Abstract

It is widely assumed that our “evidence” is at least one source of the “justification” that we have for believing things—where this notion of “justification” seems to be a normative notion. More precisely, it seems to be an agential normative notion, evaluating the different possible attitudes that are available to an agent at a time, on the basis of facts that are just “given”—that is, facts that it is not available to the agent to change through the way in which she exercises her reasoning capacities at that time. However, given the meaning that the term “evidence” has in everyday English, what is in this sense “given” to you cannot be identified with your current “evidence”. First, some propositions count as part of the “evidence” that you now have precisely in virtue of the fact that you now believe those propositions; but these beliefs are not part of what is now “given” to you—they are themselves among the attitudes whose justification is in question. Secondly, what is now “given” to you arguably includes facts about what you believed in the past—but such facts cannot naturally be called part of your current “evidence” unless you now know or believe these facts. In fact, it seems, the concept “evidence” does not really belong to individualistic epistemology, but to social epistemology: the function of “evidence” is that it is what we can use to show that something is true to the contextually relevant audience.