In Beversluis's C. S. Lewis and the Search for Rational Religion, he presents Lewis as someone who accepts the challenge posed by what later can be known as the evidentialist objection.
I think this is a very tricky claim to make out. The big problem is to try to figure out what is packed into the evidentialist objection.
Consider the classic Cliffordian statement of the evidentialist objection:
"It is wrong always, and everywhere, and for anyone, to believe anything upon insufficient evidence."
And compare it with C. S. Lewis's statement of "evidentialism":
I am not asking anyone to accept Christianity if his best reasoning tells him that the weight of evidence is against it. That is not the point at which faith comes in.
I would make note of the fact that these are two different types of claims. Clifford is talking about what is morally wrong for everyone to do, Lewis is talking about what he is asking someone to do. Clifford says there must be sufficient evidence for any belief, Lewis is just saying he won't ask for belief if a person's reasoning tells him there is evidence against the belief.
In fact Lewis's account of the ethics of belief raises questions as to whether it is a moral issue at all. He writes in Mere Christianity:
I used to ask how on Earth it can be a virtue--what is there moral or immoral about believing or not believing a set of statements? Obviously, I used to say, a sane man accepts or rejects any statement, not because he wants or does not want to, but because the evidence seems to him good or bad. If he were mistaken about the goodness or badness of the evidence, that would not mean he was a bad man, but only that he was not very clever. And if he thought the evidence bad but tried to force himself to believe in spite of it, that would be merely stupid.
Well, I think I still take that view.
So here we don't have Clifford's moral thunderbolts against those who believe for insufficient evidence, just the suggestion that it isn't a very bright idea.
Further, Lewis seems to cast the net of evidence widely; more widely that, I think Clifford would countenance.
The man who accepts Christianity always thinks he has good evidence; whether, like Dante, fisici e metafisici argoment: or historical evidence, or the evidence of religious experience, or authority, or all of these together. ("On Obstinacy of Belief")
But surely Plantingian properly basic beliefs have experience or authority backing them up, at the very least.
One way of explaining the difference between Lewis and someone like Clifford is to make the case that Clifford is a strong rationalist, who holds that the position that "in order for a religious belief-system to be properly and rationally accepted, it must be possible to prove that the belief-system is true." Further, "prove" has to be parsed in such a way that in order to prove something true one should have evidence sufficiently strong that all rational persons ought to be convinced. Of course, there are many issues on which all persons are not convinced. For example, there is a flat earth society. But this, we suppose, is not because the evidence for a round earth is lacking; rather it is due perhaps to some emotionally-driven biases.
But it doesn't look as if Lewis thinks in this way. In very confident modes Lewis comes off sounding like he thinks he can meet the strong rationalist's criteria for rational belief. Beversluis notes points at which he says "we are forced to conclude..." But he also notes that Lewis says that the evidence may be sufficiently strong to eliminate the psychological possibilty of doubt, but not the logical exclusion of dispute.
My own view is that Lewis is a critical rationalist, who believes that "religious belief systems can and must be rationally criticized and evaluated although conclusive proof of such a system is impossible." That is not to say that his language and tone do not suggest otherwise at times. If by evidentialist we mean the he thinks one must evaluate evidence when forming religious beliefs, of course he is an evidentialist. If we mean that he thinks there is some burden of proof on religious as opposed to non-religious beliefs, or that we should only believe if we have evidence that everyone ought to be able to accept, then he isn't an evidentialist in that sense.