Hume, we shall recall, had no place in his metaphysics for the miraculous:
'...[N]o testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous, than the fact, which it endeavors to establish; and even in that case there is a mutual destruction of arguments, and the superior only gives us an assurance suitable to that degree of force, which remains, after deducting the inferior.' When anyone tells me, that he saw a dead man restored to life, I immediately consider with myself, whether it be more probable, that this person should either deceive or be deceived, or that the fact, which he relates, should really have happened. I weigh the one miracle against the other; and according to the superiority, which I discover, I pronounce my decision, and always reject the greater miracle. If the falsehood of his testimony would be more miraculous, than the event which he relates; then, and not till then, can he pretend to command my belief or opinion.
Humans always defer to the weight of experience, the credibility and number of witnesses, and other such factors in evaluating the truth of claims. A miracle, in violation of the principles experience establishes, are excluded by definition.
Lewis's objection is rooted in his supernaturalism, which holds that Reason--the lynchpin by which the Naturalist worldview is formed--exists only due to a force outside/beyond Nature that we call God. This assumes the Natural system is not closed and self-contained, which undermines Hume's basis for objecting to miracles:
If we stick to Hume's method, far from getting what he hoped (namely, the conclusion that all miracles are infinitely improbable) we get a complete deadlock. The only kind of probability he allows holds exclusively within the frame of uniformity. When uniformity itself is in question (and it is in question the moment we ask whether miracles occur) this kind of probability is suspended.
What Lewis is doing is casting doubt on the means by which we understand how the world operates. Of the reasons men believe in the Uniformity of Nature, he lists 1) like animals, we are creatures of habit, 2) we don't assume Nature will act differently because there's nothing to be done; these, he rightly notes, "would be just as effective in building up a false belief as in building up a true one." The history of discarded scientific concepts, which were arrived at through greater than normal observational rigor, tells us as much.
The third reason is that we would find a universe of irregularities "profoundly repugnant" and "detestable." Though Lewis doesn't say so, such a world would in fact be crazy, with observable physical phenomena occurring with no discernable reason or regularity. It would drive people mad, as the architects of the United States's torture policy knew all too well.
Because of his dismissal of the Naturalist's explanation for Reason--a curt disgust with the possibility of natural origin that is the root of his objection to Naturalism--Lewis says there is in fact no grounds for considering our faith in uniformity true, and that a sustained belief in natural order can only be held by admitting a God shares our "repugnance to disorder."
But this is no solution either. For Lewis's worldview requires miracles: "you have no security against it." What we are given, instead of a truly ordered system, is one that only appears so but at any moment can be rendered otherwise. Lewis analogizes God's miracles to a master artist who knows when to bend or break the rules to create a masterpiece--whereas novices don't know the rules and pedants refuse to deviate from them--but such capriciousness, employed on reality itself, comes off less like an artist and more like a tyrant who upholds the law until it becomes inconvenient.
Such a belief is detrimental to scientific inquiry. Lewis has already stated outright that,
The whole mass of seemingly irregular experience could never have been turned into sceientific knowledge at all unless from the very start we had brought to it a faith in uniformity which almost no number of disappointments can shake.
Yet he would propose a worldview in which irregular experience can be dismissed as a miracle, beyond our understanding, when the history of inquiry tells us exactly the opposite: a bright flash in the night sky is not a sign of God's wrath, it's an exploding star, whose company we'll be joining in a few million years. God didn't bring the Black Plague down on Europe, rats did. Lewis's appeal to the miraculous, so much more articulate than these primitive notions, is less an argument from ignorance than an argument for ignorance.