Sorry, made some substantial edits to my previous post in the process of your replying.

What is required is instead a different account of truth, one that rejects both sides of the anti/realist ledger in order to affirm that "to grasp a concept [like truth] is to have a practical mastery of the inferences it is involved in. It is to be aware of its role in justifying some further attitudes and in ruling out others.... For example, to learn the use of the word “red” is to learn to treat “This is red” as incompatible with “This is green,” as following from “This is scarlet,” and as entailing “This is colored.”" So too with truth - "the truth conditions of our statements are determined by... our conventional rules for predicating “is true” of them... once we conceive of learning the uses of “is true” as a matter of learning a certain practical mastery, the meaning we take our sentences to have is necessarily consistent with the use to which we put them. For we learn what it means to say [for example] that a past-tense statement p is true when we learn the [practical] criterion for judging its truth, that is, when we learn how to use the statement “p is true.” (Sara Ellenbogen, Wittgenstein's Account of Truth). — StreetlightX

I don't see how "the truth conditions of our statements are determined by ... our conventional rules for predicating 'is true' of them" is any different to Dummett's account. It's a rejection of the realist's claim that truth conditions transcend verification. The truth of "there is a chair in the next room" is (wholly) determined by linguistic conventions and the empirical contexts in which language is put to use. Seems like anti-realism in a nutshell.

And it's here where the distinction between realist and anti-realist is made; the anti-realist argues that the correctness of using statements is determined by the things we see and the things we say and the things we think whereas the realist argues that the correctness of using (some) statements (e.g. "there is a chair in the next room") is determined by something else (something verification-transcendent). — Michael

This is absurd because statements aren't made outside of the context of our experience. If we are asking about the correctness of a statement, we are always dealing with what we experience. The distinction you are worried about doesn't get off the ground in the first place.

A realist arguing a statement is true because a chair exists in the next room doesn't suggest a truth maker outside what we think and say. Indeed, it is just the opposite: if we are dealing with a true statement, then they entire point is we have said something about the world. The realist argues that truthmakers, the existence of state itself, which someone is talking about in language, is that which we are thinking and talking about.

John: Having said this, though, even in regard to verification-transcendent truth, I think we all believe there are such. If I say to you "Remember that online debate we had about realism/anti-realism last Friday, but of which there is now no record since Philosophy Forums crashed and all the posts were lost, and you say "No, we never had any debate last Friday", don't you believe that it is simply either true of false that we had such a debate, even though it can never be verified (i.e. even though it is a verification-transcendent truth/falsity we are dealing with)?


Michael: I'd either say that it's true or say that it's false. But this is just to engage with the language-game I've learnt to use. When asked a question I consider the things I've seen (or am seeing) and the things I've been told and respond with the most appropriate answer. Nothing about this requires accepting the realist's verification-transcendent truth. — Michael

But, I wasn't asking about what you would say. In the thought experiment you say that we did not have a debate last Friday and I say that we did. There is no way of verifying which of us is correct (the case is verification transcendent).

Are you claiming that since there is no way to verify which of us is right, that it cannot be the case that either of us are right, or that both of us are right? Or are you saying something else?

@Michael Sorry for the late reply, TPF has been down for me for a couple of days. Anyway...

But there's nothing to be anti-realist about. What is different are the reasons for such a rejection. Dummett rejects them because he thinks that natural language is full of sentences whose truth conditions we would not be able to recognize as obtaining - meaning that one must reject a truth-conditional theory of meaning. I reject them because the very idea of truth conditions that can or cannot be recognized makes no sense. I don't reject the realist's claim that truth conditions transcend verification because it is false. I reject it because the very idea of truth as correspondence is literally senseless. It's like asking 'what does 'dichotomy' taste like?'. The question is malformed, and one cannot take a stance on it either way.

Dummett still thinks that truth is a matter of correspondence (here's Dummett: "the correspondence theory expresses one important feature of the concept of truth... that a statement is true only if there is something in the world in virtue of which it is true." He literally elevates this to a 'principle' he christens 'principle C'). Dummett's semantic-anti realism actually consists in the 'next step' he takes, when he goes on to argue that because meaning cannot be wholly parsed according to this theory of truth, then one must reject truth-conditional theories of meaning. But Dummett, like the realists he argues against, both have truth all wrong to begin with. Correspondence is not even at issue, it never was, it never will be. Truth simply doesn't work like that.

A way to think about it is like if two people had differing opinions on whether cars can blink. One is a blinkist, and another is an anti-blinkist. A third bloke comes along and says, hold on, the whole idea of cars blinking is ridiclious. He notes that sometimes people speak of cars as having 'blinkers' which indicate left and right, but they don't blink, at least not in the way of having eyelids and so on. After which the other two immidiately tell him that he's actually an anti-blinkist. There is a sense in which they are right, but only in a way that is completely trivial and inane. The anti/blinkists are not simply wrong - they are not even wrong.

But, I wasn't asking about what you would say. In the thought experiment you say that we did not have a debate last Friday and I say that we did. There is no way of verifying which of us is correct (the case is verification transcendent).

Are you claiming that since there is no way to verify which of us is right, that it cannot be the case that either of us are right, or that both of us are right? Or are you saying something else? — John

Either it is appropriate, given the empirical situation (memory, experience, and so on) and the rules of our language-game, to answer with "it's true" or it is appropriate to answer with "it's false" (or the appropriateness is debatable). One doesn't then need to say that there is some verification-transcendent fact-of-the-matter, independent of this appropriateness. The "verification" is the empirical situation and the language-game, not something else.

So what I would say is all that matters.

OK, so you are saying that you don't believe there is any fact of the matter?

No, I'm saying that the fact-of-the-matter isn't some independent, verification-transcendent state-of-affairs. Either it's appropriate to say "it's true" or it's not (given the empirical context and the rules of our language-game), just as either it's appropriate to say "it's a game" or it's not (given the empirical context and the rules of our language-game). We don't need to introduce some "but it's really true" or "but it's really a game" as something separate to this.

A way to think about it is like if two people had differing opinions on whether cars can blink. One is a blinkist, and another is an anti-blinkist. A third bloke comes along and says, hold on, the whole idea of cars blinking is ridiclious. He notes that sometimes people speak of cars as having 'blinkers' which indicate left and right, but they don't blink, at least not in the way of having eyelids and so on. After which the other two immidiately tell him that he's actually an anti-blinkist. There is a sense in which they are right, but only in a way that is completely trivial and inane. The anti/blinkists are not simply wrong - they are not even wrong. — StreetlightX

I don't get this "not even wrong" thing. Surely either cars blink or they don't. Those who say that cars blink (the blinkers) are wrong and those who say that cars don't blink (the anti-blinkers) are right. The fact that "the whole idea of cars blinking is ridiculous" is not that "the whole idea of cars not blinking is ridiculous" so your rejection of anti-realism on the grounds that realism is nonsense just doesn't make any sense to me.

Are you a flyerist (believe that people can fly (without technological aid)) or an anti-flyerist (believe that people can't fly)? I'm an anti-flyerist, and I think it perfectly sensible (and in accord with the mechanical facts). How is this any different to the case of blinking cars or verification-transcendent truths? Or is it the same and you will maintain your "not even wrong" position and neither claim "people can fly" nor claim "people can't fly"?

I reject them because the very idea of truth conditions that can or cannot be recognized makes no sense.

This seems inconsistent with your claim that "In other words, we learn to predicate truth of propositions in the same way learn to predicate color of things. By learning in what circumstances it is normally considered appropriate to make such predications." Surely the circumstances in which it is (in)appropriate to predicate truth of a statement is a truth-condition? It is because of these circumstances that the claim "X is true" is the right (or wrong) thing to say.

As usual, we're at an impasse, and I've lost my appetite to go on. I will say that I literally meant cars blinking, as in what eyelids do - which is what I specified. I meant it as pure nonsense - like foreclosing a door. Otherwise, I've little desire to continue.

The way I understand it, in any case of the type I outlined where two people remember events differently: "debated on Friday", "did not debate on Friday" there are only possibilities; 'debated on Friday' or 'did not debate on Friday'. It could not be the case that we both debated and did not debate on that day.

Do you agree that one of the memories of the events, if they exclude the possibility of the other, must be wrong?

Do you agree that in (at least) certain cases there is no possibility of verifying which of the memories of events is right, and that such situations qualify as 'verification transcendent'?

Do you agree that, in such cases, the rightness or wrongness of the memories of events cannot be established and thus cannot be dependent on anything that is said?

There are countless examples like this, not necessarily involving more than one person. For example, on any day I perform countless actions that I have no memory at all of. Does the fact that I cannot remember those actions entail that they did not occur? If someone is murdered and the murderer is never caught, does that mean that there is not someone out there who committed the murder, even if they don't remember it themselves?

As I see it your anti-realism can have no coherent answers to such questions.

Streetlight, I'm not quite sure what you are saying here. Are you saying that the truth of sentences is dependent on, or independent of, human discourse.

For example, say X murders Y, but no one knows it for sure, including X (due to amnesia, say). People suspect it though, and the sentence 'X murdered Y' is uttered. The truth status of that sentence is "verification transcendent", it cannot be established; does that entail the sentence is not either true or false?

This seems inconsistent with your claim that "In other words, we learn to predicate truth of propositions in the same way learn to predicate color of things. By learning in what circumstances it is normally considered appropriate to make such predications." Surely the circumstances in which it is (in)appropriate to predicate truth of a statement is a truth-condition? It is because of these circumstances that the claim "X is true" is the right (or wrong) thing to say. — Michael

But the understanding of the conditions which amount to a statement being true aren't separate to that.

In the you are using it, there is no act of verification. The person who knows "X is true" doesn't need some separate thing verifying it is "really true." They just need the understanding the condition in question is true.

The idea truth conditions can or cannot be recognised is nonsensical because, in any situation where truth is known, where someone knows about a truth condition, they have recognised the truth condition by definition.

I know, for example, that it is true I am writing this post. The question of whether or not a recognise this truth condition is moot. Given my knowledge, my understanding, I must. I can't know: "Willow is writing this post" is a true statement if I haven't recognised the truth condition in question.

You are making the very distinction, and asking for the event thing, the support for something being "really true," which you are claiming is nonsensical.

Or is it the same and you will maintain your "not even wrong" position and neither claim "people can fly" nor claim "people can't fly"? — Michael

The problem is the question in the first place. You are asking someone who knows (analogous to the truth condition of a statement) the answer to that question (e.g. people can't fly), whether it is true people can or can't fly. And then when the person states the know people can't fly, you are ignoring their knowledge and asking the question again. (e.g. "Ah, but I need "verification" you know people can't fly. We still don't know whether people can fly or not. Please show me people not flying is "really true" ).

But the understanding of the conditions which amount to a statement being true aren't separate to that.

In the you are using it, there is no act of verification. The person who knows "X is true" doesn't need some separate thing verifying it is "really true." They just need the understanding the condition in question is true.

The idea truth conditions can or cannot be recognised is nonsensical because, in any situation where truth is known, where someone knows about a truth condition, they have recognised the truth condition by definition.

I know, for example, that it is true I am writing this post. The question of whether or not a recognise this truth condition is moot. Given my knowledge, my understanding, I must. I can't know: "Willow is writing this post" is a true statement if I haven't recognised the truth condition in question. — TheWillowOfDarkness

So you accept that there are truth-conditions. That was the point.

You are making the very distinction, and asking for the event thing, the support for something being "really true," which you are claiming is nonsensical.

I'm not claiming that being "really true" is nonsensical. I'm claiming that being "really true" as something independent of whether or not it is appropriate, given the empirical context and the rules of our language-game, to predicate truth of it is nonsensical.

The problem is the question in the first place. You are asking someone who knows (analogous to the truth condition of a statement) the answer to that question (e.g. people can't fly), whether it is true people can or can't fly. And then when the person states the know people can't fly, you are ignoring their knowledge and asking the question again. (e.g. "Ah, but I need "verification" you know people can't fly. We still don't know whether people can fly or not. Please show me people not flying is "really true" ).

This is completely mistaken. My point was that it is no more nonsensical to claim "cars can't blink" or "truth is not verification-transcendent" than to claim "people can't fly".

You have a tendency to read the strangest things from my posts.

Do you agree that one of the memories of the events, if they exclude the possibility of the other, must be wrong? — John

I agree that either it is appropriate, given the rules of our language-game and the empirical context (e.g. my memory), to claim that "it happened" is true or to claim that "it didn't happen" is true. That's all there is to truth.

Do you agree that in (at least) certain cases there is no possibility of verifying which of the memories of events is right, and that such situations qualify as 'verification transcendent'?

I agree that sometimes we don't know whether or not it is appropriate to predicate truth of some statement, and also that it is appropriate to predicate truth of the statement "either it is true or it is not". But I don't agree that there is some verification-transcendent condition in virtue of which the former statement is true (or false).

I agree that either it is appropriate, given the rules of our language-game and the empirical context (e.g. my memory), to claim that "it happened" is true or to claim that "it didn't happen" is true. That's all there is to truth.

I agree that sometimes we don't know whether or not it is appropriate to predicate truth of some statement, and also that it is appropriate to predicate truth of the statement "either it is true or it is not". But I don't agree that there is some verification-transcendent condition in virtue of which the former statement is true (or false). — Michael

Yes, but if you admit that you don't know which claim is appropriate, but that one of them must be, then you are admitting that there are unknown conditions which would (if only we could know them) determine which one is appropriate.

Which claim is appropriate cannot be determined by "the rules of our language-game and the empirical context" (but of course if they were to be determined or could be determined, they could only be determined within "the rules of our language-game and the empirical context", which is a different
matter entirely). I think you are failing to see this logical distinction, basic to our intuitive understanding of truth conditions, between being determined by and being determined within.

So, there are "verification-transcendent condition(s) in virtue of which the former statement is true (or false)" but there are no verification transcendent conditions in virtue of which it could be determined to be true or false, because determinations of truth and falsity are possible (if at all) only by virtue of verification immanent conditions. The point is that even verification immanent conditions are not created by "the rules of our language-game and the empirical context" but our understanding of them is merely mediated by the latter.

The anti-realist position you seem to be advocating cannot support a coherent distinction between truth and belief, that is its fatal flaw.

Yes, but if you admit that you don't know which claim is appropriate, but that one of them must be, then you are admitting that there are unknown conditions which would (if only we could know them) determine which one is appropriate. — John

Not at all. I'm not saying either that "X" is true by virtue of some verification-transcendent condition or that "X" is false by virtue of some verification-transcendent condition. I'm saying that "X" is either true or false by virtue of the axiomatic laws of thought (excluded middle and non-contradiction).

The point is that even verification immanent conditions are not created by "the rules of our language-game and the empirical context"...

Of course they are. The truth of "bachelors are unmarried men" is 'created by' the rules of our language-game and not by some verification-transcendent condition.

The anti-realist position you seem to be advocating cannot support a coherent distinction between truth and belief, that is its fatal flaw.

Of course it can. "X" is true iff it is appropriate, as per the rules of our language-game and the empirical context, to predicate truth of "X". "X" is believed to be true iff one believes that it is appropriate, as per the rules of our language-game and the empirical context, to predicate truth of "X",

What leads you to believe in the LME?

What about truths other than tautologies?

And how do you differentiate between "appropriate" and "believed to be appropriate"?

What leads you to believe in the LME? — John

LME?

What about truths other than tautologies?

Still about the rules of our language-game and the empirical contexts in which it's put to use. I see this and so it's appropriate to say that.

And how do you differentiate between "appropriate" and "believed to be appropriate"?

Well, the semantic difference is apparent to me as a native speaker of the English language. I know that "'X' is appropriate" and "'X' is believed to be appropriate" mean different things. I know that there are contexts in which it is appropriate for me to say "John believes that 'X' is the appropriate thing to say but actually 'not X' is the appropriate thing to say".

Law of the Excluded Middle.

You have not answered the questions adequately.
I don't believe you are interested in discovering the weaknesses of your position, but rather want to gloss them.

The law of diminishing returns...

Law of the Excluded Middle. — John

Ah, so LEM rather than LME. ;)

Well, what leads anyone to believe it? Perhaps it's just an axiom. Perhaps it better describes the structure of our language. I'm not entirely sure how the reason for accepting it is relevant to the discussion anyway.

You have not answered the questions adequately.
I don't believe you are interested in discovering the weaknesses of your position, but rather want to gloss them.

Well, I think I have answered them adequately. If you disagree then perhaps you could explain what's problematic.

So you accept that there are truth-conditions. That was the point.


I'm not claiming that being "really true" is nonsensical. I'm claiming that being "really true" as something independent of whether or not it is appropriate, given the empirical context and the rules of our language-game, to predicate truth of it is nonsensical. — Michael

But that's the problem your approach. You fail to understand how language is of the world, that awareness of the truth condition is embedded within language, such that the standard or "really true" never make sense. Any knowledge of a truth condition is given in a person's language. Their can be no "verification" of this knowledge from outside their language. The standard of "really true" doesn't make sense because any instance of empirical confirmation is found within how is thinking or speaking. Any justification of "really true" is merely within a person's own language and experience. It doesn't get outside their experience, their judgement, to demonstrate what is true free of the "taint" of their perception. Things themselves, which may be around when not perceived, are of a nature as spoken by the appropriate language game and that which is seen in empirical observation.

And because of this position you are strawmanning the (direct) realists. They've never asserted that things are independent of language games and empirical contexts. Indeed, their point is this is never true: things of the world are as they are perceived, as they are appropriately spoken about in language.

This is completely mistaken. My point was that it is no more nonsensical to claim "cars can't blink" or "truth is not verification-transcendent" than to claim "people can't fly". — Michael

The problem isn't claiming cars can't blink. On can do that perfectly well, and be right or wrong, depending what type of blinking someone is talking about. Rather, the problem with the question you are posing, which ignores that people know the truth, which views knowledge as a question of "Proof outside your language" rather than a matter of language embedded in the world which says something true.

And because of this position you are strawmanning the (direct) realists. They've never asserted that things are independent of language games and empirical contexts. — WillowOfDarkness

Yes they have. That's what it means to be a realist.

But we've been over this countless times before.

Yeah, and you've been wrong countless times...

But crucially, misunderstanding of direct realism aside,with respect to understanding language, you are still making the split between the world and language. You are treating it is if language doesn't talk about the world by its definition (i.e. it's first and foremost only semantic), as if we could have language which talks about the world which was in the first instance, only semantic.

But crucially, with respect to understanding language and misunderstanding of direct realism aside, you are still making the split between the world and language. You are treating it is if language doesn't talk about the world by its definition — TheWillowOfDarkness

Except I haven't and I don't. Again, you read the strangest things from my comments.

You've outright claimed it here:

The truth of "there is a chair in the next room" is (wholly) determined by linguistic conventions and the empirical contexts in which language is put to use. Seems like anti-realism in a nutshell. — Michael

Here you are saying that, for a chair to be in the next room, all we need is for someone to speak of the relevant empirical and linguistic context. Supposedly, the "semantic" is enough to define the truth.

But it's not. We may have a person who speaks of the empirical context of the chair, but not a chair. The use of language may not talk about the world, even profess that semantic meaning. Equally, their could be a chair in the next riot which no one is speaking about.

The truth of "there is a chair in the next room" is NOT (wholly) determined by linguistic conventions at all. It takes a state of the world for that statement to be true. One which is not given by the language alone. For "there is a chair in the next room" to be a true statement, there needs to be relevant state of the world and, specifically, a language which talks about that state of the world. There is more to the truth condition of "there is a chair in the next room" than whether someone utters words with that meaning.

Here you are saying that, for a chair to be in the next room, all we need is for someone to speak of the relevant empirical and linguistic context. — TheWillowOfDarkness

I didn't say that we need to speak of the empirical and linguistic context. I said that the empirical and linguistic context is what makes our talk of other things – like the chair in the next room – appropriate.

The truth of "there is a chair in the next room" is NOT (wholly) determined by linguistic conventions at all. It takes a state of the world for that statement to be true.

Does this state of the world transcend verification? Realism requires "yes" and anti-realism requires "no".

Ah, so LEM rather than LME. ;)

Well, what leads anyone to believe it? Perhaps it's just an axiom. Perhaps it better describes the structure of our language. I'm not entirely sure how the reason for accepting it is relevant to the discussion anyway.

Well, I think I have answered them adequately. If you disagree then perhaps you could explain what's problematic. — Michael

Law of Middle Excluded??? :-}.



How and why do you think the LEM is self-evident to you? Do you think the fact that contradictions cannot obtain merely reflects the "structure of our language" or is it not rather that the structure of our language reflects the nature of experience and the experience of nature?

You haven't said what it means to know "when you see this it's appropriate to say that", as distinct from merely believing it is (out of habit, convention, convenience or whatever).

Saying that you must understand the distinction because you know how to use the two phrases is a cop out; all it shows is that you know the definitions of the words in the phrases, not that you can explain the logical distinction, as it consists in your position (if it indeed does), between knowing something and believing something.

If you can explain that distinction in terms of anti-realism then do so; if you cannot then admit it.

I didn't say that we need to speak of the empirical and linguistic context. I said that the empirical and linguistic context is what makes our talk of other things – like the chair in the next room – appropriate. — Michael

But its not. We need more. In this case, as the language is about the world (and of the world), we need the worldly context. We need, in the world, the existence of language which talks about something in the world. What is at stake is more than the definition of language, for than what constitutes a statement which means.

You are missing the critical description that the empirical and linguistic context is worldly.

Does this state of the world transcend verification? Realism requires "yes" and anti-realism requires "no". — Michael

This question is a misstep. No state transcends verification. It is possible to verify any state subject to empirical verification.

But it is also true that states, even known states (e.g. the rising sun tomorrow), frequently are unverified. So in the sense that states of the world are frequently unconfirmed, they do "transcend" verification.

You are missing the critical description that the empirical and linguistic context is worldly. — TheWillowOfDarkness

Where am I missing it? I have never said that they're not worldly. In fact, I've repeatedly said that they are.

No state transcends verification.

And that's exactly why realism fails.