Chapter 1: The Consistency of the Concept
This thesis was originally written by Keith Parsons in partial fulfillment of requirements for the degree Master of Arts in the College of Arts and Sciences at Georgia State University, 1982.
Chapter 1: The Consistency of the Concept
The most potent sort of objection that could be made against the concept of the miraculous would be one that showed the concept to be self-contradictory. If it could be shown that the concept of the miraculous contradicts itself, the apologist could no more invoke a miracle-claim than he could produce a round-square as evidence for his case. In the recent literature on the topic of miracles the best-known statement of such an objection is found in Alastair McKinnon’s article “‘Miracle’ and ‘Paradox.'” McKinnon’s argument will be examined in this chapter, as well as attempts by two theistic philosophers to show that his criticisms fail. It will be concluded that the theistic rejoinders to McKinnon do not succeed, but that McKinnon’s argument is inadequate for other reasons. According to McKinnon, the belief that a natural law can be violated rests upon an erroneous analogy to civil laws. Civil codes compel or prohibit certain behavior and clearly can be broken without being annulled. Natural laws on the other hand, are neither prescriptive nor proscriptive but rather descriptive:
Natural law is not, as has been widely supposed, a kind of code for nature having legislative and, perhaps particularly, prohibitive force. This is an outdated, untenable, and completely unscientific view. Natural laws bear no similarities to civil codes and they do not in any way constrain the course of nature. They exert no opposition or resistance to anything, not even to the odd or exceptional. They are simply highly generalized shorthand descriptions of how things do in fact happen.
If “natural law” is thus defined as “a description of whatever happens” (and McKinnon claims that this definition accords with standard scientific usage) it follows that there is a contradiction in describing any event as a violation of natural law. A law of nature, according to this definition, is merely a description of the actual course of events. To say that such a law was violated would be to say that an event occurred other than what was in fact the actual course of events:
This contradiction may stand out more clearly if for “natural law’, we substitute the expression “the actual course of events.” “Miracle” would then be defined as “an event involving the suspension of the actual course of events.” And someone who insisted upon describing an event as a miracle would be in the rather odd position of claiming that its occurrence was contrary to the actual course of events.
Since no event can be a violation of natural law, any phenomenon that is seen to conflict with a nomological proposition merely shows that the proposition did not express a true law. It then will become necessary to modify our understanding of the laws of nature so that we can account for that event. Thus, if a person is confronted with an ostensible miracle he has a simple choice: “He can affirm the reality of this event and repudiate the ‘laws’ which it violates. Or he can affirm these ‘laws’ and repudiate the event. What he cannot do is to affirm both the event and the ‘laws’ of which it is a violation.”
If, for example, we were to see a man walking upon water McKinnon would have us decide on one of two alternatives: either we should disbelieve our eyes or admit that what we had previously conceived to be the laws of hydrodynamics were not in fact the true laws. We cannot say that this man’ walking on water constitutes a violation of natural law, for the fact of its occurrence is a counter-instance that falsifies any ideas we might have entertained about the impossibility of walking on water. The upshot is that, according to McKinnon, the would-be claimant to a miracle faces a dilemma. Whatever he puts forward as a violation of natural law only serves to falsify the very law it was intended to violate.
The most common criticism of McKinnon is that his claim to find a contradiction in the conception of a violation of natural law is based on a misunderstanding of what a law of nature is and how it might be falsified. Richard Swinburne contends that, as scientists actually employ the term, “law of nature”, does not (pace McKinnon) mean simply a description per se, but a description of a certain type. Philosophers of science have devoted considerable effort to articulating the criteria for identifying those types of descriptions that are “lawlike.” Although there is no universal agreement about these criteria or their application, Swinburne contends that there is a reasonably broad consensus about some of the necessary conditions which a statement must meet if it is to be considered a law of nature. By outlining some of those conditions and showing how McKinnon’s definition of “law of nature” fails to satisfy them, Swinburne hopes to show that it is not incoherent to speak of a violation of natural law.
First of all, a law of nature is not a mere description of observational results, but is a generalization or formula extrapolated from those results. Secondly , from this generalization or formula it should be possible to deduce all of the past observational results as well as to make predictions about those that will be observed in the future. Finally, since for any given set of data an infinite number of hypotheses can be generated from which that data can be deduced, scientists attempt to provide ( ceteris paribus) the simplest or most economical account. A description that does not meet these criteria cannot be a law of nature even if it is a description of what has actually happened.
Swinburne concludes that “The upshot of all this is that — against McKinnon — laws of nature do not just describe what happens (‘the actual course of events’). They describe what happens in a regular and predictable way.” Only descriptions of what occurs in regular and predictable ways will begin to be able to meet the criteria of lawlikeness. This means that “When what happens is entirely irregular and unpredictable, its occurrence is not something describable by natural laws.”
This last sentence provides the key to understanding what a violation of natural law would be. A unique, non repeatable event will not fit into the category of things describable in terms of natural law. As such, Swinburne will argue, such an event can violate without falsifying a law of nature. However, before proceeding to an examination of this argument the terms “violation” and “falsification’, need to be clarified. When reduced to its simplest terms, a law of nature is a proposition of the form “All As are Bs.” As Iranian Smart points out, all propositions of this form can be contradicted by a single counter-instance:
Now the simplest variety of propositions of the form “All As are Bs” are those like “All tigers are quadrupeds” or “All humans are rational animals” or “All swans are white.” Now these have a simple logical feature, namely, that they are contradicted by propositions of the form “At least one A is not a B” — e.g., “At least one tiger is not a quadruped,” “At least one human is not a rational animal,” or “At least one swan is not white.” Clearly, if these latter are true, the corresponding propositions of the form “All As are Bs” are false, and conversely. It follows from this that in order to disprove a law of nature, all you have to do is to produce an exception, a negative instance. Then the supposed law of nature turns out not to be a law of nature.
Any sort of counter-instance, even one caused by the direct activity of a supernatural agent, will falsify a proposition of the form “All As are Bs.” This is simply a truth of logic. However, it will serve our purposes here to make a distinction between two types of counter-instances and to refer to only one of these as a “falsification” and to the other as a “violation” of a law of nature. This distinction can be best illustrated with an example: Suppose that we believe it to be a true law of nature that all swans are white and then we come across a black swan. Then it will no longer be true that all swans are white. We might account for the existence of the black swan by discovering a new law of nature of the form “all swans are white except in certain circumstances and then they are black.” On the other hand, we might come to the conclusion that there is no law of nature that can account for the existence of the black swan and that its existence is due to the agency of a supernatural being. The distinction to be made, therefore, is between a counter-instance to a law L that can be subsumed under another law L’ (where L’ is logically inconsistent with L), and a counter-instance to L that cannot be subsumed under a law of nature. In the remainder of this thesis only the former sort of counter-instance will be meant by the term “falsification” whereas the latter sort will be called a “violation.”* The importance of making this distinction is that, as the above example illustrates, a miracle could be what we have called a violation but could not be a falsification (as defined above). Swinburne’s task is therefore to define “miracle,’ in such a way that an instance of it can be distinguished from those counter-instances that are falsifications of natural law.
* Note: When it is unknown or irrelevant whether a counter-instance is a violation or a falsification (as here defined), that counter-instance will be referred to as an event “contrary to” a law of nature or a nomological proposition.
Swinburne concedes that repeatable counter-instances serve to falsify laws of nature. However, he argues that a non-repeatable counter-instance will not constitute a falsification:
. . . what are we to say if we have good reason to believe that an event E has occurred contrary to predictions of a formula L which otherwise we have good reason to believe to be a law of nature, and we have good reason to believe that events similar to E would not occur in circumstances as similar as we like in any respect to those of the occurrence of E? E would then be a non repeatable counter-instance to L . . . The evidence shows that we cannot replace L by a more successful law allowing us to predict E as well as other phenomena supporting L. For any modified formula which allowed us to predict E would allow us to predict similar events in similar circumstances and hence, ex hypothesis, we have good reason to believe, would give false predictions. Whereas if we leave the formula L unmodified, it will, we have good reason to believe, give correct predictions in all other conceivable circumstances. Hence if we are to say that any law of nature is operative in the field in question we must say that it is L.
Hence, if part of our definition of “miracle” is “a non repeatable counter-instance to a law of nature” we can coherently speak of a miracle as violating and not falsifying a natural law.
Ninian Smart arrives at the same conclusion through a somewhat different argument. He notes that scientists do not abandon a well-established law of nature when confronted with only a single counter-instance, but rather demand repeatable evidence of its failure:
It is not to be thought that the scientist does an experiment, finds a result contrary to all his previous experience, and then good-naturedly shrugs his shoulders, sighing “And there goes another scientific law.” Rather he concludes that he must have made a mistake in his experiment.. But if he and his colleagues keep on finding the same thing, then there is nothing for it but to reject the law. Or at least tinker with it in some way so that it is made to correspond with the new finding. Thus the law is scrapped, or, as we might say, changed; but it must be remembered that even if it is only changed, then strictly speaking it is, in its original form scrapped, but only, as I say, with reluctance.
Only repeatable or experimentally reproducible negative instances are capable of overturning accepted laws of nature. As Smart puts it, a falsifying negative instance must be a “small-scale” law of nature that overturns or limits a “large scale” law of nature. He means that an accepted law of nature (a large-scale law) is falsified when the set of regularities which it predicts are replaced by another set of regularities (the small-scale law) whenever certain special circumstances exist. For example, the (large-scale) law of gravitation would have to be modified if it were found that under certain conditions balls rolled down an inclined plane faster than had been predicted by the law of gravitation. Smart concludes that it is now possible to say how miracles could violate the laws of nature without falsifying them:
The relevance of all this to miracles is readily apparent. Miracles are not experimental, repeatable, they are particular, peculiar events occurring in idiosyncratic human situations. They are not small-scale laws. Consequently, they do not destroy large-scale laws. Normally, they may seem to destroy the “Always” statements of the scientific laws, but they have not the genuine deadly power of the [falsifying] negative instance.
If Swinburne and Smart are right, we can coherently speak of miracles as non-repeatable counter-instances that violate but do not falsify a law of nature. As George Landrum points out, however, there does not seem to be any reason why a miracle could not be a repeatable, even a predictable event. In fact, many of the miracles spoken of in the Bible and church doctrine have a repeatable character: “When the Israelites were fed in the wilderness for forty years by manna, the food they ate was of unnatural origin, even though it appeared with more regularity than any natural crop ever did. (It was the irregularity of natural food that was the reason God provided manna.) Transubstantiation is supposed to take place regularly, but it is supposed to be nonetheless miraculous.” We can even imagine that miracles could occur with lawlike regularity:
Let us suppose that there is a holy book of great antiquity containing some quite unambiguous statements about God and that whenever someone proclaims his disbelief in one of these statements an angel appears to him within a week and says “Grave misfortune will befall you on account of your heresy.” Then, within a week some catastrophe always happens to the heretic. Sometimes he is murdered, sometimes he is struck by lightning, sometimes he commits suicide, sometimes he is paralyzed, sometimes he is merely beset by boils.
Despite the lawlike nature of the events imagined in this example they would still be miracles because they involve the workings of supernatural agents in bringing about occurrences that would not have happened in the natural order of things. It might be objected against Landrum that we have no way of distinguishing his miraculous regularities from those that are unusual or extraordinary but entirely natural. Landrum offers a reply to this charge, but since this criticism raises an entirely different issue — I.e., the issue of the identification of miracles (and not merely the coherence of the concept) — it will be dealt with in a later chapter.
It appears that non-repeatability cannot be made a part of the definition of a miracle. And since, for Swinburne and Smart, a miracle is necessarily non-repeatable, their attempt to show that “miracle” is a coherent concept fails. However, this failure does not mean that McKinnon is right.
McKinnon and his critics all have failed to make a very crucial distinction: So far, the laws of nature have been identified with certain formulae, descriptions, and nomological propositions. However, the term “law of nature” can also mean a regularity that exists in nature independently of our formulae, descriptions, and propositions. The meaning of “miracle” that corresponds with such an understanding of “law of nature” would therefore be “an alteration or suspension of the usual regularities of nature.”
By thinking of the laws of nature exclusively as the descriptions employed by scientists, McKinnon was led to conclude that miracles do not occur because describing an event as a miracle would conflict with an assumption that is required by scientific methodology:
. . . the majority of contemporary scientists are convinced that this concept (natural law), if it is to be allowed at all, is and must be universal in its application. They assume that all events are lawlike (whatever that really means) or, at least, that they must be treated as such. They assume that every event can be shown to be an instance of some generalization, whether simple or statistical. This is why the scientist holds that there are no suspensions of natural law.
Yet even if this assumption is necessary for science, as it probably is, it cannot legislate reality and so cannot preclude the possibility that a miracle (in the sense of a suspension or alteration of one of the regularities of nature) might occur. In other words, no commitment on the part of scientists to seek naturalistic explanations for all events can rule out the possibility that God exists and might occasionally cause things to happen which otherwise would never occur. If such an event were to take place, perhaps the scientist, qua scientist; could never describe it as a miracle; but this does not mean that such an event could not take place. Furthermore, to assume that the “actual course of events” will include only such phenomena as have natural causes, is to beg the question of whether or not miracles occur. Hence, there seems to be no way that scientific methodology can rule out, a priori, the possibility of miracles. If, then, we think of the laws of nature as the actual regularities that exist in nature, and not as the formulae, descriptions, propositions etc. that express human knowledge of those regularities, there seems to be no reason why God could not change these regularities as He pleases. It may be that we can never know that a miracle has taken place, but this is a different problem from whether the concept of the miracle is consistent. And the fact that we have no reason to think that God could not suspend or change the regularities that exist in nature shows that that conception is consistent.
 Alastair McKinnon, “‘Miracle’ and ‘Paradox,'” American Philosphical Quarterly, 4, No. 4 (October, 1967).
 R. G. Swinburne, The Concept of Miracle, (London: Macmillan and Co. Ltd., 1970), pp. 23-26.
 Ninian Smart, Philosophers and Religious Truth, (New York: The Macmillan Co., 1964), pp. 24-25.
 Swinburne, The Concept of Miracle, p. 27.
 Ninian Smart, Philosophers and Religious Truth, pp. 28-29.
 George Landrum, “What a Miracle Really Is,” Religious Studies, (March, 1976), 52.
 McKinnon, “‘Miracle’ and ‘Paradox,'” p. 309.
“The Conception of the Miraculous and Christian Apologetics” is copyright © 1982, 1997 by Keith Parsons. All rights reserved.
The electronic version is copyright © 1997 by Internet Infidels with the written permission of Keith Parsons. All Rights reserved.