world being experienced through mental representations of those external objective events. Mental representation (consciousness) being subjective.

 

[Perspicuo]

Not all physical experiences are subjective experiences. Some trees fall in the middle of the forest where no one can hear or see them.