Squaring the Culture

"...and I will make justice the plumb line, and righteousness the level;
then hail will sweep away the refuge of lies,
and the waters will overflow the secret place."
Isaiah 28:17

08/13/2009 (5:19 pm)

I Don't Have Enough Faith to Be an Atheist: Are Miracles Possible? Part II

In previous installments I’ve established

  • that there’s a need for explaining why Christianity is the most reasonable position for an educated, skeptical individual to take (see the post here);
  • that there exists such a thing as absolute truth, and that truth claims may be made about religion just as they can about any other topic (see the post here);
  • that using the Cosmological Argument, the scientific fact that our universe had a beginning establishes that something like a Theistic God must exist (see the post here);
  • that using the Teleological Argument, the anthropic principle establishes that the universe was designed for life, which requires a designer something like a Theistic God (see the post here);
  • that using the Moral Argument, the fact that we all recognize that some things are more morally acceptable than others requires that a universal moral standard exists outside of ourselves, requiring a moral God (see the post here);
  • that the summation of the Cosmological, Teleological, and Moral arguments gives us a composite picture of what the theistic God must be like, and that that composite picture is remarkably similar to the God of the Christian Bible, but not quite similar enough for a positive ID (see the post here);
  • that if God exists, then miracles are neither impossible nor disproved nor violations of nature (see the post here.)

This installment will examine the most common argument modern philosophers raise against believing that miracles have occurred.

Can No Miracle Be Believed? Hume and His Friends

Though modern skeptics who have not studied philosophy tend to claim miracles are not possible, modern philosophers usually take a more clever position. They don’t claim miracles are strictly impossible; they simply say that we cannot know when one has happened. There is no circumstance, according to this argument, in which we might actually declare an event a miracle. This was the approach taken by David Hume in the 18th century, and updated more recently by Anthony Flew. This argument is the most generally-accepted argument against the possibility of miracles.

The argument is based on uniformity of experience. Hume’s version begins by establishing that humans base their knowledge on experience, and gain experience by repetition. But miracles are events that have not occurred in normal situations and are never repeated, observes Mr. Hume, so they can never be established by experience. So, when a wise person hears of a miracle, his uniform experience says that it could not have been a miracle, it must have been something else. This, Hume regards as the only wise conclusion, even as proof. Norman Geisler abbreviates Hume’s argument like this, using Hume’s own words:

  1. “A miracle is a violation of the laws of nature;”
  2. “Firm and unalterable experience has established these laws;”
  3. “A wise man… proportions his belief to the evidence;”
  4. Therefore, “the proof against a miracle… is as entire as any argument from experience can possibly be imagined.”

Hume concludes, “There must, therefore, be a uniform experience against any miraculous event, otherwise the event would not merit that appellation.”(1)

Anthony Flew produced an updated version of this argument in his entry entitled “Miracles” in the Encyclopedia of Philosophy. He observes that Hume was concerned, not with whether miracles occurred or not, but with evidence. He then observed that the very characteristics that would cause an observer to call an event miraculous, are the same characteristics that would cause an historian or a scientist to discard them out of hand. Though miracles may logically be possible, scientifically or historically they are not,

“…for it is only and precisely by presuming that the laws that hold today held in the past… that we can rationally interpret the detritus of the past as evidence, and from it construct an account of what actually happened.”(2)

Like Hume, Flew asserts that since knowledge is based on uniform experience and repetition, miraculous events, which occur only once and are not repeatable, can never achieve a level of credence that would make a reasonable observer accept that a miracle had actually occurred.

Both Hume and Flew prove far too much. If we are to take their arguments seriously, most events in the past disappear along with miracles, because most events studied in the past are singular events that have occurred only once. What if I were to say that a single, highly motivated individual not even 20 years old could conquer pretty much the entire known world in fewer than 10 years, traveling only on foot or on horseback? None of us have ever seen it done. Poof! Alexander the Great just disappeared, because uniform experience mitigates against believing that his achievements are possible. Which of us has ever seen a universe appear out of a singularity, containing time, space, matter, energy, and laws of nature? It has happened only once that we know of. Poof! The explosion that began our universe just disappeared, because uniform experience mitigates against believing that such a thing could happen. Which of us has ever seen life spring spontaneously from a pool of complex chemicals, and begin the process of evolution? Poof! Life on earth just disappeared, because if such a thing ever happened, it happened only once, and uniform experience mitigates against believing such a thing could happen. And so forth.

Ah, but miracles are even less believable than those, which are events that occurred in nature, says the Flew-Hume skeptic. The skeptic has just begged the question; there is no particular reason why anybody not already believing the presuppositions of Naturalism or some other science-promoting philosophy should regard miracles as uniquely improbable. If God exists, miracles are no less probable than any other rare event. No, Hume and Flew based their arguments on experience, and it is to experience that we must stick. All events that humans have not experienced recently are equally improbable in this view (without begging the question), and by their arguments, overwhelmed by a uniform flood of contrary examples, and not to be believed.

In order to see where these folks go wrong, consider the following numbers: 02, 05, 28, 33, 54, and 30. According to the web site of the Pennsylvania lottery, those were the numbers that won the Powerball sweepstakes on August 2, 2009. (3) The odds against those numbers appearing in that order in any particular trial of the Powerball lottery are a trillion to one; in practical terms, that is a probability of zero. We may fairly call that “improbable,” and note that uniform experience mitigates against those particular numbers appearing. Yet, none of us have any difficulty believing that those numbers actually occurred in the PA lottery on August 2, 2009. Why not?

The answer is, we have experience with official state web sites and with state lotteries, and we know they usually report accurately. Though most of us have never visited the Pennsylvania Lottery web site nor heard anything about it, we still have no trouble believing that this one-in-a-trillion event occurred because our experience with official state web sites and lotteries generally suggest that the reported results are trustworthy. If we wanted to verify the result further, we would find a copy of the August 3, 2009 edition of the Pittsburgh Post-Gazette or the Philadelphia Inquirer at the local library and look it up. The experience that we use to assess the evidence is not of our experience of the event, which in itself is highly improbable, but of the reliability of the witnesses and circumstances surrounding the event.

Hume and Flew, and many, many others, make the mistake of confusing the principles of study with the object being studied. They claim we should doubt the EVENT (the object being studied) because it is rare. In practice, we gather experiences about the way people and objects behave in ordinary circumstances, and then we apply them to the evidence supporting the reports of the events in the past (the principles of study). We have experience with coinage, for example, and we know that nations strike coins with the faces of famous men on them to identify those coins as official. We know that ancient governments did the same, and put the faces of emperors on coins. So, when we find an ancient coin with the face of Alexander on it, even if we have not heard of Alexander otherwise, we infer there was an emperor named Alexander. We also know that histories are written by ancient authorities intending to record events accurately, so when we find accounts of a conqueror named Alexander by Plutarch and Arrian, we believe them, even though the account has him conquering more than it would seem possible for a teenage general to conquer in a mere 10 years. We have experience with historians generally, we can evaluate the sources from which these historians drew their facts, and we can examine artifacts from the various parts of the world Alexander’s armies were alleged to have visited. The fact that Alexander accomplished things that were unbelievable makes us more suspicious about the accounts, but we verify them in the usual ways accounts are verified; we don’t dismiss them out of hand merely because what they report is rare. It is not the experience with the one, singular event that determines whether the event is believable or not; it is the repeatable experience with the sources and the types of evidence.

Geisler offers the additional example of the SETI project, the Search for Extraterrestrial Intelligence. SETI researchers scan random background radio waves looking for a pattern demonstrating intelligence. They are looking for other species of intelligent beings, which, I think it is fair to say, are less likely to exist than is the theistic God. And yet, even a single radio-wave transmission demonstrating a non-random pattern will be sufficient to prove that they exist. Why? Is it because we have vast experience of extraterrestrial beings? No, we have no experience of such beings whatsoever; but we have uniform experience with messages, and we know that they come from intelligent minds. Thus if we find a single message in the background radio signals, we have proved that mind exists, no matter how improbable that might seem. (4) What we evaluate is not the rarity of the event being studied, but the evidence for the event, based on our common experience with the elements in the evidence.

Persisting in the Hunt for Natural Causes

So, Hume and Flew are wrong simply to discard evidence of miracle claims because miracles are rare. Flew’s version of the argument adds another error, though, in that it observes that even if the evidence points to a miracle, researchers continue to look for natural causes rather than accept it. Flew’s argument focuses specifically on the scientist’s attitude toward miraculous events, saying that the scientist will always be looking for the natural law at work regardless of whether the event seems inexplicable or not. He claims this excludes the possibility that any event could ever be called a miracle.

In addition to the error already noted with this approach, and discounting for the moment the fact that most scientists will simply ignore a single, non-repeatable event altogether, we should note that science is a specialized and limited field that deliberately focuses solely on nature. The fact that a scientist might continue to look for a natural cause for an event that would otherwise be called a miracle is not sufficient reason for the rest of us to refuse to call that event a miracle, because by doing so the scientist is not making a judgment about the believability of the miracle. After all, the scientist does not, by omitting all facts outside of nature, make the claim that nothing exists outside of nature; he simply declares that he is not professionally interested in what does not. By looking for the natural cause, he is simply doing his job. Just because the scientist practices Methodological Naturalism, does not mean the philosopher must do so as well. Methodological Naturalism as a professional practice does not support Metaphysical Naturalism as a philosophy. Philosophers, who concern themselves with a wider set of data and seek all truth, could easily go on to call an event a miracle while the scientist continues in his professional demeanor.

In truth, the unending quest for natural causes is sometimes improper, because we can know a miracle wherever we understand the workings of nature. The Catholic Encyclopedia explains:

“… the miracle is essentially an appeal to knowledge. Therefore miracles can be distinguished from purely natural occurrences. A miracle is a fact in material creation, and falls under the observation of the senses or comes to us through testimony, like any natural fact. Its miraculous character is known:

  • from positive knowledge of natural forces, e.g., the law of gravity, the law that fire burns. To say that we do not know all the laws of nature, and therefore cannot know a miracle (Rousseau, “Lett. de la Mont.”, let. iii), is beside the question, for it would make the miracle an appeal to ignorance. I may not know all the laws of the penal code, but I can know with certainty that in a particular instance a person violates one definite law.
  • from our positive knowledge of the limits of natural forces. Thus, e.g., we may not know the strength of a man, but we do know that he cannot by himself move a mountain. In enlarging our knowledge of natural forces, the progress of science has curtailed their sphere and defined their limits, as in the law of abiogenesis.” (5)

Thus the greater knowledge of science makes the knowledge of miracles possible. The better we understand sperm and egg, the better we can establish whether a natural cause was involved in a pregnancy. The better we understand the medical side of life and death, the better able we are to assess a resurrection claim. In cases where the limits of nature are known, the continuing quest for a natural cause is often nothing more than a faith-driven assertion of Naturalistic bias.

Geisler notes many additional problems with Hume’s and Flew’s arguments against miracles. He notes, for example, that Hume evaluates the believability of evidence by adding it together rather than weighing its importance. He observes that the evidence supporting the rare event often overwhelms that for the common event. He notes both Hume’s and Flew’s eagerness to use arguments that can actually be used against their own particular worldview. And, he notes that both engage in a sort of special pleading for the naturalistic worldview.(6)

For this reason, if an event occurs that clearly fails to conform to the ordinary pattern of natural events, and if the evidence supporting the occurrence of that event is sound enough, we can believe that the event occurred. If the circumstances are right to suggest that the source of the event might be deity, we can call it a miracle.

Next time, I will examine a bit more closely under what conditions we should call an event a miracle.


(1) Geisler, Norman L., “Miracles & the Modern Mind,” in In Defense of Miracles, Geivett & Habermas, eds, Intervarsity Press, Downer’s Grove, IL, 1997, p. 74-75.
(2) Flew, Anthony, “Miracles,” in The Encyclopedia of Philosophy, ed. Paul Edwards, New York, McMillan, 1967, p. 350, as quoted in Geisler, op cit., pp 76-77.
(3) http://www.palottery.state.pa.us/news.aspx?id=124306
(4) Geisler, Norman, “Miracles and Modern Scientific Thought,” http://www.origins.org/articles/geisler_miracles.html
(5) Catholic Encyclopedia, “Miracle”, http://www.newadvent.org/cathen/10338a.htm
(6) Regarding rare events that have better evidence than common events, note that among astronomical bodies, the occurrence of bodies on which human beings can walk unaided by whole-body shields and breathing apparatus are extremely rare, but the evidence that at least one such body exists is rather overwhelming. See also Geisler, “Miracles & the Modern Mind,” op cit., chapter 6 for additional examples and arguments.

« « A Little Good News | Main | The Myth of Prevention » »


August 13, 2009 @ 10:04 pm #

This is a fascinating essay, Phil, and a good cursory review of Hume’s idea on miracles. Thanks for the hard work.

Did you know that Antony Flew has recanted, and is now a Theist?

August 14, 2009 @ 7:40 pm #

Great post on miracles, Phil.

And speaking of Anthony Flew, I read his book _There is a God_ (but it was borrowed so I don’t have it right now). It was so interesting to see him present his old case against God, miracles and faith in the first part of the book and then pick it apart in the second half– essentially rebutting his old self. He also seemed rather partial to the figure of Jesus and Christianity as a whole, though he has not embraced either (that I know of) as of yet .

August 15, 2009 @ 10:31 am #

darkhorse & dullhammer,

Yes, Flew is now a theist, in a very analytical, very non-spiritual sort of way. It looks like an interesting story, but I haven’t read much about it. Maybe dullhammer will re-borrow the book and give us a little report?

August 15, 2009 @ 12:18 pm #

The book is on my wish list at Amazon. I will order it and give you a review.

August 16, 2009 @ 4:13 am #

Sorry Phil, but there is a big difference between a “onetime event” and a “onetime suspension of the laws of nature.” Miracles involve the suspension of the laws of nature – Your examples consist of onetime events in which nothing that happens suspends the laws of nature, which remain uniformly regular to our experience.

The philosophers are arguing that any evidence that one could muster in favor of a miracle (a onetime suspension of the laws of nature) would be precisely the sort of evidence that experience (which tells us that the laws of nature are uniform and regular) would preclude us from considering. Your counter examples do not contain any events that experience tells us are impossible. So, the fact that they are one time events that we nonetheless believe, does not refute Hume.

The point was not that miracles were rare (one time events). The point was that an empiricist approach to understanding reality, which David Hume argued elsewhere is the only philosophically sustainable approach, requires our assumption that the laws of nature operated uniformly and regularly in the past (i.e., that they were never suspended). If we don’t assume this, we can’t study the past empirically, which is the only possible way to study it at all.

Additionally, in his statement that “there must, therefore, be a uniform experience against any miraculous event, otherwise the event would not merit that appellation,” Hume is saying that the very idea of a “miracle” presupposes the uniform experience premise regarding the laws of nature – otherwise the [allegedly miraculous] event would not merit the appellation [“miracle”]. But the uniform expierence premise renders miracles unthinkable for an empiricist. In other words, the very idea of a miracle is self-refuting for an empiricist.

A better approach would be to attack Hume’s empiricism. Hume’s empiricism was so extreme that he posited what is known as the “empirical criterion of meaning.” According to Hume, except for purely analytical propositions, an idea that cannot be traced back to sense date was not merely false or questionable – it was meaningless. It was nonsense (analytical and mathematical propositions are sometimes referred to as nonsensical propositions).

Successfully attack this version of empiricist philosophy and you’ll deal with his miracles argument.

I also recommend caution in concluding that you have refuted David Hume – he was probably the greatest analytical thinker in the history of the world.

August 16, 2009 @ 8:45 pm #

there is a big difference between a “onetime event” and a “onetime suspension of the laws of nature.”

A) I explained in the last installation that there is no necessity for suspension of the laws of nature (recall my example of catching the pen), and that if a God exists, the possibility of His acting similarly in nature surely exists.

B) I explained in this installment that the distinction you’re making requires that God intervening in nature be uniquely improbable — which begs the question, as it asserts the presuppositions of either Naturalism or Deism. I see nothing in your rebuttal that addresses this; you merely make the claim that miracles require “a suspension of the laws of nature” that is “impossible,” which, as I pointed out, begs the question; it’s not “impossible” if God exists.

It is certainly the case that a naturalist will never accept the evidence suggesting a miracle, but that means nothing to the objective philosopher. The naturalist is biased by his presuppositions.

I also recommend caution in concluding that you have refuted David Hume

It’s not originally my argument. Mostly, it’s Geisler’s. Geisler presented his argument in some form to Anthony Flew (before Flew became a theist) and his only response was to observe that atheists need to come up with better arguments to refute modern theists. Lewis also pointed out the circularity of Hume’s core argument (we cannot believe reports of miracles because we have no experience of them, we know we have no experience of them because all such reports are unbelievable), and this point is fairly obvious. So I do think I have sufficient basis to consider the argument sound.

Oh, yes: also by the way, Hume’s assertion that miracles are necessarily unrepeatable turns out to be false, as does his assertion that there is no contemporary experience of them. There exists at least one reported miracle, called the Miracle of the Holy Light, that has been occurring annually in Jerusalem since the 4th century. There exist others, like the healings at Lourdes, in which similar events occurred with some regularity for decades.

Miracles are like any other event in nature; if the preponderance of believable evidence points to one having occurred, then the observer has to accept the evidence. The only way one can avoid this conclusion is by assuming that no god exists.

August 16, 2009 @ 9:51 pm #


You’re not using the same definition of “miracle” as Hume, so you haven’t dealt with Hume’s argument. Moreover,the miracles we care about all involved a suspension of the laws of nature.

Also, and with all due respect to C.S. Lewis, Hume’s argument is not circular. Hume argued that: (1) an empiricist must assume the regularity of natural laws in the past, in order to study the past; (2) empiricism is the only workable philosophy of knowledge; (3) therefore, an alleged miracle – an alleged suspension of the laws of nature, cannot be considered by an empiricist.

He also argued that the definition of a “miracle” implicitly conceeds the uniformity of experience premise, and that the uniformity of experience premise makes a miracle unthinkable for an empiricist.

Nothing circular about these arguements that I can see.


September 4, 2009 @ 1:10 pm #

Hi Joe,

Curious about your statement that “empiricism is the only workable philosophy of knowledge.” What do you mean by that?



March 17, 2010 @ 10:24 am #

[…] Next, I will consider the most commonly-accepted argument against miracles, Hume’s claim that wise men can never accept that a miracle has taken place, and Anthony Flew’s modern update of that argument. Here’s the link to the next piece. […]

April 16, 2010 @ 2:10 pm #

[…] that Hume’s objection that we can never accept any event as a miracle proves too much, and leaves us questioning every event in history (see the post here); […]

RSS feed for comments on this post. TrackBack URI

Leave a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>