Yes. And in fact Dummett was a devout Roman Catholic, iirc.

I was aiming for, let's say, "strategic overstatement", but I think I wound up with bollocks instead.

Unknowability just doesn't look like a big deal in this context. People act on what they believe to be true, or even believe to be probable, and either is rational. You could even know, for a fact, that a proposition has arbitrarily high probability of being true without knowing that it is true; that's surely rational grounds to act on.

I'm still thinking about other ways to approach the OP, but this doesn't look like one ...

Why isn't this a topic about pragmatism and the latter Wittgenstein? — Posty McPostface

Let's.

Let's say something can be true whether you know it or not, whether you even can know it or not, either in principle or just as a matter of fact.

So what? If it's something we can't know, who cares? Now let's grant @The Great Whatever's claim that something can affect you even if you cannot possibly know it does.

What we can still say is this: if there is something, which we cannot know, that affects us, and we cannot know this, we cannot act in response, not in any rational way I can think of.

Even if you happened to enunciate that such a state of affairs holds, through sheer luck, you couldn't know that you had, you couldn't convince others, you couldn't learn more, form a plan of action, nothing.

Such a truth would be a truth we cannot rationally act on.

The Dredd Scott decision — Rich

Hmmmmmm.

whether someone says something true is of course dependent on the language, because what someone says is a matter of linguistic convention. but whether this latter thing is so is not — The Great Whatever

Okay, yes. Of course.

(Been at work for 11 hours now and should probably quit trying to do philosophy at the same time.)

I'm good with pretty much all of this, or something close to it. But this is the hard part.

Allowing truth to attach to sentence tokens allows a cleaner treatment of indexicals -- I think -- but then you have to make truth dependent on a language+interpretation as well as on how the world is, which I take it you're not inclined to do.

I am persuaded that the sort of equivalence you describe in your first paragraph means there must be some way of dealing with the theoretical entity issue you allude to in the second. Also hard, though.

I have been tempted to see if I can have my cake and eat it too by having equivalence classes of sentences as truth-bearers, but I have no idea if that can be made to work. Yet.

ADDED: Ignore the indexicals comment. You don't have issues with indexicals or ambiguity or ellipsis, etc., if truth-bearers just aren't linguistic at all.

gold melting at a certain pt. is also not a matter of convention; — The Great Whatever

Yes.

hence what's expressed by a sentence claiming that gold can't ever melt isn't true or false as a matter of convention / use either. — The Great Whatever

But here I'd like to slow down. Did you say "what's expressed by a sentence" rather than just "a sentence" for a reason? Is it the sentence that's true or false, or is it what's expressed by the sentence? If it's the latter, what sort of thing is that?

I think the next simple, common sense step is to say that a population using the word "gold" to refer to gold is a convention, but there being such a thing as gold is not.

And then maybe we're ready to get back to melting points and such ...

But you don't want to say that to know a word refers to gold, you have to know what gold is?

I think you're just saying this: attributing to someone knowledge of what the word for gold is, presupposes that they know what gold is.

And that seems pretty straightforward.

Maybe it's simpler to look at adults who already have a full complement of concepts and empirical abilities but are learning a new language.

we don't judge whether someone knows what 'gold' means by how good of a prospector they are. — The Great Whatever

Well I'm certainly not suggesting that. After all, we already agreed you can be competent using a word without knowing everything there is to know about either the referent or further uses of the word.

But when you teach a child "moon" they call everything in the sky "moon" and gradually zero in on restricting their utterances of "moon" to when the Moon is up there. That's when we'll start attributing competence, isn't it?

For example can I not competently say "Gold used to be found in those hills", even if I am not someone who can tell real gold from fool's gold? — Janus

That looks like a nice example. If you pushed it farther, say gold has long since disappeared, you might say things like "They used to find something here they called 'gold'." But even if you're using the word they used, doesn't this seem like, I don't know, a lesser competence than they had, who actually knew what gold was?

There's a distinction there, yes, and someone whose linguistic competence includes using the word "gold" properly, in whatever sense, may occasionally fail, but that failure we wouldn't usually describe as not knowing which word to use (though that happens too) but as not knowing whether the word applies in the case at hand.

But we immediately face an issue I'm not sure how to handle: competence using the word "gold" does require competence in recognizing gold. Neither needs to be perfect, but pretty reliable. We don't expect the congenitally blind to be able to acquire competence in using color words, for instance. And the only way we have to judge another's linguistic competence is by observing how consistently they link occasion features we recognize to words we expect. I don't want to leap to the conclusion that this is what competence consists of, but it is the criteria by which we judge it. (Just as there are criteria by which we pick out gold.)

And honestly the ideal would be high empirical competence coupled with high linguistic competence. Failure of either sort degrades the effectiveness of communication, right?

i just don't see what observations have to do with anything. — The Great Whatever

a way of mapping words to their extensions via appropriate criteria. — The Great Whatever

That's the sort of thing I mean. I just mean "observation" in the sense that, presented with a sample of gold, you would assent to "That's gold." Nothing more subtle than that.

Very roughly, you could imagine recognizing gold, looking up gold in your lexicon, and finding it maps to the word "gold", so you say "That's gold." It's simplistic, but kind of what we want, right?

Alright, so I want to see if we can work our way back toward the OP.

If we want to link meaning and truth conditions, we want not the word "gold" on its own, but the one-place predicate "... is gold". Then "Gold melts" comes out as a conditional, "If x is gold, then x melts."

Our hypothetical population correctly uses the predicate "... is gold" but incorrectly treats the conditional "If x is gold, then x melts" as false. But because they cannot (or do not) achieve temperatures at which gold would melt, then there are sentences they count as true for the wrong reason: place a golden idol over a campfire and they will say "It will not melt," which is true.

In fact, all of their statements about gold (and other metals) would be true if we substituted for "melts" something like "melts at temperatures we observe". Or maybe a better approach, more suitable for the campfire example, is "melts at this temperature". We can take "melts" as carrying with it an implied indexical.

In fact, the problem with our folks is the invalid inference from "We have only observed temperatures under 1500F" (say) to "There are only temperatures under 1500F." There may be scientists (or even philosophers!) among them who suggest they should really be saying "Gold has not been observed to melt at temperatures we have observed." (That will look like pedantry to some people, as just another way of saying, "Gold has not been observed to melt." Of course we have only observed the temperatures we have observed!)

All of these considerations seem to relate to the beliefs of our population rather than to the meanings of their words. But the sort of competence we were looking for should give us a way of mapping words onto observations, what we might describe as associating meanings with truth conditions.

The questions we are trying to get to have to do with how that function is constrained by its domain and by its range, by the domain of observations or possible observations, on the one hand, and the range of meanings that will get us from truth conditions, observed or not, to words, on the other.

I agree, but it's a complication I was putting off.

So are you inclined to say that people who think gold doesn't melt know how to use the word "gold"?

the extension/intension distinction doesn't matter here. to use 'gold,' you have to be able to pick out samples of gold – that doesn't just mean happening to point to the right samples, it might also mean using 'intensional' capacities that allow you to figure out which things are gold. to do this, however, you may very well need only a superficial sensory clue, as to its color for example. gold, of course, is infinitely complex beyond this mere sensory signal. — The Great Whatever

But whether the word "gold" has been used correctly doesn't depend on whether you just happen to be pointing at the right sample or used some sensory clues or something. You don't even have to know that it's gold.

Of course all that only holds under interpretations that map "gold" to gold.

My understanding of your view was that you're using the word "gold" right so long as you use it to pick out gold. When you talk about gold, you're talking about a substance that has properties you don't (and maybe can't) know about, but that doesn't mean you're using the word wrong.

whatever goes on beyond our recognition won't affect how we use our language — Srap Tasmaner

Yeah, that was very poorly expressed. Of course it would "affect" our language use, vaguely described. The issue I was aiming at was whether things we don't or can't know would show up in the meanings of our words. And they needn't, if we can still use the word "gold" correctly without knowing all sorts of things about gold.

For instance, suppose in the absence of temperatures at which gold melts, a population believes gold does not melt. We could say they're using the word "gold" correctly -- to pick out gold -- and have a false belief about it, about something they are successfully referring to; or we could say that "metal that does not melt" is part of the meaning of "gold" for them. In one sense, it doesn't matter, because in the circumstances in which they use the word "gold", their version of the word overlaps completely with the version that includes "melts above 1948F".

I'm not really seeing how "meaning" helps here. (Unless there are possible worlds in which gold doesn't melt, and I don't know how to figure that out.) In which case, we stick to what you said at first, that these folks are using the word "gold" correctly and have a false belief about gold. (All this hangs together: we can only say they have a false belief about gold, because we say they're using the word "gold" correctly.)

So the fact that there's a property of gold they cannot know doesn't affect whether they're using the word correctly. But the melting point of gold affects them in that they form a false belief about gold. It affects their language use in that, if asked to list metals that can melt, they won't list gold. But that's still using the word "gold" correctly, just saying something false about gold.

Are we on the same page up to here?

things we don't know about will very much affect us, and our language use — The Great Whatever

Then I misunderstood. I thought you had said all there is to using the word "gold" correctly was getting its extension right, which you can do whether you know or can know everything about gold.

I think the conflict here is inevitable, because there are intensional and extensional aspects to language.

If you approach things from the intensional side, you get a clear view of how a population uses words. If you approach things from the extensional side, you get a clear view of what it means for a statement to be true.

The intensional, use-oriented approach has trouble accounting for truth. I take this to be @The Great Whatever's point, and it is well taken.

From the other side, we have Dummett's argument that truth alone does not explain how a population uses language.

Mostly the two camps talk past each other. Dummett is attempting to extend the reach of the intensional account into the extensional camp's home territory: whatever goes on beyond our recognition won't affect how we use our language. Even TGW essentially admits this by allowing that we may have the practical ability to use a word, say "gold", without knowing everything about gold. It's also not clear if the extensional camp can provide any account for that practical ability.

But the intensional camp still has no plausible account of truth, and has trouble explaining how the meaning of the word "gold" can be extended beyond the sort of situations in which it has heretofore been used.

It's clear that the battleground is truth, and that ideally we'd be able to combine the two approaches. David Lewis tries in Convention, but I haven't made much sense of the last chapter. :-(

Whether philosophy as a whole or my approach to it are intellectually bankrupt is off-topic here. We should have ended this long ago.

the fitch argument is super duper simple — The Great Whatever

I think there's room for debate.

why do philosophy if you've decided a priori that whatever prejudices you currently have are sufficient to refute any argument? — The Great Whatever

Oh no! I wouldn't say that at all. I am intensely curious about everything I've mentioned in this thread, and open to being persuaded either way. My resistance to an argument or an approach is a hurdle it must clear, that's all.

I see your point, yes.

There were two different stands of thought there I was trying to keep separate. I even numbered them: one is Dummett's intuitionist response; one is my own sense that there's something odd going on in Fitch's argument.

The first time Dagfinn Føllesdal saw the slingshot argument in Word and Object (iirc), his immediate reaction was, roughly, "There's no way that's right." People do feel that way about Gettier, and I feel that way about Fitch's paradox. All three have produced cottage industries attempting to refute them.

I don't have handy a refutation of any of those arguments, but I still don't trust them.

That's not to say there isn't much to be learned from arguing about these, and about the Liar, for that matter. But I do not feel compelled to accept Fitch's argument as a refutation of anything.

I don't know what to tell you.

It's a thread largely about Michael Dummett. I've been doing my best to make sense of his position. If you're asking which side I'm on or something, I'm really not inclined to commit at the moment. Why should I?

It's their choice not to speak anymore if that is the case, no? — Agustino

Sure. The question is whether we as forum members give people reasons to continue speaking or reasons not to.

For example -- note I am not attributing anything to you here -- the "logical content" of the following is the same:

"I respectfully disagree."

"Only a moron would think that!!!"

Both are ways of saying "That's false" but one contributes to the health of the forum and one really doesn't.

My point has only been that we should be mindful of not only the philosophical import of our words, but their effect on the health of the forum. It's a matter of faith, perhaps, that the latter would also lead in the long run to better philosophy.

i think there's some confusion in thinking the way logical systems work is that you can simply 'choose' to use whichever system you like to validate or invalidate any proof. — The Great Whatever

Then it's a good thing this is not what Michael Dummett did.

there's no persuasion to be done, it's a proof — The Great Whatever

I'll say two things about Fitch's:

1. If you use intuitionist rules of inference and interpret the logical constants along intuitionist lines, you might be okay, as Dummett is, saying "p→~~Kp" but that's not saying "everyone is omniscient."

2. I put it in the same box with the slingshot argument and Gettier cases. They're fascinating, but I am far from alone in feeling that a logical fast one is being pulled.

if you can't imagine that there might be something you can't know, — The Great Whatever

I don't really know what to say about this, so I'm going to go away and think about it for a while. It's an interesting question I honestly haven't thought about, so you have my thanks for raising the issue.

"Less inclined to speak" which you put entirely on them if you want, but the circumstances matter.

I don't find Fitch's persuasive at all.

just imagine any situation where you can't know something, & it will still be able to affect you whether you can know it or not — The Great Whatever

Does that rule out talk about the future as our example?

If there are or are not such possibilities, how would we figure that out? Examples don't seem to be doing the trick.

construct the example such that there's no way for you to know — The Great Whatever

I think statements about the future might fit the bill.

Thoughts?

making predictions among your friends or placing a wager on the outcome of the show is not the same sort of activity as stating facts — Srap Tasmaner

And here's the thing: everything else I've been doing around here lately is pushing me toward thinking these actually are very similar activities. But all that's still up in the air ...

Dude, I may be a little further along than you, but don't think I don't struggle to understand this stuff! Totally worth it though, so keep at it.

We are talking about something like the antinomy of democracy as a model for doing philosophy: open, honest exchange of ideas should lead to better philosophy. But if some of those expressions drive people away or silence others, that's not what we wanted.

If GRRM had that as part of his outline for how the series ends, then yes. Otherwise, the truth value becomes determinate in the future. — Marchesk

I think the most that could get you is "GRRM planned to have Jon Snow sit on the Iron Throne" but that can be followed by all sorts of stuff -- who knows what the producers actually end up doing. Jesse was not supposed to be a major character in "Breaking Bad".

I wouldn't say "equated". There's an "if and only if" between knowing one and knowing the other ...

He does explicitly reject what's here called the "strong" view, and that means rejecting the idea that truth conditions explain meaning. I know one of the various arguments he tries is, roughly, that you can't possibly know whether the truth conditions of a sentence obtain if you don't already know what the sentence means. (Recognition again.)

But then he himself has to come up with these sort of Quinean feature report sentences to get the linguistic ball rolling -- to learn the first words of your language there have to be sentencelets you can count as true ("true"?) like "Grass!" "Ball!" and so on that depend only on recognizing salient features of your environment. Otherwise you can never break into language.

I've only muddied the waters, haven't I?

Game of Thrones — Michael

As I clarified, I didn't mean to make an argument about statements about the future, but maybe I should have!

Now to more muddiness ...

Dummett may very well reject bivalence, if not across the board, then at least for this domain, statements about the future. He will not say, " 'Jon Snow will sit on the Iron Throne' is either true or false." But Dummett also rejects truth-value gaps, so he will not say, " 'Jon Snow will sit on the Iron Throne' is neither true nor false."

Well?

You might conclude that "Jon Snow will sit on the Iron Throne" is not really a statement, somewhat like a logical positivist. It's normally meaningful words put together to look like something you could assert but it isn't really.

I don't think Dummett actually says that sort of thing anywhere, and Wittgenstein sort of quit saying it, and instead suggested wrapping it in a context where it could make sense -- making predictions among your friends or placing a wager on the outcome of the show is not the same sort of activity as stating facts. A prediction or a wager is not an assertion. Something like that anyway.

I think Dummett's view must be near there, but it's never been clear to me.

It does relate to the issue of how the meaning of a sentence is determined, and whether recognition-transcendent truth conditions can play a part. If the truth conditions of the assertion "Jon Snow will sit on the Iron Throne" are inaccessible, I think you conclude that this sentence cannot be used as an assertion in the usual way -- that you must be doing something else. And "what you mean" is tied to that.

Dummett never accepted Davidson's view that truth conditions give you an account or an explanation of meaning. But they do run together, "agreement in judgements" and all that.

If, as Dummett says, the meaning of a sentence is exhaustively determined by its use, I think you conclude that recognition-transcendent truth conditions can play no part in determining the sentence's meaning. That feels like it's allowing truth conditions to determine meaning, but only if you ignore the bit about recognizing that they obtain or not, which gets you back to how a population uses the sentence.

Does that make sense? If that's right, then the answer is "yes".

I'm not sure it's a sound doctrine though.

Statements about the future look like they would be a clearer test case, since we're not inclined to allow knowledge of the future. But then there's a question about when such a statement is true: at the time of utterance, or at the time of evaluation?

Against my previous post, it could be claimed that, in each case we think of, it is only contingent that a given truth could be known, that I haven't shown the knowability of the truth is necessary.

We are now essentially debating verificationism, which isn't to @Michael's question: whether LW's approach to meaning leads, as it did for Dummett, to some form of anti-realism. I'm still not sure; it depends on how you take the connection between meaning and truth conditions.

Something @Nagase mentioned is helpful here, the idea of a designated value. With the 3-position light switch, there are two obvious ways to do this: entirely on, or one of the others; entirely off, or one of the others.

Dummett uses the comparison of conditional bets to conditional commands. If I tell you not to leave without saying goodbye (i.e., if you leave, ...), not leaving (and not saying goodbye) counts as compliance. But if I bet you that if the Cubs make it to the World Series they'll win, and then they're knocked out in the NLCS, you don't owe me a thing.

Then the question is, which one is assertion like? If you take "truth" as the designated value, you can allow various ways of not being true and lump them together. (More like a bet.)

As for bivalence versus excluded middle: your standard switch can be in one of two positions; whether that also turns on the lights depends on whether there's power. Switch position is syntactic; lights going on or off is like adding an interpretation to your system that assigns truth and falsehood -- semantics. The first is LEM, the second PB.

But the terminology is so confusing, it's best just to be explicit about what you're doing, even making up terms as Nagase did with "strong" and "weak" bivalence. It's the ideas that matter not the terminology.

At Titus Andronicus shows, Patrick Stickles makes this little speech about how not everyone has fun the same way, and asks the crowd to be aware of people around them if they want to get slammy. (I'm paraphrasing.) You can be punk and still be sensitive.