To understand a word is just to be able to make use of it. — Banno

This cannot be right. We know from many medical conditions that some people can understand words without being able to make use of them.

Broca's aphasiacs, for example, retain use of a limited vocabulary, but this does not hinder their understanding of a much broader vocabulary:



Notice how he understands the word "sixty" perfectly well, but cannot use it.

One of my favorite strands in metal lyrics is the paranoia over immanent nuclear destruction. It's died down in recent years, but it was great.

OK, I see.

I agree, that is very much at odds with Chomsky. If I understand correctly, Chomsky does see language as an individual, biologically ingrained cognitive capacity, which is only 'accidentally' externalized for communication and social functions.

So reading your post, is your position, and Dor's, that language had some core initial evolutionary function that involved this kind of transmission of experience, and then only later took on its other functions? Or is there something other than evolutionary priority that makes this function the 'core' one in some way?

I haven't read the book, but I'm confused in what sense this is a new approach. Hasn't this idea been the stock view since at least Aristotle?

With all due respect, this quote belies the first claim:

His objection appears to be that the definite description 'Godel' in the sentence 'Godel was a great mathematician' has an ambiguous referent, and that is a fatal flaw in the descriptivist approach. — andrewk

There is no way even a cursory reading of the book (or a summary of it!) could give you anything like this impression, so either you didn't read it or have no memory of it. The basic terminology isn't even right: 'Godel' is not a description, but a proper name (a description is a determiner followed by predicative material, like 'the boy' or 'a cat').

As far as arguing in good faith, continuous refusal to look at the examples given you after you ask for them, then a wholesale dismissal of an entire approach, is not a token of good faith.

Anyway, you're right, this isn't going to lead anywhere, but I thought this shouldn't go unaddressed.

I'm sorry to hear that you feel that way. — andrewk

Don't "apologize" to me – read the book, or else inquire in good faith, when discussing a topic in the future. The advice is for your benefit, not mine.

I concede that Kripke's approach may well be useful in metaphysics. Where I don't see it as being useful is in relation to language. To me it seems to bear almost no relation to the way people actually speak to each other, or how people learn to speak. — andrewk

Nobody is going to stop you from having whatever opinion you want, but your opinion is ill-informed, and it's frustrating that when confronted with this fact, you continue to choose to double down on your misinformation and ignorance of the topic rather than engage with the material you're attempting to criticize. This makes me think the criticism is not entirely in good faith.

There can't be any criticism without understanding – your current qualms don't even rise to the level required to have a problem with Kripke, since all evidence from your posts points to you not knowing what he said. You don't know what a rigid designator is, and you seem not to know what the function of possible worlds in the modal logic is. You've also repeatedly asked for examples of the notion's empirical application, and when given them, have chosen to double down and ignore. It's not great sportsmanship to waste other people's time when you have no intention of trying to figure out the material.

Ah, I see. I mistook you as meaning that the point was to 'foster understanding' in the sense of a linguistic community coming into some kind of self-knowledge as to its own practices. That, I think, is also silly – philosophy of language is an abstruse discipline of no more general interest to the populace than biology, as it should be.

The purpose of theorizing isn't to provide a blueprint for perfect use. It's rather to foster understanding. Hence, that theories about language use are bound to be incomplete means no more and no less than that our self-understanding, qua language users, is bound to be imperfect. — Pierre-Normand

It's not to provide a blueprint for use at all! It isn't as if we are trying to teach people how to use language 'right.' Language is a natural phenomenon, like digestion, that can be objectively described as to its workings.

That a theory is bound to be incomplete is not an injunction against theorizing. That is a very silly thing to think.

Do you think there is some sort of opposition between describing language as it's used, and describing it using notions like rigid designation?

As if we did not use names as rigid designators!

I'm not seeing the point. This is precisely what we're talking about anyway.

The same way they're able to digest food without being able to offer one. Theory about how something that humans do works has nothing to do with whether they themselves have such a theory or understand it.

If I don't know the author of the book or Trump's wife then I won't know if referring to them as "Steve" or "Melanie" is a mistake or not, — Michael

Whether you know it is a mistake or not has nothing to do with whether it is. Your view of language is solipsistic. The fact is, it is a mistake. That is not their name.

Surely you can refer to my brother using the name "Andrew" even if his name isn't Andrew. — Michael

If I did that, I would be using the wrong name, and so making a mistake. I would fail to refer to him conventionally, but if someone understood what I meant, they might recover what I meant (and in this sense failed) to say, and so I might succeed in referring in the sense of conveying who it was I wanted to talk about.

It doesn't make a difference if the wrong name is intentional or mistaken. — Michael

That's right – it's an objective linguistic mistake, either way. Hence, the wrong name.

It depends on what you mean by 'refer to.' Clearly we can reconstruct who a person means to refer to using a word, and so in that colloquial sense, since their intention succeeds, and we know what they intended to convey, if we are charitable they do in fact convey this, and so in this sense they referred to someone.

However, to leave it there is to be overly coarse – there is another sense in which they failed to refer to anyone, as can be seen when an interlocutor does not choose to be charitable, and says 'who the hell is Melanie/Steve?' If you do not make this distinction, then you can in fact make no sense of the simple fact that the speaker used the wrong name. For if we refer to whoever we intend to refer to simpliciter, in what sense are there ever wrong names, so long as the speaker's intensions are clear to himself?

Not quite. My claim is that if they said that because they believed Godel did the Incompleteness Theorems and that's all they knew about Godel then their intention was to praise the person who wrote the incompleteness theorems. — andrewk

Of course their intention is to praise the person who wrote the theorems. That is why they pick a name that refers to a person who wrote them. The point is, they're wrong about who wrote them, and so accidentally refer to the wrong person.

Trying to classify natural language statements into two boxes - right and wrong - is way too crude. — andrewk

In some cases, maybe. But not in this one: clearly, I was wrong. I don't think this is reasonably disputable.

What about the real life version of this? Substitute Shakespeare for Godel and Francis Bacon for Schmidt. WHat do I mean when I say I love Shakespeare. Do I mean I love whoever wrote the plays attributed to S? — andrewk

Presumably, what you say when you say that you love Shakespeare, is that you love Shakespeare. This is the most obvious and best hypothesis; why you find the alternative, that when you say you love Shakespeare you say that you love someone other than Shakespeare, is a bit mystifying.

If you think that Adam's pen name is 'Steve,' and you try to refer to Adam using 'Steve,' then you have tried to refer to Adam, but messed up. Anyone who knows that this is not the pen name, but who knows who you were trying to refer to, can be charitable and recover your intention.

It’s the very thing being discussed. — Michael

It is not. Would you like to discuss, instead, the issue of how to determine, when using a name, which bearer of that name is meant?

It isn’t obvious that I’m referring to the author’s brother. — Michael

The position that, when you say Steve is the author, you are referring to someone other than Steve, is ludicrous.

You may, of course, have meant to refer to someone else, and made a mistake. And other people, on understanding your mistake, can understand what you meant to say, and so catch your drift. But that is neither here nor there; in such cases, that a mistake has been made (such as using the wrong name) is as much a fact as anything else.

Why him and not any other Steve? — Michael

Do you want to talk about this now instead? This is an orthogonal issue.

Which Steve? There are lots of Steve’s in the world. — Michael

The one you posited in the very example you gave.

So somehow your intention when you tell me that Steve is the author fixes the referant of the name “Steve” when I then tell someone else that Steve is the author? How does that work? — Michael

No, "Steve" refers to Steve. This really is not hard. There's no transmissions of intention-fixing. The name has a conventional referent.

And what if at the same time someone who knows that Adam’s name is Adam and that Steve’s name is Steve but who falsely believes that Steve is the author tells me that Steve is the author? When I then say that Steve is the author am I committing your mistake of referring to Adam using the wrong name or the other person’s mistake of referring to the author’s brother? — Michael

If you say Steve is the author, you have said that Steve is the author, not someone else. This is obvious.

But how have I come to refer to someone other than the author of the book? — Michael

Because you referred to Steve, who isn't the author of the book.

What is it about the name "Steve" that makes its referent someone other than the author? — Michael

Because "Steve" refers to Steve, who isn't the author of the book.

This is why my follow up question was regarding the situation where you know that Adam is the author but incorrectly believe that his name is Steve. In that situation, are you and I referring to the author (using the wrong name) or the author's brother when we say "Steve is the author"? — Michael

In that sort of situation, I'd say you attempted to refer to Adam, but made a mistake, but if anyone catches your mistake, they can realize what you did and recover what you meant to say.

When I say "Steve wrote this book", am I referring to Steve or to Adam (albeit using the wrong name). I believe andrewk is saying it's the latter. — Michael

Steve, obviously. You have a mistaken idea about who wrote the book, and so said something about the wrong guy. You intended to say something about the author of the book, which is why you used the name of the man you thought was that author. But that doesn't mean that, because you were mistaken, the universe magically rearranges so that you said something else. What you said was, that Steve wrote the book.

To suggest otherwise would be, ludicrously, to imply that 'Steve wrote this book' was a necessary, and so trivial, truth – of course the guy who wrote the book wrote the book. But that is not what you said. You said that Steve did, and he didn't.

Is your claim that if someone says 'Gödel was a brilliant mathematician,' but if it turns out that Schmidt came up with the theorems, then what they said was that Schmidt was a brilliant mathematician?

Suppose Gödel was a fraud, and I say the above sentence. It turns out he is a terrible mathematician, and stole all his work from Schmidt. Was I right or wrong about what I said?

My response to that is firstly that Witt's approach, which I prefer, would be to look at the whole speech act and its context, who is speaking, what they know about Godel and the theorems that are attributed to them, and infer from that whether the person was referring to the author of the Incompleteness theorems (Schmidt), or to the Godel that ended up at Princeton, and to whom those theorems are attributed. — andrewk

The point is that 'Gödel' refers to Gödel, not to Schmidt. It doesn't matter who did the completeness theorems.

Have you read Naming and Necessity?

Teachers already know that errors or falsehoods are rarely found in homework — StreetlightX

I wish!

The question 'what do the words "Donald Trump" mean' is malformed, because meaning depends on context. — andrewk

This simply doesn't follow. There is no inference from 'meaning depends on context' to 'there is no answer to the question of what words mean.'

One cannot usefully discuss what the words mean without a context. — andrewk

Yes, one can; words have context-independent conventional meanings as well, which include in part schemata for determining what particular use they will be put to in a context. It is in fact only because they have such meanings that they can be usefully be employed in context. It is not as if in every context we begin ab initio, beginning with zero context-independent knowledge about the language, trying to figure out what someone means. This would be Humpty-Dumptyism. What's meant in a context is always constrained by stable meaning in the language.

Both R's and W's theories recognise context-dependence. — andrewk

This is irrelevant, since literally no one that I'm aware of does not recognize context-dependence.

Indeed there is no apparent difference between the example here and Russell's one about 'Bismarck' vs 'the current chancellor of Germany' (or 'Walter Scott vs the author of Waverley'). — andrewk

The point is that there is a huge difference between proper names and definite descriptions because one is generally rigid while the other is generally not, which has a number of consequences for all sorts of modal environments.

Put simply, what philosophical problem or question is Kripke trying to solve? — andrewk

I've already talked about this, but have you read Naming and Necessity?

I'm afraid I'm still not seeing how Kripke's approach aids understanding of the use of language in those linked examples. — andrewk

What do you not understand? I can't help unless you pinpoint some area of difficulty. You asked, and I performed; if you are not arguing in bad faith, it is up to you to engage rather than dismiss.

Is it perhaps just a matter of preference — andrewk

No.

where R's or W's doesn't, and vice versa for others? — andrewk

There is no 'Kripke's approach' as opposed to 'Russell or Wittgenstein's approach.' Wittgenstein didn't know anything about the issues Kripke was talking about, so the comparison is anachronistic. As to the specific issue being discussed here, where Russell discussed (non-logical, i.e. ordinary) proper names at all, he seemed to take them to be abbreviated descriptions. This is wrong for the sorts of cases Kripke talks about, and examined in the post I linked you to. For instance, if 'Donald Trump' means 'the 45th president...' etc., or some other description, then the predicated behavior of the name in all of these modal environments is wrong, empirically, not as a matter of personal taste.

Can you provide an example where Kripkean analysis helps to understand a speech act in a useful way, that is not available by a different approach, in particular a psychological approach, which is what I see Wittgenstein's as being.. — andrewk

See this post:

https://thephilosophyforum.com/discussion/comment/211811

Kripke's hypothesis explains the behavior of proper names in a wide variety of intensional contexts, in a way that theories that he was repudiating do not. This was just at the time that compositional semantics was getting off the ground, and so prior to this (including in Wittgenstein's day), people weren't sensitive to these kinds of issues. It's part of a real advance in our understanding of language that we now are.

I don't agree that our ability to express is limited by the literal meanings of words. We all frequently make interpolations to cover when people say things that literally make no sense, but where we are able to divine by context what the aim of their speech act was. — andrewk

The latter does nothing to illustrate the falsity of the former.

Our brains conduct an amazingly rapid search of all these possible patterns of connections when we hear a speech act, to come up with an interpretation of it in a fraction of a second. — andrewk

I am suggesting to you that the processes by which this happens are something worth studying, rather than throwing our hand sup in the air and saying 'we don't need to worry about meaning – it all happens somehow.'

Part of this will include thinking about the meaning of proper names. And empirically, proper names are rigid designators (in these sorts of positions – elsewhere they're not, but Kripke didn't think about those cases). I don't care what you ultimately end up saying. That names behave like rigid designators is an observable fact, and the language use is such that we can just arbitrarily declare them not to.

Sure, when we analyse a speech act from a Wittgensteinian perspective, we will pay close attention to any parts that look like a proper name. But we don't need to get hung up on consideration of the metaphysical meaning of the proper name. All we need ask ourselves is why did the speaker use that word in that part of the act, and what effect were they aiming to achieve. — andrewk

Because the way people use speech acts and intend to do things with them is importantly tied to the conventional meanings of the words involved. It's not as if we just string together words in any random order and mean whatever we want by them. What we do and can mean is limited by the literal meaning of the words. And most often, we are competent with using words in this way with their literal meanings, without having any clue of the vastly complex mechanisms by which we do so – so of course 'our intentions' do not just transparently control whatever we want to say, and it's a silly enterprise to just 'all-at-once' divine the meaning of speech acts as if there were no systematic mechanisms by which they operated.

Rigid designation is a useful concept because this is how proper names behave with respect to modal operators. It's just a fact about the world, re: language use.

Do you see the meaning of a term as having some kind of abstract existence? Or do you see proper names as having a different kind of meaning than other terms? — Dfpolis

There are any number of things you could be talking about with 'meaning.' I take the literal meaning of an expression just to be the role it plays in composing the truth conditions of the sentence of which it's a part. Words have conventional literal meanings in virtue of their use, and there's no trouble with people using words whose meaning is in part opaque to them. Words do not mean whatever people think they mean, etc. If I say that John is a fool, but didn't realize that John was Mr. Smith, I called Mr. Smith a fool. I did this by mistake, but nevertheless this is what I said.

Note also that when we don't realize that two words co-refer, there is a restricted sense in which we don't fully use them correctly. Presumably there is some vantage point from which, viewing the individual, we'd refuse to appropriately apply the name. Again, this is due to our ignorance.

I would say that, in normal intercourse, the contingent meaning is the literal meaning. In fact, until the contingent meaning is grasped, the trivial necessary meaning cannot be grasped. — Dfpolis

To someone who knows that 'Donald' and 'Mr. Trump' are co-referential, and has no need to access this other meaning, saying something like 'Donald is Mr. Trump' will indeed sound bizarre because it is trivial. The literal trivial meaning is therefore perfectly accessible – but we can also employ this trivial literal meaning for other ends.

Note also that literal meaning and 'primary' meaning, in the sense of the foremost meaning we intend to convey, are not the same thing. It can perfectly well be that the primary purpose of our utterance isn't to assert the literal trivial meaning. The literal meaning is the conventionalized meaning of the sentence in virtue of how the literal meaning of the words compose. This literal meaning can then be employed for any number of purposes, only one of which is to inform the addressee of that literal meaning. In cases where the literal meaning is trivial, we generally have other purposes for stating such sentences.

You have not commented on my claim that proper names need not refer to individuals simpliciter, but to individuals as known -- because covert guises do not elicit the idea expressed by the name. — Dfpolis

I don't see any reason to believe this. It's true that we are sometimes unaware of what names mean, and can't tell that two names denote the same person. But this is already taken care of if names denote individuals – it's not as if the names themselves bake into their meaning facts about whether you know who is who. That we can be confused about who is who, and which words refer to what, is already a trivial fact about language use when names refer to individuals anyway.

So, I do not see how it is possible to say that what we learn "outruns" the literal semantic content, when what we learn (the intentional state elicited) we learn from the literal semantic content. What cpuld you possibly mean by "literal semantic content" other than the intentional state that a literal reading elicits? I — Dfpolis

Consider the following example. If I know some Spanish, but am not fluent, and I hear someone say "El gato está durmieno," clearly referring to a cat, then I might learn from the fact that this sentence was truly uttered that the word "gato" means the same as the English "cat."

Does this sentence literally mean that "gato" means "cat?" No – it just means "the cat is sleeping." However, from the true utterance of that sentence in that context, I learn something other than the literal content of the sentence, viz. something about how the words used to express its literal content are used.

Likewise, even if "Donald is Mr. Trump" literally expresses a necessary truth, the contingent proposition that I learn (that these two words refer to the same individual) is as a result of realizing that this necessary proposition is literally expressed by those words. And indeed in saying such a thing, my primary intention may to to impart this information, not the (trivial) necessary proposition.

Certainly part of what is communicated by the statement, in the broadest sense, is that the man has a certain name, or that the two names refer to the same individual. But this does not mean that there are two readings of the sentence. All we need is the ordinary reading, on which it states a necessary identity. However, the cognitive significance of a sentence, i.e. what we're capable of learning from the fact that the sentence expresses a true proposition, outruns its literal semantic content. From the fact that this necessary proposition is true, as expressed by these words in this context, we learn a contingent truth, e.g. that the two words refer to the same thing. For if they did not, then the proposition expressed by those words could not have been true. Therefore accepting that the sentence is true teaches us a contingent truth, but not because the literal content of the sentence expresses a contingent truth.

The question, then, is whether or not there is an analogous de dicto reading of "water is H2O" such that a counterfactual chemical composition of water is not a contradiction. Given that you've accepted that it's possible that scientists are mistaken in their claim that water is H2O, isn't that an acceptance of such a de dicto reading (else it would be as incoherent as suggesting that we could be mistaken in thinking that Donald Trump is Donald Trump)? — Michael

The mistake would amount to thinking falsely that 'H20' and 'water' refer to the same thing. Admittedly H20 is a terrible example, since it's not a canonical proper name, but a kind of description of the chemical composition of a compound.

So take a proper name example. Suppose you know about Donald, and about Mr. Trump. As far as you know, they could be the same guy, or they could not be. Scenario 1: they are, which means that 'Donald' and 'Mr. Trump' mean the same thing, and in learning that Mr. Trump is Donald, you learn that the words corefer (and that the sentence 'Mr. Trump is Donald' conveys a necessary proposition). Scenario 2: they aren't, which means that the words don't mean the same thing, and the sentence expresses a necessary falsity. Now you don't know which scenario you're in, so it 'could be' either. But this has to do with whether the words co-refer or not, which is a contingent matter. Given that they co-refer or not, they express necessarily true or necessarily false propositions. That you are unaware of which it is, and that this depends on the meaning of the words, is where the feeling of contingency comes from. For it is contingent whether the sentence expresses a necessarily true or necessarily false proposition.

Here are some environments you can use to test for rigid designation across possible worlds:

Attitude contexts, such as belief reports: with non-rigid designators, these tend to have de dicto/de re ambiguities. With rigid designators, there tends to be no such ambiguity.

So if we say 'John believes that the president is a fool,' then this has two potential ways of being construed. First, it means that the actual president, Trump, is such that John believes him to be a fool. This may be true, even if John does not know that Trump is the president.

On the other hand, John might believe that the president, whoever it is, is a fool – he may believe this even not knowing that the president is Trump. The idea here is that if you look at all the worlds 'compatible' with John's belief, in each such world the president at that world, whoever it is, is a fool. But Trump is not necessarily a fool in all of them.

Notice that the same construction with a proper name lacks this second reading: 'John thinks that Trump is a fool.' This can only mean that John thinks of some individual, namely Trump, that he is a fool. Which individual he attributes the quality to does not 'shift' across worlds compatible with his belief.

–––––

Counterfactuals: 'If the president were a fool....' has a reading on which we're to imagine that the president, whoever it is, not necessarily Trump, is a fool. Compare 'If Trump were a fool...' no such reading. We are only bid to imagine scenarios in which the one individual picked out is a fool.

–––––

Modals: 'The president must be a fool.' This can be said even if one isn't sure who the president is, but that whoever it is, potentially different people in different scenarios, that person is a fool. 'Trump must be a fool' can't be used this way – it can only mean that this one guy, Trump, must be a fool (whether he is president or not).

–––––

Questions: 'Is the president a fool?' This may be asked felicitously even if the asker does not know who they are asking about. They simply want to know whether the president, whoever that happens to be, is a fool, and so not necessarily any information about a single individual. 'Is Trump a fool?' cannot be read this way.

We also refer to the same thing using the words "Donald Trump" and "the 45th President of the United States" but don't say that if Donald Trump wasn't the 45th President of the United States then he wouldn't be himself. — Michael

That's exactly the point. 'The 45th president....' etc., is not a rigid designator.

So how do we determine which words are rigid designators – which words refer to the same thing in all possible worlds – and which don't? — Michael

It's an empirical question – you run the arguments, and see whether the denotation of the term shifts beneath modal operators, for example. So, 'Donald Trump might not have been the 45th president' is unobjectionable, but 'Donald Trump might not have been Donald Trump' sounds like a contradiction. This is because, so the argument goes, 'the 45th president' picks out relative to a world whoever is the 45th president, whereas 'Donald Trump' just picks out the same guy regardless.

A possible world where the 45th president (in that world) is non-identical to Donald Trump is unobjectionable, but what we're supposed to imagine as to a world where Donald Trump is not Donald Trump, i.e. not himself...this is less clear.

Your answer before was that it's just a matter of stipulation. We stipulate that the 46th President in a counterfactual scenario is "the same" Donald Trump who is the actual 45th President, so why can't we stipulate that H2O2 in a counterfactual scenario is "the same" water which is actually H2O? There's this implicit premise that some properties are "essential"1 to a thing's identity and others contingent, but how do we determine which properties are which? You say that the chemical composition of water is an essential property, but perhaps that the material of the Taj Mahal isn't (and so it could have been made of wood, even though when we refer to the Taj Mahal we're referring to a building made of things other than wood)? — Michael

No, we are not stipulating anything – that is how the language works, independent of our desires. If we say 'suppose Donald Trump were...' then because 'Donald Trump' is a rigid designator, we are already talking about how that very guy would be in some alternate situation. There is no question of 'how to determine' which guy in another world is him. The point is that proper names already pick out the same person 'across' possible worlds, so to talk about people using proper names in counterfactual scenarios is already to talk about one person as they might have been in various scenarios.

This is not what happens with definite descriptions. Again, we don't stipulate this – this is simply how definite descriptions work, independent of what we say about them. 'Suppose the 45th president were...' does not entail that we're talking about the same person across situations (on the relevant de dicto reading).

Because the fact that we might not know what words mean already solves this problem, and the Fregean sense, if it is something more than this, adds nothing to the explanation. The literature on Fregean senses is very confused, and the notion is ill-defined and has led to a lot of pointless back-and-forth.

There are in fact at least three things that Frege might have meant by a 'sense,' including Carnapian intension, Kaplanian character, and Stalnakerian information about the meaning of a word given that it was uttered in some context or other. We already have notions for all these things; introducing 'Fregean senses,' which have never been adequately delineated anyway, doesn't help.

Yes, but that much is obvious from the fact that words mean things, and what they mean might be opaque to their users. Invoking a dubious notion like Fregean senses is probably not a good idea, unless this obvious fact is all one means by it.