The Argument Against Miracles
The excellent article below is reproduced (by permission) from the Baker Encyclopedia of Christian Apologetics.
Miracles, Arguments Against. Most modern thinkers who reject miracles trace their reasoning to the Scottish skeptic (see Agnosticism), David Hume (1711–1776). Hume provided what many believe to be the most formidable of all challenges to a supernaturalist perspective: Miracles are incredible.
Hume laid out three arguments against miracles: philosophical, historical, and religious. The first argument is an argument in principle, based on the incredibility of claiming natural laws are ever contravened. The second is an argument in practice, which challenges whether miracles have ever had credible witnesses (see New Testament, Historicity of). The last is from the self-canceling nature of similar miracle claims that abound in all religions.
The Incredibility of Miracles. Building on his empirical epistemology, Hume launched his attack on miracles with the comment, “I flatter myself that I have discovered an argument . . . which, if just, will, with the wise and learned, be an everlasting check to all kinds of superstitious delusion, and consequently will be useful as long as the world endures” (Hume, An Enquiry Concerning Human Understanding, 10.1.18). Hume’s reasoning goes like this (Inquiry, 10.1.18, 120–23):
- A wise person proportions belief to the evidence.
- An event that can be established on infallible experience can be, with full assurance, expected to reoccur in the future.
- The reliability of evidence derived from witnesses and human testimony establishes proof or probability, as it is corroborated by other reports and evidence.
- All circumstances should be considered in judging probability, and the ultimate standard is how the reports comport with personal experience and observation.
- Where personal experience is not the same, the person should keep a contrary judgment and subject the question to thorough argument.
- Any contradictions among witnesses should be regarded with suspicion. Suspicion should also arise if the witnesses are few in number, of “doubtful character,” have a vested interest in what they affirm, hesitate in their testimony, or assert it too vigorously.
- “But when the fact attested is such a one as has seldom fallen under our observation, here is a contest of two opposite experiences; of which the one destroys the other as far as its force goes, and the superior can only operate on the mind by the force which remains.”
- A miracle violates the laws of nature, which have, by “firm and unalterable experience” been established.
- Therefore, “the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined.”
- Experience is direct and full proof against the existence of any miracle.
Hume’s argument can be abbreviated:
- A miracle is a violation of the laws of nature.
- Firm and unalterable experience has established these laws of nature.
- A wise person proportions belief to evidence.
- Therefore, the proof against miracles is overwhelming.
Hume wrote, “There must, therefore, be a uniform experience against every miraculous event. Otherwise the event would not merit that appellation.” So “nothing is esteemed a miracle if it ever happened in the common course of nature” (10.1.122–23).
Alternatives in Hume’s Argument. There are two basic ways to understand Hume’s argument against miracles. We will call these the “hard” and “soft” interpretations. According to the “hard” interpretation, Hume would be saying:
- Miracles, by definition, violate natural laws.
- Natural laws are unalterably uniform.
- Therefore, miracles cannot occur.
Now, despite the fact that Hume’s argument sometimes sounds like this, it is not necessarily what he has in mind. If this is his argument, then it clearly begs the question by simply defining miracles as impossible. For if miracles are a “violation” of what cannot be “altered,” then miracles are ipso facto impossible. Supernaturalists could easily avoid this dilemma. They could refuse to define miracles as “violations” of fixed law and simply call them “exceptions” to a general rule. Both premises are deniable. Natural law is the regular (normal) pattern of events. It is not a universal or unalterable pattern.
This would be an easy way out of the problem. Actually, Hume’s position contains an argument that is much more difficult to answer, one that addresses a “softer” view of natural law. It is not an argument for the impossibility of miracles, but for their incredibility:
- A miracle is by definition a rare occurrence.
- Natural law is by definition a description of regular occurrence.
- The evidence for the regular is always greater than that for the rare.
- Wise individuals always base belief on the greater evidence.
- Therefore, wise individuals should never believe in miracles.
Notice that this “soft” form of the argument does not rule miracles out of the question; they are held to be incredible by the nature of the evidence. Wise people do not claim that miracles cannot occur; they simply never believe they happen. Sufficient evidence never exists for belief.
In this “soft” interpretation of the argument, miracles are still eliminated, since by the very nature of the case no thoughtful person should ever hold that a miracle has indeed occurred. If this is so, Hume has seemingly avoided begging the question and yet has successfully eliminated the possibility of reasonable belief in miracles. Variations of these arguments are still held to be valid by some widely respected contemporary philosophers.
Evaluation of Hume’s Argument. Since the “hard” form of Hume’s argument clearly begs the question and is easily answered by redefining the terms, we will concentrate on the “soft” form. The key to unlocking this attack rests in Hume’s claim for uniform experience.
Hume’s “uniform” experience either begs the question or is special pleading. It begs the question if Hume presumes to know the experience is uniform in advance of the evidence. How can one know that all possible experience will confirm naturalism, without access to all possible experiences, past, present, and future? If, on the other hand, Hume simply means by “uniform” experience the select experiences of some persons (who have not encountered a miracle), this is special pleading. Others claim to have experienced miracles. As Stanley Jaki observes, “Insofar as he was a sensationist or empiricist philosopher he had to grant equal credibility to the recognition of any fact, usual or unusual” (Jaki, 23). As C. S. Lewis observed,
Now of course we must agree with Hume that if there is absolutely “uniform experience” against miracles, if in other words they have never happened, why then they never have. Unfortunately we know the experience against them 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. [Lewis, 105]
The only alternative to this circular argument is to be open to the possibility that miracles have occurred.
Further, Hume does not really weigh evidence objectively; he really adds up the evidence against miracles. Death occurs over and over; resurrection occurs rarely. Therefore we must reject the latter. In Hume’s own words, “It is no miracle that a man, seemingly in good health, should die on a sudden, because such a kind of death has yet been frequently observed to happen. But it is a miracle that a dead man should come to life; because that has never been observed in any age or country.” Hence, “it is more probable that all men must die” (Enquiry, 10.1.122).
There are other problems with Hume’s concept of adding up events to determine truth. Even if a few resurrections actually occurred, according to Hume’s principles, one should not believe them. However, truth is not determined by majority vote. Hume commits a kind of consensus gentium which is an informal logical fallacy of arguing that something is true because it is believed by most people.
This argument really equates “evidence” and “probability.” It says in effect that one should always believe what is most probable, what has the highest “odds.” One should not, therefore, believe that the rolled dice came up with three 6’s on the first roll. The odds against that happening, after all, are 216 to 1. Or, one should not believe that he was dealt a perfect bridge hand (which has happened) since the odds against this happening are 1,635,013,559,600 to 1! Hume overlooks that wise people base beliefs on facts, not odds. Sometimes the “odds” against an event are high (based on past observation of similar events), but the evidence for the event is very good (based on current observation or testimony for this event).
Hume’s concept of “adding” evidence eliminates belief in any sort of unusual or unique event. Richard Whately satirized Hume’s thesis in his pamphlet, Historical Doubts Concerning the Existence of Napoleon Bonaparte. Since Napoleon’s exploits were so fantastic, so extraordinary, so unprecedented, no intelligent person should believe that these events ever happened. After recounting Napoleon’s amazing and unparalleled military feats, Whately wrote, “Does anyone believe all this and yet refuse to believe a miracle? Or rather, what is this but a miracle? Is not this a violation of the laws of nature?” If the skeptic does not deny the existence of Napoleon, he “must at least acknowledge that they do not apply to that question the same plan of reasoning which they have made use of in others” (Whately, 274, 290).
Finally, Hume’s argument proves too much. It proves that a person should not believe in a miracle even if it happens! For it argues, not that miracles have not occurred, but that we should not believe they occurred because the evidence for the regular is always greater than that for the rare. On this logic, if a miracle did occur—rare as it may be—one should still not believe in it. There is something patently absurd about claiming that an event should be disbelieved, even if one knows it happened.
Uniformitarian Denial of Miracles. Can one eliminate belief in present events based on evidence for past events? It would seem that Hume wants each wise person always to believe in advance that miracles never have, do not now, nor ever will occur. Before examining the evidence, one should be prearmed with the uniform and “unalterable” testimony of uniformitarianism. Only if one approaches the world with a kind of invincible bias against anything that has not been personally perceived in the past can all claims for the miraculous be discounted.
Hume recognized the fallacy of this reasoning when he argued that, based on past conformity, nothing can be known as true concerning the future. We cannot even know for sure that the sun will rise tomorrow morning (An Abstract of Treatise on Human Nature, 14–16). Hence, for Hume to deny future miracles based on past experience is inconsistent with his own principles and is a violation of his own system.
If it were true that no present exception can overthrow “laws” based on our uniform experience in the past, there could be no progress in our scientific understanding of the world. For established or repeatable exceptions to past patterns are precisely what force a change in scientific belief. When an observed exception to a past “law” is established, that “law” is revised, if possible, to account for the exception. A new “law” replaces it. This is precisely what happened when certain outer-spatial but repeatable “exceptions” to Newton’s law of gravitation were found, and Einstein’s theory of relativity was considered broader and more adequate. Exceptions to “laws” have a heuristic (discovery) value; they are goads to progress in our understanding of the universe. Now what is true of repeatable exceptions that call for a natural explanation is also true for unrepeatable exceptions that point to a supernatural explanation.
Lack of Credible Witnesses. Hume also argued against the testimony for miracles in practice. We have shown that the a priori attempts to eliminate miracles fail, so we are left with a posteriori arguments. Hume objects that there is not enough evidence to establish New Testament miracles. He enumerates several arguments which, if true, would exclude the credibility of the New Testament witnesses.
Hume says, “there is not to be found, in all history, any miracle attested by a sufficient number of men of such unquestioned good-sense, education, and learning as to secure us against all delusion in themselves.” Nor are there enough witnesses of “such undoubted integrity, as to place them beyond all suspicion of any design to deceive others.” Neither are they “of such credit and reputation in the eyes of mankind, as to have a great deal to lose in case of their being detected in any falsehood.” Finally, neither have the alleged miracles been “performed in such a public manner and in so celebrated a part of the world as to render the detection unavoidable” (Abstract of a Treatise, 124).
“The strong propensity of mankind to the extraordinary and marvelous . . . ought reasonably to beget suspicion against all relations of this kind.” And “if the spirit of religion join itself to the love of wonder, there is an end of common sense,” wrote Hume (ibid., 125–26).
Miracles and the Ignorant. Hume believes the case for miracles is damaged because “they are observed chiefly to abound among ignorant and barbarous nations.” Those that have found believers in civilized countries, he added, usually got them originally from “ignorant and barbarous ancestors.” Further, “the advantages are so great of startling an imposture among ignorant people that . . . it has a much better chance for succeeding in remote countries than if the first scene had been laid in a city renowned for arts and knowledge” (ibid., 126–28).
“Upon the whole, then, it appears that no testimony for any kind of miracle has ever amounted to a probability, much less to a proof.” Further, “even supposing it amounted to a proof, it would be opposed by another proof derived from the very nature of the fact which it would endeavor to establish” (ibid., 137).
Evaluation. Even though Hume implies that he is open to actual evidence for a miracle should it meet his standards for purity, one quickly suspects that the rules of evidence have been tampered with, so as to rule out the credibility claims for any miracle.
Hume at one point candidly admits that no number of witnesses would convince him of a miracle. Speaking of what he acknowledged to be highly attested Jansenist miracles of his day, Hume wrote: “And what have we to oppose to such a cloud of witnesses but the absolute impossibility or miraculous nature of the events which they relate?” Such impossibility, he adds, should be sufficient “in the eyes of all reasonable people” (ibid., 133, emphasis added). No matter how many witnesses one provides for these “absolutely impossible” events, no “reasonable person” will believe them. If this is the case, then Hume is still approaching every miraculous event, no matter how well it is attested, from an incurably a priori naturalistic bias. All the talk of testing the credibility of the witnesses is poorly concealed antisupernaturalism.
This bias shows that his argument cuts in two directions. Knowledge of human nature also reveals biases against accepting miracles.
Hume’s position also is inconsistent. He would not allow testimony for miracles, yet he would allow testimony from those who had seen frozen water, in preference to the testimony of those who never had. But why allow testimony for one event and not the other? He cannot reply that it is because others have seen water frozen, for this begs the question. The problem is that a tropical tribe has never seen it, so why should they accept the testimony of an outsider who says he has, regardless of how often he has seen it? Miracles have happened more than once. Further, according to Hume’s own principles, even if one saw water freeze only once and he walked or slid on it, that would be sufficient to know that it happened. But, the same applies to a miracle. Only an antisupernatural bias would hinder a person from honestly considering reliable testimony about its occurrence.
Hume is apparently unaware of the strong historical evidence for the reliability of the biblical documents and witnesses (see Bible, Evidence for; New Testament, Historicity of). At least, he overlooks it. But biblical miracles cannot be dismissed without a closer look. For no one should rule out the possibility of these miracles in advance of looking at the evidence for them.
New Testament Witnesses and Hume’s Criteria. Hume outlined the basic criteria that he believed necessary for testing the credibility of witnesses (ibid., 120). These are discussed in the article Witnesses, Hume’s Criteria for. They can be summarized as four questions:
- Do the witnesses contradict each other?
- Are there a sufficient number of witnesses?
- Were the witnesses truthful?
- Were they prejudiced?
Witnesses do not contradict each other. Hundreds of alleged contradictions in the Gospels have been weighed and found wanting by scholars, including Gleason Archer, John Haley, William Arndt, and others (see some of these defenses in the list of sources for this article). The error is not in the Gospel but in the procedure used by the critic. For a study of sample charges, see Bible, Alleged Errors in. The testimonies of the New Testament witnesses are never mutually contradictory (see Bible, Alleged Errors). Each one tells a crucial and overlapping part of the whole story.
To be sure, there are minor discrepancies. One account (Matt. 28:2–5) says there was one angel at the tomb on the morning of Jesus’ resurrection; John says there were two angels (John 20:12). It should be noted about these kinds of discrepancies that they are conflicts but not irreconcilable contradictions. Matthew does not say there was only one angel there; that would be a contradiction. Likely at one point there was one angel, and at another a second angel was about. Conflict in details is what one would expect from authentic, independent witnesses. Any perceptive judge who heard several witnesses give identical testimony would suspect collusion (see Gospels, Historicity of).
The number of witnesses is sufficient. Twenty-seven books in the New Testament were written by some nine persons, all eyewitnesses or contemporaries of the events they recorded. Six of these books are crucial to the truth of New Testament miracles, Matthew, Mark, Luke, John, Acts, and 1 Corinthians. All of these books bear witness to the miracle of the resurrection. Even critical scholars now acknowledge that these books are first-century documents, most written before a.d. 70, while contemporaries of Christ were still alive. Virtually all scholars acknowledge that 1 Corinthians was written by the apostle Paul around a.d. 55 or 56, a little over two decades after the death of Christ. This is a powerful witness to the reality of the miracle of the resurrection: First, it is a very early document. Second, it is written by an eyewitness of the resurrected Christ (15:8, cf. Acts 9:3–8). Third, it refers to more than 500 eyewitnesses of the resurrection (15:6), stressing that most of these witnesses were still alive (vs. 6). Any immediate reader of 1 Corinthians could check out the reliability of the evidence for the resurrection.
The witnesses were truthful. Few challenge the fact that the New Testament provides a great standard of morality based on love (Matt. 22:36–37) and inner piety (Matthew 5–7). Jesus’ apostles repeated this teaching in their writings (for example, Romans 13; 1 Corinthians 13; Galatians 5). Their lives exemplified their moral teaching. Most died for what they believed (2 Tim. 4:6–8; 2 Peter 1:14), an unmistakable sign of their sincerity.
In addition to teaching that truth is a divine imperative (Eph. 4:15, 25), it is evident that the New Testament writers were scrupulous about expressing it. Peter declared, “We did not follow cunningly devised fables” (2 Peter 1:16). The Apostle Paul insisted, “Do not lie one to another” (Col. 3:9).
Where the New Testament writers’ statements overlap with the discoveries of historians and archaeologists, they have proven to be exactingly accurate (see Archaeology, New Testament). Archaeologist Nelson Glueck concludes, “It may be stated categorically that no archaeological discovery has ever controverted a Biblical reference. Scores of archaeological findings have been made which confirm in clear outline or exact detail historical statements in the Bible” (31). Millar Burrows notes that “more than one archaeologist has found his respect for the Bible increased by the experience of excavation in Palestine” (Burrows, 1). There is no hint that the New Testament writers ever falsified facts of the case. Their testimony would be accepted as valid by any unbiased jury. As the great Harvard legal expert Simon Greenleaf concluded, their testimony shows absolutely no sign of perjury.
The witnesses were not prejudiced. There is every reason to believe that New Testament witnesses of the miracles of Christ, particularly of his resurrection, were not predisposed to believe the events to which they gave testimony. The apostles themselves did not believe the when the women reported it (Luke 24:11). Even some disciples who saw Christ were “slow of heart to believe” (Luke 24:25). Indeed, when Jesus appeared to ten apostles and showed them his crucifixion scars, “they still did not believe it because of joy and amazement” (Luke 24:41). And even after they were convinced by Jesus’ eating food, their absent colleague Thomas protested that he would not believe unless he could put his finger in the scars in Jesus’ hands and side (John 20:25).
Jesus also appeared to unbelievers, in particular his skeptical half-brother, James (John 7:5; 1 Cor. 15:7), and to a Jewish Pharisee named Saul of Tarsus (Acts 9). If Jesus had only appeared to those who were believers or had a propensity to believe, there might be legitimacy to the charge that the witnesses were prejudiced. But the opposite is the case.
Witnesses to the resurrection had nothing to gain personally by their witness to the resurrection. They were persecuted and threatened (cf. Acts 4, 5, 8). Most of the apostles were martyred. Yet they proclaimed and defended it in the face of death. Nor should witnesses be dismissed simply because they have an interest in what occurred. Otherwise, we should not accept testimony from the survivors of the holocaust, which we do. The question is whether there is evidence they were telling the truth.
Self-Canceling Claims. Hume claims that “every miracle, therefore, pretended to have been wrought in any of these religions (and all of them abound in miracles) . . . so has it the same force, though more indirectly, to overthrow every other system.” However, Hume believes, these miracles do not accomplish their task. Rather, “in destroying a rival system, it likewise destroys the credit of those miracles on which that system was established” (Hume, 129–30). Since all religions have the same sorts of miracles, none of them establish the truth of their doctrines. They cancel one another out as witnesses to truth.
There are, however, several significant problems with Hume’s argument from the self-canceling nature of miracle claims.
All Miracles Claims Are the Same? Hume wrongly assumes that all alleged miracles are created equal. This is contrary to fact. Some obviously refer to natural anomalies or psychosomatic cures. Particularly in the Eastern and New Age religions, supernatural occurrences generally can be shown to be tricks (see Miracles, Magic and). In the case of prophecies, their accuracy is too low to be taken seriously. There is a big difference between walking on hot coals, a feat that anyone can be taught to do, and walking on water, as Jesus did (John 6). There is a difference between healing someone of migraine headaches and healing a person born blind, as Jesus did (John 9). Faith-healers in all religions raise up the sick, but Jesus raised the dead (John 11).
All Witnesses Are Equally Reliable? Hume’s reasoning assumes that the credibility of the witnesses for the miracle claims in all religions is the same. The New Testament miracles are attested by contemporary eyewitnesses. Islamic miracle stories appear generations later (see Muhammad, Alleged Miracles of). Some have credible witnesses; others do not. The reliability of each witness to a miracle must be evaluated on its own merits. They are decidedly not equal.
Evaluation. Rather than disproving New Testament miracles, Hume’s third argument that all religion’s miracle stories are equally (un)reliable, supports the authenticity of biblical miracles. For the superiority of the Christian witnesses is a sound argument against all non-Christian miracle claims. We may restate the argument this way:
- All non-Christian religions (which claim miracles) are supported by similar “miracle” claims (in both their nature and witnesses).
- But no such “miracles” have strong enough testimony to maintain evidential value, so they are self-canceling.
- Therefore, no non-Christian religion is supported by miracles.
If this is so, then we can argue that only Christianity is divinely confirmed as true.
- Only Christianity has unique miracle claims confirmed by sufficient testimony.
- What has unique miraculous confirmation of its claims is true (as opposed to contrary views).
- Therefore, Christianity is true (as opposed to contrary views).
Jesus’ miracles were instantaneous, always successful, and unique. So-called miracle workers who claim partial success effect only psychosomatic cures, engage in trickery, perform satanic signs, or other naturally explainable events. No contemporary healer even claims to heal all diseases (including “incurable” ones) instantaneously, with 100 percent success. Jesus and his apostles did. This is unique, and it sets these miracles against all competing claims by other religions. If biblical miracles are unique, then they alone confirm the truth-claims connected with them (Exod. 4:1f.; 1 Kings 18:1f.; John 3:2; Acts 2:22; 14:3; Heb. 2:3–4). All other so-called miracles are, as Hume’s argument shows, self-canceling.
Arguments from Analogy. Ernst Troeltsch (1865–1923) laid down the rule of analogy: The only way one can know the past is by analogy in the present. That is, the unknown of the past is arrived at only through the known in the present. On the basis of this, some argue that, since no miracles occur in the present such as are alleged to exist in the past, it follows that proper historical method eliminates the miraculous.
Troeltsch used “the principle of analogy” and Antony Flew a similar principle of “critical history” against miracles. These theories are extensively examined in the article Troeltsch, Ernst, so they will be covered only in general terms here.
Troeltsch’s “Principle of Analogy.” This principle of analogy, according to Troeltsch, asserts that “Without uniformity in the present, we cannot know anything from the past” (Historicism and Its Problems). On the basis of this principle, Troeltsch and others have insisted that no evidence or witnesses are adequate to establish miracles (Becker, 12–13).
This argument does not insist that no such miracles as are reported in the Bible occurred. The claim, rather, is that they are historically unknowable, whether they occurred or not. Most would agree that no such miracles as a virgin birth, walking on water, or raising the dead are occurring today, so by Troeltsch’s analogy, such events cannot be known to have happened ever.
Flew’s “Critical History.” Similar is Antony Flew’s “critical history.” Flew asserts that the remains of the past cannot be interpreted as historical evidence unless we presume that the same basic regularities obtained then as do today. The historian must judge the past evidence by personal knowledge of what is probable or possible (350).
Flew concluded that the critical historian dismisses stories of a miracle out of hand, ranking them with the impossible and absurd (ibid., 352). The impossibility, Flew adds, is not logical but physical. Miracles are possible in principle, but in practice they break natural laws that are simply never broken.
Evaluation of the Historical Argument. Troeltsch and Flew attempt to rule out knowability by what Flew calls “critical history.” Further, the argument (as Flew admits) follows the basic form of Hume’s antisupernaturalism, critiqued above. All of these arguments assume that to be critical and historical one must be antisupernatural. By this view, a closed mind is prerequisite to doing “critical” historical study.
The principle that the present is the key to the past, or the past is known by analogy to the present is valid. This is so since those living in the present have no direct access to the past. The kind of causes known to produce certain kinds of effects in the present can be assumed to produce similar kinds of effects in the past.
But this principle does not rule out a credible belief in miracles in the past, even if no such miracles exist in the present. Fallacies are involved in the historical argument.
Uniform or uniformitarian?. Troeltsch and Flew confused principles of uniformity (analogy) and uniformitarianism. They assumed that all past events are uniformly the same as today’s. This is not only an assumption, but it doesn’t fit what even naturalistic scientists believe about origins. All scientists believe that the origin of the universe and the origin of life are singular and unrepeatable events (see Origins, Science of). But if the past can be known only in terms of processes now at work, then there is no scientific basis for knowledge about them. Another problem with uniformitarianism is that processes change. Geological uniformitarians fail to account for catastrophes, climatic changes, landmass shifts, and other factors that might have altered geological forces.
Uniformitarianism illogically assumes that there have been no past singularities. While knowledge of the past is based on analogies in the present (uniformity), the object of this knowledge can be a singularity. Archaeologists may know on the basis of analogy that only intelligent beings can make projectile points. However, the making of one unique spear point by a particular craftsman in a particular tribe can also be studied in itself. What can be learned about this singular past event can become present knowledge—a basis for analogy when other spear points are discovered. By analogy scientists have learned that certain levels of specified complexity originate only in intelligent beings.
Analogy, properly understood, supports as credible the possibility that some past events had a supernatural intelligent cause. Even without analogy to the present, there is good evidence that the universe began (see Big Bang), and that it had an intelligent supernatural cause.
Special pleading. The Historical argument against miracles makes a special pleading that evidence for individual events cannot be allowed unless the events are repeated. This weighs the evidence for all regularly occurring events, rather than for the particular event(s) at issue. This is not a standard rule of evidence. Further, it pleads that no miracle either has occurred, can, or ever will in today’s world. Flew and Troeltsch are simply not omniscient enough to know this is true.
Begging the question. Flew also commits the fallacy of petitio principii. In practice, he begs the question when he asserts that miracles are “absolutely impossible” and that the critical thinker will dismiss them “out of hand.” But why should a critical thinker be so biased against the historical actuality of a miracle? Why should one begin with a methodology loaded against certain past events, before looking at the evidence?
Hindering scientific progress. Uniformitarian views have hampered the progress of science. The big bang theory is an example. Astrophysicist Arthur Eddington spoke of this special, explosive beginning of the universe as “repugnant,” “preposterous,” and “incredible” (Jastrow, 112). Albert Einstein made a mathematical error, so sure was he that the big bang was “senseless” (ibid., 28).
The evidence is so compelling that many scientists now believe that the basic hydrogen atoms of the universe were created in milliseconds. Most astronomers today accept the reality of a great initial explosion. Here is a singularity, which by its nature cannot be repeated. Yet it is a viable theory of origins and the proper object of science, though scientists had to be dragged to it because it does hold definite theistic implications.
Appealing to the general to rule out the particular. A strange sort of logic works in the historical argument. One must judge all particular (special) events in the past on the basis of general (regular) events in the present. Why not use special events in the present as an analogy for special events in the past? There are unique and particular “anomalies.” From a strictly scientific point of view a miracle is like an anomaly. Here the historical argument uses special pleading. Neither Troeltsch nor Flew allows evidence to count for particular events, in lieu of the evidence for general categories of events. There are far more regular and repeatable events than the unrepeatable kind. there is no evidence for the unrepeatable. It is like refusing to believe that someone won the lottery, because thousands more lost it. Along these same lines, the contemporary philosopher Douglas K. Erlandson argues that scientific law, as such, is concerned with general classes of events, whereas the supernaturalist is concerned with events that do not fit general classes. A belief in the latter does not upset belief in the former (Erlandson, 417–).
Proving too much. The historicist arguments prove that much of what naturalists believe about the past cannot be true. As Richard Whately showed in his famous satire on Hume’s naturalistic skepticism (Whately, 224, 290), if one must reject unique events in the past because there is no analogy in the present, then the incredible history of Napoleon must be rejected.
Not critical enough. Actually, “critical history” is not critical enough. It does not criticize the unreasonable acceptance of presuppositions that eliminate valid historical knowledge. Far from being open to evidence, its naturalism eliminates in advance any miraculous interpretation of events in the past. It legislates meaning rather than looks for it.
Arguments from Science. Since the origin of modern science it has been common to claim that miracles are not scientific. Some critics object to miracles because they are said to be contrary to the very nature of the scientific procedure for handling irregular or exceptional events. They insist that when scientists come upon an irregular or anomalous event they do not posit a miracle. They broaden their understanding of natural processes to take in that event. To do otherwise would be to forsake the scientific method. Some individual arguments include:
Ninian Smart. Ninian Smart reasons that nothing in nature can be out of bounds for exploration. Otherwise it would stultify scientific research. But a belief that certain events are miraculous erects a bar against science. Hence, acceptance of miracles violates the proper domain of science (Smart, chap. 2). The argument can be summarized.
- A miracle is an exception to a natural law.
- In science, exceptions are goads to find a better explanation, not an indication to stop research.
- Hence, accepting miracles stops scientific progress.
Therefore, a miracle cannot be identified ever as an irregular event or anomaly. Rather, it calls for further research. When one natural law does not explain an exception, scientists do not throw in the towel; they look again, more deeply. What is an exception to one scientific description (L1) can be included within a broader description (L2).
Patrick Nowell-Smith. The supernaturalist’s claim that an event is a miracle because it cannot be explained in terms of scientific laws bothers Patrick Nowell-Smith. “We may believe him [the supernaturalist] when he says that no scientific method known to him will explain it. . . . But to say that it is inexplicable as a result of natural agents is already beyond his competence as a scientist, and to say that it must be ascribed to supernatural agents is to say something that no one could possibly have the right to affirm on the evidence alone (Nowell-Smith, 245–46).
However strange an event, he argues, it must not be ascribed to the supernatural, because future scientists may very well explain it. At one time the bumblebee’s flight was unexplained by natural law. However, principles of this very natural occurrence have come to light in the discovery of power packs in the bee’s cells called mitochondria, which make flight by rapid wing motion possible. The argument can be described:
- What is scientifically unexplained is not necessarily scientifically unexplainable.
- Miracles are scientifically unexplained.
- Miracles are not scientifically unexplainable.
An explanation qualifies as scientific, according to Nowell-Smith, if a hypothesis from which predictions can be made can afterwards be verified (ibid., 249). Further, the explanation must describe how the event comes about.
In this definition, “lawful” miracles should be explainable by laws that can be stated. If not, the event can be explained. “If we can detect any order in God’s interventions, it should be possible to extrapolate in the usual way and to predict when and how a miracle will occur” (ibid. 251). Nowell-Smith challenges supernaturalists to consider whether the notion of explanation does not necessarily include hypothesis and prediction and thought about whether the “supernatural” could play any part in it (ibid., 253).
Should it be objected that he is simply redefining the “natural” to include miracles, Nowell-Smith replies: “I will concede your supernatural, if this is all that it means. For the supernatural will be nothing but a new field for scientific inquiry, a field as different from physics as physics is from psychology, but not differing in principle or requiring any non-scientific method” (ibid.). This may be summarized:
- Only what has predictive capabilities can qualify as an explanation of an event.
- A miracle explanation cannot make verifiable predictions.
- Therefore, a miracle explanation does not qualify as an explanation of the event.
The implications of this reasoning are that miracle explanations must become scientific or they cease to be explanations at all. So a miracle is methodologically unscientific. It is contrary to the scientific means of explaining events, a way that always involves the ability to predict similar events. Further, Nowell-Smith denies that rational agency is necessary to account for any anomaly in nature. Ultimately, all that happens will be shown to result from natural law.
Alistair McKinnon. Another opponent of miracles, Alistair McKinnon (see another McKinnon argument in the article Miracle.) put the scientific law argument this way:
- A scientific law is a generalization based on past observation.
- Any exception to a scientific law invalidates that law as such and calls for a revision of it.
- A miracle is an exception to a scientific law.
- Therefore, any so-called “miracle” would call for a revision of the present scientific law.
In McKinnon’s view, a miracle would be assumed to be a natural event under a new law that incorporates it into its natural explanation. Laws are like maps, and maps are never violated; they are revised when found to be in error.
Malcolm Diamond. Others have attempted to argue against miracles as opposition to scientific methodology. For example, Malcolm Diamond, professor of philosophy at Princeton University, insists that it is disastrous to accept miraculous exceptions to scientific laws. If one accepts some exceptions as supernatural, “scientific development would either be stopped or else made completely capricious, because it would necessarily be a matter of whim or whether one invoked the concept of miracle” (Diamond, 317).
Diamond sees two problems with supernaturalism. First, exceptions should not stop scientific research. They are, in fact, goads to further study. Second, exceptions should not necessarily be called miracles. Does the odd prove God? If not, how does one distinguish the unusual from the supernatural?
According to Diamond, “Allowing for the possibility of supernatural explanations of naturally observable occurrences is something that would, in effect, drive working scientists to opt right out of the scientific enterprise. . . . These scientists would not be able to investigate [the miracle]. . . . As scientists they would not be able to determine whether the exception was supernatural” (ibid., 320). Scientists must operate with autonomy. They must set their own rules and referee their own games. Therefore, although nothing logically would prevent a scientist from accepting a supernatural interpretation for an utterly extraordinary, the scientists would be selling out science.
Diamond concludes: “The answer that I shall offer on behalf of the naturalistic interpretation is pragmatic. It recommends reliance on the scientific explanations without pretending to be a conclusive refutation of supernaturalism” (ibid.).
The outline of this argument is pragmatic, based on the autonomy of the scientific method:
- Scientists, as scientists, cannot give up looking for naturalistic explanations for every event.
- To admit even one miracle is to give up looking for a natural explanation.
- Therefore, to admit miracles is to give up being a scientist.
Evaluation. Unlike other arguments against miracles, the scientific objection does not try to prove that miracles are impossible or even incredible. If successful it would show that miracles are not identifiable by the scientific method. It leaves open the possibility that there are other ways to identify a miracle. If by definition the scientific method deals only with a certain class of events (the repeatable), then singular events such as miracles cannot be identified by the scientific method. But what such an argument does not prove is that miracles do not occur or that there is not some other way to identify them. Nor does it show that there is no other way to identify the scientific method by which a miracle could be identified, at least in part.
Anomalies and the scientific method. Even the scientific procedure that deals with regular repeatable events allows for exceptional events that do not call for the explanation of another natural law. A scientist who encounters an anomaly does not automatically revise previously held laws. If the exception is not repeatable, there is no right to use it as the basis for a new law. It is inappropriate to demand that all exceptional events be naturally caused, but only that repeatable events be explainable. So in the nonrepeatable miracle, there is no violation of a scientist’s right to do science.
Science in the commonly understood sense deals with regularities, not singularities. One cannot expect a method geared to deal with regularities to eliminate the scientific viability of a miracle.
A scientific approach to the world is not limited to regular events. There are legitimate scientific approaches that deal with singular events, as even supernaturalists claim.
Even the scientific method admits exceptions or anomalies, and no scientist revises existing natural laws based on a single exception. Unless the scientist can show that it is a regular, repeatable part of nature, he has no basis on which to make a new natural law. There is no reason a miracle cannot fall into the broad category of the anomalous, even within the general sense of the scientific method.
Of course, there is more to a miracle than a mere anomaly. There are “divine” earmarks. However, even from a strictly scientific approach that deals with regularities, one cannot legitimately eliminate the possibility of identifying a miracle. To argue that every exception to a known natural law demands another natural explanation, simply begs the question. Such an argument goes beyond science and reveals a naturalistic bias (see Materialism; Naturalism).
As theists have long insisted, if there is a God, then he cannot be locked out of his creation. If he had the ability to create the universe, he has the power to produce occasional but naturally unrepeatable exceptional acts within his world. The only effective way to disprove miracles is to disprove God (see God, Alleged Disproofs of).
Confusion of categories. Even some naturalists have admitted that this argument is an a priori argument that can be refuted by noting that a supernaturally caused exception to a scientific law would not invalidate it. Scientific laws express regularities. A miracle is a special and nonrepeatable exception (Diamond, 316–17). One nonrepeatable exception does not call for revising a natural law. More likely it would be credited to faulty observation anyway. From a strictly scientific view, a nonrepeatable exception remains just that—an exception to known scientific laws. If, under specified conditions, the anomaly recurs, then a scientist has the right to call it a natural event. In this case, anomalies would be pointers to the development of a more general natural law.
Miracles, however, are not the result of natural laws. They were caused by the willed actions of rational agents, God and his representatives. That action of will is what cannot be repeated and therefore places miracles outside the realm of scientific observation. A miracle takes place because God wants it to. One cannot arrange for God to “want it” again so that scientists can watch. Miracles do not change our view of scientific laws, they simply step outside of them.
Since miracles are unrepeatable exceptions to known laws, they leave natural laws intact and therefore are not unscientific. Smart wrote, “Miracles are not experimental, repeatable. They are particular, peculiar events. . . . They are not small-scale laws. Consequently, they do not destroy large-scale laws.”
Begging the question. If scientific objections are understood to eliminate the acceptance of miracles by a rational person, they are unsuccessful. They clearly beg the question by insisting that every event in nature must be considered a natural event. For if whatever happens—no matter how unrepeatable—must not be considered a miracle, miracles are eliminated in advance by definition. Even if a resurrection from the dead occurred, it could not be a miracle.
Despite the fact that he claims the problem must be attacked with an open mind (ibid., 243), Nowell-Smith shows an invincible bias in favor of naturalism. His standards mandate that any event will be declared to be a natural event. He is, in fact, open only to naturalistic interpretations, not to the supernatural. That he begs the question is evident. He defines “explanation” in such a narrow way as to eliminate the possibility of a supernatural explanation. He arbitrarily insists that all explanations must be naturalistic in order to be counted.
The supernaturalist does not insist that “an event no matter how strange must have been due to a supernatural agent.” It does seem likely that most strange events are natural. But the supernaturalist does object when Nowell-Smith says that supernatural agency cannot be part of the report of a strange event. The supernaturalist says that one should look at the evidence on its merits.
Nowell-Smith simply assumes that all phenomena ultimately admit a natural explanation (ibid., 247). He cannot know this as a scientist. There is no empirical proof. This assumption is simply a matter of naturalistic faith. Even if he were presented with empirical evidence of a miracle, he makes it clear that he would never admit it to be supernatural. Pending discovery of a naturalistic explanation, he will persist in believing that an explanation can be found.
Neither is it necessary that all true explanations have predictive value. There are events he would call natural that no one can predict. If the naturalist replies that he cannot always predict an occurrence in practice but can in principle, the supernaturalist can make that level of prediction. In principle we know that a miracle will occur whenever God deems one necessary. If we knew all the facts, including the mind of God, we could predict precisely when the miracle would happen. Further, biblical miracles are past singularities. Like the origin of the universe, they are not currently being repeated. But no prediction can be made from any singularity; they can only be made from patterns. The past is not known by empirical science, but by forensic science. It is misdirected to ask for predictions forward. Rather one attempts to make retrodictions backward.
The supernaturalist can agree with Nowell-Smith that “the breakdown of all explanations in terms of present-day science does not . . . immediately force us outside the realm of the “natural’ ” (ibid., 248). The two part company when Nowell-Smith requires natural causes for miracles. Such a position goes beyond what is warranted by the evidence. The naturalist demonstrates a faith commitment that rivals the religious dedication of the most ardent believers in miracles.
One problem behind this kind of scientific naturalism is the confusion of naturalistic origin and natural function. Motors function in accord with physical laws, but physical laws do not produce motors; minds do. In like manner, the origin of a miracle is not the physical and chemical laws of the universe, even though the resulting event will operate in accord with natural law. While natural laws regulate the operation of things, they do not account for the origin of all things.
Methodological naturalism. Scientific arguments against miracles are a form of rigid methodological naturalism. The very method chosen does not admit the possibility that any event will ever be identified as a miracle. Explanations that cover regular events do not necessarily apply to singularities. Rounded stones in a river are produced according to describable natural forces. But no natural law can account for the faces on Mount Rushmore. Here a non-natural, intelligent cause is appropriated (see Evolution, Chemical; Teleological Argument).
When a singularity is not known to be attributable to natural causes and demonstrates signs of divine intervention, then there are positive reasons to accept it as a miracle. The following are discussed with more detail in the article Miracle:
- They have an unusual character as an irregular event.
- They have a theological dimension as an act of God.
- They have a moral dimension, since God is an absolutely perfect moral Being. One moral mark of a miracle is that it brings glory to God.
- They have a teleological dimension. They are purposeful events.
- They have a doctrinal dimension. Miracles are connected, directly or indirectly, with “truth claims” (Heb. 2:3–4; see Miracles, Apologetic Value of).
When an irregular, unrepeatable event, not known to be produced by natural causes, is accompanied by other marks of intervention, there is reason to identify it as an act of a theistic God (see God, Evidence for).
Too restrictive a definition of science. The science arguments against miracles are based on an overly restrictive definition of science, one that deals only with repeatable events. Science also deals with singularities. True, the scientific method only tests regular, repeatable events. But scientists also recognize origin science, which is largely a study of singularities. The big bang origin of the universe is a radical singularity. The history of our planet is a singularity, yet it is the object of research. We would regard it as both strange and foolish for a geology teacher to rule out anything but a natural cause for the sculpted faces on Mount Rushmore. It would seem odd if an archaeologist were limited to natural causes for projectile points and pottery. Insisting that one who does not insist on natural causes cannot be scientific is to improperly restrict science.
Miracles and the integrity of science. We are now in a position to evaluate the charge that belief in miracles is unscientific. Diamond’s comments make evident his belief in the absolute autonomy of the scientific method. He assumes as a matter of faith, with only pragmatic justification, that the scientific method is the method for determining all truth. Indeed, it is not just the scientific method, but one aspect of the scientific approach—the search for natural causes—that is assumed to be the only approach to truth. Diamond’s arguments are vulnerable to several criticisms:
First, it is wrong to presuppose that the scientific method necessarily entails naturalism. Scientists, as scientists, need not be so narrow as to believe that nothing can ever count as a miracle. All a scientist needs to hold is the premise that every event has a cause and that the observable universe operates in an orderly way.
Second, it is wrong to assume that natural laws have dominion over every event, rather than every regular event. To assume that every irregular, unrepeatable event has a natural explanation is not science but metaphysics. Natural laws do not account for the origin of all events any more than the laws of physics alone explain the origin of an automobile. Natural laws account for the operation of these things.
Third, it is unscientific to be closed to reasonable explanations. If a God caused the universe to exist and cares for it, it is not unreasonable to expect that he can perform some regular activities and also some special events. The only way to effectively disprove this possibility is to disprove the existence of such a God, which most atheists agree is impossible to do (Geisler, Miracles and the Modern Mind, chap. 12). The truly scientific and open-minded person will not dismiss in advance, logically or methodologically, the possibility of identifying some miraculous events in the defense of scientific autonomy.
Fourth, when the argument against miracles reduces to its basic premises, it becomes:
- Whatever actually occurs in the natural world is a natural event.
- Some so-called “miracles” have occurred.
- Therefore, these miracles are really natural events.
This formation lays bare the circular reasoning of the naturalist’s argument. Whatever happens in the natural world is, ipso facto, a natural event. Whatever occurs in nature was caused by nature. Even Michael Polanyi seemed to fall into this trap when he wrote, “If the conversion of water into wine or the resurrection of the dead could be experimentally verified, this would strictly disprove their miraculous nature. Indeed, to the extent to which any event can be established in terms of natural science, it belongs to the natural order of things” (Jaki, 78). This, of course, assumes what is to be proven, that there is no supernatural Being who can act in nature. Just because an event occurs in the world, does not mean it was caused by the world. It may have been specially caused by a God who transcends the world.
The preservation of the scientific method. If miracles are allowed, how can one retain the integrity of the scientific method? If some events are ruled out of bounds to the scientists, then has not the supernaturalist closed the door on rational examination of some events? Positing a supernatural cause for the origin of some rare events in no way affects the domain of science, assuming science is based on a regular pattern of events. Operation science is naturalistic and has every right to demand explanatory control over all regular events. But science, as such, has no right to claim that it alone can explain singularities.
Science has unlimited authority in the classification of regular events. The scientist has a right, even an obligation, to examine all events, including anomalies. However, the singular, unrepeated event that is not part of a regular pattern must be classified among the “not yet explainable as natural events.” Within this class are events that may have a supernatural cause. To assume that all not-yet-explained events are naturally explainable moves beyond science into philosophical belief in naturalism. Indeed, it rules out the possibility that there is a supernatural God who can intervene in the world he created. But this is contrary to the evidence (see God, Evidence for).
Summary. Hume offered a forceful argument against miracles. But, strong as it may seem, the evaluation indicates that he was overly optimistic to believe that this argument could be “an everlasting check” and “useful as long as the world endures” to refute any credible claim for the miraculous. In fact, Hume’s argument is not successful. In the “hard” form he begs the question by assuming that miracles are, by definition, impossible. In the “soft” form of the argument, Hume ignores contrary evidence, begs the question, proves too much (for example, that Napoleon did not exist), is inconsistent with his own epistemology, and makes scientific progress impossible. In brief, to eliminate miracles before looking at them seems prejudicial. A wise person does not legislate in advance that miracles cannot be believed to have happened; rather he looks at the evidence to see if they did occur. So, for the rational mind, Hume’s efforts to eliminate miracles must be considered unsuccessful.
Hume was right to demand that witnesses meet criteria of trustworthiness. Indeed, courts of law depend on such criteria to determine life and death issues. However, unbeknown to Hume, his tests for the truthfulness of witnesses, which he believed would eliminate the credibility of miracles, actually verify the trustworthiness of New Testament witnesses, particularly the miracle of the resurrection.
Hume’s self-canceling-witness argument fails because it is based on false presuppositions which, when corrected, boomerang into a proof for the uniqueness of Christianity. His argument is based on the premise that all alleged miracles are created equal. But this is not true, either of the nature of the alleged miracle or of the number and reliability of the witnesses.
In evaluating the historical argument against miracles it must be noted that there is a crucial difference between the principle of uniformity (or analogy), on which all valid inquiry is based, and the principle of uniformitarianism. The latter is a naturalistic dogma which rules out in advance by its very methodological principle the credibility of the miraculous. Troeltsch’s principle of analogy, used to reject miracles, is an example of historical uniformitarianism. A form of historical naturalism, it assumes that all events in history are naturally explainable. This bias, however, is contrary to both rational thought in general and scientific thought in particular.
Various attempts have been made to prove that belief in miracles is contrary to scientific explanations or to scientific methods. Some argue that miracles, contrary to natural laws, are unpredictable; others contend that miracles are unrepeatable or would sacrifice the autonomy of science. Such arguments beg the question in favor of naturalism. They assume the scientific method must be defined in such a way that excludes acceptance of miracles. The central, though hidden, premise is that every event in the world must have a natural cause. If one does not now have that explanation, it must be believed to ultimately exist. The supernaturalist points out that one does not have to be incorrigibly naturalistic to be scientific. Properly speaking, the domain of scientific law is the realm of regular, not all, events.
Miracles do not destroy the integrity of the scientific method. Science is possible so long as theists believe that the world is orderly and regular and operates in accordance with the law of causality. If the origin of the world can have a supernatural cause without violating the laws by which it operates, such a God can also cause other events without violating the regular natural operation. Since empirical science deals with the way things operate, not how they originate, the origination of an event by a supernatural cause in no way violates natural law. As physicist George Stokes observed, a new effect can be introduced into the natural world without suspending the ordinary operation of the world (Stokes, 1063).
- L. Archer, Jr., Encyclopedia of Bible Difficulties
- F. Arndt, Bible Difficulties
———, Does the Bible Contradict Itself?
- Barbour, Issues in Science and Religion
- Becker, “Detachment and the Writing of History,” in P. L. Snyder, ed., Detachment and the Writing of History
- H. Bradley, The Presuppositions of Critical History
- Burrows, What Mean These Stones?
- L. Diamond, “Miracles,” Religious Studies 9 (September 1973)
- K. Erlandson, “A New Look” in Religious Studies (December 1977)
- Flew, “Miracles,” in The Encyclopedia of Philosophy, ed. P. Edwards
- L. Geisler, Answering Islam
———, Christian Apologetics
———, Miracles and the Modern Mind
———, When Critics Ask
- Geivett and G. Habermas, In Defense of Miracles
- Glueck, Rivers in the Desert: A History of the Negev
- Greenleaf, The Testimony of the Evangelists
- W. Haley, An Examination of the Alleged Discrepancies of the Bible
- Hawking, A Brief History of Time
- Hume, An Abstract of a Treatise on Human Nature
———, An Enquiry Concerning Human Understanding
———, Treatise on Human Nature
- Jaki, Miracles and Physics
- Jastrow, God and the Astronomers
- S. Lewis, Miracles
- Nowell-Smith, “Miracles,” in A. Flew, et al., eds., New Essays in Philosophical Theology
- Smart, “Miracles and David Hume,” in Philosophers and Religious Truth
- Stokes, International Standard Bible Encyclopedia
- Swinburne, The Concept of Miracle
- Troeltsch, Historicism and Its Problems
———, “Historiography” in Encylopedia of Religion and Ethics
- Whately, “Historical Doubts Concerning the Existence of Napoleon Bonaparte,” in H. Morley, ed., Famous Pamphlets, 2d ed.
- N. Whitehead, The Concept of a Miracle
- Wilson, Rocks, Relics, and Biblical Reliability
- P. Yockey, “Self-Organization, Origin of Life Scenarios, and Information Theory,” JTB (1981)
 Norman L. Geisler, Baker Encyclopedia of Christian Apologetics, Baker Reference Library (Grand Rapids, MI: Baker Books, 1999), 457–468. ©1999 by Norman L. Geisler. All rights reserved. No part of this article may be reproduced, stored in a retrieval system, or transmitted in any form or by any means—for example, electronic, photocopy, recording—without prior written permission of the publisher. The only exception is brief quotations in printed reviews. The article above was reproduced with specific written permission from Dr. Geisler for use on this website, http://faithsaves.net. For more of Norman Geisler’s material, one can visit http://normangeisler.com. Much of Dr. Geisler’s apologetic material is very useful, although he is a neo-evangelical Baptist, not a historic Baptist separatist.