What’s Good and Bad in In Defense of Miracles (1999, 2005)
Richard Carrier
[Part 2 of a more comprehensive Review of In Defense of Miracles.]
Many of the chapters in the book In Defense of Miracles: A Comprehensive Case for God’s Action in History (1997) make arguments that need to be addressed, or at least acknowledged. For instance, Richard Purtill‘s essay on “Defining Miracles” (61-72) makes a much-needed point. All too often the theist’s position is straw-manned by setting forth a definition of miracle that many theists do not really hold, and then arguing for the impossibility or difficulty of that definition. This can no longer be done in response to this book or any miracle claims made by the many authors involved, many of whom still engage in public debates.
These authors have made it their first point to define what they mean by a miracle, and it is not what many critics will expect. In Purtill’s words, “a miracle is an event brought about by the power of God that is a temporary exception to the ordinary course of nature for the purpose of showing that God has acted in history” (72). Though this definition creates certain problems for recognizing a miracle, it should be clear that this is not something that is logically impossible, nor is it something that is unable to be a subject of scientific investigation, two claims usually made by critics.
However, in the process of escaping these two criticisms, the theists have run right into the camp of the empiricists and are forced to defend themselves on another front: how can anyone recognize a miracle even if they see one? Winfried Corduan‘s chapter aims to address this (99-112), but it is so feeble it’s astonishing to me that his essay wasn’t dumped altogether. He shows no signs of having read Purtill’s chapter, and thus does not address problems specific to Purtill’s definition of a miracle. Moreover, his argument, even as stated, is too weak: he only aims to show that it is “possible” that miracles could be recognized. He never attempts to show exactly how we are to recognize them. Certainly, it may be possible to travel faster than light, too, but how exactly would one propose to do that? Corduan thus fails to make any useful point, and his is a major weak link in the book’s deliberately “cumulative” argument. I devote a separate chapter to Corduan’s Failure.
On the other hand, Geisler’s chapter successfully refutes part of Hume’s “argument against miracles,” which he says is largely responsible for the modern mindset against the miraculous. Geisler rightly notes that Hume’s argument was actually against belief in miracles, not against their possibility, but he does miss one nuance of Hume’s argument: Hume only addresses miracle stories, and never even touches upon the question of direct observation. On that question Hume is silent. But Hume’s argument is still too strong for one important reason: two hundred years have passed in which our understanding of the universe, and our ability to examine it, has magnified a thousand times, thus making possible what Hume in his own time thought impossible. Antony Flew explains how Hume was perfectly justified to hold the position he did at that time, by analogy with a similar case involving Herodotus (51-2). Moreover, so long as the state of the evidence remains as it is, we are still justified in sharing the same conclusion as Hume, even though we can now say it may yet be possible to change this conclusion, if new evidence were to arise, something Hume, unaware of the future, did not concede. This is what Geisler’s contribution establishes.
In short, Geisler adequately flushes out the fallacy in the claim that since miracles are unrepeatable but natural laws repeatable, therefore evidence of the latter will always outweigh the former. This is not sound as a logical principle, since it is possible for miraculous events to be common enough that, even should each one be unique, the preponderance of cases would stand as repeatable evidence that miracles, in general, do occur. Likewise, history is filled with the unrepeatable. We shall never be able to repeat Julius Caesar’s crossing of the Rubicon, but nevertheless we believe we can prove it happened. Although Flew and Hume argue that this sort of claim is believable because we know such an event to be possible, given the natural laws as we know them, this is not the whole story. For example, there was a long time when we could not adequately account for precisely how the pyramids were built, yet we still regarded the evidence that it had in fact been done by human hands as compelling enough to believe it.
The question is thus not just one of mere repeatability, but a combined issue of feasibility and evidence: we disbelieve what we currently regard as unfeasible, and what we lack sufficient evidence to believe in the first place. And since God may yet be proved to exist, it follows that God’s action in history, in the form of miracles, would in that case become feasible. Repeatability would come into the picture only when it came time to prove that God can work miracles (by Purtill’s Definition). Then all we would need is repeatable proof of his general ability, not the repetition of any specific event–just as all we need to believe that Shakespeare could write Othello is a general proof that men can write plays. Of course, although this makes belief in miracles logically possible, the current state of things might not make such a belief reasonable. But other authors were left to address that issue.
Another essay that was important to read, and yet very typical of contemporary apologetics, is Ronald Nash‘s chapter on how a worldview is a crucial consideration in any argument about what is possible or impossible or what we can expect (115-131). The upshot is that when the world is seen from a Christian point of view, it is not unreasonable to expect miracles to be possible and recognizable, even if it is justifiably unreasonable from a skeptic’s point of view. Of course, the problem here is that the worldview in question must first be justifiable. If it is unreasonable to adopt a Christian point of view, then there is no point in arguing about what is reasonable from that point of view, because it will not be reasonable to expect that point of view to correlate with reality. Nash does not succeed in making this essential case.
Of course, Nash argues that “naturalism” (a particular kind of atheism) is unreasonable (I address this attack in a separate chapter on Nash’s Argument), but he does not even attempt a case for the reasonableness of Christian theism. This creates a disconnect between his introduction, where he argues that Christian theism makes miracles plausible, and his conclusion, where he claims that naturalism (which makes miracles implausible) is unreasonable. We are left to wonder, even if naturalism is unreasonable, has he made the case that believing in miracles is reasonable? It might be, if Christian theism were reasonable, but he does not argue that. Of course, the rest of the book is somewhat structured to make that case, but primarily through the argument from miracles, and that cannot even get off the ground if you do not first establish that it is reasonable to believe in miracles in the first place. In other words, Christian theism must be reasonable without appealing to miracles before you can use it as a justification for believing in them. Otherwise, all you have is a circular argument, which is not an argument at all, but a mere assertion. This is not to say that you must show Christian theism to be reasonable before you can be justified believing in miracles, but you cannot use “Christian theism” as the basis for that justification unless you have first justified Christian theism.
A better point is made by J.P. Moreland in his essay on theistic science (132-148), although he drops the ball by leaving most of his argument in other sources which he does not quote, wasting space here to weakly argue for libertarian freewill. Though he uses this approach in a novel way, it is only of doctrinal relevance, since libertarian freewill is not required for miracles to exist or to be studied scientifically (I discuss his argument further in a separate chapter on Moreland’s ‘Christian Science’). Here I will discuss the underlying, key point–the purpose of this chapter in the book’s editorial strategy: miracles can be a subject of scientific investigation. In principle, I agree. But he does not make a good case. There is also something sly about his argument: he chooses the weakest case for the reasonableness of Christian theism, rather than what would amount to the best case if Christian theism were actually true.
For example, Moreland argues that a “god-of-the-gaps” approach is not unscientific, which is a difficult position to defend, although it can be true under certain conditions. But “theistic science” would be better defended the same way all sciences are: by showing the actual positive contributions of that science. For instance, if we could study the attributes of God the same way we can study those of human beings, such as by asking Him to do things and observing His capabilities, and observing His personality as reflected in His choices, moral as well as casual, and so forth, then it would make perfect sense to include theism as a branch of science, since its claims could then be tested experimentally, and that is one key feature of sound science–one of the differences between science on the one hand, and philosophy and theology on the other. It is also the one thing that would actually improve our knowledge about God and the universe. This would also follow for other features of Christian theism, for example the claim that certain people can heal others by the grace of God.
But Moreland can’t make this case, because it would betray the fact that there is no way such a science could ever take off, for God does not make himself available to be tested. Neither do faith healers for that matter. There is, to put it bluntly, nothing to study. This makes his case for theistic science look as ridiculous as it really is. No wonder he avoids it. Instead, he argues for his theistic science by only talking about null hypotheses (though he does not use that term, this is in fact what he is talking about), as if that were sufficient to make an investigation a science. After all, a science consisting of nothing but null hypotheses would never make any progress. It is true that it can be a valid scientific endeavor to experimentally refute a claim, but a science which did nothing but refute, which never experimentally tested a positive theory, would not be a very useful science.
Most importantly, however, Moreland seems to assume that debunking naturalist theories is the same thing as proving theism. But his “theistic science” will never really touch on the idea of a god, and so it can hardly be called “theistic.” It would only be able to show that one particular natural explanation for something was untrue. But there will always be another one around the corner, and his theistic scientists would be like hamsters treading a wheel. A real “theistic science” would actually seek to experimentally prove that a god exists, and as Beck defines God, that would be a feasible scientific endeavor. But that cannot be done by merely debunking alternatives–for there will always be a new alternative. And since science has repeatedly failed to experimentally prove that Beck’s god exists, the concept of a “theistic science” is ludicrous–even more ludicrous than the idea of a “psychic science” (and for a critique of how an exclusive focus of null-hypotheses has doomed parapsychology, see Dr. Susan Blackmore’s autobiography In Search of the Light: The Adventures of a Parapsychologist, 1996). I am left wondering whether Moreland knows this, and thus is being devious by avoiding the issue, or if he has somehow missed this essential point.
After Moreland, some contributions were obviously necessary to make their cumulative case for miracles, but are not really worth reading. For example, W. David Beck‘s essay on “God’s Existence” (149-162) is a rather poor summary of tired arguments for the existence of god that we have all seen refuted before. The one useful feature of his essay is that he actually defines “God.” Moreover, his definition may surprise you: “God is a being powerful enough to produce events in space/time” and thus doesn’t have to be omnipotent, “God is an intelligence with a capacity to frame the convergence of events in space/time” and thus doesn’t have to be omniscient, and “God is a personality with the moral concern to act in history” and thus doesn’t have to be omnibenevolent (149). Of course, most Christians believe God is all these things and more, and the book itself does not make a consistent case, since Stephen Davis gives a different definition of God later on (163). But Beck at least sets himself a task that is much less ridiculous. Even so, he fails to make a case, and in a separate chapter I address Beck’s Failed Attempt.
Perhaps the most useless essay included is that of Stephen Davis, who argues the seemingly inane point that a god acting in history is plausible (163-177). In particular, is it possible for God to be “immaterial” and yet still affect matter and energy? Does the idea of god being “timeless” still allow for discrete actions in time? Does it make sense for a divine genius to keep tinkering with a universe that he should have been smart enough to make tinker-free in the beginning? Does it make sense to even believe in a miracle-working God? These sound like impressive questions to tackle, but Davis only addresses them in the most cursory and predictable fashion. The only answer Davis gives to each of them is “it’s possible” and yet that hardly needs to be argued. This, then, is not an argument for belief, but a defense against alternative theologies, like Deism, which deny that a God acting in history is logically possible, and against atheists who like to bombard theists with boring, nitpicking arguments.
Davis’ contribution is essentially like that of John Feinberg (226-246), who desperately defends the notion of the trinity (and the dual-nature of Christ as God and man) by arguing vigorously that it is logically possible. Both authors acknowledge that being possible is nowhere near the same thing as being true, yet they say nothing about why we should believe that their solutions to these problems have anything to do with the truth. Consequently, both these essays are of very little value–unless you are one of those who think the answer to any of those questions is an unqualified “its impossible.” Since I am not one of those, I will not discuss these chapters further.
All the remaining chapters deal in one respect or another with history or historical method, and so they will be discussed in more detail in separate chapters. One thing they all have in common is historical incompetence. It is often supposed that all historians do is read and write, and since anyone can do that, anyone can do history. This mistaken idea has captured virtually every Christian apologist, and these authors are no exception. Beckwith, Geivett, Clark, Newman, and Craig and Habermas each make at least one, if not several, mistakes of fact or method which no competent historian would likely make, much less get away with.
For there is a lot more to doing history than just reading and writing. I have not spent eight years of my life just learning how to read. Although I have devoted many of those years to learning how to read in four different languages (in addition to my own), something essential for any real historian, especially a historian of antiquity, the great bulk of my time, and my professors’ attention, has been spent upon two tasks which take years to perfect and are essential for history to be done competently:
- First, it is necessary to become so immersed in a time and culture that its features become familiar, and connections and comparisons readily made, and importing modern assumptions into ancient contexts easily avoided
- Second, it is necessary to master all the means and methods of error and deception in every form of historical evidence, from material evidence to inscriptions to literature, by constantly exposing yourself to these materials and studying the common features of obvious, and not-so-obvious, lies and mistakes.
The skill of the historian rests not only in his ability to recognize what he is looking at and to understand what it would have looked like, what it would have meant, to the people of the time and place being studied, and what it signifies to us about them and what they wrote, but it rests also in his ability to avoid being duped. As we will see, none of the authors in this book possess any appreciable skill as an historian.
Finally, Geivett and Habermas compose the book’s introduction, which, among other things, surveys the modern history of historical thinking in basically a single page (15-6). But they forget to provide a similar summary of ancient historical thinking. Since the central focus of the book is on miracles recorded in antiquity, not to address how historical thinking in that period would affect the transmission and reception of stories is a serious fault. It is, however, indicative of how apologetics gets done: Christians think only within the box of their own time, culture and literature. Rarely do they read widely in the non-Christian literature of ancient times, much less tackle in any depth the critical issues of ancient history. Instead, they often import modern assumptions when examining ancient claims. This, too, will become clear in my more detailed analysis of their treatment of historical issues in this book.
Return to this review’s Table of Contents to read more detailed critiques of specific chapters in In Defense of Miracles.