This doesn't make any sense. You can't just say that something is true or not without knowing what's being said and what the non-linguistic situation is like, and you haven't given an example of something you can say, since you've only said "X," and there is no explained situation, either.

If I say "It's raining, but John thinks it's not," I can't just say that's true or not – it depends on whether it's raining, and whether John thinks it's not. Surely you agree that can be true, in such a situation, right? Now suppose we're in such a situation: it really is raining, and John really thinks it's not.

Now suppose John hears me say this and says, 'that's right!' He is agreeing with what I just said, no? And so affirming that it was true? But it's weird for him to do that, even though it wasn't weird for me to say it.

Note that this example illustrates the same point without John himself having to make any such statement of the form you cite.

I don't know what more to say. If the statement "It's raining but John thinks it isn't" is true, then it's raining but John thinks it isn't. If John's statement "It's raining but I don't think it is" is true, the John must think it's raining and think that it's not raining; or perhaps know it's raining and think it isn't raining. In what circumstances would that be true?

John didn't say that he thinks it's raining. What he said was that it is raining, but he thinks it isn't. And that is true when it's raining but he thinks it isn't.

The point is that the sentence is still odd because he commits to believing that it's raining in virtue of asserting that's it's raining. But that isn't part of what he said.

So if John said, for whatever weird reason, 'It's raining but I don't think it is,' then if it really is raining and he really doesn't think it is, then I'll say, yeah, he's right, despite the fact that what he said was really bizarre. To say that what he said was true is the same thing as to say that 'It's raining but John thinks it isn't' is true, because that was what he said.

Yes, it would be bizarre for him to say that. That is the point. But what he said was just that it's raining but he doesn't think it is. I can say that just fine, and it can be true, no problem. Yet when he says the same thing, even in the same circumstances, it's bizarre. That is the point.

I've been keeping in my back pocket an example of this that comes straight out of speech act theory.

Suppose the US has information that some very high level Canadian official, who's traveling through a US airport, is connected to a terrorist group, and for whatever reason Canada can't change his status in time for the US to apprehend him. You could imagine the President declaring Canada a terrorist state to provide a legal basis for the action, and announcing it by saying, "I do not in fact believe Canada to be a terrorist state and will revoke this declaration in about three hours, but for legal reasons I am officially declaring Canada to be a terrorist state."

According to Moore, in the first case there's a mistake. In the second case, there is no mistake. — Ciceronianus the White

Well, Moore is dead wrong. He offers an example of one being mistaken and another pointing that out, while it's happening, and then goes on to ask why one cannot say the same things about themselves, which is to ask why someone cannot point out their own mistake while it's happening.

That is exactly what "It's raining, but I do not believe it's raining" is doing(would be doing anyway) when one is pointing out their own mistakes from afar, like another.

Aside from this, I think that you and I are largely on the same page. The statements about the weather and my belief are true(if it's raining but I do not believe it) when spoken by another, but they are self-contradictory whenever they are spoken by me(the mistaken one).

Self contradictory statements are not true, cannot be true. They are meaningful. They must be. Contradictory statements cannot both be believed at the same time. They can be reported upon, shown as such, but they cannot both be believed at the same time. In order to report upon our own lack of true belief about the weather like another does, we must point out the mismatch between what's going on, and the fact that we are oblivious to what's going on; those facts/events/happenings/ongoings/etc.

If we are oblivious, and we certainly are during such times, then we cannot report upon that because there's no difference, in our own minds, between what's going on and what we believe about that. That's why they cannot both be believed.




When viewing a visual recording of ourselves being unexpectedly surprised by rainfall, we'll watch what others did at the time it actually happened. We'll watch ourselves walk right past the coat closet. Yep. We did not believe that it was raining outside. That's where we keep our umbrella.

Hanging on a hook securely attached to the outside of the sidewall. We walked right past. From our vantage point, we can know with utmost certainty that we did not believe it was raining outside.

The difference between the two accounts of our own mistakes, one being reported by another, and the self-reporting, is that another could inform us of the weather outside ensuring that we grabbed our umbrella thereby successfully avoiding unexpected rainfall.

But we cannot convince ourselves to grab the umbrella out of the coat closet on the way out the front door. That is true regarding both timeframes, now and then! It's also true at each and every individual point in time in between...

If you have the patience for it, I'd like to set aside my position and just examine yours for now. My goal at the moment is just to understand your position better than I do now. I'm not even looking for arguments against it, though I will have questions.

Here's how I understand your position:

(1) The sentences 'Dewey has defeated Truman' and 'I believe Dewey has defeated Truman' mean the same, have the same use.

(Thus Moore's paradox is only an apparent paradox; the speaker who utters 'Dewey has defeated Truman, and I do not believe Dewey has defeated Truman' is uttering a contradiction.)

The idea behind (1) is that 'Dewey has defeated Truman' is really 'I believe that Dewey has defeated Truman' but elides or suppresses the 'I believe ...' It is nevertheless an expression of belief, or of purported belief, if we have to account for insincerity at some point.

This is so because we can only talk about the world as we understand it, and we understand it entirely in terms provided us by the mental model of the world our brain constructs and continually updates.

Our words refer to artifacts of this model, not to objects in the world.

Everything we say is, in a sense, a 'report' on the latest iteration of the model we are aware of.

If I have this right so far, this is where I'm a little unsure. Certainly, as you've said, there are those occasions where we might speak about our beliefs as an outsider would, observing ourselves as an object, but this is not the usual case, so 'report' there sounds a little wrong. We want a clear way of saying that the sentences we utter are informed by the mental model, entirely dependent on it, but for speech acts to be something other than comments about it.

Certainly we'll continue from here with a pragmatist account of speech acts, but we have to backtrack a little because 'belief' is being used in two ways here. On the one hand, it's kind of kind of pragmatist shorthand -- 'Jim believes there's beer in the fridge' just unifies descriptions of actions Jim is taking or might take, things he says or won't say, and so on. A belief is nearly a theoretical posit -- what we're really interested in is behavior. But above we claim that 'Dewey has defeated Truman' is something a lot like a description of my mental state, that it's about my beliefs in a sense not too different from the way people generally understand beliefs, just somewhat broader. That claim doesn't seem to rely directly on some idea of utility as a pragmatist account might; it's an argument about what the semantic content of our speech acts must be.

So my question is roughly this: do we argue here that the pragmatist account actually kicks in a little earlier, in the model taken as a sort of Bayesian inference engine, that this is where we find the idea of utility? But then are we also looking for pragmatist account of semantics? Do we even need one, or do we sort of get it for free by focusing on how the model (whence all our speech acts originate) works?

Not too clear, but I hope you get the gist of what I'm asking. If not I'll take another run at it.

There's two elements to your summary which I feel I need to clarify.

1)

Moore's paradox rests on the understanding of two processes - the selection of a referent for an utterance, and the truth-valuation of that utterance. Each of these may change depending on the context, the words alone do not give us the processes being used in any given case. We do, however, require a judgement of both to obtain the paradox, reference alone won't do it, it's also that the utterance appears to be true (or capable of being true) when uttered by another, so we have assumed both reference and truth evaluation.

I don't see a problem, as we discussed with regards to stories, with the referent of an utterance being treated as an imaginary object (Santa Claus could be the referent of "Santa Claus wears a red outfit"), so if we wanted to talk about our shared world (socially constructed), then we can have it that in that language game the referent is the convenient fiction of weather and rain - our (hopefully) shared model.

The problem in Moore's paradox is that someone's belief about the weather is being treated as evaluable by an outsider with an accuracy that the first person somehow is assumed to lack. IF A and B are the only people in the context, whether it's raining (as a socially constructed fiction) is a fact by virtue of A and B agreeing on the matter. A cannot be mistaken simply because B thinks it isn't any more than B can be mistaken simply because A thinks it isn't. So B's saying "A doesn't believe it's raining but it is" can amount to no more than "A doesn't believe it's raining but I do" which then means that when A utters the sentence "It's raining but I don't believe it is", he's similarly saying "A doesn't believe it's raining but I do" - only here 'I' refers to A whereas for B's sentence 'I' refers to B so they're not the same sentence.

So it's not quite a matter, for me, of saying that the referent can only be our model. I have no problem with a language game in which the referent is a fiction, a shared model, a social construction...whatever. It's just that the truth evaluation has to match the nature of the referent, otherwise we end up with saying something like "Santa Claus wears a red outfit, but he doesn't wear a black belt because he doesn't exist and so he can't" - either we're talking about Santa Claus the fiction, or Santa Claus the (lack of) reality, we can't just mix both and expect to make any sense.

Which leads to the second part...

2)

We could interpret (though abnormally) "I believe..." as having the shared model of states of mind as a referent. We can model our own thinking, even though we use some of our thinking to construct such a model - I don't see any intrinsic problem with that. We can share that model with others and come to some mutual agreement about it such as to make a socially constructed reality out of it. But then A is simply saying something different to B. A is saying "I believe (as a description of my state of mind) it's raining, but (I believe - meaning my model of the weather is such that) it isn't", which is a clear contradiction. B is saying "A believes (as a description of A's state of mind) it's raining but (I believe - meaning my model of the weather is such that) it isn't"

Either way, B is not saying the same thing (in terms of truth evaluation) as A.

I hope this hasn't been too long-winded. Basically I'm saying that all we can ever judge about the external world is our beliefs about it, and we can judge them by evaluating the response of the world to our actions upon it (in fact I go as far as to say that it the reason we act upon it, but that's another story), we update our beliefs such that acting upon them minimises the likelihood of surprise. So truth only really makes sense to me as an assessment of belief. That belief can be about a shared idea (belief about what is in other people's minds) like Santa Claus, or even a belief about my own beliefs like neuroscience or psychology - either way the truth evaluation comes from treating it as if it were the case and measuring the degree of error from doing so.

So a simple answer to your question would be that we can have whatever we want as the referent within a language game, but when it comes to truth evaluation (in a pragmatic sense), I see no way to meaningfully achieve this without having our belief as a referent.

Self contradictory statements are not true, cannot be true. — creativesoul

Yes. Nor would they be made in "ordinary life." I think the further we depart from the real, the less sense we make.

What Canada is, and how it can be treated under a particular law, need not be the same. So there is no contradiction. In Moore's example, "raining" isn't one thing when we speak of another, but another thing when speaking ourselves.

Reporter: 'So Canada is a terrorist state?'
President: 'Yes.'
R: 'But you don't believe Canada is a terrorist state.'
P: 'Of course not.'

It's just a counterexample to the claim that no one would ever say 'P but I don't believe P.' It's not a restatement of Moore's paradox.

We've established that all sorts of different utterances 'following' that form can be stated.

None of them can be believed by the speaker at the time - aside from misuse of verb tense when reporting upon what's already happened, such as Isaac's example. Even then, it's a misuse. That's not how we're supposed to use those words.

The form is utterly inadequate for taking proper account of all of the particular meaningful language uses that count as being of that form.

Isn't that a huge problem for the accounting practice itself?