Even if some cases of knowledge seem to fit the evidentialist model well, it is a stretch to extend the position to other cases, such as cases involving memory knowledge. What is my evidence, for example, that I ate eggs for breakfast, or that my car is parked in the driveway? There might be some vague phenomenology involved in remembering these things. My memory beliefs might be accompanied by a characteristic sort of feeling--a kind of confidence, or perhaps a kind of attraction. But it is a stretch to think that these states are evidence--to think that these states "support" my memory beliefs, or that my beliefs are evidentially based on such states. (pp. 5-6)
I get a bit irked about this, because the answer seems so obvious as to make the evidentialist suspect that the externalist is just being obtuse. I have three comments, first an argument that there is some sort of evidence on which memorial beliefs are based, then one strategy for explaining the nature of this evidence, and then an alternate strategy.
1. Earl Conee argues in "Seeing the Truth" that seeing the truth of obviously true propositions like
P Every golden trumpet is a trumpet
is, contra Alvin Plantinga's account in Warrant and Proper Function is not merely a matter of believing a proposition in a certain way. Seeing the truth, is, Conee thinks, a separate matter from believing what one sees to be true. Consider P - there will be, I think, if you attend to the experience of coming to believe P to be true, a separate experience apart from believing P, which believing P follows. This may be described as a "seeming that P" or "seeing that P", a kind of mental experience such that one bases a belief on it. One reason for thinking that this is a separate experience is that it is imaginable that it occur without believing P. Suppose you had good reason to think that you were liable to deception about P, perhaps you'd been assured that you'd been given a drug that would make things seem obvious in just this way and despite how seemingly true P is, you refrain from belief. Further, you could imagine simply experiencing it, but before forming the belief, stubbing your toe and being distracted by the intense pain. So, at least for me, I agree with Conee, there is a separate experience of seeing the truth of obvious propositions.
Likewise, I think that there is just this sort of experience with memory. I wonder what I had for breakfast, and it seems strongly to me that I had a clif bar and I believe that I had a clif bar. I can imagine this experience being separated from belief too. I might get distracted - stubbing my toe might do it. I might have a defeater. Suppose I also remember that as a favor to a friend today, I participated in a psychological experiment and was given a false memory that I ate a baloney sandwich for lunch. I am aware of this deception, but it still seems to me that I remember eating a baloney sandwich for lunch. It seems conceivable that I might have an experience as of seeming to remember having a baloney sandwich, but not also believe that I did, because I know that my memory to that effect is false. So, just as in the case of seeing the truth of an obvious proposition, there is an event prior to belief and separable from belief that seems (to me at least) to be a ground for my so believing.
Thus, there is some reason to think that evidence exists that may justify my memorial beliefs. The burden then, for the anti-evidentialist is to show that there are cases in which one genuinely remembers and should be justified in believing, but in fact, one has no evidence or insufficient evidence. I am inclined to think that such cases will not exist - the problem of forgotten evidence seems to me to be a non-starter too, but is a problem for another blogging.
2. So, supposing that there is evidence, what strategies are available for the evidentialist? I think that one obvious case is "seeming evidentialism" or "phenomenal conservatism". Michael Huemer has defended the view that seemings that P may play a role in justifying P. I find this view plausible, though it (of course) has its detractors. If seemings are evidence, the evidence mentioned above (separable from one's memorial belief) may be a seeming of this sort, and if so, then that seeming counts as evidence for belief in the remembered proposition. The following is true in cases of remembering that P: it seems to me (often quite strongly) that P is true (or was true). So, if seemings are evidence, then seemings with a particular character or phenomenology may provide the evidence in memory.
3. Suppose one's not a fan of seemings. Other accounts seem available to the evidentialist. Suppose that P seems to be true in that way in which memories present themselves to us. If justification involves some proposition being the best explanation of our evidence, then it may well be the case that, given our background information, the best explanation of our memorial experiences are that the events we seem to recollect in fact did occur. The idea would be similar to the use of best explanation in justifying our perceptual beliefs. In fact, the situation is exactly parallel, just the class of propositions justified by these experiences are different. In one situation they are present-tense propositions about the external world, in the other, they are propositions what was the case - or at least about times earlier than the present (if one likes things put into B-theoretic terminology).
It's possible that one thinks that neither of the two ways I've offered are sufficient for justifying memorial beliefs. I think that it's plausible that either may work (they may not even exclude one another), or even that some other way might work (e.g. a Chisholmian approach where more finely grained epistemic principles govern justification of beliefs on certain sorts of experiences). But if these are controversially true, they warrant discussion rather than quick dismissal by anti-evidentialists (as do similar approaches for reductive strategies for justifying the acceptance of testimony).
