Science in Christian Perspective



The Logic of Miracles
Philosophy of Religion
Denver Free University
Denver, Colorado

From: JASA 33 (December 1981): 145-152.

Christians who require strong evidence for their beliefs would have little to lose should miracles be found inadequate to establish them. Other grounds for belief are, I believe, quite sufficient.1 However, because we have good reason to believe that Jesus as well as the whole of the Hebrew and early Christian tradition accepted miracles as adequately verifying religious claims, Christians particularly should be hesitant to write off miracles completely.

With this as my own motivation for looking into this question, I would like to examine the logic of miracles as evidence for metaphysical and religious claims. I believe that only if such foundational matters are closely analyzed can we adequately understand miracles and answer many of the common objections to the claim that they possess evidential value.

Following this lead of going back to the ancients, I believe that contemporary apologists have basically gone wrong by building on modern concepts, such as that of natural law. It has become common to try to provide evidence for God's existence and activity by attempting to present phenomena which indicate a suspension of such natural laws. By making such a move in defense of miracles, apologists have left themselves open to the most devastating critiques.2

Another late concept, that of the nature/supernature duality, has not had the same disastrous results and, indeed, has greatly contributed to our understanding and defense of miracles. However, being a late concept, we will see that it is not strictly necessary for understanding the logic of miracles. We will examine the concept of supernaturalism later in this paper.

What is strictly necessary is to find a way to justify the claims of the agent of the miracle. But how can these claims be established as true? How can a miracle be identified which possesses evidential force?

Similar questions can be found in contemporary critiques of miracles. So it is of value to look at some common objections. Antony Flew maintains that one must have a strong sense of natural law in order to claim miracles: "it is only and precisely insofar as it is something really transcendent- something so to speak, which nature by herself could not contrive-that such an occurrence could force us to conclude that some supernatural power is being revealed.3 I He goes on to make his second point: "We certainly cannot say, on any natural (as opposed to revealed) grounds, that anything that actually happens is beyond the power of unaided nature any more than we can say that anything which any man has ever succeeded in doing transcends all merely human powers."4

The same objection can be found among orthodox Christians: the non-theist could quite justifiedly face a miracle

How can we identify a miracle such that particular religious claims are verified?

and respond, "Well, strange things do happen after all," according to presuppositionalist Cornelius Van Til.5 Even evangelical apologist Norman Geisler claims that miracles have evidential value only after the theistic world view is in some other way established. Otherwise, proponents of any of the several world views-natural, supernatural and pantheistic-could justifiedly interpret the same phenomena to fit their respective world views.6 At one point or another the contention is made that something other than a high theism can account for any phenomenon in this world. Thus the most important question remains: how can we identify a miracle such that particular religious claims are verified?

I define a miracle as "a non-normal occurrence in the world, perceived through normal sense experience and giving evidence for the existence of a being of greater than normal human intelligence and/or power, and giving evidence for the claims of the apparent agent of the phenomenon." 7

As was stated earlier, with this definition the distinction between nature and supernature is not necessarily presupposed; indeed, any of the several major world views may be evidenced depending on the claim of the agent of the miracle. What follows is an analysis of the logic of miracles as, I believe, it has been assumed from antiquity. Concepts such as that of natural law and the distinction between nature and supernature have not always been with us; yet the intrinsic evidential value of miracles has been accepted for ages.

Characteristics of Miracles

The first necessary characteristic of a miracle is that it is an unusual and rare phenomenon; it does not fit "the order commonly observed in nature," as Aquinas puts it. Secondly, a miracle must involve a propositional claim which is either inherent in the phenomenon or precedes it as a prediction that the phenomenon will occur.

There are several possible interpretations of a miracle as distinguished by the above characteristics. One can accept a naturalistic world view and claim that nature is absolutely uniform and admits no exceptions. Potentially, this strange phenomenon will eventually be fully explained through scientific examination.

One can take a naturalistic world view and claim that nature is not absolutely uniform; exceptions may be admitted. Causal factors such as forces or even existent objects may come into being from no prior cause at all. The laws of nature may have exceptions. The strange phenomenon called a miracle is claimed to be a true ontological anomaly of nature that science will never be able to explain.

A second world view claims that the "miracle" is merely a manifestation of God-who is the universe-into a new form. According to the pantheist, God sometimes appears simply as the world, sometimes as Krishna, sometimes as Jesus, sometimes as a burning bush which is not consumed; all are God, Atman, the One, or whatever the particlar metaphysic may wish to claim.

The third world view is supernatural theism. Here it is claimed that the miracle is the direct result of the power of the divine or the supernatural acting in nature.

All of these world views are equally justified in their interpretations if we grant the first phenomenal characteristic alone, the miracle's unusualness or rarity. But if the second characteristic, the propositional element, is also present, then a particular claim and/or one of these particular world views may be evidenced. This now needs to be demonstrated. A miracle gives evidence for the existence of a being with greater than normal human power and/or knowledge and it provides evidence for the claims of the apparent agent of the miracle. The evidence for each is different and so each will be considered separately.

Evidence of Superhuman Power or Knowledge

There are one thousand white marbles and one black one in an urn. A man randomly selects marbles from the urn. Before he does so, I make a prediction that at some seemingly random point-say the 813th pick-the black marble will be selected. On the 813th pick, the black marble is selected. The chances are 1/ 100 1 that I could make the right guess-not very good odds to say the least. But if I had some kind of knowledge of the arrangement of the marbles and the exact way the man would select each marble, or if I have some kind of control over the arrangement of the marbles and over the man's choices in selecting marbles, then there would not be any improbability whatsever of making the right guess as to when the black marble would be picked.

Likewise, if Jesus predicted his bodily resurrection from the dead, the chance would be extremely improbable that it would occur unless he had some control over the situation and/or some knowledge people do not normally have.

Though we cannot know for sure, it seems very unlikely that a physical resurrection from the dead could ever occur by chance, no matter how many eons of time we allow. But let us imagine that eventually such an event could take place. Imagine that the chemical arrangements of the cells of the body just happened to reform in a manner such that life would again be possible. (Many biblical miracles are such that we can imagine them happening by chance at some time or another if we are given enough time.)

So let us imagine that given enough time, a physical resurrection could occur. Say the chance is one out of some very large number like 101000 (1/101000) that it would occur to some particular individual. If we have 101000 persons, it will likely occur eventually. This is the same principle we use when we say that we have 1/1001 chances of picking the black marble at any given time but that we have 1/1 chance of picking it eventually.

Now suppose Jesus never predicted his resurrection. Suppose he was crucified and died and after two nights and one day in the tomb, he came back to life totally healthy. He could not now claim this event as evidence for the existence of some being with superhuman power or knowledge who effected the event (or that he himself had this power or knowledge). He could not do this any more than I could claim special knowledge or powers for myself if I wait until after the black marble was picked to predict its selection. 

My point here is that a prediction is needed for us to have evidence of greater than normal human knowledge or power. This is because the event is far too unlikely to occur by chance at the time predicted. The alternate explanation of special power or knowledge is the only alternative. We face no improbability whatsoever for the latter explanation. A wise man apportions his belief according to the evidence according to Hume. It is immensely improbable that this event could be caused by chance. It is not at all improbable that it had been caused by the intervention of a being with greater than normally assumed human power and/or knowledge. Therefore, this second explanation is the more reasonable and the wise man will assume it.

Now a fulfilled prediction is not the only means of giving evidence of special knowledge or power. The miracle of the burning bush (Exodus 3) is such a case. Here the phenomenon (the bush burned without being consumed)
involved the propositional claim within itself (the Lord speaking from the bush). The situation was such that it
seemed as though the speaker actually had control over the phenomenon, the burning.

The distinction between prophecies and most other biblical miracles (and miracle claims of other religions) consists in the fact that the miracles usually involve an event that cannot be accounted for by science-or at least by con
temporary science. A prophecy may be of a normal natural occurrence. But as we have seen above, a miracle is essentially of the same nature as a prophecy. Special power or knowledge is evidenced by both.

The problem we now face is to move from our present conclusion-the existence of a being with special power
and/or knowledge, whether it is the apparent agent of the miracle or a claimed other-to the more particular claims
of the agent. That is, if Moses or Jesus made certain claims about the nature of the being they stated enacted these
signs, it seems that they should be believed. So far, we know only that special power and/or knowledge has been evidenced. We do not know if this power/knowledge is possessed by the one who predicted the event would occur-or if it is carried out, as was usually claimed, by another, such as God. We do not even know if the special knowledge/power consists merely of a slight-of-hand trick or some other more sophisticated deception. It may consist of technological knowledge possessed by an advanced  civilization such as the "Chariot of the Gods" hypothesis claims.

Human trickery can be somewhat eliminated by close examination of the phenomenon or the historical evidence. So the historical evidence does have a bearing on removing some interpetations of the phenomenon. But other alternatives cannot as easily be eliminated. How do we know that the resurrection of Jesus gives evidence for the existence of a good creator God, for instance? Might the one (either Jesus or another) who effected this event be evil? Is anything evidenced by this event other than special power or knowledge?

In fact, any of the three major world views could still be claimed to be compatible with the phenomenon. Naturalists could claim that extraterrestrial aliens caused the event. Pantheists could claim that this is a manifestation of a God who has the power to so manifest himself. The alternative that is definitely eliminated is the claim that the event is a result of pure chance, one of the more important naturalistic interpretations. We have evidence of the power and/or knowledge of a being who is either in and part of the physical universe or nature (naturalism), outside of nature (supernaturalism) or identical with nature (personalistic pantheism).

Can we move from these conclusions to justified belief in the claims made by the agent of the miracle?

Dennis Jensen received his M.A. in Philosophy of Religion from Trinity Evangelical Divinity School (Deerfield, Ill.). He has worked with InterVarsity Christian Fellowship for a number of years, both through college and more recently as an associate staff member. He is currently teaching Philosophy of Religion at the Denver Free University.

Evidence for the Claims of the Agent of a Miracle

Regardless of one's world or life view, it must be admitted that a greater than normal human power and/or knowledge is evidenced by a miracle. Normally the leap from this claim to belief in the claims of the miracle worker, the prophet, or the object of the vision involves something very like the scientific process of hypothesis verification, with the claims constituting the hypothesis.

The claim may confirm any one of the major world views with variations ranging from spiritism to panentheism. The miracle worker may even claim to be a very accomplished and capable conjurer and that the power and knowledge displayed is solely his own.

The claim, that is, the hypothesis under consideration, is relatively verified by the sign or miracle. For example, the claim that an omnipotent, omniscient, omnibenevolent creator exists is relatively verified by the occurrence of a phenomenon that requires an extremely great knowledge and/or power. If God exists, then this event could occur; the event occurs; therefore, this God exists. Thus we have the logic of hypothesis verification.

Obviously the syllogism is not valid since it involves the fallacy of affirming the consequent. Our problem-as is commonly the problem in scientific hypothesis verification-is that other hypotheses may equally well be able to account for the occurrence. A less than infinite mind or power, a pantheistic god, an intelligent creature from a distant galaxy, perhaps even a good human conjurer may be able equally well to account for the occurrence. Why pick the particular hypothesis claimed by the miracle worker? How can one hypothesis be found to be more adequate to the evidence than another?

Most people would admit that if they confronted a miracle, such as any of the several presented in the Hebrew and Christian Scripture, they would believe the claims of the apparent agent of the miracle. The most antitheistic professor I had as an undergraduate, an avid Humean believer in the epistemic worthlessness of miracles, admitted (perhaps inconsistently to his Humeanism) that he would "believe" if he saw the burning bush (Exodus 3), so long as he had a chance to check for mirrors, wires etc. What exactly is this evidential force that most people ;ould admit to?

I believe that the answer ties in several factors combined. First of all the claim, tht hypothesis, fits and thus is relatively verified by the now clearly evidenced state of affairs that there exists a being of greater than normal human intelligence and/or power. The involved claim is more evidenced than it would be without the miracle. As we have just seen, however, this is not enough in itself to establish the evidential value of miracles. Secondly, it is clear that there is an agent of the miracle who either knows that the claim is true or does not know that it is true. Thus the agent of the miracle must either be definitely lying or definitely speaking the truth. Thirdly, because a claim should normally be believed unless there is good reason to doubt the credibility of the claimant and because of the first factor we have just considered, it follows that the claim of tht ultimate agent of the miracle should be believed (and the proximate agent, if there is a distinction between agents) Allow me to clarify this argument further through a critique of this view.

Stephen Wykstra in a critique of John Montgomery's evidentialism suspects that Montgomery may accept the same criteria for miracles we have just presented:

. . . the fact that a person has a certain extraordinary power neither entails that he know, nor, if he knows, that he is truthful about, the true explanation for this power. If it were factually established that Jeanne Dixon could prophecy, or that Uri Geller could bend spoons across a room, would we be rationally obliged to accept any explanation of their powers they proffer, simply because other humans cannot do what they can do? By the only criterion Montgomery provides, Uri would have to be accepted as an Agent of Revelation if he explains to us: "Only by the special power of God Himself, the Omnipresent One, can I bend spoons across an empty room.8

Wykstra suggests that this criterion is obviously specious. I would argue that once the more ludicrous features of such miracle examples are removed, the criterion should be admitted as valid. Bending spoons across an empty room is, in this case, much in the same category as miraculously growing hair on billiard balls (Montgomery's example). Both lack a "fitness" to the claims of the agent of the miracle, at

least if the claim is notable (i.e., if, say, it is claimed that the phenomenon were carried out by an infinite God rather than by an apprentice magician). Clearly a greater than normal power and/or knowledge is evidenced; but such a display of knowledge or power is not quite what we would expect of the Omnipotent One.9

So would one otherwise, "be rationally obligated to accept any explanation of their powers they proffer, simply because other humans cannot do what they can do?" One would, because the claim is relatively verified by the phenomenon and because we should in any normal circumstances trust a person's non-exceptional claims.

Should such a phenomenon be given, it would be a display of power or intelligence as would be expected of theomniscience or omnipotence of God. We trust a person's normal claims in normal circumstances without full proof of their statements but only with minimal indication of their truthfulness. Likewise we should trust one's religious claims so long as a similar yet proportionately greater minimal indication of their truthfulness is given by the miracle. Because the religious claim requires more verification than a normal claim, we need a display of power/ intelligence as would be expected were the religious claims true.

Miracles were called "signs" in biblical times because they were pointers evidencing the religious claims made. They did not contain within themselves a full display of the infinite power, knowledge, goodness or creatorhood of God; rather they gave sufficient evidence for an experient to justifiedly infer the truth of the claims of the apparent agent.

It should now be said that some miracle claims are evidentially very strong even without any inference to the
full claims of the apparent agent. Certain prophecies, for instance, which involve very long periods of time and exact
historical concurrences give evidence of so great a source of intelligence and power that even hypotheses of super-
intelligent space travelers cannot explain the data." Thus the hypothetical alternatives may diminish even more than
we had suggested earlier.

With this analysis of the evidential value of miracles we can now look at some common objections.


Miracles were called '"signs" in biblical times because they were pointers evidencing the religious claims made they gave sufficient evidence for an experien t to justified ly infer the truth of the claims of the apparent agent.

Let us first consider the arguments against miracles presented at the beginning of this paper.

Flew claims that we need a strong view of natural law in order to have evidence that nature is truly being transcended by a miracle. But on our present argument, we do not nothing more than "the order commonly perceived in nature" to evidence a miracle.

Such an order need not imply exceptionless "nomologicals." We are not verifying religious claims by a phenome
non which is an exception to natural law. Rather we areverifying them by establishing the existence of one of
greater than normal human intelligence and/or power and by establishing the truthfulness of their claims. The non
normalcy of the phenomena need consist of nothing more than a fulfilled prediction or some other manifestation of
power or knowledge.

Though we need not accept inviolable nomologicals, we could just as easily accept them with no harm to our argument. On the one hand, we could reject nomologicals: A miracle may be a "breaking" or "overriding" or "suspension" of natural law. Lewis argues that general but not absolute or inviolable laws or uniformity of nature are
philosophically acceptable." Modern advocates of exceptionless laws, such as Flew, are oblivious to the
metaphysical leap they are taking. One cannot establish exceptionless nomologicals following the normal criteria for
establishing natural law (i la Flew, Hume's Philosophy of Belief, p. 206) since these criteria can in no way show
whether or not there can ever be exceptions to such laws. An intraversable "broad ugly ditch" separates experience
from inviolable laws.

On the other hand, inviolable nomologicals; may be accepted by the proponent of miracles. Jesus' walking on the
water or his ascension to heaven need not be seen as break ing a natural law any more than our sending a rocket to the moon should be seen as breaking the laws of gravity. Both merely indicate special activity in the normal course of nature by human or non-human agency and both can fit the laws of nature. Augustine represents this view by claiming that, "a portent is not contrary to nature, but contrary to our knowledge of nature."12

In Flew's second argument, he asks how we can know that anything which does occur is beyond the power of an
unaided nature. We can know by the claim of the agent of the miracle which is established as trustworthy. Our reply to Geisler is that the several world views cannot equally interpret an occurrence if that occurrence displays a greater than normal power and/or intelligence and if the apparent agent of the phenomenon makes specific claims concerning  the truth of one world view. If no such claim is made, then only what is stated is verified.

Flew: The Arbitrariness of Miracle Choices

We may be able to glean from Flew another important objection." If we admit a biological miracle such as the resurrection on the basis of historical evidence, why not admit another type of miracle which could equally well explain the evidence? Instead of a biological miracle, why not posit a psychological miracle such as mass hallucinations of Jesus and the empty tomb. Or suppose that two or three hundred years after the death of Jesus (or any other arbitrary number of years up to the present) individuals and groups all over the civilized world began quite spontaneously to believe in the resurrection and deity of this Jesus (who otherwise would not even be remembered). NonChristians likewise assume the existence and history of this Christian community. Aged documents relating stories of Jesus' teachings, life, death and resurrection as well as the teachings and exploits of his immediate followers spontaneously appear in the possession of some of these Christians.

This state of affairs would explain the physical evidence as we now have it. And such possible alternate miracle scenarios could be thought of ad infinitum. Why choose one miracle over the other? Why believe the biological miracle of a resurrection over the psychological miracle of the body being stolen by the Jewish leaders or the tomb guards?

It may be that this is not such a decisive problem after all. Don't we merely end up with several possible miracles, any one of which justifies the religious claim made? Does it matter which miracle we choose as long as the same religious conclusion is reached? Suppose, for instance, that we posit our late-spontaneous-belief hypothesis. We have a definitely non-normal phenomenon and we also have a claim that a non-normal phenomenon will evidence the Messiahship of Jesus. The claim is now in written documents which were created ex nihilo several hundred years after the time of Jesus. The claim is apparently made by the agent of the miracle.

The only problem with this answer (and the problem is substantial) involves the truthfulness of the speaker, the agent of the miracle. In this case, the stated claim is that the miracle is that of a resurrection and that this is stated by a pre-mortern Jesus. If this were actually a miracle of created documents and implanted beliefs, then the speaker would be lying. In such a case, the speaker could not be trusted in any of his claims and the miracle would evidence little more than that a being of greater than normal human intelligence and/or power has acted.

However, as we saw when we examined the evidential value of miracles, a person's religious claims should be trusted if under normal circumstances we would trust their non-religious claims and if some further minimal indication of their truthfulness is given by the phenomenon fitting the claims (as a resurrection could be expected of an omnipotent God). As we should trust one's normal claims, so we should trust one's non-normal but proportionately more attested religious claim.

Hume's Argument From Probability

David Hume is believed by many to have conclusively put an end to any possible argument from miracles to religious belief.14 C. S. Lewis has refuted Hume's central argument from probability more thoroughly and in more detail than I would hope to attempt.15 But the general form of Hume's argument, or at least a necessary element in it, which has been much repeated by other critics can be briefly considered here.

Hume claims that human experience can establish probability. Thus uniform human experience can supposedly establish the infinite improbability of the occurrence of a miracle. If all of our past experience does not include a particular type of experience, then no perceptual or testimonial evidence, according to Hume, should rightfully persuade us.

But this is surely inadequate. Indeed, Hume is inconsistent with his own principles which establish this claim. He says that, "it is a miracle that a dead man should come to life, because that has never been observed in any age or country. . . . And as a uniform experience amounts to a proof, there is here direct and foolproof, . . . against the existence of any miracle."16 Yet elsewhere Hume admits that such unprecedented events could be known to occur. A total darkness covering the earth for eight days, to use Hume's example,17 is surely an event fitting his above definition of a miracle.

Wykstra's Argument from Probability

Wykstra draws out of Hume's critique a much more credible presentation of the problem of probability." He points out that since we establish through experience those probabilities we assume for nature's order, the comparative probabilities of Jesus' resurrection to his "swooning" (a mistaken diagnosis of death by the Roman execution team) would establish the former as infinitely less likely to have occurred. ". . our experience concerning what happens to physical bodies following death is much more extensive-one might even venture 'infinitely more extensive'-than is our experience of what happens when Roman soldiers attempt to do their job."19 Wykstra is definitely correct if this is all that is involved.20 Indeed, we had earlier suggested some arbitrary degree of probability, say 1/101000, that a predicted physical resurrection could occur to some particular person. But as we further noted in that discussion, that improbability would decrease enormously if some superhuman intelligence and/or power were involved in the events considered. Since a theistic world view is not antecedently less likely than any other, a resurrection would in this case be quite possible and far more probable than any alternate explanation such as that of the "swoon" theory.21

It is in our modern application of natural law to miracles-such as in the definition of a miracle as a breach of natural law-that modern apologists have gone wrong.

A Third Argument from Probability

There is another problem the proponent of miracles must face concerning probability.22 The proponent will admit that a miracle is an unusual event. Thus it would be improbable that a miracle would occur at any particular time. This is not to say that it is improbable that a miracle will ever occur. Though it is improbable that one will pick the first time the one black marble in an urn filled with one thousand white marbles, it is not improbable that it will be picked eventually.

But if it is improbable that at any particular time a miracle will occur, then would this not count against any particular claim that a miracle has occurred? This question does give us reason to require stronger than normal testimonial evidence. An uncommon event should obviously be more attested than a normal event. (At this point we should recall that we have seen that we cannot assume, as Hume did, that miracles are infinitely or even extremely improbable.)

The event is admittedly uncommon, but not in principle more uncommon than, say, the prospect of being called up by a casual high school acquaintance after a separation of forty years. If the phone call does occur, we do not normally (nor should we) wonder whether the event did actually occur simply because it was extremely improbable that it should occur on that particular day. And if we should tell someone else that this ancient acquaintance has called, we would likely be doubted only if our veracity is for other reasons questionable.

Incompatible Miracle Claims of Conflicting Religions

This objection, classically stated by Hume," involves the claim that conflicting religions are supported by miracles which, as such, mutually destroy whatever evidential value they would otherwise have. The answer to this will be only summarily stated because of space.24 Generally, the following criteria question whether the conflicting miracles both equally evidence their proffered claims or are equally evidenced to have occurred. This objection fails unless there is equal evidence for conflicting miracles.

If a true conflict of miracle claims occurs, we should first compare the testimonial evidence to determine whether both miracles can justifiedly be claimed to have occurred. Secondly, we need to compare phenomenal characteristics. Is one miracle more unlikely to occur without superhuman power than the other? Third, is the propositional element in proper relationship to the phenomenon? A normal phenomenon must be adequately predicted, for example. Fourth, we have a test of more limited value. In some cases, one miracle may display a source of power greater than that of the other, as when the serpent of Aaron devoured the serpents of Pharoah's magicians (Exodus 7).

A fifth criterion for comparative analysis is a moral test. Any miracle that involves a moral evil or the advocation of evil in the proposition should be rejected in its claim. This must be our conclusion unless there is other very conclusive evidence-such as that of religious experience, I would claim-that the agent of the miracle and the claim has the right to cause or command this evil. If the ultimate agent of the miracle is evidenced to be evil, then he cannot be trusted to give us the truth. A greater than normal intelligence or power is evidenced, but nothing more.

Should all of these tests fail or be inapplicable for comparative analysis then Hume's critique would have force. This would not, however, exclude the possibility of adjudication by some other means such as philosophical or other arguments, or religious experience. With the exception of the fourth criterion, all of the other criteria should be considered for evaluating a miracle's evidential value even when not used for comparative purposes.

Supernatural Miracles

It is helpful to see how the relatively modern nature/supernature duality fits the concept of miracles as we have just considered it. The nature/supernature duality is a valuable hypothesis or model for explaining how science would be unable to examine the ultimate origin of a miracle. This would be important for theists who maintain that God cannot be found out or "examined" by human effort alone or that God should not be confined to the created universe. If by whatever means such theistic claims are verified, then a miracle would have to involve "an interference with Nature by supernatural power" as Lewis defines a miracle," or a similar model would have to be constructed. The following model presents a miracle not as a suspension or overriding of natural law but as an event which fits natural law.

Imagine nature as a complicated nexus of causes and effects (Figure 1). Like billiard balls, one factor "hits" another which hits another, which hits another, etc. A miracle involves the introduction of a causal factor from outside of nature into nature (Figure 2). By "outside of nature" I mean something analogous to the state of 11spaceland" being outside of "flatland" in Edwin Abbott's Classic Flatland. 16 Indeed, at present I see no need to deny that the analogy could be quite correspondent: supernature could be merely a fourth spatial dimension that we would have no possible way of entering or perceiving without substantial changes in our very nature, which in turn we could not effect on our own. As the flatlander cannot even conceive of spaceland, so we cannot conceive supernature without specially given vision or except by analogy. On the other hand, our three-dimensional world would be readily open to entrance by beings in the four-

dimensional world. "In calling them miracles, " says Lewis, "we do not mean that they are contradictions or outrages; we mean that, left to her own resources, she (nature) could never produce them."27

The most ideal science conceivable would be able to trace causal factors to their original sources within the system thus explicating all of the causal origins of all events. If science were thus to become omnicompetent, it could not in principle trace back the ultimate origin of a miracle. The onmicompetent scientist would trace back causal factors to a point beyond which he could not go. At that point at the border of nature, it would seem as though the last examined event were caused by apparently nothing. We would have effects with no apparent causes.

A miracle may have a long series of natural causes which accumulate in the non-normal phenomenon itself. Indeed, all claimed miracles may have been set into motion at the origin of the universe. If this is so, then an ideal science would trace such a miracle back through normal causal factors in nature until it has reached the origin of the universe. The biblical miracle of the parting of the Red Sea (Exodus 13) is one that apparently has a long series of intermediate natural causes. Scripture itself speaks of a strong wind causing the parting. The fact that natural causes such as these can be pointed out does not diminish the evidential value of a miracle, as we have seen. On the other hand, it may be that a particular miracle involves very few, if any, intermediate natural causes between it and the causal entrance into nature.

Flew: Miracles as a Roadblock to Scientific Inquiry

We have been considering supernatural miracles primarily because of their limitation on the power of science. So, of course, objections arise to this claim. Flew complains that it would be a roadblock to inquiry to claim that some things are or may be inexplicable scientifically.28 According to a supernatural model of miracles, the introduction of causal factors into the causal nexus of nature from outside of nature forms a path that even an ideal science could not trace backwards beyond the point of introduction. Flew should not be so close-minded as to assume that there cannot be admitted to be any area inscrutable to science. As Grace Jantzen points out, he should rather be willing to allow the scientist to investigate to determine whether this is the case and be open to evidence as it presents itself.29

We should follow science as far as it can go. If it comes to a dead end in any particular investigation, we should not presuppose either that science will eventually be able to investigate this matter further or that science cannot in principle investigate any further. We must simply wait and see what science with time can discover. In the meantime, we should keep an open mind to either possibility. Of course, a long term failure to get beyond such a dead end may dispose us to believe that there may be no way science can do so. However, it would be a mistake to claim that this alone would give us grounds for believing that we have a miracle evidencing God's activity (as Jantzen claims). This approach is incorrect, as Geisler would point out, because other world views can just as adequately adapt the phenomenon.30


It is in our modern application of natural law to miracles-such as in the definition of a miracle as a breach of natural law-that modern apologists have gone wrong. Accordingly, the dominant consensus in philosophy that miracles cannot evidence religious claims has been held with good justification. We can only begin to establish a defensible criterion for identification of miracles and to reverse this consensus once we come to understand the rationale with which mankind has for ages accepted "signs" as having evidential value. I believe that such a criterion has been presented and that miracles have been shown to definitely verify religious belief.


1I believe that religious experience can be defended as the Christian's grounds for belief. Indeed, this essay was originally written as an appendix to a Master's Thesis on "The Evidential Value of Religious Experience," (Trinity Evangelical Divinity School, 1979), pp. 108-123, Other than religious experience, I believe some theistic proofs as well as historical evidences have value.

2A very recent and exceptionally forceful critique has been given at this point of issue by Stephen J. Wykstra in "The Problem of Miracle in the Apologetic from History," Journal of the American Scientific Afflliliation 30 (December, 1978): 154-163.

3Antony Flew, "Miracles," Encyclopedia ofPhilosophy, 5:348.

41bid., p. 349.

5Cornelius Van Til, The Defense of the Faith, 3rd ed., rev. (Philadelphia: Presbyterian and Reformed Publishing Co., 1967), p. 240.

6Norman L. Geisler, Christian Apologetics (Grand Rapids, Mich.: Baker Book House, 1976), pp. 95-97.

7This conforms roughly to Aquinas' definition; "those things ... which are done by divine agency beyond the order commonly observed in nature." (Summa Contra Gentiles 3. 100.) In my definition, however, I am not assuming that divine agency is necessarily involved.

8Wykstra, "Problem of Miracle," p. 157.

9The criterion of fitness is one we have referred to when discussing the problem of incompatible miracle claims of conflicting religions in "Evidential Value," p. 121.

10In "Daniel's prophecy of the seventy weeks" (Daniel 9), it is predicted that the Messiah would come and be manifested at a particular time. The time stated by the writer of Daniel very exactly corresponds to the time of Christ. Some scholars claim to have determined to the day that the predicted date of the prophecy is the same day that Jesus entered Jerusalem earlier in the week before his crucifixion, viz., the date of the "triumphal entry." See e.g., Harold W. Hoelmer, Chronologicaf Aspects of the Life of Christ, (Grand Rapids, Mich.: Zondervan Publishing House, 1977), pp. 115-139.

11C. S. Lewis, Miracles (New York: Macmillan Co., Macmillan paperback. 1947), p. 109.

12City of God 21.8.

13This objection was pointed out by John Montgomery in his critique of Flew in Faith Founded on Fact (Nashville: Thomas Nelson, Inc., 1979) pp. 57-58.

14David Hume, An Inquiry Concerning Human Understanding, in On Human Nature and the Understanding, Anthony Flew, ed., (New York: Collier Books, 1962), ch. 10 pp. 115-136.

15Lewis, Miracles, ch. 13, pp. 103-111. Flew admits the force of Lewis' argument at the point to be questioned (Hume's Philosophy ofBelief, p. 205).

16Inquiry, pp. 119, 120.

17Ibid., p. 133.

18Wykstra, "Problem of Miracle," p. 160.


20Wykstra misreads Lewis as contesting this statement (Wykstra, p. 160). Wykstra seems to give the impression that Lewis' basic contention in chapter thirteen involves a mistaken distinction between "antecedent" and "historical" probabilities. Whether or not there is such a distinction is virtually insignificant to the argument of this chapter. Indeed, Wykstra admits to Lewis' conclusion that for some events, both past and future (Lewis, p. 108), normal probability determination procedures would not be applicable (Wykstra, p. 161). Wykstra rather maintains that the more important question yet remains unanswered: "How can we know when we face such an event?" It is this question we have been concerned to answer in this paper.

21Our comparison of the "swoon" and resurrection hypotheses may suggest the problem of the arbitrariness of miracle choices. If the improbability of a resurrection is decreased by the possibility of intelligent intervention, might not the "swoon" theory be established as even more probable than a resurrection on the possibility that intelligent superhuman intervention is involved with it? Here I must refer directly to the section of this paper dealing with the arbitrariness of miracle choices. Clearly the answer must be no.

22This argument is implicit in Hume's critique. See, for example, Norman Geisler's discussion in Apologetics, pp. 266, 267.

23Inquiry, p. 125.

24See Jensen, "Evidential Value," pp. 120-123 for a more detailed explication of this answer.

25Lewis, Miracles, p. 10.

26Flatland (New York: Dover Publications, Inc., 1952). See the writings of Karl Heim for a more sophisticated development of the flatland theme. For an interesting discussion on this theme, see "UFOs: Is Science Fiction Coming True?" by Mark Albrecht and Brooks Alexander, SCP Journal 1 (August 1977): 21 particularly. Published by the Spiritual Counterfeits Project, P.O. Box 4308, Berkeley, Calif. 94704.

27Miracles, p. 63.

Hume's Philosophy of Belief, pp.
169, 193.

29Grace Jantzen, "Hume on Miracles, History and Politics," Christian Scholars Review 8 (1979): 324.

30Not only can pantheistic systems adapt it but also naturalistic systems. It may be claimed that this is a true anomaly of nature which in principle cannot be reduced to natural law. Or, it may be claimed, the phenomenon fits a law of nature which science will never be able to discover because of human limitation. Why should any of these alternate world view explanations be considered less likely than the theistic view?