Wolfgang Schwarz

Bashing "Two Dogmas"

Christian has to write an introductory paper on Quine's "Two Dogmas". I wouldn't like to do this. I think "Two Dogmas" is excessively overrated, and should only be read in courses on the history of American philosophy. Unfortunately, Christian seems to agree with most of my misgivings. Maybe I find some opposition here.

"Two Dogmas" consists of three parts: §§1-4, §5 and §6. In §§1-4 Quine argues that there is no distinction to be drawn between analytic and synthetic statements. His argument appears to be as follows:

1. If we cannot give a reductive definition of "analytic", no distinction can be drawn between analytic and synthetic statements.
2. A reductive definition of "analytic" must not use the notions of "meaning", "necessity", "synonymy", "semantical rule", and other notions I don't like, unless these in turn are reductively defined.
3. The common definitions of "analytic" discussed in §§1-4 all violate premise 2.
4. Therefore no distinction can be drawn between analytic and synthetic statements.

This argument is sound but for the fact that the conclusion doesn't follow from the premises and all premises are false: As for premise 1, I agree that it would be strange if we had to accept "analytic" and "synthetic" as irreducible, but since there are clear cases of both, this would still be better than denying the distinction. As for premise 2, I can see no reason at all to accept the restriction. As for premise 3, Quine's discussion in §§1-4 actually showed that some proposed definitions do not violate premise 2, namely the definitions of logical truths and of truths by stipulative definition. These do not comprise all ordinary examples of analytic truths, but still they illustrate that a distinction that at least partly matches the intuitive distinction can be drawn even under Quine's restrictions.

In §5, Quine argues against a view he calls "reductionsism", which says that all statements are translatable into statements about experiences. The argument is:

1. Carnap hasn't succeeded in constructing such a translation manual.
2. Therefore there is no such manual.

He then uses this conclusion to argue against the idea that there is a 1-1-mapping between statements and experiences that would confirm/infirm these statements. Here the argument is:

1. The most naive motivation for this idea is reductionsism.
2. As we've just proved, reductionism is false.
3. Therefore the idea is false.

To me these arguments look rather silly. At least this time I do believe the conclusion. But not every argument for a true conlusion is a good argument. In fact, I'm not even sure if Quine is not himself committed to the views he intends to refute. See below.

In §6, Quine suggests that the unit of empirical confirmation (and infirmation) is total science. This doesn't really follow from the fact that it's not single sentences, let alone from any considerations about analyticity. But that's okay, since Quine doesn't tell us that it does. He merely presents an alternative view.

The view is however quite unclear to me, so I don't really know what to say about it. The basic idea seems to be that if there is a conflict between theory and experience, any part of the theory may be revised, but no (proper) part of it must be revised to restore consistency. I would like to know how such a conflict is supposed to arise: Is it that the theory contains (or implies) statements about experience which may contradict incoming experiences (transformed into propositions)? I guess not, since then it wouldn't be true that "any statement can be held true come what may": If theory says I now have a reddish visual experience, and experience says I don't, how can I keep holding true that statement of the theory? Besides, can there be a conflict even when the theory has jettisoned the law of non-contradiction? After all, Quine says that the laws of logic themselves are nothing but statements in our theory. This also generates a well-known problem: Why is believing { A, not B, if A then B, Modus Ponens } any worse than believing { A, not B, if A then B }?

Even granting the notion of conflict between theory and experience, I wonder if Quine's sketch isn't utterly trivial: Did anyone ever doubt that many different sets of statements are consistent with the same empirical data? Of course I can hold true any statement come what may if irrationality doesn't bother me -- so what? It would be non-trivial to suggest that generally many different good theories are equally well supported by the same data. If this is what Quine suggests, I can't even find the beginning of a hint of a reason for believing it in "Two Dogmas".

Finally, Quine also appears to defend a kind of holism about meaning. This is somewhat at odds with his holism about confirmation: If to rescue my believe that p I have to change other bits of theory, and these other bits partly determine the meaning of "p", I don't really rescue my belief that p. Maybe we mustn't even ask about the meaning of single statements and beliefs? Anyway, now we can carry out the "reductionist" programme: let T be any theory. Let's assume for simplicity that only finitely many experiences E_1,...,E_n are consistent with T (whatever that means). Then we can simply translate T into the disjunction of E_1,...,E_n.