Do you see the problem? — Pierre-Normand

If one believes that the person is drinking champagne, then the description represents the belief. The belief refers to the person the speaker believes to be drinking champagne.

The truth of the DD is irrelevant to successful reference. The speaker believes it to be true. That's relevant.

The truth of the DD is irrelevant to successful reference. The speaker believes it to be true. That's relevant. — creativesoul

Actually no, I stand by my prior statement. All that is necessary is for a sufficient number of the descriptive properties to be identified as referring to the same person. It doesn't matter whether the speaker believes them true either.

For example, it is a common technique in black communities to deliberately lie about descriptions which is known to others. As an overly simple example, they will say 'don't insult my brother like that.' The person who is not his brother then nods in agreement and raises a fist.

If one believes that the person is drinking champagne, then the description represents the belief. The belief refers to the person the speaker believes to be drinking champagne. — creativesoul

This is common ground. The issue is to explain how the speaker's belief comes to be about the speaker's intended referent in the world. @andrewk's account of the reference of the speech act relies on the assumption that the mental act thereby expressed is about the very same object she is thinking about. What accounts for the reference of the belief in the first place?

The issue is to explain how the speaker's belief comes to be about the speaker's intended referent in the world. — Pierre-Normand

Again, it is a common technique in black communities to deliberately lie about descriptions which is known to others. As an overly simple example, they will say 'don't insult my brother like that.' The person who is not his brother then nods in agreement and raises a fist.

I've tried to help understand the issue but I do have to rest.

Well I regret I must agree with andrewk. As you are interested in externals only, the belief doesnt matter. All that matters is that the two identify a sufficient part of the descriptive properties as referring to the same person. Thats the point of the theory. It doesnt matter how many of the descriptive properties are true or false, or if some of them could truthfully apply to others too. — ernestm

In order to establish that the speaker and his intended audience share (a sufficient part of their) beliefs about the same individual, to whom they are thinking about, one must first be able to say who it is that their beliefs are about. However, even if we leave this circularity problem aside, Kripke made a good case that shared beliefs aren't necessary. The audience would normally know who the speaker is thinking about even if the audience knew most or all of the descriptive content used by the speaker not to be true of her intended target. @andrewk got at least that right.

The audience would normally know who the speaker is thinking about even if the audience knew most of all the the descriptive content used by the speaker to be wrong of her intended target. — Pierre-Normand

That isnt true either. Black communities love to make fun of white people by talking about, for example, invented people who dont exist, when actually talking about the white person. And Ive heard them do it many times without the white person realizing it and some black teenager sniggering out of sight. There's alot more forms of communication than are obvious from the blithe statements of simple truths and falsehoods that people for whom English is not a first language figure out, and then deliberately connive to humiliate native English speakers together without the native English speakers realizing it.

And I do have to go to bed now, good night.

Again, it is a common technique in black communities to deliberately lie about descriptions which is known to others. As an overly simple example, they will say 'don't insult my brother like that.' The person who is not his brother then nods in agreement and raises a fist. — ernestm

That the descriptive content can represent false beliefs about the intended target is common ground. What we're inquiring about is the positive account for the reference of the thought (or of the speech act expressing this thought) being what it is in spite of the fact that the definite description by means of which the speaker thinks of the individual is false or mostly false.

I've tried to help understand the issue but I do have to rest.

Rest in peace :wink:

That the descriptive content can represent false beliefs about the intended target is common ground. — Pierre-Normand

hence belief is removable by occam's razor. Its irrelevant to the theory. You too.

hence belief is removable by occam's razor. Its irrelevant to the theory. You too. — ernestm

That's not an argument against Kripke, neither does it help @andrewk who does appeal to the content the speaker's belief in his account of reference.

That isnt true either. Black communities love to make fun of white people by talking about, for example, invented people who dont exist, when actually talking about the white person. And Ive heard them do it many times without the white person realizing it and some black teenager sniggering out of sight. There's alot more forms of communication than are obvious from the blithe statements of simple truths and falsehoods that people for whom English is not a first language figure out, and then deliberately connive to humiliate native English speakers together without the native English speakers realizing it. — ernestm

That's interesting, but what I meant to say is that for reference to be successful, on Kripke's account, it's not a requirement that the descriptive content associated with the speaker's 'idea' (i.e. the de re sense of a demonstrative or proper name) of her intended reference be mostly true. I did not mean to imply that making use of misleading descriptions can't possibly, indeed, mislead outsiders to a conniving community.

The issue is to explain how the speaker's belief comes to be about the speaker's intended referent in the world. — Pierre-Normand

Simply put...

As a result of the speaker knowing how to use language to draw an other's attention to the 'object'.

This account presupposes that your belief about that person indeed is about that person and not about someone else — Pierre-Normand

I do not understand this. How can it be the case that I have a belief about somebody that is in my field of view, and yet the belief is not about that person? Isn't that a bare contradiction - "I have a belief that is about X and not about X"?

Would it help to break it up? My belief is about the person at 12 o'clock (so in the above sentence we can replace 'X' by 'the person at my 12 o'clock'), and the belief is that that person is a young man and has a glass of champagne and has winked at me. As far as I am concerned 'the person at my 12 o'clock' is enough to identify the person. But talking to somebody else, I probably feel a bit more info is needed to avoid confusion - for instance my 12 o'clock may be Sabrina's 10 o'clock. So in my verbal statement I replace the position reference by my belief about the champagne and the age and sex, and the belief about the wink becomes a question rather than a part of the DD.

As a result of the speaker knowing how to use language to draw an other's attention to the 'object'. — creativesoul

Yes, that's sketchy but basically right. It also takes us out of the realm of Kripke's descritivist targets, and dovetails with his own account.

I do not understand this. How can it be the case that I have a belief about somebody that is in my field of view, and yet the belief is not about that person? Isn't that a bare contradiction - "I have a belief that is about X and not about X"? — andrewk

I was only considering the set of persons (person-1, ..., person-4) who are perceptually present and can reasonably be thought of by the intended audience to be the person being talked about (and demonstrated) by the speaker. One of them might be drinking champagne unbeknownst to the speaker. Of course, the speaker herself would know who the person is that she is looking at and thinking of, but she would not know that by description (or so would I be prepared to argue on Kripke's behalf).

Would it help to break it up? My belief is about the person at 12 o'clock (so in the above sentence we can replace 'X' by 'the person at my 12 o'clock'), and the belief is that that person is a young man and has a glass of champagne and has winked at me. As far as I am concerned 'the person at my 12 o'clock' is enough to identify the person. But talking to somebody else, I probably feel a bit more info is needed to avoid confusion - for instance my 12 o'clock may be Sabrina's 10 o'clock. So I add in the belief about the champagne and the age and sex, and the belief about the wink becomes a question rather than a part of the DD.

In that case, when you add "...at my 12 o'clock" to the description, you are relying to the content of the true part of the description to secure reference in spite of the falsity of the other parts of the description. When you are thus relying on a true descriptive core (however small) in order to account for the determination of the reference, you move back into the target area of Kripke's objections to descriptivism, which you had attempted to evade by means of your account of reference by means of (potentially false) descriptions that merely match up with the speaker's (potentially false) beliefs about her intended reference.

This account presupposes that your belief about that person indeed is about that person and not about someone else who might actually be, unbeknownst to you, drinking champagne, (or about nobody, if nobody is having champagne). What is this account of the reference of your belief on the basis of which the truth of the predicative content of the DD can be evaluated as matching up with this belief? — Pierre-Normand

Not to speak on behalf of andrewk, but rather on my own behalf...

The above criticism is based upon a misunderstanding of belief and how it works. False beliefs are not true. What's said about the referent in a false description is about the referent. It need not be true in order to refer.

Jane believes Joe killed Bob. She refers to Joe as "the man who killed Bob". Joe did not kill Bob. Allen did. When Jane says "the man who killed Bob", she is not expressing a belief about Allen even if and when it is the case that he satisfies the description(that he matches up with the description).

That alone shows us that satisfying the description is not necessary for successfully reference. To talk about "matching up with this belief" is to talk about whether or not the description is true. That is irrelevant to successful reference.

As a result of the speaker knowing how to use language to draw an other's attention to the 'object'.
— creativesoul

Yes, that's sketchy but basically right. It also takes us out of the realm of Kripke's descritivist targets, and dovetails with his own account. — Pierre-Normand

Well you asked a question that could only be answered with such ambiguity. I also do not see how it takes us out of the realm of descriptivist accounts. I'm ok with agreeing with Kripke. I'm ok with not. I don't think I understand his position well enough to know which is the case...

Not to speak on behalf of andrewk, but rather on my own behalf...

The above criticism is based upon a misunderstanding of belief and how it works. False beliefs are not true. What's said about the referent in a false description is about the referent. It need not be true in order to refer. — creativesoul

Of course. It only needs to be true in order to refer descriptively, in case the intended reference would be singled out descriptively by the predicative content of the definite description. If the intended reference is singled out demonstratively, for instance, and we can account for demonstrative reference non-descriptively, then it's possible to express a false belief by means of a false definite description of this demonstratively referenced individual.

Jane believes Joe killed Bob. She refers to Joe as "the man who killed Bob". Joe did not kill Bob. Allen did. When Jane says "the man who killed Bob", she is not expressing a belief about Allen even if and when it is the case that he satisfies the description.

That alone shows us that satisfying the description is not necessary for successfully reference.

To talk about "matching up with this belief" is to talk about whether or not the description is true. That is irrelevant to successful reference.

This is all common ground between Kripke, you and me. It is @andrewk who relied on this matchup in order to make his descriptivist account work.

False beliefs are not true. What's said about the referent in a false description is about the referent. It need not be true in order to refer.
— creativesoul

Of course. It only needs to be true in order to refer descriptively, in case the intended reference would be singled out descriptively by the predicative content of the definite description. If the intended reference is singled out demonstratively, for instance, and we can account for demonstrative reference non-descriptively, then it's possible to express a false belief by means of a false definite description of this demonstratively referenced individual. — Pierre-Normand

This seems odd to me.

On the one hand you agree that false descriptions can successfully refer. On the other, you seem to be implying that they cannot refer 'descriptively'. How else do descriptions refer if not descriptively?

Have I misunderstood or is there a bit of nuance here? I'd appreciate - and do appreciate - your participation here.

If the intended reference is singled out demonstratively, for instance, and we can account for demonstrative reference non-descriptively, then it's possible to express a false belief by means of a false definite description of this demonstratively referenced individual. — Pierre-Normand

You're saying that false description does not pick out the referent, but rather that it has/had already been picked out by true description or demonstratively(pointing, showing).

Is that about right?

On the one hand you agree that false descriptions can successfully refer. On the other, you seem to be implying that they cannot refer 'descriptively'. How else do descriptions refer if not descriptively? — creativesoul

Remember Kripke's explanation that he intended to use the phrase "reference of the description" in order to match up with the descriptivist logical tradition. (That was on page 25, if I remember). That's how referring descriptively works. You supply a definite description of the item you intend to refer to, and you intend this item to be whatever uniquely satisfies this description. (That's what makes the description definite). By definition, such a description is about the item (if there is any) that uniquely satisfies the description. Another way for a definite description to refer would be as a reference fixing rather than a reference determining device. In that case, it might serve to disambiguate among several items that a speaker could be making reference todemonstratively, or by means of a shared proper name, while accounting for the fact that the content of the description could be false and merely believed to be true by the speaker.

You're saying that false description does not pick out the referent, but rather that it has/had already been picked out by true description or demonstratively(pointing, showing).

Is that about right? — creativesoul

That the intended reference might already have been determined by a true description (e.g. andrewk's "...at twelve o'clock") might be a requirement of a purely descriptivist account. Kripke is arguing that the reference can be singled out entirely non-descriptively but that, in some cases, in order to enable the reference to be communicated non-ambiguously, a definite description (either true or merely believed to be true of the intended reference) can be supplied to other people in order to draw their attention to the intended item and thereby enable them to refer to it under the same (or deferred) mode of demonstrative reference. (I am somewhat adapting his account of proper names to demonstrative reference since, in spite of the obvious differences, they both are de re modes of reference that rely on the possibility of knowledge by acquaintance).

When you are thus relying on a true descriptive core (however small) in order to account for the determination of the reference, you move back into the target area of Kripke's objections to descriptivism — Pierre-Normand

Where in N&N do you find that?

If all that Kripke is saying is that, where every single belief that a person has about a person, including that he is standing at 12 o'clock, or that I was introduced to him yesterday in a meeting, or that my grandmother told me a story about him, is false, then one cannot give an account of how the person can be referred to, then the situation he is using is so rare that it is ridiculous to use it as an objection to any theory of anything.

To which I would add that, if every belief I have about a person is false, the question of whether I 'can refer' to that person is moot - it becomes unclear what 'referring to' would mean in that situation. At best the person is 'referring to' an element of a hallucination they had.

The only situation I can think of in which every belief somebody has about somebody is false is where I am undergoing hallucinations or have false memories, so that I am thinking of the person over there but there is no person over there, or I am remembering my grandmother telling me of this person called Nixon, but that memory is false and my grandmother never told me of Nixon.

I am fairly sceptical of Kripke's theory but I don't believe it is as trivial as such an interpretation would suggest. Certainly that's not what I get from N&N.

On the one hand you agree that false descriptions can successfully refer. On the other, you seem to be implying that they cannot refer 'descriptively'. How else do descriptions refer if not descriptively?
— creativesoul

Remember how Kripke explained how he intended use the phrase "reference of the description" in order to match up with the descriptivist logical tradition. (That was on page 25, if I remember). That's how referring descriptively works. You supply a definite description of the item you intend to refer to, and you intend this item to be whatever uniquely satisfies this description. (That's what makes the description definite). — Pierre-Normand

I do remember that. It's been the source of a bit of confusion on my part.

Another way for a description to refer would be as a reference fixing rather than a reference determining device. In that case, it might serve to disambiguate among several items that a speaker could be making reference demonstratively, or by means of a shared proper name, while accounting for the fact that the content of the description could be false and merely believed to be true by the speaker.

See if I have this right...

Here the difference between reference fixing and reference determining would be that the former makes use of an otherwise inadequate description(one that is incapable of successfully picking out an individual), whereas the latter is making use of a purportedly adequate description, according to one who argues in favor of definite descriptions.

If all that Kripke is saying is that, where every single belief that a person has about a person, including that he is standing at 12 o'clock, or that I was introduced to him yesterday in a meeting, or that my grandmother told me a story about him, is false then one cannot give an account of how the person can be referred to, then the situation he is using is so rare that it is ridiculous to use it as an objection to any theory of anything. — andrewk

Kripke isn't arguing that it isn't rare that all of our beliefs about an item that we are making reference to non-descriptively (e.g. by means of a demonstrative device or of a proper name) are false. That might indeed be extremely rare. What he's arguing only is that however big or small the core of our true beliefs about this item might be, it's not this true core of beliefs that determines what the reference is. He's arguing against descriptivist theories that make some core of beliefs about an item necessarily true of this item in order that the believer might be making reference to it, or that make it necessary that there be such a true core.

See if I have this right...

Here the difference between reference fixing and reference determining would be that the former makes use of an otherwise inadequate description(one that is incapable of successfully picking out an individual), whereas the latter is making use of a purportedly adequate description, according to one who argues in favor of definite descriptions. — creativesoul

That sounds about right, with the caveat that there is nothing wrong or inadequate about referring to an item purely descriptively (and thus ensuring that our description of it is true, if the item uniquely exists) if this is what one intends to do. (Mathematicians often refer to mathematical objects purely descriptively). Kripke's main point is that reference by means of proper names doesn't work like that.

What he's arguing only is that however big or small the core of our true beliefs about this item might be, it's not this true core of beliefs that determines what the reference is. — Pierre-Normand

Where do you believe he argues that?

I don't see any argument on p25-6 to that effect. The only support he offers to his claim that a descriptivist must say that the reference is to the person uniquely satisfying the description (somebody of whom the speaker is presumed to be unaware) is another claim:

"This is the sense in which it's been used in the logical tradition."

I am dubious of that second claim (and he offers nothing to support it) but, even if it were true, that would not mean that it is essential to a descriptivist theory that one takes that interpretation.

I am dubious of that claim (and he offers nothing to support it) but, even if it were true, that would not mean that it is essential to a descriptivist theory that one takes that interpretation. — andrewk

Well, for sure, you yourself attempted to supply a more nuanced account. But there is some unfinished business above since I have claimed that your own proposed account still is vulnerable to Kripke's objections. Do you propose to amend it as the claim that, whatever set of true beliefs a speaker happens to have about the item she intends to refer to, this set determines descriptively what this item is? Or maybe you want to phrase it differently?

Where do you believe he argues that? — andrewk

It's been more than 15 years since I've read Naming and Necessity back to back. So, I must acknowledge that I may now tend to conflate some of Kripke's original arguments with those of other pragmatists/externalists who have followed in (some of) his footsteps, and expanded on his views, such as Gareth Evans, Hillary Putnam, Michael Luntley, Gregory McCulloch, John McDowell and David Wiggins.

Yes, my first stab at summarising my notion of a robust descriptivist position is something like this:

When a speaker uses a verbal description to refer to an object, they will usually* use properties that they believe belong to the object and that they believe are sufficient to uniquely identify it amongst the set of objects that might reasonably be considered candidates, given the context.

When a speaker uses a proper name to refer to an object, they will have in mind a description with the above properties.

Whether a listener understands the speaker and picks up the same object as the speaker intended will depend on:

- in the description case, whether that description uniquely picks out the same object as the speaker intended, given the listener's beliefs and the context;

- in the proper name case, whether the proper name leads the listener to pick out the same object as the speaker intended, given the listener's beliefs and the context.

I find the descriptivist account incomplete, but not for the reasons Kripke suggests. Rather, I believe it is often unhelpful to break up a speech act into parts and insist on identifying referents. It is often the case that the speech act has an intent that does not depend on the identity of the presumed referent, or even on whether there is a referent. In this case I think Wittgenstein is the one with the most realistic and helpful account.

*An exception would be when the speaker believes that the listener has a different set of beliefs about the object. In that case it would make sense for the speaker, in order to be understood, to use properties that the listener believes holds, even though the speaker does not believe they hold.