How is meaning-holism a problem for the analytic-synthetic distinction?

Question

I read Quine's Two Dogmas, and I also read two of Harman's papers:'Doubts about Conceptual Analysis,' and 'Analyticity Regained?' I have a couple of observations: Both Quine and Harman adopt a kind of holism. Quine thinks that analyticity is not to be understood in terms of verificationism, and his argument from this involves confirmation-holism (I'm not sure what this amounts to either, but that's a separate question). Harman, on the other hand, disagrees with quine along with pretty much everyone else that 'meaning' can be made sense of. But Harman doesn't think that analyticity can be made sene of in terms of meaning because meaning-holism is true. So I have two questions:

1. What is theory of meaning-holism that Harman endorses and how does it differ from Quine's confirmation holism?
2. How would meaning-holism be a problem for the analytic-synthetic distinction?

Answer 1

Answer to #2. Quine's view is controversial and so here is how I understand his position. The central thrust of "Two Dogmas" is (i) there is no coherent notion of analyticity and (ii) arguments for reducing the meaning of each term to its immediate experience fail. These two tenets Quine contends are dogmas of modern empiricism, and their failure to withstand critical scrutiny entails (i) a blurring of the boundary between speculative metaphysics and natural science and (ii) a shift toward pragmatism and holism.

Quine does not say that the analytic-disinction is impossible; rather, he argues: (A) that there is no coherent and adequate account of analyticity as analytic truths fall into two classes: (i) logical truths, and (ii) those that rest on some other notion (e.g. synonymy, definition, explication, paraphrase, interchangeability, semantic rules) AND (B) that a reductive verificationism cannot save the notion of analyticity. [Insert Quine's arguments for all of these positions here]

His conclusions are these: (i) without a clear demarcation account of analyticity, the supposition that there are some statements whose truth whose factual component is null is not defensible. For Quine then, the blurring of the analytic and synthetic boundary implies that all statements have some factual component to them.

Quine wrote a follow-up work to "Two Dogmas" called “Two Dogmas in Retrospect”, originally published in The Canadian Journal of Philosophy (1991), which is a revised version of the paper he presented at the University of Toronto in December of 1990; it is reprinted in Quintessence: Basic Readings from the Philosophy of W.V. Quine (2004).