In this post, I want to illustrate a methodological assumption that is sometimes made in epistemology.The idea is that facts about the linguistic appropriateness/inappropriateness of an assertion can count as evidence for or against an epistemic theory

Here’s an example as to how someone might make use of this assumption.

The idea is that there is something inappropriate about assertions in contexts where the lack of certainty is made salient.

For example, suppose George says, “I have two hands,”

Now suppose Peter says, “Oh, come on George! For all you know you could be in the Matrix right now. Surely, you don’t know you have two hands.”

George reiterates, “I have two hands.”

Some think that there is something inappropriate about George’s second assertion.

One way to explain this inappropriateness is to say that George doesn’t know he has two hands during his second utterance.

This could be carried a step further. A skeptic might say that George doesn’t know he has two hands (and never knew).

Someone like Hawthorne or Stanley might say that this case better supports a kind of subject-sensitive invariantism. Since George’s first assertion seems appropriate, but George’s second assertion does not…it may be that the standards for knowledge shifted once it was made salient to George that he lacked certainty. He had knowledge during his first assertion, but lacked it during his second assertion.

There are two assumptions at play here. The first is some kind of assertion norm like the following.

Knowledge is a Norm of Assertion
Assert P only if you know P
(E.g. it is inappropriate to assert P if you don’t know P)

The second assumption is something like the best explanation of the inappropriateness of George’s assertion is best explained by the violation of the above assertion norm.

I have a lot of issues with this kind of reasoning. I’ll get to those later. For now I just want to have an example so that people know what I’m talking about.

(Posted from my Android phone)