For the viewers, Tim apparently asserts that language is governed by conventions. The best rebuttal of that of which I am aware is Davidson's essay. I've used it before, it has been discussed at length.

Tim and Leon prefer to pretend it doesn't exist.

How is it cheap? Even charitable reviews of Davidson on this point allow that:

It is easy to balk at such a provocative statement. If there were no such thing as a language, why would there be linguistic discrimination
and even persecution, as Dummett (1986) remarks? Onthe face of it this seems grounds for an unconditional dismissal; all the more because
Davidson has fulfilled Lewis’s (1975) prophecy that only a philosophercould deny the role of convention in communication.1

However, there clearly is a qualification to this claim that demands attention

Andreas van Cranenburg - No Such Thing as a Language?

Which is exactly my point:

Prima facie, that's a ridiculous claim unless one runs back from the motte to the bailey in order to massively caveat it so as to make it an entirely different claim.

I can't really fault Davidson for doing this, because it's very much a "thing" in modern scholarship, stating radical theses for effect and then caveating them into something else, but it's still a pet peeve of mine.

Tim apparently asserts that language is governed by conventions.

Sure, what do you think people do when studying foreign languages or grammar?

Again, why don't you start with clarifying:

P1: Any phrase could be used as a password.
P2: ????
C: Therefore there are no languages to learn and linguistic conventions don't determine what words mean.

Why should I clarify an argument I haven't made?

Lewis' definition of a convention is like this:

Lewis analyzes convention as an arbitrary, self-perpetuating solution to a recurring coordination problem. It is self-perpetuating because no one has reason to deviate from it, given that others conform. For example, if everyone else drives on the right, I have reason to as well, since otherwise I will cause a collision. Lewis’s analysis runs as follows (1969, p. 76):

A regularity R in the behavior of members of a population P then they are agents in a recurrent situation S is a convention if and only if it is true that, and it is common knowledge in that, in any instance of S among members of P,

(1) everyone conforms to R;

(2) everyone expects everyone else to conform to R;

(3) everyone has approximately the same preferences regarding all possible combinations of actions;

(4) everyone prefers that everyone conform to R, on condition that at least all but one conform to R;

(5) everyone would prefer that everyone conform to R′, on condition that at least all but one conform to R′,where R′ is some possible regularity in the behavior of members of P in S, such that no one in any instance of S among members of P could conform both to R′ and to R. — SEP

Is that what you had in mind? Or were you thinking of convention as being the same as a dictionary?

But what if we actually spoke about assertions rather than circumlocutions that may or may not indicate assertion? What about:

a) "The cat is on the mat."
b) "I assert the cat is on the mat." — Leontiskos

They mean different things and have different truth conditions.

(a) is true if and only if the cat is on the mat
(b) is true if and only if I assert that the cat is on the mat

(b) can be true even if (a) is false.

Tim apparently asserts that language is governed by conventions. The best rebuttal of that of which I am aware is Davidson's essay. I've used it before, it has been discussed at length. — Banno

Quoting from here:

Davidson denies that conventions shared by members of a linguistic community play any philosophically interesting role in an account of meaning. Shared conventions facilitate communication, but they are in principle dispensible. For so long as an audience discerns the intention behind a speaker’s utterance, for example, he intends that his utterance of “Schnee ist weiss” mean that snow is white, then his utterance means that snow is white, regardless of whether he and they share the practice that speakers use “Schnee ist weiss” to mean that snow is white.

Assuming that this is an accurate summary, it seems to me that Davidson believes both that words and phrases have conventional meanings and that words and phrases can be used to mean anything a speaker intends. So yours and Count Timothy's positions might not be incompatible.

Thanks. That quote risks giving the impression that the meaning of some utterance is to be found in the intent of the listener or the speaker. That's not Davidson's approach. For Davidson, an utterance functions in a context by the listener interpreting that utterance in such a way as to discern the beliefs and intentions of the listener, on the presumption that the speaker and listener share much the same beliefs and are both rational.

Hence Davidson's account provides an explanation for how we are able to understand malapropisms, which by their very nature run contrary to the conventions of language.

Davidson might well say Tim's "words do have a stipulated, conventional meaning that relies on limited context, that is accessible to all speakers" is a useful fiction, but no more.

For me the interesting thing here is the comparison with Searle, who gives an excellent account of how conventions function in the construction of social reality. I don't see a strict incompatibility between Davidson's account of interpretation and Searle's account of the construction of social institutions. Paying that out would make an interesting thread.

As for Tim and Leon, from previous discussions I suspect they share a simplistic view of meaning as speaker's intent, although I may be mistaken. We won't find out, until they find a way to move past merely scoffing and actually address the discussion.

I don't see a strict incompatibility between Davidson's account of interpretation and Searle's account of the construction of social institutions. Paying that out would make an interesting thread. — Banno

I don't, either.

And it'd be interesting to try and combine the notions.

Thanks for that.

What is interesting here is the compatibility between Lewis and Davidson. They agree that the shared context includes both action (the utterance, in the main) and belief. For Davidson this is seen in adopting the Principle of Charity - thinking of your interlocutor as rational and mainly in agreement as to how things stand. Lewis depends on convention for this task. Both think of language as a way to assign truth functions to sentences. Both triangulating speaker behaviour and context to determine what must be true if the utterance is true.

The difference is perhaps that for Lewis convention is presupposed, while for Davidson it is secondary to the interpretation.

It is worth asking how Lewis might have dealt with malapropisms, were convention is often reversed.

A good rule of thumb is that no sooner is a convention proposed than that someone will show it's antithesis.

The key seems to be that conventions cannot be the basis of interpretation, but that this does not mean there are no conventions.

There's the somewhat trite analogy of Gödel sentences, that cannot be produced by the rules of a system.

No fixed set of rules (conventions) can capture all actual or possible uses of language.

Lewis' ideas are Quine-approved, especially the arbitrariness of conventions. I don't know how he dealt with malapropisms.

Yep. He and Davidson have that common background. They would presumably agree on some form of the indeterminacy of translation.

(I take that back - turns out Lewis was critical of indeterminacy... A new rabbit hole! Thanks again!)

The second prong of Davidson's triangulation requires ascribing intent to the speaker charitably assuming rationality and logic to the speaker.

Does ChatGPT satisfy #2?

If not, must we smuggle in internal state talk to maintain the distinction between humans and AI?

The second prong of Davidson's triangulation requires ascribing intent to the speaker charitably assuming rationality and logic to the speaker. — Hanover

Yes, although the point made above concerning the IEP quote applies here, too. Somewhat perfunctorily, the goal is not to expose the intent of the speaker, but to note the circumstances under which their utterances would be true.

So we have the meaning of a Chat reply if we have the circumstances under which it would be true.

In this regard, no intent need be attributed to Chat in the process of working out what it meant.

(This appears to be another argument against the speaker's intent theory of meaning...)

Somewhat perfunctorily, the goal is not to expose the intent of the speaker, but to note the circumstances under which their utterances would be true. — Banno

Is this correct though? I took the truthfulness of the statement to be the 3rd prong, not the 2nd. As in the "cat is on the mat" has meaning if (1) I believe the cat is on the mat, (2) I charitably infer your intent is to communicate the cat is on the mat based upon my assumption you are rational and logical, and (3) the cat is in fact on the mat.

The way a lack of intent affects meaning can be seen by imagining that you see a handwritten note with poem written on it, stuck on a wall in a bar. You ponder the meaning of the poem, but then someone tells you it was computer generated. That's when you realize you have a reflexive tendency to assume intent when you see or hear language. You may experience cognitive dissonance because the poem had a profound meaning to it, all of which was coming from you.

The problem with using ChatGPT is that it's processing statements that were intentional. It's not just randomly putting words together.

The problem with using ChatGPT is that it's processing statements that were intentional. It's not just randomly putting words together. — frank

I'm not sure what you mean here. The sentences created by ChatGPT are truly compositional it would seem. That is, they are not just the random slamming together of simple words into sentences or the combination of preset sentences into paragraphs. Davidson often refers to "concatenation" which identifies the ability to create infinite sentences from finite words.

Explaining how oncatenation comes to be is a major part of his project. That is, how does meaning emerge as a sum greater than its parts.

I think it's a hard argument to make that ChatGPT is just an arranging finite elements into finite sentences. It appears to compose, to concatenate.

This ties into Davidson"s resistance to convention being the primary driver of meaning. Intent of the speaker is demanded, which pulls ChatGPT out from producing meaningful statements.

If that is the result, I wonder if AI disproves triangulation. AI under his theory speaks without meaning, yet I feel I understand what it means. But, should I say its lack of intent erases its meaning, am I not just demanding the secret sauce of consciousness into the equation? If that, he becomes just another dreaded metaphysician.

I think it's a hard argument to make that ChatGPT is just an arranging finite elements into finite sentences. It appears to compose, to concatenate. — Hanover

I agree. I think it's crunching data that's made out of intentional human content.

This ties into Davidson"s resistance to convention being the primary driver of meaning. Intent of the speaker is demanded, — Hanover

He's saying that the expectation of intent goes into calculating meaning. He's not saying the listener actually knows the speakers intent. @Pierre-Normand do you agree with that?

He's saying that the expectation of intent goes into calculating meaning. He's not saying the listener actually knows the speakers intent. — frank

I agree with that. The interpreter applies the principle of charity to assume the speaker's rationality and logic, which assumes consistency of usage, but he's definitely not admitting to a tapping into the speaker's internal state.

But assumption of intent is demanded, else it would be a simple conventionalism.

But assumption of intent is demanded, else it would be a simple conventionalism. — Hanover

Right. As with the computer generated poem, realizing there's no intent undermines meaningfulness.

Davidson intends his approach to be extensional, so he avoids intentional contexts. Hence intention plays little part in his approach.

The three points of the triangulation are speaker, interpreter and shared world.

The interpreter surmises a sentence S such that the utterance of "p" by the speaker will be true if and only if S. S is confirmed or adjusted on the basis of ongoing empirical evidence.

There's no appeal to internal meaning or intention - doing so would result in circularity.

Intent might be inferred post hoc.

there's no appeal to internal meaning or intention - doing so would result in circularity. — Banno

We must charitably assume the speaker is rational and presents his statement accurately to intent. This makes no demand upon deciphering internal thoughts, but if we dispense with linking what he meant with what I understand, it deflates to Wittgensteinian meaning is use.

"To understand the speech of another we must interpret in a way that makes most of his utterances true and rational, given the totality of what we take to be his beliefs, desires, and intentions."
“Radical Interpretation"

I take this as requiring us to construct intent from behaviors but also coupled with an assumption of internal coherence and rationality.. We're not getting into the speaker's head, but we are assuming intent.

Sure, if you like. We can drop "intent" from that without much loss.


Added: we don't much need the bit about inferring some intent on the part of the speaker. We can do so, but it's not needed. Meaning here is not the intent of the speaker. Speaker meaning is something else.

That'll cause some folk no end of confusion. It shouldn't. It does not imply that the speaker does not have an intent.

we don't much need the bit about inferring some intent on the part of the speaker. We can do so, but it's not needed. Meaning here is not the intent of the speaker. Speaker meaning is something else.

That'll cause some folk no end of confusion. It shouldn't. It does not imply that the speaker does not have an intent. — Banno

Nor does it imply that there aren't cases where speaker intent is very important. I think the Chatbot example is such a case. The program itself can't be said to have intentions, thought the sentences it produces have meaning. But the intention of its programmers, as best we know, is to impersonate intention on the part of the program. This of course takes "intention" to a different level, but that's my point.

Hence Davidson's account provides an explanation for how we are able to understand malapropisms, which by their very nature run contrary to the conventions of language. — Banno

I’m a little confused. If malapropisms “by their very nature run contrary to the conventions of language” then there are conventions of language. So the very existence of malapropisms is proof that there is a (conventionally) “correct” way of speaking (else nothing could be a malapropism).

In responding to @banno and your comments here, is that intent is a necessary component in Davidson's triangulation theory. This does not mean that we look into the heads of the speakers to decipher intent, but we have to ascribe it to the person based upon our assumption that they are rational and logical. "Ascribe" is the operative word, where we assume it and place it upon the speaker, but we don't pretend to know specifically what the intent is, but we do know there is an intent, but it's a black box.

Should Davidson not hold that way, he woudl lose the foundational element for meaning to exist and he would blur into a "meaning is use" position. His position is different than Wittgenstein, although he very much rejects private language and mentalese sorts of claims. I get why there is pushback against anyone who tries to oversstate the intent requirement and tries to turn Davidson into a metaphysician when it comes to understanding meaning, but I think the opposite problem arises when someone tries to ignore the importance of the intent for his theory.

I’m a little confused. If malapropisms “by their very nature run contrary to the conventions of language” then there are conventions of language. So the very existence of malapropisms is proof that there is a (conventionally) “correct” way of speaking (else nothing could be a malapropism). — Michael

I can't imagine a language that lacks some degree of conventionalism and I'm not sure anyone holds that. There must be rules to a langauge even if you have full buy in to an internal mentalese. The question is whether it's entirely just a rules based language game or whether you're trying to find some other foundational structure. That's my point directly above related to Davidson's need to rely upon ascribing intent else he would just be a conventionalist.

This does not mean that we look into the heads of the speakers to decipher intent, but we have to ascribe it to the person based upon our assumption that they are rational and logical. "Ascribe" is the operative word, where we assume it and place it upon the speaker, but we don't pretend to know specifically what the intent is, but we do know there is an intent, but it's a black box. — Hanover

Yes, this is all fine, though "black box" might be overemphasizing the inscrutability.

That's why the Chatbot example seems relevant. We do not have to "ascribe intention to the [person] program based upon our assumption that they are rational and logical." Such a (false) assumption is the "impersonation" I'm instead ascribing to the programmer. This seems right in line with Davidson, because even by ascribing no intention to the program, we're able to explain the meaningfulness of its outputs by deferring that ascription back to the programmer -- again, without needing to be able to say specifically what these intentions are.

Is this analogy too simple?: It's like holding up a puppet and pretending it's "talking" and "having intentions." Every child knows this only a game, an impersonation.

This seems right in line with Davidson, because even by ascribing no intention to the program, we're able to explain the meaningfulness of its outputs by deferring that ascription back to the programmer -- again, without needing to be able to say specifically what these intentions are. — J

I don't think that works because Davidson speaks often of concatenation, which is the placement of a finite number of words into an infinite number of sentences. That is, we compose sentences of different meanings based upon the words used. AI composes from its database, which means the sum is greater than the parts. There is no programmer out there, for example, that went through and intentionally answered whatever question you might pose to ChatGPT. In fact, AI can create a program, which can create a program, which can create a program, etc. Suggesting you can, through the principal of charity, assume a rational and logical intent based upon a programmer's program 20 generations ago who had no idea of the data within the massive internet database seems quite a stretch to define Davidson as defining AI speak as meaningful language .