Question My question concerns the 20th Century doctrine of "logical postivism" and its apparent refutation. Its distinction between analytic and synthetic statements seems to me straight forward and an important one. Wittgenstein's quote seems appropriate: "On what cannot be spoken of one must remain silent." I understand that logical positivism has been successfully refuted by Quine and others. I cannot grasp that refutation. One of those arguments seems to be the "indeterminacy of translation"); an argument I understand and accept. I also understand that ALL language has different connotations to different people. However, it seems impossible to make an understandable "synthetic" statement about metaphysics. That is, if we cannot verify the existence of something empirically, such as a concept (God, for instance), we cannot come to any agreement about it. In other words what I find valuable about logical positivism, as a materialist, is that metaphysics is simply speculation and cannot be understood similarly across a human population. It has is some way, "no cognitive content" that we can grab hold of and evaluate. It seems to be indeed, what logical positivists called it: "non" sense. The practical result from this appears to me that we can debate claims like the the "moon is made of cream cheese" or that "Washington crossed the Potomac" by means of bringing scientific and historical evidence in favor or not if favor of the proposition. However, the claim that "God wants us to be happy" is a claim without any meaning. Consequently, nonsense. Is it not true that there is a difference in "kind" between those two statements? I certainly see a difference. Now, is it the case that the refutation is only about "language" and not about the "real world." If so, I understand that language only approximates, or pictures reality and is not reality itself. But that's true of ALL statements in language. ??? Thank you.

Accepted:
June 28, 2010

Comments

Yes, many of the logical positivists drew a sharp line between analytic truths and synthetic ones, respectively, those that owe their truth merely to the rules of language that determine meaning and those that also owe their truth to how the world is. The distinction seems to turn on acknowledging that sentences have determinate meanings in the first place - in some cases, those meanings settle the truth of the sentence (the analytic ones) and in other cases they do not (the synthetic ones). Quine's thesis of the indeterminacy of translation claims that sentences do not have such determinate meanings: in addition to facts like how many moons the Earth has, there are no facts about what some string of words means. (This can sound outrageous and much care needs to be taken about the thesis being advanced and the reasons for it - no time for that here!) And so the thesis of indeterminacy rejects a presupposition of the distinction between analytic and synthetic truths.

You say there's clearly a difference in kind between certain metaphysical assertions and the claims of science because one can bring evidence to bear in the latter case but not the former. But on Quine's view, we can't directly bring evidence to bear even on scientific claims. Individual sentences don't really have any empirical import and what really gets tested against the evidence is a theory, a large body of inter-related propositions. (He called this position holism.) Perhaps you're struck by the thought that no testable predictions follow from the claim "God wants us to be happy", while they do from the claims of natural science. But that's not so: ask yourself which testable predictions follow just from, say, Newton's Second Law of Motion. Of course, Quine accepts Newton's Laws, but not the claim about God. That's because the one, but not the other, is integrated into a body of theory that has great empirical support (and not because the one, but not the other, can be directly confirmed by the evidence).