Reply to W. M.'s Response to my views on Opacity
by Dennis J. Darland
June 27, 2009
Last revised 27.06.2009 13.46 time
Copyright © 2009 Dennis J. Darland

WM says, "The difference between opaque and non-opaque propositions seems to hinge on the degree of omniscience governing the proposition." I think this is not quite what he means to say. The believers of propositions have varying degrees of knowledge - not the propositions themselves. Often facts are stated in a sort of impartial third person form, but any actual occurrence of belief in these propositions would be by a person with limits on his knowledge. It is in statements of belief that this limited knowledge reveals itself in apparent opacity. When we describe the situation, we include facts which are left out in the argument for opacity. I will repeat WM's restatement of that argument:

1) Cicero = Tully.
2) S believes that Cicero denounced Cataline.
3) S does not believe that Tully denounced Cataline.
4) from 1 and 2 (and sensitivity) it follows that S believes that Tully denounced Cataline.
3 and 4 are contradictory.

WM then pretty much correctly states my views but goes on to say: "Darland seems to believe that the 'real' referent in the symbol_r relation is not capable of being substituted for the symbol in belief_r. I don't see why not, as at least from a practical standpoint. The problem of opacity still will not arise because it is made clear that for S Tully and Cicero refer to different things, and so the substitivity of those names fails on an ontological level when the referent of those names is used instead of the mere names."
I am not suggesting a (practical) change for ordinary speech. Sometimes there is confusion in practice when two different people have the same name, or one person goes by more than one name, but we are able to sort these things out most of the time - there may be some mistakes which never are corrected.
My problem was to express these facts is a way which would satisfy Quine, and preserve substitivity of identicals.
WM proposal amounts to suggesting that 1 be replaced by:

1b) S believes Cicero ~= Tully

But from 1 and 1b and subtitivity we cab infer:

1c) S believes Cicero ~= Cicero

The crux of the problem is to serve its purpose, language has to be about the world, not just our beliefs about the world.
We want to say "Cicero = Tully" - not just "S believes Cicero = Tully".
But there is very little that is not subject to error - so there are many cases where we want to assert "p", and not merely "I believe p", but in which it may turn out that "p" is false.
Logic is useless if we limit ourselves to belief.
From I believe (x)(fx ⇒ gx) and I believe fa it does not follow that I believe fb.
But from (x)(fx ⇒ gx) and fa it does follow that fb.
Hence the usefulness of logic.
And the reason for WM's mention of omniscience I believe.
In a way, we have to pretend we are omniscient, even though we know we are not.
My analysis of belief permits this to be understood as needed, and not as something to be applied constantly.
We must generally assert what we believe, keeping in mind that both our and other's beliefs (and consequentially assertions) are subject to revision.

Other remarks:

My analysis is intended for belief in general not just for where opacity occurs, and I expect other propositional attitudes as well, although I haven't worked on them (one thing at a time.)
I think the H2O - water problem is a bit different and needs to be explained better.
There is no 'general failure of substitivity' - my purpose was to save substitivity without abandoning belief as Quine would have us do.
Cases where there is no referent for a symbol can be handled with a definite description.
(BTW: I don't quite know what to make out of WM's references to my 'idealism'.)

Back to Top