HUME’S ARGUMENT AGAINST RATIONAL BELIEF IN MIRACLES
John M. Fraiser
May 2005

Since its appearance over 250 years ago, David Hume’s essay “Of Miracles” remains a convincing and powerful argument for many who reject Christian testimony of miracles. At present, any claim to understand Hume’s exact argument is highly suspect. His essay appears prima facie straightforward enough, but as scholars press in to pinpoint the argument, their opinions become only more divergent.¹ Rather than attempt to argue for a particular reading of Hume which would likely be suspect, I will demonstrate that in cases where there are two (or three) possible interpretations of Hume, his argument fails on both (or all) accounts.

Hume begins by laying out his epistemic criterion for rational beliefs. “A wise man proportions his belief to the evidence.”² Furthermore, “the ultimate standard, by which we determine all disputes that may arise concerning them, is always derived from experience and observation.”³ Hume admits though that human experience is not infallible and “in some cases is apt to lead us into errors.”⁴ Still, according to Hume, experience is all one has to go on. Applying this standard to miracles, Hume asserts that “a miracle is a violation of the laws of nature; and as a firm and unalterable experience has established these laws, the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined.”⁵ All that one knows about Christian miracles comes through the evidence of testimony. “Our evidence, then, for the truth of the Christian religion is less than the evidence for the truth of our senses.” Since “weaker evidence can never destroy a stronger,” we are left with more evidence against the occurrence of a miracle than we have for it.⁶ In fact, past experience always outweighs testimony for a miracle unless

the testimony be of such a kind that its falsehood would be more miraculous than the fact which it endeavours 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 deducing the inferior.⁷

The argument outlined above comprises the first half of Hume’s essay and is commonly referred to as his a priori argument. The second half builds off of this general reasoning about the evidence of testimony by examining particular testimonies to miracles made in history and the reasons one should reject them. It is not necessary to outline this section since Hume’s main thesis is in the first section and the a posteriori section serves primarily to support it.

Hume on the Possibility of Miracles

There are several passages in Hume’s essay that seem to permit several differing interpretations. It is unclear whether Hume believes miracles are metaphysically impossible or whether they are possible but simply lack epistemic warrant for believing in them on the basis of testimony.⁸ On the one hand he says,

There must, therefore, be a uniform experience against every miraculous event, otherwise the event would not merit that appellation. And as a uniform experience amounts to a proof, there is here a direct and full proof, from the nature of the fact, against the existence of any miracle; nor can such a proof be destroyed, or the miracle rendered credible, but by an opposite proof, which is superior.⁹

On the other hand, Hume says, “For I own, that otherwise, there may possibly be miracles, or violations of the usual course of nature, of such a kind as to admit of proof from human testimony; though, perhaps, it will be impossible to find any such in all the records of history.”¹⁰

The more charitable reading allows that Hume believed in the possibility of miracles, since if he dismisses them a priori on the basis of uniform human experience then his argument is defeated on account of begging the question. If Hume says that there can never be a miracle because of the absolute uniformity of past experience then he has pre-judged the testimonial evidence, which would, in effect, mean that he rejects any testimony to the miraculous by claiming that no one has ever witnessed anything miraculous. As C. S. Lewis states it, “We know the experience against them [miracles] to be uniform only if we know that all the reports of them are false. And we can know all the reports to be false only if we know already that miracles have never occurred. In fact, we are arguing in a circle.”¹¹ Thus, Hume’s argument could be dismissed as being a mere tautology—any witness to miracles is false because no one has ever witnessed a miracle.¹² 

Laws of Nature

But if Hume is not arguing that miracles are a priori impossible then miracle claims must have a certain probability to them which would allow them to be tested against past experience just as any other claim of experience. But one cannot assume this too quickly. If Hume does, in fact, argue that miracles are possible it does not follow that he grants a miracle can be likened to other extraordinary phenomenon let alone mundane experiences. Hume distinguishes between events which are marvelous and events which are miraculous. Marvelous events are simply not conformable to our experience, while miraculous events are contrary to it.¹³ Hume argues that something may be extraordinary and not violate a law of nature, but a miracle must by definition violate a law of nature. Once again, it is not immediately obvious how Hume intends this. There are several interpretational options. Some understand Hume to mean that a law of nature is an exceptionless regularity that would prevent any kind of anomaly.¹⁴ This interpretation finds some support from certain statements in Hume’s essay. For example, Hume states that “a miracle is a violation of the laws of nature; and as a firm and unalterable experience has established these laws, the proof against a miracle…is as entire as any argument from experience can possibly be imagined.”¹⁵ Others, however, seeking to fit Hume’s language about laws of nature with his other writings interpret him to mean that laws of nature are habits of the human mind that are built on custom, or psychological associations.¹⁶ This view fits well with such statements as:

Let the course of things be allowed hitherto even so regular, that alone, without some new argument or inference, proves not that the future will continue so. In vain do you pretend to have learned the nature of bodies from your past experience. Their secret nature, and consequently all their effects and influence, may change…This happens sometimes, and with regard to some objects: Why may it not happen always, and with regard to all objects? What logic, what process of argument secures you against this supposition?¹⁷

Others are not so sympathetic to Hume and consider him to be simply caught in a contradiction with his previous statements regarding the a priori impossibility of miracles.¹⁸ Of course, how one reads Hume on laws of nature is related to how one reads him on the possibility of miracles discussed above. If a law of nature is an inviolable uniformity and he considers a miracle to constitute a violation of an inviolable uniformity then clearly Hume is rejecting miracles a priori.¹⁹

However, taking him to mean only psychological associations when he speaks of laws of nature opens the door for the possibility of miracles, since this law is only an apparent law, or one based on past experience. While this view is not subject to the charge of circularity, it invites a different kind of problem, namely, the problem of induction. How can one be confident of the regularity of the future when this confidence is grounded solely on the regularity of the past? One may think that there are absolute laws governing the universe but—as many have understood Hume to be saying—these are only psychological associations. Having observed past regularity, humans operate according to a convention of what seemingly appears to be a law or at least law-like. But how can Hume be certain that the future will behave like the past. The answer, of course—by Hume’s own admission in Treatise IV—is that he cannot be. The possibility always remains that the future could be different than the past, no matter how uniform.

But certainly the statistical probability is low, perhaps close to zero, that the amazing predictions of scientific theories such as Einstein’s theory of relativity, or the prediction of the magnetic moment of the electron by QED are not discoveries of scientific laws that govern the behavior of those things. As Karl Popper states, “It cannot be just due to an improbable accident if a hypothesis is again and again and again successful when tested in different circumstances, and especially if it is successful in making previously unexpected predictions….If a theory has been well-corroborated, then it is highly probable that it is truth-like” (emphasis Popper’s).²⁰ Furthermore, as Colin Howson points out, “To think otherwise is not ‘exercising the faculty of reasonable doubt’. It is not even scepticism. It is paranoia.”²¹ Surely, the near impossibility of predicting a future regularity without a correct theory indicates that one has discovered an absolute law of nature, or so the claim goes.

But all successful predictions of science are only known to be successful after the event has occurred, so that all one knows about their ability to successfully predict events pertains to what is now in the past. Predictions may have been made about events that were at one time in the future, but their success is only realized when the events become past-futures. It remains to be seen whether those predictions will hold true for what are future-futures, and since the future is always before us no theory can guarantee its prediction.

Furthermore, any theory of regularity developed on the observation of a collection of data is always a generalization made from a limited amount of data in a particular time. One can never possess all data relevant to the discovery of a natural law because she does not have access to future data. The statement, “All crows are black” is a conclusion drawn only from a collection of black crows. Thus, one cannot have absolute confidence that she has spoken correctly about the whole class of crows when the conclusion is based only on a limited amount of data.²²

On this interpretation of Hume, he is clearly aware of the problem of induction and does not understand a “law of nature” in an absolute sense but as an apparent law “as entire as any argument from experience can possibly be imagined.”²³ It is a law in as much as we have observed it to be so, and since “experience be our only guide in reasoning concerning matters of fact,” it is against this that one must judge testimony to the miraculous.²⁴

There are then at least three possible readings of Hume’s view of the possibility of miracles. 1) He rejects testimony to miracles a priori on the grounds that they violate inviolable laws of nature. 2) He allows for the possibility of miracles yet contradicts himself by believing in inviolable laws of nature. 3) Hume allows for the possibility of miracles and believes laws of nature to be customs based on past experience that cannot produce certainty.

If Hume argues anything like the first two interpretations then one can answer him easily by charging him with circular reasoning on the first interpretation and contradiction on the second. It is unlikely (though not impossible) that such an eminent philosopher as Hume should make such a vapid and careless argument. Since, therefore, the third interpretation makes for a more responsible argument it warrants more attention than the first two.

Hume on the Probability of Miracles

Hume’s understanding of probability rests on observation. He writes, “All probability, then, supposes an opposition of experiments and observations, where on the one side is found to overbalance the other, and to produce a degree of evidence, proportioned to the superiority.”²⁵ Following this criteria of weighing the number of observations on one hand with the number of observations on the other, Hume discusses several scenarios in which it might be acceptable to believe in a unique event or an anomaly of some sort. First, Hume relates the story of an Indian prince who rejected testimony to the effect of freezing water on the grounds that it did not conform to his uniform experience. While ice is not an anomaly on a normative level, it is an anomaly for the individual who has never experienced it. Thus, the Indian prince was justified in his disbelief since he lacked a sufficient level of testimony to overcome his uniform experience.²⁶ Though Hume does not tell the reader specifically what would constitute sufficient testimony, he nevertheless believes that there could be a level sufficient to obligate the Indian prince to believe in ice apart from any experience of it. Hume’s second scenario involves a widespread, cross-cultural, consistent report of eight days of darkness over the whole world allegedly occurring 150 years before his writing. Hume contends that in this case one should believe the testimony.²⁷ Hume points out that neither scenario counts as a testimony to a miracle, because while these testimonies are not conformable to a recipient’s experience they cannot be said to be contradictory. In the case of the Indian prince, he has only seen water in his own winter climate but never in a Muscovy winter, and cannot be sure that it would not freeze. The only problem in this case is that he lacks enough witnesses to make it credible. In the case of eight days of darkness, “the decay, corruption, and dissolution of nature, is an event rendered probable by so many analogies, that any phenomenon, which seems to have a tendency towards that catastrophe, comes within reach of human testimony, if that testimony be very extensive and uniform.” But, according to Hume, a testimony to, say, someone rising from the dead is not just non-conformable—it is contradictory to all of one’s experience and thus contradictory to apparent laws of nature.²⁸ As such, testimony to a resurrection could never achieve the status of a proof no matter how much testimony there was, because the evidence of past experience when stacked against it would always be greater than the evidence favoring testimony. So then, a testimony to a miracle always proves evidentially impossible. Marshaling support for a miracle can at best only weaken the evidence for a law of nature, but can never prove a miracle.²⁹ Thus, according to Hume, the problem is not that a miracle claim lacks probability but that its probability can never outweigh the counter-evidence if we follow the claim that something’s probability is based on the weighing of occurrences. Testimony does not carry the weight that past experience does because “it be more probable that this person should deceive or be deceived” than that one’s past experience should be fallible.³⁰ While Hume admits that our experience is sometimes fallible, we nevertheless have plenty of evidence to the fallibility of human testimony, and this is why when receiving a report of a miracle one should only believe it if “its falsehood would be more miraculous, than the fact, which it endeavours to establish.”³¹

Hume on the Actuality of Miracles

Hume concludes that a religious miracle has never occurred.³² But for the sake of argument, Hume discusses what one should think if there was ever a testimony sufficient to count as full proof of a genuine miracle. In such a case, according to Hume, one would have proof against proof, and the stronger proof would have to prevail. Whichever proof survives, one would have to subtract the weaker proof from it and hold only to the residual as a belief.³³ Any rational belief should be quite weak, since the surviving probability of the miracle’s occurrence is quite weak as well.³⁴ Thus, one could never be certain that a miracle has occurred once one subtracts the opposing evidence. So while a religious miracle is possible, testimony is never sufficient to establish it.

 

Evaluating Hume’s Argument

How worthwhile is Hume’s project of evaluating testimony to the miraculous? Hume’s argument for all of its insights and developments in the field of probability theory offers no compelling reasons to reject miracles. One need not base this conclusion on a particular interpretation of Hume; his argument fails regardless of the interpretation. This evaluation of Hume’s argument will address the more responsible reading of Hume that holds that: 1) miracles are metaphysically possible; 2) a law of nature is a psychological association based on past experience that cannot produce certainty, 3) miracles are different than an extraordinary/anomalous/marvelous event because the former is contrary to past experience, while the latter is simply non-conformable to it; 4) no religious miracle has ever actually occurred; 5) it is evidentially impossible that testimony could ever establish a religious miracle.

Uniform Experience and Laws of Nature

Any belief in a law of a nature is based upon a posteriori judgment of the uniformity of past experience, that is, uniformity can never be decided deductively. If Hume admits that a law of nature is only formed inductively and that the law is only a habit or custom of the mind, one has to wonder what would be the harm in an event violating this inductively formed habit. Why, if God exists, would this habit of the mind be so inviolable? Perhaps, Hume would respond that while a custom could be violable, one would never be rational in believing it if in fact humanity has a confirmed uniform experience. If in the history of human experience (or in one’s own experience) there is evidence of variation, then there are scenarios in which it is rational to accept a report to a miracle. But once again, this argument fails from the same problem of circularity as before, namely, how can one arrive at a belief in uniform experience and reject testimony to miracles without first rejecting testimony to miracles in order to arrive at a belief in uniform experience? In essence, Hume’s argument, while it may not be circular at the a priori level becomes circular on the a posteriori level. Hume’s question-begging, while probably not be based on a deductive belief in inviolable law of nature, may be based on inductively forming a belief that all of human experience has a certain uniformity to it. Regardless of whether it is a normative law of nature or a subjective habit of the mind, one can only be irrational in believing a miracle-claim if experience is in fact uniform. But this assumption can never be proven, and thus, Hume’s conclusions of regularity based on inductive grounds is wholly inaccurate. Even when scientists do believe in uniform experience they do not conclude that there is no room for doubt, otherwise it would be hard to account for their ongoing efforts to search for exceptions.³⁵

Furthermore, if there are governing laws of nature, they would be only that, laws of nature. They would govern only nature and could not be said to govern anything beyond nature. Perhaps one could reply that laws of nature govern only the natural order because nature is all there is. But this argument is once again smuggling in naturalistic assumptions. The fact remains that all anyone can know about the jurisdiction of natural laws is limited to the physical order and thus, should there be anything beyond nature, they could not govern it.

Moreover, on Hume’s view, a miracle is not a naturally occurring exception to the laws of nature (or “customs” in his specialized sense of the word); such a concept would be contradictory. Rather, a miracle is a supernatural violation of the laws of nature. There is nothing about a supernatural cause being free with respect to natural laws, interfering with an element of nature that calls into question the law itself. Only an exception to the law could undercut the law, but in calling the event a supernatural interference one actually upholds it and defends it against the charge that it has an apparent exception by identifying that apparent exception as a miracle. Miracles affirm laws of nature since unless the law is in place there could be no interference.³⁶ It is precisely because there is a law that one can call it an interference, otherwise it would have to be called an exception, which would only invalidate the law.

The possibility that a law could be violated raises an interesting question, one discussed by John Stuart Mill as an appendix to Hume’s argument. If a miracle did occur, there is still an alternative besides attributing it to a supernatural being. The event could be the result of an unknown natural cause. In other words, the event may simply reveal that there is an error in human knowledge of the laws of nature or that observers lack relevant information, but it need not require a supernatural explanation.³⁷ Maybe the event could be attributed to an unknown natural cause.

Even if this were the case, it would still not remove God from the picture. A miracle need not be defined only as an interference with a law of nature. As far as biblical miracles go, a high number of them are perfectly explainable in terms of natural causes. Even an event as amazing as the parting of the Red Sea does not require a violation of any natural laws.³⁸ A strong wind dividing the sea and drying the land is perfectly consistent with natural causes, but this does not mean that God could not ultimately be back of it. While not all biblical miracles are explicable in terms of natural causes, any that are or could be would only prove that God’s action in the world does not always require him to interfere with natural laws.

On the matter of laws of nature, Hume appears to be trapped. If there are objective laws that govern the universe then these laws would only govern nature and would be incapable of governing supernatural action. But if Hume argues that laws of nature are only apparent laws, then miracles are only apparent violations. In order to constitute a real violation, a miracle must violate a real law.³⁹ Thus, Hume cannot rationally object to miracles on grounds that it violates a law of nature. He can no longer separate out miracles from other experiences on the grounds that it is contrary to uniform experience. Testimony to miracles is in a place where it can now be evaluated in terms of probability.

Probability of Miracles

Just because a miracles cannot be dismissed on grounds that they are supernatural interventions into the natural world, it does not follow that it is rational believe in them. While miracles may be probable, they surely have a low degree of probability in and of themselves—that is what qualifies them as miracles. If the probability was extremely high (say, 0.9) then miracles would be quite common place and undisputedly considered natural. What is it that gives miracles any probability at all? Miracles are probable because (as has already been argued): 1) Hume cannot assert uniformity without assuming it. 2) If uniformity is not absolute then miracles cannot be said to be contrary to uniformity. 3) If they are not contrary to uniformity, then they are probable. Now one may add a fourth conclusion; the probability of miracles in themselves is low, or extremely low. As long as the probability remains so low or extremely low, according to Hume, it is not rational to believe in testimony to miracles.

Many have sought to overcome Hume’s argument by comparing testimony to miracles with other statistically improbable events. Thus, Robert Hambourger compares testimony to miracles to state lottery statistics.⁴⁰ In a lottery with one million participants, on the day after the drawing , a reliable newspaper such as the Chicago Tribune reports Mr. Smith as the winner. The probability of Mr. Smith’s winning is one in a million, and the probability of the Tribune misreporting the story is lower than Mr. Smith’s winning, say, one in ten thousand. Following the statistical evidence and comparing the probability of the Tribune misreporting on the probability of Mr. Smith’s winning the lottery, it would be absurd for one to disbelieve that the Tribune accurately reported Mr. Smith’s winning. The improbability of an event is naturally irrelevant when one picks up the newspaper and believes it report.

But what relationship does this illustration have to miracle-claims? Hume would certainly think that it is unrelated, and he would be correct. Consider that in the example of the lottery 1) it is certain that someone will win it, so that the Tribune accurately reports Mr. Smith as the winner is no surprise.⁴¹ 2) The Tribune is not making in any metaphysical claim about the world. 3) There is no law of nature—apparent or otherwise—that argues against Mr. Smith winning the lottery, and 4) his winning does not invoke supernatural intervention, while the same cannot be said in the case of religious miracles. Therefore, any probability comparison to a miracle-claim is going to have to be a similar comparison, and the only comparison similar enough to a miracle-claim is a miracle-claim. All other comparisons break down at some level. One should not accept all claims to miracles and Hambourger’s lottery comparison comes close to arguing that one should do so. A closer comparison than Hambourger’s would be if the Tribune reported that evidence of aliens was found on Mars and that their religion was to worship Oprah Winfrey.⁴² One would be much more skeptical about this story than Hambourger’s lottery story. There are many more likely reasons for why an alien report ended up in the paper: maybe it slipped past the editor as a joke; maybe someone is trying to sabotage the newspaper’s reputation; maybe the editor feels they need to keep up with the tabloids, etc.

Peter Millican offers an example that would counter Hambourger’s.⁴³ Suppose there is a medical test that can diagnose a rare and fatal genetic disorder that only expresses itself in middle age, and the odds of having it are one in a million. The test has an extremely high accuracy rate of 0.999, with the probability of misdiagnosis being a mere .001, and it is never inconclusive. Fred is a hypochondriac nearing his fortieth birthday and fears he may have the disorder. He goes to a doctor for testing, and to his and the doctor’s shock tests positive for the disorder. What is the probability that Fred has the disorder? The answer may at first seem simple—99.9% of course, since the probability of having the disease corresponds directly to the test’s rate of accuracy. This conclusion would be inaccurate though. If 55 million middle-aged men were tested for the disorder, one would be expected only about 55 to have it given its rate of infliction. It is 99.9% likely that these 55 will receive a correct diagnosis. But what about the other 54, 999,945 test subjects who do not have the disorder? While a huge majority of them will test negative, the test’s 0.1% inaccuracy while seemingly insignificant, becomes quite significant for the 55,000 men who test positive. So out of 55,055 positive returns only 55 actually have the disorder. Thus, the probability that Fred has the disorder is less likely than that the test revealing him to have it is accurate.

There is no reason to think that one can establish miracle-claims in terms of bare statistical probability. Even if miracles do not violate laws of nature their probability is still extremely low and thus one would not be rational in believing in them and ignoring the evidence against them. This may seem, at first, to deal a blow to belief in miracles. However, it only matters for any old claim to miracles. If the miracle has more than simply testimony to substantiate it, then the fact that common probability comparisons are invalid is far less relevant.

As much as one may believe Hume’s project to be detrimental to religious integrity, it is equally detrimental to religious integrity to believe that testimony alone could establish religious miracles. As Hume states, the testimonies invalidate themselves when they are all advanced as evidence for contrary religions across the world.⁴⁴ If testimony alone can establish a miracle, then one is left more religiously confused than she was before testimony.

Just because a witness is usually reliable does not mean that one does not have good reason to question it when making claims that are metaphysically hard to swallow. Those who believe, for example, in the miracles of Jesus Christ do not do so simply on the grounds of statistical probability that the gospel writers were telling the truth. Surely, this is a significant piece of evidence—perhaps the most significant—but it is not enough on its own to establish belief in miracles. Christians believe in miracles for many reasons.

Rationality and Belief in Miracles

So far it has been argued that Hume’s essay is subject to a variety of interpretations, and that his argument is unsuccessful following any of them. Here is a diagram of the reasonable interpretations of Hume’s argument:

 

While some interpretations appear to be more viable than others, all interpretations eventually lead to the same conclusion, namely, that it is irrational to believe in miracles. Hume’s conclusion is the point at which he makes his greatest leap of logic. But so what if it is irrational to believe in miracles on the basis of testimony alone. There are many reasons that one may believe in miracles, not the least of which is that he simply finds himself believing them.⁴⁵ On evidentialist assumptions such as Hume’s, one may say that there is not enough evidence for belief in the testimony to miracles, but why must anyone else commit to Hume’s assumptions?

While many theists in Hume’s day used miracles to argue for God’s existence,⁴⁶ what if one flipped the argument on its head and used God’s existence to argue for miracles? Surely, in a naturalistic universe miracles are improbable, but in a theistic one their probability increases exponentially. Hume is only comparing the evidence of past experience with the evidence of testimony, and though it is surely irrational to believe in miracles on these grounds, were he to allow more factors than this into the argument, it might turn out that he is the irrational one for rejecting miracles. What if one added to testimony, an omnipotent God who created the universe and rules all that is in it, a purpose for his miraculous acts, and a personal experience with God? While none of these factors would constitute absolute proof for a miracle one would certainly be rational in believing in miracles on these grounds, and it is only the rationality of miracles that Hume argues against.⁴⁷ Hume has cleverly staged the issue in such a way that his evidentialist assumptions would be the criteria by which belief in the testimony to miracles is rational. However, those who believe in miracle-claims are under no obligation to accept his assumptions. In fact, as Plantinga rightly argues, evidentialism fails to meet its own criteria.⁴⁸ What evidence is there to support the proposition that one should always, “proportion his belief to the evidence.”⁴⁹ Furthermore, as Plantinga points out there are a host of beliefs which it is quite rational to hold apart from considering any evidence.⁵⁰ Memory beliefs are just one example. No one ever believes in what happened five minutes ago on the basis of the evidence, one simply finds herself trusting in her memory. If pressed to prove how she knows her memory is accurate she might at first appeal to the fact that it has always proven reliable, but no evidentialist should accept this as evidence since she must rely on her memory in order to defend its reliability. There is no conclusive proof that she could offer and yet one could hardly argue that she is irrational in generally trusting her memory. It is simply how she is designed to function and thus her memory belief constitutes a properly basic belief.

Not needing evidence to sustain belief in miracles does not mean, however, that there is not good, solid evidence for belief in miracles, nor that such evidence is of no value. Even given Hume’s criteria it is possible that belief in testimony to miracles could be sustained. Take for example the consummate Christian miracle—and the one that Hume himself attacks—the resurrection of Jesus Christ. There are only two conclusions that one can draw about the response of the disciples if the resurrection did not happen: 1) they were deceived, or 2) they were lying. And yet the statistical probability for either of these is extremely low when one considers that there were many would-be messiahs in that day who boasted a collection of disciples and that many of them were executed for being revolutionaries. In each case, the disciples either found a new messiah or abandoned the revolution. Jesus’ disciples are the only exception to this norm.⁵¹ Now if one is required to base her belief on uniform past experience, she faces a peculiar dilemma. Which collection of past experience should she follow? At least this past experience is undisputedly uniform, which cannot be said for Hume’s apparent laws of nature. As difficult of a proposition as it might be, one has trouble making sense of many pieces of historical evidence if Jesus did not rise from the dead. As Wright says, “Once you allow that something remarkable happened to his body that morning, all other data fall into place with astonishing ease. Once you insist that nothing so outlandish happened, you are driven to ever more complex and fantastic hypotheses to explain the data. For the historian, as for the scientist, the answer should be clear.”⁵²

The disciples’ response is part of what troubles Hume about belief in miracles. Hume believes that one’s skepticism of testimony should rise when the witnesses have something to gain in what they affirm, and one would have to agree that Jesus’ disciples had much to gain in what they affirmed.⁵³ But it is precisely this factor which adds further credibility to the Christian witnesses’ testimony. Given that a disciple knew how consequential the truth or falsehood of an event would be, he would likely consider the event more carefully than he would if its occurrence was insignificant. If the miracle was as inconsequential as, say, his pen instantly disappearing from the surface of his desk, he might rub his eyes for a moment and likely consider himself to be hallucinating or afterward report that the strangest thing happened to him that day. However, he would not likely sit in careful reflection over it the way he would if he felt he had experienced an event which, if true, would be the most powerful event of all time. Thus, Hume’s argument is, at best, suspect when tested in terms of evidence, and yet one is not even required to go that far in order to reject his argument.

Conclusion

Hume’s argument is based on comparing past experience with testimony to miracles alone. Following this criteria, it would be hard to make a case for rational belief in miracles (although as argued above there are several reasons to think that belief in miracles could succeed even on an evidentialist model), but theists are under no obligation to accept Hume’s criteria. He is right inasmuch as he demands more evidence for a miracle than human testimony, but wrong to think that testimony is all that miracles have going for them.

If Hume is arguing that one should dismiss miracles a priori because they violate inviolable laws of nature, then he is simply presupposing naturalism and begging the question. If he is arguing that miracles violate only apparent laws of nature that are developed from the custom of human experience, then there is nothing irrational (on these grounds) about believing that God override them. Furthermore, if human custom is absolutely uniform, then Hume is again begging the question since he has already discounted testimony to miracles in order to argue for uniform experience. Thus, Hume’s argument fails on all accounts.

 

                1 As Alvin Plantinga notes, “This striking divergence is testimony to the fact that Hume is a black enigma: a certain surface clarity masks a deep underlying murkiness that makes confident interpretation impossible” (Alvin Plantinga, Warranted Christian Belief [New York: Oxford Press, 2000], 218).