On Contextualism: It's irrelevance to skepticism

Intro
First I will briefly discuss the notion of context sensitivity and one main reason for thinking that “knows” (and the relevant set of cognates) is context sensitive. I will question this reason and then present the case for thinking that the context sensitivity of knowledge attributions provides a basis for an answer to skepticism. I will argue that it *cannot* do any such thing.

What context-sensitivity is
A term is context-sensitive, roughly, when it expresses a different property or in this case relation in different contexts of utterance. Equivalently, they could be thought of as expressing the same relation but as being tacitly indexical or indexical-like in that they contain an argument place whose value is set by context. A sentence can be context-sensitive when it contains a context-sensitive term. Context-sensitive sentence types vary in truth-value from token to token due to semantic features of terms they contain.

Epistemic contextualism
Epistemic contextualism focuses on how the context-sensitivity of the term “knows” causes knowledge ascription sentences to be context sensitive. It asserts that the truth-value of knowledge ascriptions can vary according to context. “knows” is said to be context-sensitive in that context of utterance can vary the standards about what must obtain for knowledge ascriptions to be true. In high-standards contexts, for example, it might be that an extreme amount of evidence is required for true beliefs to count as knowledge and thus the correctness of a knowledge ascription to a case of true belief. In low-standards contexts, less evidence can be required. Standards might vary with stakes. As the stakes rise and fall, so might the standards.

There are other, less plausible accounts of how contexts shift, but I will not consider them. What I have to say about contextualism is independent of what features of conversation cause context shifts. Usually, it is the context of the attributor which is said to govern the attribution, not the attrubutee, though there is logical space for such a proposal.

The case for epistemic contextualism
The datum
The argument for epistemic contextualism is that it explains a facet of our reactions to skeptical arguments and in doing so solves the skeptical puzzle. The datum from which this explanatory argument for contextualism starts is that we seem to have at times very firm convictions which seem to conflict. We ordinarily take our commonsense external world beliefs to be well-enough justified to constitute knowledge. What external world proposition could be more obvious than that I have hands? Yet, when we consider skeptical arguments, some are quite strong, they seem to show that we don’t know such things after all, for that we do know them seems to imply what is obviously false: that we know we are not in the Matrix, that we have ruled out all possibility of error, that we wouldn’t believe what we do if it weren’t true, or what have you. The case for skepticism seems to undermine all our common sense external world beliefs.

The Explanation
Contextualism purports to explain these conflicting judgements. When we are not discussing or considering or worrying about skeptical arguments, the requirements for knowledge are lax and so we really do know true propositions based on simple perceptual experiences. But when we *are* discussing, considering, or perhaps worried about skeptical arguments the context is different and the requirements for knowledge are much higher: high enough that we can’t meet them and the skeptic is right in this context. So not only does epistemic contextualism purport to explain the opposing pulls of common sense and skepticism, it also shows that skepticism is false for most of us most of the time.

The Case Against Contextualism
There is little reason to believe contextualism and some reason to doubt it.
Ambiguity alternative simpler
The explanatory argument is undercut by the simpler theory of ambiguity: that there are just two (or possibly a few) different senses of “knowledge” an ordinary one that we use in ordinary life and one invented by Descartes or other pre-Enlightenment epistemologists in which knowledge entails the elimination of all logically possible doubt.

The Feldman Factor
General statement
From the right angle, it is not hard to see that contextualism does not and cannot solve the problem of skepticism. This is pointed out by Feldman 1999 which I generalize a but further. Essentially, epistemic contextualism asserts that there is some requirement R on knowledge such than in some contexts that requirement is met easily and in another it is not. But for any requirement R on knowledge (or at least any serious proposal for one or at least any serious proposal I’ve ever heard of) there is some standard skeptical argument A such that A seriously calls into question our knowledge by exploiting R.

Skeptical models
Let a skeptical model be an ordered triple áR,C,Añ consisting of a requirement R on knowledge, a context C which generates R, and a skeptical argument A which exploits R in C. For any given skeptical model—or at least many—there will be legitimate questions concerning the success of A. Since skeptical models already take contexts into account (they are like world-indexed properties and thus essentialism is true of them) it is just impossible for contextualism to address them. The relevant question in any skeptical model is whether A is a successful argument.