Miracles and Natural Explanations
In a discussion of miracles in this journal, Robert Larmer attacks the contention that "it is always more rational to believe that an event could be explained naturalistically, if only we had the requisite scientific knowledge, than to believe that a miracle has occurred." It is perfectly justifiable, he acknowledges, to seek initially for a natural explanation for a physical event. And it is logically possible that some revision of scientific law might enable us some day to offer a natural explanation for any given event which cannot presently be explained naturally. But it is also logically possible, he continues, that no revision would enable us to offer a natural explanation for some such events. Accordingly, "one's decision concerning which of these alternatives is most probably true must be based on an assessment of evidence." And there are conceivable situations, he argues, in which the evidence would not make it most reasonable to hold out for a natural explanation. The example he offers is the following:
Suppose one hears of a man who claims to perform miracles of healing through the power of God. Upon investigating, one learns that this person has not only an exemplary character, but also an apparent ability to perform remarkable cures. One is able to capture on film occasions when, immediately following the prayers of this man, fingers lost to leprosy were regrown to their original form and length in a matter of seconds, and occasions when eyes severely burned by acid were immediately restored to sight. One finds further not only that this man appears to have the power to heal any kind of disease or injury, but also that no interposition of lead screens or strong electromagnetic fields or the like has any effect on his apparent ability to heal. Indeed, it is observed that his power is apparently independent of distance, since people in distant countries have experienced dramatic healing after this man prayed for their cure.
Why would it be so hard in this case "to postulate a natural explanation rather than the occurrence of a miracle?" For one thing, Larmer tells us, it would force us to adopt a position of radical skepticism concerning the claims of science. "If one has good reason to accept the fact that such extraordinary events have occurred, and if one insists that all explanation of physical events must be natural, then one must be prepared to reject the laws which led one to expect different results." But the defender of miracles, we are told, does not have this problem. He or she can acknowledge the occurrence of the event and "yet retain [his or her] faith in what seem to be basic, well-evidenced, and accurate statements of natural law." Accordingly, he continues, it is difficult to see how a religious interpretation could be considered any less reasonable in this case. And this fact in turn leads Larmer to conclude that "surely only a somewhat dogmatic and uncritical metaphysical assumption that nature is in fact an isolated system can explain the insistence of some thinkers that, no matter what the event and no matter what the context in which it occurs, it is always more rational to live in the faith that such an event has a natural explanation rather than believe it to be a miracle."
Larmer's argument, it seems to me, contains two basic confusions. First, is it true that if someone grants the occurrence of the type of extraordinary event Larmer cites and yet still wants to hold out for natural causation, she "must be prepared to reject or revise the laws which led (her) to expect different results"? I think not. It is true that scientists must assume as a working hypothesis that current natural laws express what will always happen under certain conditions. And it is also the case that any apparent counterinstance to a current law necessitates a reevaluation of the 'law' in question. But, as philosophers like Richard Swinburne have repeatedly pointed out, natural laws are not merely summary statements of what does in fact happen. Nor do they describe in some a priori fashion what in fact can or cannot happen. Natural laws, rather, only describe events in so far as they take place in a predictable and regular manner. Accordingly, there is no reason to grant that, when faced with a seeming counterinstance, we are under some mandate to either admit that the event occurred as reported and abandon the relevant law or affirm the relevant law and deny that the event actually occurred. If a seeming counterinstance proves to be repeatable, then the current laws in question certainly must be abandoned (or modified). But, otherwise, we can, in principle, affirm both the counterinstance and the law in question.
To do so will not, as Larmer implies, render the laws in question non-functional or turn science into a quasi-metaphysical guessing game. For as long as the recalcitrant event is not repeatable, it furnishes us with no basis upon which to build an alternative to the current natural laws in question. Such laws can remain (indeed must remain) the working hypotheses which the scientist utilizes. But what of Larmer's more important claim: that to hold that it is always most rational to deny supernatural causal intervention is a dogmatic, uncritical, question begging assumption. He makes this claim, remember, because he believes that there are conceivable situations in which the 'evidence' makes belief in supernatural causation most reasonable. But is this so? Are there such conceivable cases?
There may well be situations which, when considered in isolation, make divine intervention a very plausible causal hypothesis. But, when assessing a causal hypothesis, one must consider all the relevant data. Applied to Larmer's example, this means that we must not only determine, as Larmer implies, whether divine causation is the most plausible explanation for the 'healing' alone. We must also determine whether the existence of an interventive God is compatible with all other experiential data. We must, for example, consider whether the existence of an interventive God is compatible with all the evil we experience.
Now let us assume that in comparing the plausibility of affirming that God has healed the individuals in question (and, thus, that God's existence is compatible with the amount the types of evil in the world) with the plausibility of affirming that God's existence is not compatible with such evils (and, thus, that God did not heal the individual in question) someone decides that God's nonexistence is more plausible overall. In other words, let us assume that the prima facie evidence for God's existence which has been generated by the 'healing' in question is not of sufficient weight in the mind of a given individual to make the theistic perspective the most plausible overall. Could Larmer justifiably accuse such an individual of assuming an uncritical, dogmatic, question begging stance?
I believe not. To substantiate such a charge, he would need to stipulate some set of objective criteria initially accepted by both the theist and nontheist in relation to which it could be shown that, given the 'healings' in question, nontheism could no longer be considered an acceptable world-view for any rational person. But to what mutually acceptable set of criteria could he appeal? What set of criteria would allow him to demonstrate objectively that the evidence for God's existence generated by the remarkable healings must be seen by all rational people as outweighing, for example, the seeming evidence against God's existence generated by the amount and types of evil we experience?
I am not convinced that such criteria exist. Both theists and nontheists normally agree that the affirmation of an inconsistent set of beliefs is irrational. But there is no agreement on criteria for 'weighting' various experimental perspectives. Some nontheists contend that the 'evidence' against God's existence — e.g., evil — is such that no amount of conceivable additional 'evidence' for God's existence could ever counterbalance it. Other nontheists are more tentative at this point. Some theists contend that their 'personal experience' with God is so self-authenticating that no amount of conceivable 'evidence' against God's existence could 'outweigh' it. Other theists are more cautious in their assessment of the situation. Each is willing to consider the evidence; each considers his or her perspective to be most rational. I, personally, can think of no objective, non-question-begging basis for resolving this metaphysical stalemate. I must agree with Alvin Plantinga that the probability with respect to belief or disbelief in God (which is necessary for belief in divine causation) is ultimately relative to one's noetic structure.
Accordingly, it seems to me that there is little basis upon which to claim that all proponents of solely natural causation are guilty of dogmatic, uncritical, question-begging reasoning. To claim emphatically that there is in fact no God (and thus no divine causal intervention) may be an unwarranted metaphysical contention. But the nontheist need not be making any such ontological claim. She can simply be saying that, while this epistemological contention is debatable, its affirmation is not necessarily any more dogmatic or question begging than the belief that the 'total' evidence makes theistic belief (and thus the possibility of divine intervention) most reasonable.