A behaviourist wouldn't necessarily deny belief or opaqueness, though? — bongo fury

To support that, you'd need to explain how a substitution failure cashes out in terms of behavior.

right — flannel jesus

Not convinced?

Seems that folk here are solving a problem they have yet to specify. Oh, well.

right
— flannel jesus

Not convinced? — Banno

Why does me saying "right" lead to you saying "not convinced?"? "That's right" is the kind of thing someone says when they are agreeing with something...

If I misread your lack of further comment, that'd be pleasing.

Is there then any issue here that is left hanging, once the scope is sorted?

The guts of Davidson's article is the difference between "Superman is Clark Kent" and "Lois believes that Superman is Clark Kent". The former is a relation of identity between two characters, the latter a belief on the part of a third character. The two are very different things.

The guts of Davidson's article is the difference between "Superman is Clark Kent" and "Lois believes that Superman is Clark Kent". The former is a relation of identity between two characters, the latter a belief on the part of a third character. The two are very different things. — Banno

Sorry, I was vague about that. The OP is from an IEP article on referential opacity. I was struggling to understand all the aspects of it. Davidson's article is called Rational Animals. I felt like I needed to understand opacity better before I digested his argument.

If I misread your lack of further comment, that'd be pleasing. — Banno

I read your comment (the one I replied right to) sounded like it was explicitly agreeing with me. That's all

Yeah, all clear. A good approach.

I've looked, but have not been able to locate a good account of opacity. The IEP article is framed around Frege, and so leads in to Russell an descriptions. The SEP article on Quotation might be both clearer and of greater generality. See especially the list of five possible responses. For Davidson, the Demonstrative theory is key.

right.

I've looked, but have not been able to locate a good account of opacity. — Banno

The IEP article is terrible. I'll try the SEP article.

A behaviourist wouldn't necessarily deny belief or opaqueness, though?
— bongo fury

To support that, you'd need to explain how a substitution failure cashes out in terms of behavior. — frank

Well, as assent to contradictory sentences?

Well, as assent to contradictory sentences? — bongo fury

So we put two flashcards in front of Lois with contradictory sentences on them, and if she nods her head to both, we have referential opacity.

As with Ralph accepting sentence (11) here?

https://www.uvm.edu/~lderosse/courses/lang/Quine(1956).pdf

Yes.

How could you tell from watching Ralph's behavior that he's "ready enough to say" something? With behaviorism, we're trying to avoid a God's eye view on Ralph's psyche. In fact, for our purposes (as faithful behaviorists), Ralph has no psyche.

How could you tell from watching Ralph's behavior that he's "ready enough to say" something? — frank

Let alone that the readiness is to say it "in all sincerity"!

Not sure I see an inherent problem. But I don't know if anyone (e.g. Davidson) has to be fundamentalist about the behaviourism, anyway? I think it's plausible that we describe the psychology of a linguistic animal in terms of its dispositions to assent and dissent to sentences?

Let alone that the readiness is to say it "in all sincerity"!

Not sure I see an inherent problem. — bongo fury

You were asking if it would be more surprising if propositional attitudes did not result in substitution failure. I was saying that this would depend on how one understands belief. A behaviorist might have a hard time accepting substitution failure because beliefs reduce to behavior. Evidence for substitution failure would be scarce and unreliable.

You're identifying belief with the readiness of a subject to say S with sincerity. So it's

a. Clark Kent = Superman
b. Lois is ready enough to say S1 with sincerity, where S1 is "Superman can fly."
c. Therefore, Lois is ready enough to say S2 with sincerity, where S2 is "Clark Kent can fly."

This is a misapplication of the Identity Elimination Schema, which goes

Major: t1 = t2

Minor: ϕ(t1)

Conclusion: ϕ(t2) — IEP

The reason it's a misapplication is that neither t1 nor t2 show up in the b sentence. You can't substitute t1 for a mention of t1. If you do that, you'll end up with:

a. Istanbul is Constantinople.
b. “Istanbul” has eight letters.
c. ∴ “Constantinople” has eight letters. — IEP

In other words, if we use your scheme, we'll end up applying Leibniz' Law to the speech of a parrot. Parrots can be ready enough to say S1. What does that have to do with laws of identity?

Edit. Actually, I think I'm wrong about calling that a mention, it's more just an utterance. It's just sound that Lois is making.

a. Clark Kent = Superman
b. Lois is ready enough to say S1 with sincerity, where S1 is "Superman can fly."
c. Therefore, Lois is ready enough to say S2 with sincerity, where S2 is "Clark Kent can fly."

This is a misapplication of the Identity Elimination Schema, — frank

I dare say. More to the point, it's referential opacity. :smile:

More to the point, it's referential opacity. — bongo fury

No, you only get referential opacity if there are mental states involved... specifically of a kind a parrot wouldn't have.

The problem of referential opacity is to explain why a certain inference rule of classical logic sometimes produces invalid-seeming inferences when applied to ascriptions of mental states. — IEP

Well, it doesn't say the ascriptions shouldn't be in terms of dispositions to assent and dissent? Seems bizarre if it did, given that would rule against Quine, who introduced the phrase. (I think?)

Well, it doesn't say the ascriptions shouldn't be in terms of dispositions to assent and dissent? — bongo fury

In this case, the misapplication is so bad you don't even have a substitution failure. You at least have to have t1 in the b sentence.

Comport yourself so that t1 shows up in the b sentence, and we can evaluate for referential opacity.

And isn't Davidson saying the parrot wouldn't recognise the opacity? — bongo fury

I haven't gotten back to Davidson. I'm still trying to figure out opacity. You're helping. :grin:

And isn't Davidson saying the parrot wouldn't recognise the opacity?
— bongo fury

I haven't gotten back to Davidson. — frank

Nor have I, which is why I edited out that remark. :wink:

The guts of Davidson's article is the difference between "Superman is Clark Kent" and "Lois believes that Superman is Clark Kent". The former is a relation of identity between two characters, the latter a belief on the part of a third character. The two are very different things. — Banno

:up:

Major: t1 = t2

Minor: ϕ(t1)

Conclusion: ϕ(t2)

Here t1 and t2 are expressions which refer to entities (for example, proper names of people or cities). ϕ(t1) is a sentence containing at least one occurrence of t1, and ϕ(t2) is a sentence that results from replacing at least one occurrence of t1 in ϕ(t1) with an occurrence of t2, eliminating the “=” of t1 = t2. Recurring ti presumes that ti is univocal throughout, and recurring ϕ presumes that the sentential context ϕ is not altered, syntactically or semantically, by the replacement. — IEP

None of that need be about ‘Lois believes that..’. It is all about clarifying the identical.

Referential opacity occurs between contexts. Indeed, it can be considered part of what defines a context. — Banno

‘Opacity’ points to ‘difference’ (one shielded by opacity from the other). Difference is the line drawn ‘between contexts’.

So identity is about sameness and difference. And all of the valid or erroneous permutations that follow, and that can be translated into analytic terms.

The guts of Davidson's article is the difference between "Superman is Clark Kent" and "Lois believes that Superman is Clark Kent". The former is a relation of identity between two characters, the latter a belief on the part of a third character. The two are very different things. — Banno

Yes they must be treated one at a time. Because the two are ‘different things’ as you say.

But, though they are two different things, the process of the third person believing, only occurs once Lois forms some concrete identity in the form of ‘what’ she believes; once she has a ‘what’, she can believe what she believes. So they are different things, but when discussing belief, identity must be part of the discussion. (Which is what Davidson seems to think.)

The ‘what’ in this case is ‘Superman can fly.’ Lois has identified the character of Superman as the particular flying man. She identifies the man ‘Superman’ and believes “he can fly”.

So again, although ‘identity’ and ‘believing’ are different, belief requires there be the identity of what in particular is believed.

We analyze “what” she believes the way we analyze referential opacity, sameness, difference, identity; we analyze believing differently.

We analyze “that” she is believing X with newer/additional terms. These may refer to referential opacity, but again, that is its own issue (or can be treated as its own issue).

Whatever sort of thing that belief is, it doesn't allow the sort of substitution we are envisioning. — Banno

Now we are just getting into the nature of the permutations between sameness and difference in references to Clark and Superman, and a way error and correctness can occur when misapplying the analysis of what is identified and what is not (Clark is the same as super; Clark is different; Superman can fly; Clark cannot fly;)

‘Lois believes that’ narrows the multiple permutations among Clark and Superman’s sameness and differences, down to one particular instance. So substitution found in the full story of Clark/Superman may fail without incorporating Lois particularly intentional, identity awareness, purpose in speaking (she may be lying about her belief).

the latter a belief on the part of a third character. — Banno

And we, the readers and analysts are a fourth party, necessary to account for the coherence of a belief (Davidson’s communication needed for objectivity.)

Comport yourself so that t1 shows up in the b sentence, and we can evaluate for referential opacity. — frank

But isnt Quine saying, let it show up in a belief context and transparency will be sacrificed quite as much as if you put it in quotes?

But isnt Quine saying, let it show up in a belief context and transparency will be sacrificed quite as much as if you put it in quotes? — bongo fury

No, you can't put it in quotes. It has to be:

t1 = t2
Lois believes t1 can fly
therefore Lois believes t2 can fly

I think what you're suggesting is a readiness to make certain sounds. That's what you're calling belief. That's fine, it's just that you won't get opacity that way.

Somehow you have to involve t1.

No, you can't put it in quotes. — frank

Not without losing transparency, no, exactly. But Quine says, do this instead:

t1 = t2
Lois believes t1 can fly
therefore Lois believes t2 can fly — frank

and things are no better. The " t1" in "believes t1 can fly" won't have the same reference as the one in line 1. So substitution not ok. Quantifying in (from outside) not ok. Lois' t1 not our t1.

So belief ascriptions are (if we're not careful) as chaotic in their logical consequences as quotations.

So you do get opacity that way. You don't necessarily get belief, no. You may just have lines of questionable logic about belief ascriptions, yes.

The " t1" in "believes t1 can fly" won't have the same reference as the one in line 1. — bongo fury

Why not?

Quantifying in (from outside) not ok. Lois' t1 not our t1. — bongo fury

”Lois believes that” narrows the multiple permutations among Clark and Superman’s sameness and differences, down to one particular instance. So substitution found in the full story of Clark/Superman may fail in the story of Lois’ beliefs — Fire Ologist

Why not? — frank

Because if it did, we'd be able to substitute?

Because if it did, we'd be able to substitute? — bongo fury

T1 is Superman. It's a rigid designator, which is identical to Clark Kent, also a rigid designator. The b sentence tells us that Lois believes Superman (the rigid designator) can fly.

a. Superman = Clark Kent
b. Lois believes Superman can fly
c. therefore Lois believes Clark Kent can fly — frank

Sentence c, with a de dicto reading, isn't true.

If we do a de re reading of sentence c, we would claim that Lois really does believe that Clark can fly, she just doesn't know she believes that, or maybe she just wouldn't put it that way? So with the endless shenanigans people have played with this problem, there are some views that say it's not a substitution failure.

The normal, everyday, commonly held attitude is that Sentence C is wrong, and as as a result of an invasion of propositional attitudes, we have referential opacity.

The normal, everyday, commonly held attitude is that Sentence C is wrong, and as a result of an invasion of propositional attitudes, we have referential opacity. — frank

Yes. Quine agrees. Maintain the common attitude by not quantifying in, and hence not trying to reconcile Lois' Superman with ours. Not cashing in on the (future!) "rigid" rhetoric. Not substituting. Not going de re. Not concluding sentence c.

Yes. Quine agrees. Maintain the common attitude by not quantifying in, and hence not trying to reconcile Lois' Superman with ours. Not cashing in on the (future!) "rigid" rhetoric. Not substituting. Not going de re. Not concluding sentence c. — bongo fury

Ok. You just can't do that if you're defining belief as a readiness to make certain sounds. If Quine contradicts me on that, I'd have to read what he says to see what it is I'm misunderstanding.

The thread is now very messy.

Quine showed that Frege's solution didn't work, and told us not to try to substitute in such circumstances. Not really an answer so much as a statement of the problem.