From the archives: Review of Craig Keener’s Miracles

I really, really, wanted to give you all a new book chapter draft today, but I’m having even worse trouble in that department than usual, so I’m going to let some schedule slip occur. For now, enjoy one of the main things I’m trying to rework to create the chapter, my review of Craig Keener’s book Miracles, originally published back in January. Feedback on this post may help shape the book!

This is my promised review of Craig Keener’s book Miracles. It’s actually a two-volume set, but I’m going to call it a book, for simplicity’s sake. Now my verdict is that I don’t know how to express how mixed my feelings are about this book. I’ll start with the good.

Modern miracle stories

For a while now I’ve been quite aware that there are a lot of Christians who like to tell miracle stories about things that allegedly happened very recently. These are not all friend-of-a-friend type stories. Sometimes it’s things people claim to have seen themselves, or that someone they know very well has seen. So for example, a year or two ago I was listing to an interview with Evangelical biblical scholar Mike Licona, and about a third of the way through the interview Licona trots out this story that supposedly happened to an unnamed Yale-educated friend of his that involves an encounter with a demon while he was in China.

Now Mike Licona is one of those evangelicals who claims that the resurrection of Jesus can be shown to have happened with historical evidence (in fact, the resurrection of Jesus was the main topic of the interview). But my reaction to hearing that in the interview was to think that these modern reports of the supernatural are way, way more interesting than the alleged evidence for Jesus’s resurrection, because with these modern stories there is, at least in principle, the possibility that you could go track down the witnesses, do a real investigation, and potentially—if the thing really happened—get together quite a bit of documentation (and if it didn’t happen, uncover reasons not to take the story at face value).

Now based on what I know about the history of paranormal investigation and some of the adventures of the Society for Psychical Research, I’d quite confidently predict that if Christians ever did that kind of investigation, they’d eventually realize that they’re not going to find good evidence for supernatural phenomenon with those kinds of stories. Still, you could do an interesting investigation.

Enter Craig Keener. The main thing he does in Miracles is collect lots and lots of stories of seemingly miraculous happenings, most of them healings. He doesn’t really try to do any in-depth investigation of the stories he reports, but he’s up-front about that. He talks about his limitations, like a lack of funding for investigation, lack of time off from his teaching duties, and his own lack of medical knowledge, and suggests that maybe in the future other people will be able to build on his work and do an investigation that doesn’t suffer from those limitations. I think that if Keener’s book inspires other evangelical Christians to spend some real time and effort scrutinizing these kinds of stories, then the book will have done some good.

But now the bad. The problems with Keener’s book begin on the very first page when he states the book’s thesis:

The book’s primary thesis is simply that eyewitnesses do offer miracle claims, a thesis simple enough but one sometimes neglected when some scholars approach accounts in the Gospels. The secondary thesis is that supernatural explanations, while not suitable in every case, should be welcome on the scholarly table along with other explanations often discussed (p. 1)

This is what I call a weaselly thesis statement because it clearly says much less than what Keener wants to say. It lets him that hint at some very controversial claims, but because he’s officially only defending these seemingly banal claims, it gets him off the hook from really having to defend his views. So the primary thesis is something that I agree with and I agreed with before I even began reading the book, and it’s really the kind of thing you would defend the an article, not a two-volume set.

And the role of the secondary thesis, in practice, ends up being to allow Keener to spend a lot of time making ad hominem attacks against those big nasty skeptics who want supernatural explanations off the table and then Keener can fight the good fight to have the explanations on the table. (Whatever that means—part of the problem here is that “on the table” is vague, so it’s not even clear with thesis is.)

Regrown limbs

Now when I accused Keener of making ad hominem attacks, what do I mean that? Ad hominem is a phrase that I think is that horridly overused for any time someone is mean to someone else. But I mean it in the strict sense of substituting attacks on a person’s character for arguments in a situation where the person’s character is irrelevant.

So for example, let’s look at the issue of claims of regrown limbs. There’s a website called WhyWon’, (formerly known as that makes an argument:

For this experiment, we need to find a deserving person who has had both of his legs amputated. For example, find a sincere, devout veteran of the Iraqi war, or a person who was involved in a tragic automobile accident…If possible, get millions of people all over the planet to join the prayer circle and pray their most fervent prayers. Get millions of people praying in unison for a single miracle for this one deserving amputee. Then stand back and watch.

What is going to happen? Jesus clearly says that if you believe, you will receive whatever you ask for in prayer. He does not say it once — he says it many times in many ways in the Bible.

And yet, even with millions of people praying, nothing will happen…

What are we seeing here? It is not that God sometimes answers the prayers of amputees, and sometimes does not. Instead, in this situation there is a very clear line. God never answers the prayers of amputees. It would appear, to an unbiased observer, that God is singling out amputees and purposefully ignoring them.(LINK)

What’s the point of this thought experiment?

How do we know, for sure, that God does not answer prayers?… we simply pray and watch what happens. What we find is that nothing happens. No matter how many people pray, no matter how often they pray, no matter how sincerely they pray, no matter how worthy the prayer, nothing ever happens. If we pray for anything that is impossible — for example, regenerating an amputated limb or moving Mt. Everest to Newark, NJ — it never happens. We all know that. If we pray for anything that is possible, the results of the prayer will unfold in exact accord with the normal laws of probability. In every situation where we statistically analyze the effects of prayers, looking at both the success AND the failure of prayer, we find that prayer has zero effect. Prayers for amputees never work. Medical prayers never work. Prayers for “good people” never work. Battlefield prayers never work. That happens, always, because God is imaginary. Every time a Christian says, “The Lord answered my prayer,” what we are seeing instead is a simple coincidence or the natural effects of self-talk.(LINK)

Now I have to admit that my gut reaction to this argument is that this is a horribly unsophisticated argument. But I think the truth is that this is an argument that any idiot can see is correct, and the part of me that instinctively dislikes this argument is the part of me that’s terrified of being mistaken for any idiot. The fact that the only prayers God “answers” are prayers for things that have a chance of happening anyway is powerful evidence that God never actually answers prayers

If you wanted to be a little more charitable towards the “this argument is so unsophisticated” line, you might say that religious people must have good responses to obvious arguments like this one, or else there wouldn’t be any religious people, but I don’t think that’s true. I think most religious people are just good at not thinking too hard about problems for their religious beliefs like this one. Deep down, most of them have to know that prayer doesn’t really ever work, which is why they only pray for things that have a chance of happening anyway. This is a good example of how religious people compartmentalize.

Now, here’s what Keener says about regrown limbs, and this is where the ad hominems come in:

Some skeptics about healing argue (beyond the evidence) that almost anything can be psychosomatic, whereas clearly organic restorations of limbs are never reported. Certainly there are not many such reports (including the Bible), but they do appear occasionally; in one extraordinary report, for example, a leg severed beneath the knee grew back. [A footnote cites a book by televangelist Pat Robertson--Hallquist] Elsewhere, useless or shriveled limbs have become functional and filled out miraculously quickly. Those committed to disbelief that such miracles can happen will, of course, dismiss such claims; but while the rareness of such claims (hence limited possible analogies) does invite caution, one might also get the impression that some skeptics’ demands for particular kinds of evidence become stricter whenever evidence of the demanded sort appears. (p. 747)

Now Keener is completely missing the point here. The significance of the regrown limb issue is that if regrown limbs happened, they’d avoid a lot of problems you get with other kinds of healing claims. You eliminate the possibility that it could be a coincidence you, elliminate the possibility that maybe the doctors made a mistake. If someone’s leg really regrew it’d be pretty easy to document conclusively, if it happened under the right circumstances. If the limb regrows almost instantaneously, it’s going be hard to be mistaken about witnessing that.

So we shouldn’t expect false reports of regrowing limbs to happen very often. It’s going be hard to get away with making up a story like that, and we should expect that to deter people from making up stories about regrowing limbs. However, people do sometimes tell outrageous lies. So the fact that there is a story of a regrowing limb in a book by Pat Robertson doesn’t prove anything. It doesn’t change the fact that the lack of evidence of regrowing limbs is suspicious, and the fact that skeptics aren’t impressed by such stories isn’t evidence of closed-mindedness.


Now that I’ve talked about an example of a claim that people rarely make, let’s look at a claim that’s more typical of claims people do make:

Even solid medical documentation is not adequate by itself to surmount strongly held presuppositions, because one may insist in every case (even if there are thousands of them) that another explanation is possible. My colleague in Hebrew Bible, Emmanuel Itapson, was told that his third child had “the death chromosome” and would likely die before birth if not aborted. The family prayed, and the boy is now nine years old. Because 1 percent of those with this chromosome are known to live beyond infancy, one cannot prove beyond any doubt that prayer is the factor that helped him to live so long; yet I am prepared to grant that likelihood in view of the significant number of extraordinary answers to prayer in Emmanuel’s circle, including one mentioned in chapter 9 and another in chapter 12 (p. 666–I did not notice this page number until after choosing the quote).

The first thing to notice about this is that this story is evidently being filtered through people who don’t have a lot of medical knowledge. The “death chromosome” presumably refers to a lethal chromosomal abnormality, but since there are many lethal chromosomal abnormalities, there’s no such thing as “the” death chromosome. Either someone misunderstood the doctor, or the doctor was dumbing down the diagnosis for the benefit of the parents. But whatever the case, it makes this story a little harder to evaluate.

More importantly, the way Keener introduces this case suggests he thinks it illustrates how unreasonable skeptics are. That’s frankly ridiculous. In fact, setting aside for a moment the other “extraordinary answers to prayer,” this case doesn’t provide any evidence at all for the efficacy of prayer. By definition, for every 100 times someone is faced with 100 to 1 odds, one person will beat the odds. In more religious parts of the world, including the United States, I’m sure that most people, maybe an overwhelming majority of people, pray when they or their children are faced with a serious illness. In that case, most odds-beating recoveries will happen after prayer. Because stories like this aren’t surprising even if you don’t think miracles happen, these stories aren’t evidence of anything miraculous.

This is why science is neat. At the most basic level, when we’re talking about the scientific study of prayer, we’re talking about checking to see if prayer leads to beating the odds more often than not praying. We’re also checking for things like bias among people recording the data and the placebo effect. (The placebo effect is when something that wouldn’t normally do anything, like a sugar pill, leads to people doing better merely because they think they’re getting treated.)

What about the fact that this guy’s circle of friends supposedly has had a whole bunch of remarkable recoveries? Is that evidence of something supernatural? Again, no. The problem with saying “Oh it looks like we’ve got this really improbable cluster of cases,” without doing rigorous statistical analysis, is that humans are really bad at eyeballing probability. We have a tendency to see patterns in randomness, and we even sometimes judge rigged events as more random than really random ones. To give just one of many examples, psychologist Steven Pinker describes one experiment which found that “people think that genuine sequences of coin flips (like TTHHTHTTTT) are fixed, because they have more long runs of heads or tails than their intuitions allow, and they think that sequences that were jiggered to avoid long runs (like HTHTTHTHHT) are fair” (Pinker, The Better Angels of our Nature, p. 204).

This is something that’s actually not all that surprising, once you think about what randomness means. Random doesn’t mean being distributed evenly. There’s nothing about randomness that prevents events of a certain kind from clumping together just by chance, so it’s going to happen some of the time. Yes in some cases it’s going to be tempting to say “this clump is just too improbable to have happened by chance,” but except in the very most extreme of cases it’s just not something you can say without careful statistical analysis.

Furthermore, even in cases that seem extreme, what might be happening is that inaccurate reporting is taking events that were only somewhat improbable and blowing them up into something extremely improbable. There are a number of reasons that could happen. One of them is lack of medical knowledge, which I’ve already pointed out in the death chromosome story. And the point of taking a rigorous scientific approach is to avoid those kinds of problems.

Keener does discuss scientific studies of the efficacy of prayer briefly. He mentions studies with positive results, but does so only very briefly, vaguely saying in one case that “many have questioned the study” without discussing the criticisms or trying to determine whether the criticisms are valid. This reflects a general problem with the book: Keener’s approach to important questions is often to say, “some people say X, some people say otherwise, moving on…”

He does devote a paragraph to discussing the results of a 2006 prayer study funded by the John Templeton Foundation, a foundation that funds academic research related to religion. The study was the largest such study to date, and according to the authors it tried to make up for shortcomings in previous studies. It found no evidence of any benefit from prayer. Keener says a number of things to try to minimize this result, including asking, “Would God favor someone or not because they belonged to a control group?” (pp. 708-709).

Well maybe not. But you could also ask similar questions about prayer in general—why an omnipotent, omniscient God would need our input on how to run the universe. And whatever you think of those theological questions, they don’t negate the value of science, nor do they negate the problems with using collections of stories as proof of the supernatural.

Keener does at one point given very brief argument for why we can’t study the supernatural scientifically:

Since science depends on observation and experimentation, and since a “miracle is by definition an irreproducible” experience, even documented miracle cures by definition cannot fit precisely the expectations of science as it has been most narrowly defined. While affirming miracles, one scholar warns that “miracles cannot be investigated by the usual scientific methods since we cannot control the variables and perform experiments” (p. 608).

This is pretty clearly wrong. If God gave one man the power to work a certain limited kind of miracles at will, that would be reproducible, and subject to scientific experimentation. In particular, he could submit to a test under conditions designed to rule out fraud and delusion, and then we could see if he could still produce the apparent effects under those conditions. There are many people who would be happy to arrange such a test, including the James Randi Educational Foundation, which offers a $1,000,000 prize to anyone who can demonstrate paranormal abilities under controlled test conditions.

You might want to argue that God would never grant miracle-working power in that manner, but consider this famous passage from the book of Exodus (Exodus 4:1-9):

1 Moses answered, “What if they do not believe me or listen to me and say, ‘The LORD did not appear to you’?”2 Then the LORD said to him, “What is that in your hand?”

“A staff,” he replied.

3 The LORD said, “Throw it on the ground.”

Moses threw it on the ground and it became a snake, and he ran from it. 4 Then the LORD said to him, “Reach out your hand and take it by the tail.” So Moses reached out and took hold of the snake and it turned back into a staff in his hand. 5 “This,” said the LORD, “is so that they may believe that the LORD, the God of their fathers—the God of Abraham, the God of Isaac and the God of Jacob—has appeared to you.”

6 Then the LORD said, “Put your hand inside your cloak.” So Moses put his hand into his cloak, and when he took it out, the skin was leprous—it had become as white as snow.

7 “Now put it back into your cloak,” he said. So Moses put his hand back into his cloak, and when he took it out, it was restored, like the rest of his flesh.

8 Then the LORD said, “If they do not believe you or pay attention to the first sign, they may believe the second. 9 But if they do not believe these two signs or listen to you, take some water from the Nile and pour it on the dry ground. The water you take from the river will become blood on the ground.”

If you’re an orthodox Christian who thinks this story from Exodus really happened, as far as I can tell the only thing you can say here is that the reason God doesn’t empower prophets in this manner today is that he doesn’t want to make the evidence for miracles too clear.

I’ve actually heard Christians say something like this. What they’ll say is that God has given us clear enough evidence, but he’s avoided giving us too much so that closed-minded skeptics can continue being closed-minded (because we all know that skeptics are wicked and need to be set up to be punished for their wickedness). There are two problems with this. First of all I don’t think the premise that skeptics are typically closed-minded and wicked is really true. But perhaps more importantly, I just don’t think explanations that suppose that the universe is in some way conspiring to avoid giving us very good evidence are generally the best explanations.

For example, you can say that the reason people who claim to be psychic are never able to demonstrate under controlled test conditions that are designed to rule out cheating is that the presence of skeptics somehow disrupts psychic powers, but I think the more plausible explanation is that nobody really has psychic powers and precautions against cheating are doing exactly what they’re supposed. Or, a UFO organization once claimed that 2% of Americans have been abducted by aliens. In response, Carl Sagan quipped, “It’s surprising that more of the neighbors haven’t noticed” You could suppose that the aliens have various kinds of super-technology that allows them to hide almost all of the evidence, but a better explanation is that people who claim to have been abducted by aliens are suffering from hallucinations, false memories and so on (see The Demon Haunted World, pp. 64, 181). Likewise, I think the best explanation for the lack of evidence for miracles is that there aren’t any.

So it can’t be disputed that the evidence for miracles is less than perfect. That’s enough to disprove Keener’s insinuation that skeptics of miracles wouldn’t be persuaded by any evidence. The vast majority of skeptics would have no trouble believing in the power of prayer if there were as much evidence for it as there is for the power of penicillin. But there isn’t.


Another problem with stories of miraculous healings is the problem with doctors making mistakes. Consider this story:

In 2006, I interviewed Dr. Douglass Norwood; during the time of most of the testimonies he recounted, he was a Moravian pastor. He mentioned several dramatic healings but explained two in the greatest detail. The first case, which took place in Suriname, I have recounted earlier. The other case, more relevant for this chapter, involved his wife, Sarah. Her neck was broken and her spinal cord severed in a car crash on December 14, 1982; she remained paralyzed at the Rusk Institute for six months. Despite the medical impossibility of her walking with a severed spinal cord, she began walking within twenty-four hours of being “anointed with oil,” leading to a number of conversions among the hospital staff. Doug notes that the healing is only 90 percent complete, though it is a medical miracle; she walks with considerable effort and requires medicines, but that she walks at all still astonishes those who examine her (p. 438-439).

I picked up this story because it’s listed on a table at the end of the book as one where Keener had personally talked to the guy who supposedly witnessed this, and it was also listed as a case that Keener was especially confident was really miraculous. Personally, I just don’t see it. Given that the woman in this story could only walk “with considerable effort,” I don’t understand the mindset of someone who would look at this and say “this is an amazing miracle.” My guess is that what happened is that this woman really was badly injured and she just wasn’t quite as badly injured as the doctors initially thought. There’s nothing difficult to explain here.

Now Keener, once again has something to say about the issue of doctors making mistakes and once again it involves an ad hominem:

Those who question supernatural healings often attribute the more convincing cases to an initial misdiagnosis. Although genuine misdiagnosis does occur at times, this approach sometimes has been used as a means to explain away extranormal healings retroactively, and sometimes the initial evidence is too firm to aver a misdiagnosis… To simply dismiss every cure as a case of prior misdiagnosis is to allow one’s presuppositions to determine the outcome, especially when it involves many cases and the prior diagnoses involve multiple physicians. One healing evangelist reasonably complains that if critics really believe that so many hundreds of healing cases result from initial misdiagnosis, they should be raising an outcry against such widespread misdiagnosis instead of divine healing.

And once again this is all beside the point. The issue is not whether skeptics are closed-minded, the issue is that if the case is going to be touted as powerful evidence of miraculous healing, it needs to be possible to show with some degree of certainty that the doctors didn’t make a mistake. Keener claims that misdiagnosis can sometimes be ruled out, but he supports this claim with just a footnote. As happens all too often in the book, there’s no discussion of an absolutely central claim.

On top of that, there’s other silly rhetoric here. The thing about dismissing every cure as a case of misdiagnosis misses the point because a possible explanation doesn’t need to explain every case to be a serious concern. This is just like how there’s no single cause of UFO sightings that turn out not to be extraterrestrial spacecraft. Also, the complaint about hundreds of misdiagnoses is silly because in a world where millions upon millions of people seek medical care every year, a few hundred mistakes isn’t all that much. Doctors aren’t perfect.

A final point

I don’t know if you’re getting sick of this post by now, but I am, so one last point: Keener tries to explain the lack of medical documentation for alleged miraculous healings by proposing that God has seen fit to mainly work healing miracles in the context of missionary efforts in the Third World, and that makes them difficult to document (see i.e. p. 662-704-705). Again, while this is a possible explanation, I don’t think it’s the best explanation. Alleged miracles not happening under circumstances where they can be well documented is just what we would expect if no miracles were happening all.