Miracles is perhaps the deepest book of Lewis’ that I have yet read. This is because it challenges philosophical presuppositions and explains particulars of certain Christian doctrines in ways that both stay true to traditional teachings and, where possible, accord with the way we know things work in nature. In the first two chapters he sets the tone by explaining that he’s not attempting to prove the miracles of the Christian faith actually occurred; he is not attempting to assess the actual historical evidence. However, he does split the camps into naturalists and supernaturalists by defining a naturalist as one who believes that nature is all there is (305). Further, he defines a miracle as an “interference with Nature by supernatural power” (305). Thus, if Nature is all there is, then plainly miracles do not occur, whereas Lewis sets out to show that if it is possible that supernaturalism is true, then miracles cannot be ruled out as a possibility.
The next step in Lewis’ argument is to show that naturalism has a serious problem. This problem concerns the seed of what is closely related to the argument from reason (cf. Victor Reppert’s C.S. Lewis’ Dangerous Idea and Alvin Plantinga’s Evolutionary Argument Against Naturalism). First, he argues that all of human knowledge relies necessarily on reasoning being correct, or at least generally true (313). Here’s how he summarizes his thinking: “A theory which explained everything else in the whole universe but which made it impossible to believe that our thinking was valid, would be utterly out of court” (313). Due to this, it seems naturalism cannot account for bringing rational thought from the sub-rational; it is always possible that naturalism should have made it to where we only reasoned to the pragmatic truth of survival rather than truth itself (318). If that is so, then naturalism undercuts the entire basis for reasoning, and thus is self-refuting.
Lewis then deals with a few red herrings to the whole argument, including that early Christians accepted miracles like the virgin birth because they simply did not scientifically know better—a thought which Lewis takes to be ludicrous (342). He then moves to more serious objections, including the idea that miracles are violations of the necessary laws of nature, and so cannot occur. His argument is that the laws describe what happens in nature—all else being equal. Of course, whether or not all things really are equal depend on whether or not something intervenes from the outside; essentially, it depends on whether or not miracles are possible (352). Thus, this objection does not succeed.
Lewis then moves to a couple of chapters where he argues that even if early Christians did not have perfect understanding of particular truths, it does not follow that the core truth of what they defended was inaccurate (370-71). Once having laid this foundation, he attempts to adjudicate between different types of religion (after all, if he only establishes that miracles are possible, but not which religion or type of religion is most plausible, the question does not fully resolve). In the end, he decides that we can at least know some of what God positively is, and if that is so, miracles cannot be ruled out (384).
Lewis then moves to a discussion on probability. If, after all, miracles are improbable, then they should not be believed (according to Lewis). But what is it that makes it improbable? If naturalism is true, then one cannot say that it is probable that nature is uniform (i.e., actually obeys laws and will not suddenly change in the future); this suggests that naturalism is not true (395)! He finishes the work with a few chapters on the important miracle of the Incarnation and Virgin Birth, and helpful appendices on the soul and providence.
Lewis, I think, accomplishes his stated goal when he establishes serious problems for naturalism and suggesting miracles are at least possible. His goal is not to show that the historic details of Christian miracle stories are true, but even still, he takes pains to show that the Incarnation, Resurrection, and Ascension of Jesus Christ are all quite interrelated and plausible. One may not leave the book convinced that the claims of Christianity are true, but neither can he honestly leave convinced miracles are impossible—not without dealing thoroughly with this work.
There are many strengths to be found in this work, both on “offense” and on “defense.” On the offensive side, the argument from reason against naturalism is powerful. If any theory we think up results in the invalidity of thinking, then that theory counts as self-refuting (313). This important (and I think self-evident) principle guides Lewis in his argument. He showed, successfully, that materialistic versions of naturalism (which are the most prominent versions) entail this kind of self-refutation. This is because if everything is nothing but material, and this material obeys fixed laws, then reasoning is nothing more than how the material in our brains is acting and reacting, and is not true reasoning (312-13). But Lewis does not stop here, showing that even other forms of naturalism can suffer from this type of malady. Beliefs are merely “psychological events” (315), and simply because they are caused is no guarantee that they have any justification (we are asked to consider the madman whose mental condition surely has a cause, but his delusions are certainly ungrounded or unjustified). So these psychological events have causes, but seem to lack the ability to have sufficient grounds for thinking them to be rational.
On another offensive move, Lewis takes pains to defend the idea that we have what Plantinga would later call “properly basic beliefs.” These beliefs cannot be justified by any further beliefs, and just seem to us to be true. Lewis uses the uniformity of nature and the problem of induction to talk about this, and thus solves this problem with properly basic beliefs. He writes, “In advance of experience, in the teeth of many experiences, we are already enlisted on the side of uniformity” (394). For Lewis, if these properly basic beliefs are trustworthy, it remains hard to see how so on naturalism (but quite easy on supernaturalism—we were designed to have such beliefs!).
On the defensive side, Lewis defends the possibility of miracles against common objections. The first kind of objection has already been mentioned: that the early Christians were simply ignorant, and this is why they accepted the accounts of miracles. However, as Lewis points out, Joseph certainly knew the normal course of the births of babies, even if he could not articulate precisely why. As he writes, “St. Joseph obviously knew that” (342). The second kind of objection was the a priori ruling out of miracles due to the necessity of the physical laws. Interestingly, Lewis did not argue much against the necessity of the laws. This is a strength because it is the natural place to go. However, if one grants the contention of his opponent, but shows it does not entail the desired result, one has a much stronger position.
Despite the fact that I just praised Lewis for granting the discussion on the necessity of physical laws, it does seem that he should have pushed back more against it. For instance, why should any Christian (or even non-naturalist) grant that the physical laws are necessary (and hence any violation is “self-contradictory,” cf. 351)? There does not seem to be the same type of incoherence (if any) going on in the statement “Something exceeds the speed of light” as in the statement “There are married bachelors.” While Lewis’ approach can be applauded, the Christian need not grant a logical necessity condition attached to the natural laws. This is especially true when it is considered that the laws are empirically oriented (that is, they are not the result of pure, a priori reasoning with inescapable conclusions).
Another potential weakness was in the area of historical investigation. While it is true Lewis was not concerned to investigate the historical claims of miracles (which was entirely correct, in my view), it seemed he overstated his case. He thought it impossible for history to adjudicate miracle claims, since in order to do so, one must decide “whether miracles are possible, and if so, how probable they are” (304). Incidentally, this is why his discussion of probability comes far too late in the book (his concept of probability plays such a crucial role in what follows that its absence at the beginning was notable).
It is not clear to me that one must know (or, at any rate, be able to show) that miracles are possible in order to know that one has occurred. At least, one must know that miracles are not a priori ruled out, and to that extent Lewis is right. But it seems that Lewis also seems to think there is an intrinsic probability assigned to miraculous events, and that historical evidence does not serve to discuss probability with respect to miracle claims. It seems to me this is not correct. Mike Licona, for an example, maintains that one ought to adopt a stance of credulity with respect to historical claims, and then assess the evidence according to various criteria (cf. The Resurrection of Jesus). If this is right, and if one has done the spadework to defend against allegations of impossibility, then one can proceed even if for any given miraculous event E, E is highly improbable with respect to enumerative probability. For on the basis of historical evidence, one may find that E is not only probable, but the highest probable explanation of the facts of the matter there is!
A final note is not necessarily a weakness, but an implicit bias that runs throughout that may affect how a reader views some of Lewis’ contentions. He seems to be writing from at least a partially Thomistic perspective. He speaks of the form of the body, how the body is needed for the human person, how God is timeless, and even discusses more extreme forms of Thomism that speak of knowing God only by what he is not (in a critical sense). Being aware of this Thomistic influence will help the reader better understand where Lewis is coming from, and allow him to take what he will from it based on his experiences with Thomism.
This book comes recommended for both naturalists and supernaturalists, though I would not recommend it for the average layman. It is more difficult to follow the overall argument than Mere Christianity, and some readers who do not understand the connections may give up. However, for those facing challenges to their faith, this is a highly recommended read.