How does Quine effectively shift from points he has made about knowledge acquisition to conclusions he makes about knowledge simpliciter?

Question

Quine propagated forward an approach to epistemology wherein there was no need of any sort of justification "beyond observation and the hypothetico-deductive method" (Quine 1981).

Quine is going after a foundationalist view of knowledge acquisition then, wherein one can attain knowledge only in a context of possessing some prior, more certain knowledge.

But while any anti-foundationalist might agree with him here, they might not agree with Quine that there is no certain knowledge period.

That is to say, there seems to be a gap between admitting that all knowledge is based on experience, or "observation and the hypothetico-deductive method", and admitting that all knowledge is uncertain.

How does Quine effectively cross that gap?

Answer 1

It's not entirely clear what you're asking or what gap you're referring to. W. V. Quine's approach to epistemology, known as naturalized epistemology, focuses on the role of science and observation in the acquisition of knowledge. According to Quine, knowledge is not acquired through a priori justification or some sort of foundational set of beliefs, but rather through the scientific method and the observation of the world around us. In this sense, Quine's approach is anti-foundationalist, as it denies the existence of any kind of certain, foundational knowledge.

Answer 2

That is to say, there seems to be a gap between admitting that all knowledge is based on experience, or "observation and the hypothetico-deductive method", and admitting that all knowledge is uncertain

The answer to that would come from his doctrine of meaning holism, first expounded in the last section of his famous essay "Two Dogmas of Empiricism", Empiricism without the Dogmas.

Meaning holism holds that the distinction between analytic and contingent propositions is a matter of degree. Analytic propositions, including the theorems of logic, are simply propositions that are recalcitrant to the test of further experience, but are in principle revisable. Contingent propositions, of which there are many gradations, are closer to the tribunal of experience.

Quine uses the analogy of a "web of belief" to illustrate this view, in which beliefs hold together with some measure of coherence. Beliefs closer to the center are more dependent on other beliefs, whereas beliefs closer to the boundary of experience are more dependent on experiential/empirical input. To give an example: If I believe that there are three trees in my backyard, that statement is more easily revisable from experience than my other belief that matter consists of fields of force. The former statement contains terms that get their meaning more from perceptual input, whereas the latter statement contains terms that are lodged in complex theoretical schemes.

That is to say, both statements get their meanings from both experience and other terms (intesionality), but one is by gradation more easily revisable than the latter because it's meaning is less theoretically dependent than the latter. Both are, in principle, revisable but to revise the first we need to adjust very few other beliefs, while to revise the second we'd have to adjust many other beliefs that hold theoretically together.

Logic, in this picture, is more or less a formalization of broad-scale invariances that are also indirectly dependent on experience/empirical input. That is, everything hinges on the cleavage of the web of belief to the perceptual boundary, but the deeper in the web beliefs are lodged, the harder to revise they are, because they depend on chains of beliefs and interdependent conceptual schemas.

You may find this characterization of logic unsatisfactory. If so, tell me why and I can try to explain how Quine would answer to the best of my ability and to the extent of my knowledge of Quine.

I hope this somewhat answers your question!

Answer 3

That is to say, there seems to be a gap between admitting that all knowledge is based on experience, or "observation and the hypothetico-deductive method", and admitting that all knowledge is uncertain.

I don't think there's a gap. Traditionally uncertainty is aligned with empirical knowledge. Since empirical knowledge is in principle revisable, it cannot ever reach the mark of certainty. General principles arrived at empirically, namely by some form of induction, are by definition incomplete or rather extrapolations from samples to the population. As such, they can never be certain, simply recalcitrant to the test of further experience. This is why induction is said to be ampliative, as it goes beyond the available evidence.

Philosophers and mathematicians reserve certainty for things like proof and deduction. These seem to hold come what may. Namely, it doesn't matter what happens empirically. However, Quine tried to attack this haven of certainty but saying that even proofs rest indirectly on the way the world is. The principle of contradiction and identity are simply so general at our scale of experience that they are not likely to be revised, but they break down, for example, at the quantum scale. The point Quine is making is that the difference between empirical propositions and analytic propositions is not sharp along Aristotelian genus-differentia, but rather a matter of gradation and degree.