The more I read this paper, the more frustrated I become with it. Here’s my latest frustration. Lewis cites four rules aimed at distinguishing relevant from irrelevant possibilities, where an irrelevant possibility need not be ruled out by one’s evidence in order for one to have knowledge. One such rule is the Rule of Resemblance, which Lewis formulates as follows:
Suppose one possibility saliently resembles another. Then if one of them may not be properly ignored, neither may the other.
Lewis then points out that the Rule of Resemblance yields an explanation of lottery puzzles (“It is the Rule of Resemblance that explains why you do not know that you will lose the lottery, no matter what the odds are against you”) and Gettier cases (“The Rule of Resemblance also is the rule that solves the Gettier problems…”). Note the verbs here: “explains why” and “solves”. Strong language indeed. And surely unjustified, for notice the caution that immediately precedes these strong claims:
We must apply the Rule of Resemblance with care. Actuality is a possibility uneliminated by the subject’s evidence. Any other possibility W that is likewise uneliminated by the subject’s evidence thereby resembles actuality in one
salient respect: namely, in respect of the subject’s evidence. That will be so even if W is in other respects very dissimilar to actuality–even if, for instance, it is a possibility in which the subject is radically deceived by a demon. Plainly, we dare not apply the Rules of Actuality and Resemblance to conclude that any such W is a relevant alternative–that would be capitulation
to scepticism. The Rule of Resemblance was never meant to apply to this resemblance! We seem to have an ad hoc exception to the Rule, though one that makes good sense in view of the function of attributions of knowledge. What would be better, though, would be to find a way to reformulate the Rule so as to get the needed exception without ad hocery. I do not know how to do this.
By now my frustration should be obvious. If you can explain away lottery cases and solve the Gettier problem with such devices, then we’ve had adequate explanations and solutions to these issues long before the publication of “Elusive Knowledge.” Moreover, the plausibility that suitable refinement of the Rule of Resemblance will cash out this promissory note depends centrally on the notion of salience, and here I honestly have no idea what Lewis means. If salience is a psychological notion, about what is conspicuous and what isn’t, the proposal is hopeless for addressing typical Gettier cases (since, e.g., in the stopped clock case, one doesn’t even consider the stopped clock possibility, and so resemblance to actuality can’t be salient). Moreover, if salience isn’t a psychological notion, but means instead something like a prominent feature of the case, I have no idea why skeptical hypotheses are not salient.
My suspicion here is that Lewis is thinking of all of this from the standpoint of the attributor and just carelessly talking about knowledge instead of knowledge attributions. If so, however, here is a place where not being careful about the difference between knowledge and ‘knowledge’ has a cost. The Gettier problem isn’t a problem at some level of semantic ascent, and so no theory about conditions under which it would be correct to say “S knows that p” can provide a solution to that problem (though of course it could be an important part of a solution.