Miracles

views updated

MIRACLES

The term "miracle," like the word nice, is often used to refer primarily to the responses of the user. In this usage, a miracle is merely some event that astounds the speaker, with perhaps some presumption that others will or should react to it in the same way; just as in the parallel case nice means simply "agreeable to me," with perhaps again some suggestion that all right-minded people will feel the same. But the senses of "miracle" that are of philosophical and methodological interest are stronger and less subjectively oriented. Although they include the idea that wonder is called for as at least part of the appropriate response, the crux as well as the ground for the wonder is that a miracle should consist in an overriding of the order of nature. A miracle is something that would never have happened had nature, as it were, been left to its own devices.

This idea of overriding is essential; however, it is certainly subject to various variations and additions. Some writers, for instance, insist that the word miracle should be used in such a way that it becomes necessarily true that a miracle can be worked only by God or by his specially deputed agents. Others even build into their very definition of miracle some reference to the purposes for which Authority is supposed to be prepared to consider making such an exception. Certainly, most theist theologians are also at great pains to maintain that a miraculous event could not properly be considered a violation, since it would not really represent any infringement, of the fundamental hierarchical order. "It is not against the principle of craftsmanship (contra rationem artificii ) if a craftsman effects a change in his product, even after he has given it its first form" (Thomas Aquinas, Summa contra Gentiles, III, 100). But these very labors to show that and how such "violations" need involve no ultimate irregularity still admit and presuppose the essentially overriding character of the miraculous. There would be no point in trying to show in this way that a miracle must ultimately be no violation of regularity unless it were taken for granted that it apparently is such a violation.

This point is fundamental, and it needs to be stressed more heavily today than in the past. For in addition to the traditional theist reluctance to ascribe to the Deity anything savoring of unseemly irregularity, it is nowadays usual to encounter a certain shyness about any apparent repudiation of scientifically accepted modes of explanation. Thomas Aquinas, earlier in the chapter referred to above, gave a perfectly clear and unequivocal definition of miracle that makes no bones at all about the crux of the matter, namely, that "those things are properly called miracles which are done by divine agency beyond the order commonly observed in nature (praeter ordinem communiter observatum in rebus )." Again, in the twentieth century, Dr. Eric Mascall, remaining in the same forthright tradition, insisted in his article in Chambers' Encyclopaedia that the word miracle "signifies in Christian theology a striking interposition of divine power by which the operations of the ordinary course of nature are overruled, suspended, or modified."

Miracles and Natural Order

To seize the fundamental point that a miracle is an event that violates the "ordinary course of nature" is to appreciate that the notion of a miracle is logically parasitical on the idea of an order to which such an event must constitute some sort of exception. This being so, a strong notion of the truly miraculous—a notion involving something more than the notions of the merely marvelous, the significant, or the surprising—can only be generated if there is first an equally strong conception of a natural order. The inevitable tension between the ideas of rule and of exception thus gives concepts of the miraculous an inherent instability. It is perhaps relevant to notice how this tension has been felt in the history of ideas. Where there is as yet no strong conception of a natural order, there is little room for the idea of a genuinely miraculous event as distinct from the phenomenon of a prodigy, of a wonder, or of a divine sign. But once such a conception of a natural order has taken really firm root, there is a great reluctance to allow that miracles have in fact occurred or even to admit as legitimate a concept of the miraculous.

An interesting early case of this is provided by Benedict de Spinoza in his Tractatus Theologico-Politicus, in which he tried to reconcile his vision of a natural order (Deus sive natura ) with an acceptance of the Bible as in some sense a privileged document. He did this partly by admitting the limitations of observatory powers of the men of biblical days, but mainly by urging that conventional interpreters of the Bible read far more miracles into it than it contains, because they constantly read poetic Hebrew idioms literally. Today, more and more theologians seem to be noticing the exact words used by the New Testament writers in describing the sorts of alleged events that, in more scientific ages, have been characterized (and perhaps dismissed) as miraculous. These words are τερατὰ ("wonders," or "prodigies"), δυναμει̑ς ("powers"), σημεία ("signs"); and, particularly in St. Paul, χαρισματὰ ἰαμάτων ("graces of healing") and ἐνεργήματα δυνάμεων ("effects of powers"). None of these words seems to carry any entailments about the overriding of a natural order. On the other hand, once a really strong conception of natural order has arisen, its adherents tend to dismiss out of hand all stories of putative occurrences in the belief that if they allowed that these occurrences had taken place at all, they would have to admit them to have been miraculous. One may refer here to R. M. Grant's recent Miracle and Natural Law in Graeco-Roman and Early Christian Thought (Amsterdam, 1952) and to William E. H. Lecky's classic study History of the Rise and Influence of Rationalism in Europe (London, 1890). The former summarizes its own thesis as follows: "Credulity in antiquity varied inversely with the health of science and directly with the vigor of religion" (p. 41). This, however, was later qualified by the important observation that "at least in some respects Christians were far less credulous than their contemporaries, at least in the period before Augustine" (p. 120). Lecky traced a development in which stories of the ostensibly miraculous, from being accepted as a chief guarantee of the authenticity of the Christian revelation, become instead "a scandal, a stumbling block, and a difficulty" (Vol. I, p. 143). In the nineteenth century the radical biblical critic David Strauss announced in the introduction to his Das Leben Jesu (2 vols., Tübingen, 1835; translated by Mary Ann Evans as Life of Jesus Critically Examined, London, 1848), "We may summarily reject all miracles, prophecies, narratives of angels and demons, and the like, as simply impossible and irreconcilable with the known and universal laws which govern the course of events." And in the twentieth century there was even a bishop of the Church of England capable of saying of the author of Mark, "He was credulous inasmuch as the miracles, as they are narrated, cannot, in the light of our modern knowledge of the uniformity of nature, be accepted as historical facts" (F. W. Barnes, The Rise of Christianity, London and New York, 1947, p. 108).

dilemma of holding strong rules while admitting exceptions

The spokesman for the occurrence of the miraculous faces a dilemma that arises from the very essence of the concept he espouses. It is tempting, but wrong, for the believer in the miraculous to think that he can afford to gloat over any little local difficulties and embarrassments that may from time to time beset the forward march of science. But insofar as a miracle involves an alleged overriding of a law of nature, he too is committed to showing the subsistence of a natural order. Exceptions are logically dependent upon rules. Only insofar as it can be shown that there is an order does it begin to be possible to show that the order is occasionally overridden. The difficulty (perhaps an insoluble one) is to maintain simultaneously both the strong rules and the genuine exceptions to them. The oscillations in the history of thought are to be understood by reference to this tension (amounting perhaps to a contradiction) that is inherent in the concept of the miraculous, and it is on this same tension that the various logical and methodological problems also center.

Logical and Methodological Problems

It is with logical and methodological problems that we are primarily concerned. The classical, and by far the best, approach is by way of the notorious section X, "Of Miracles," in David Hume's Enquiry concerning Human Understanding (1748). This and Section XI of this Enquiry, both of which were parts of a single coordinated case, constitute Hume's answer to what was, in his day, the stock program of Christian apologetic. This program had two stages: the first was an attempt to establish the existence and certain minimal characteristics of God by appealing only to natural reason and experience, the second was an attempt to supplement this rather sketchy religion of nature with a more abundant revelation. This program, in its characteristically eighteenth-century form, received its archetypal fulfillment in Archdeacon William Paley's Natural Theology (London, 1802) and also in his Evidences of Christianity (London, 1794). In the eighteenth-century form, the weight of the first part of the case was borne primarily by the Argument to Design. If from a watch we may infer a watchmaker, then the orderliness of the universe entitles us to infer, by parity of reasoning, a Maker of the universe. The second part of the case rested on the claim that there is ample historical evidence to show that the biblical miracles, including the crucial physical resurrection of Jesus bar Joseph, did in fact occur, and that this in turn proved the authenticity of the Christian revelation.

Paley's style of systematic rational apologetic has no doubt gone out of fashion, at least among Protestants. But Hume's challenges to the whole idea of a substantial natural theology and to the project of establishing the authenticity of any alleged revelation by proving that its claims have been supported by miracles are not, and are not likely to become, dead issues. For in 1870 the third session of the First Vatican Council defined as constitutive dogmas of the Roman Catholic religion both of the positions that Hume had challenged. The relevant passage of the canon dealing with the second reads, "If anyone shall say … that miracles can never be known for certain, or that the divine origin of the Christian religion cannot properly be proved by them: let him be cast out" (si quis dixerit … aut miracula certo cognosci numquam posse nec iis divinam religionis christianae originem rite probari: anathema sit ; H. Denzinger, ed., Enchiridion Symbolorum, 29th ed., Sec. 1813, Freiburg im Breisgau, 1953).

problem of supernatural revelation

Hume's main contention was thus, in his own words, that "a miracle can never be proved so as to be the foundation of a system of religion." For him, all other questions about the miraculous were, officially at least, merely incidental to this basic tenet. He defined a "miracle" as "a transgression of a law of nature by a particular volition of the Deity, or by the interposition of some invisible agent." This definition has been attacked on various counts, but the criticism is misconceived, for two reasons. First, this was in fact the way in which the opponents whom Hume had in mind defined the term "miracle." Thus, Dr. Samuel Clarke, in his famous Boyle lectures (The Works of Samuel Clarke, Vol. II, London, 1738, p. 701), had defined "miracle" as "a work effected in a manner … different from the common and regular method of providence, by the interposition either of God himself, or of some intelligent agent superior to men." Second, if, as Clarke and the orthodox tradition would have it, the occurrence of a miracle is to serve "for the proof or evidence of some particular doctrine, or in attestation of the authority of some particular person," then surely a miracle must be conceived in this way. It is only and precisely insofar as it is something really transcendent—something, so to speak, that nature by herself could not contrive—that such an occurrence could force us to conclude that some supernatural power is being revealed.

In this context it would be worse than useless to appeal to revelation for criteria by which genuinely miraculous events may be identified, and thus distinguished from the unusual, the untoward, or the merely ordinary. For if the occurrence of a miracle is to serve as the endorsement of a revelation, then we have to find some means entirely independent of that revelation by which the endorsement itself may be recognized. Exactly the same point applies, of course, if, with what is now a rather fashionable school of apologetic, it is urged that miracles are not essentially overridings, but signs. If a sign is to signify to the unbeliever, then there must be some means independent of the doctrinal system itself by which the signs may be identified and read. As has been suggested already, there is much to be said for trying to interpret the records of τερατὰ and σημεία in the New Testament in terms of some notion of sign, rather than as miracle stories proper. But it is necessary to insist on two facts that seem to be often overlooked—namely, that part of the price that must be paid for this method of interpretation is the sacrifice of the use of these stories as independent evidence of the genuinely revelatory character of the doctrines; and that such a sacrifice presumably entails the rejection of at least one defined dogma of the Roman Catholic Church, and hence of the truth of Roman Catholicism as a theological system.

A similar but different point applies if a relativistic definition of "miracle" is adopted, as was done, for instance, by John Locke. In his Discourse of Miracles (written 1702, published posthumously), he defined the word miracle as "a sensible operation, which, being above the comprehension of the spectator, and in his opinion contrary to the established course of nature, is taken by him to be divine." It was also done, in a slightly different way, by St. Augustine, who insisted that "nature is the will of God" (Dei voluntas rerum natura est ), and hence that "a portent is not contrary to nature, but contrary to our knowledge of nature" (Portentum ergo fit non contra naturam, sed contra quam est nota natura; De Civitate Dei, XXI, 8). To operate with a relativistic notion of this sort is necessarily to be deprived of the possibility of arguing that a miracle is a miracle regardless of whatever anyone may happen to know or to believe about it, and hence to rob the attempt to base an apologetic on the occurrence of miracles of whatever initial plausibility it might otherwise possess. For the occurrence of events that are merely inexplicable to us, and at present, provides no good ground at all for believing that doctrines associated with these occurrences embody an authentic revelation of the transcendent. There is, of course, no particular reason why Locke himself should have been disturbed about this. The case of Augustine, however, is more interesting, for he is a recognized saint and one of the four great doctors of the church. And yet insofar as he held to a relativistic notion of a miracle, he was safeguarding the vital doctrine of the total dependence of the whole creation—but at the price of subverting a sort of apologetic which it has since become essential for Roman Catholics to believe in as a possibility.

problem of identifying an event as miraculous

Up to this point it has been insisted that if the occurrence of a miracle is to serve—as Clarke and the orthodox tradition would have it—"for the proof or evidence of some particular doctrine, or in attestation of the authority of some particular person," then in a traditional sense, miracles must be conceived of as involving the overriding of some natural order that is at least partly autonomous. The importance of this crucial point is often overlooked. Another immediately consequential point, however, is overlooked perhaps even more often, namely, that if an occurrence that is miraculous in the traditional sense is to serve as evidence for anything, it must be possible to identify it as being miraculous. Furthermore, as was urged above, if its occurrence is to serve as an endorsement of some doctrinal system, the method of identification must be logically independent of that system. The difficulty of meeting this last requirement is often concealed by the acceptance of what seems, for many people, to be an almost unquestionable assumption. Protagonists of the supernatural, and opponents too, take it for granted that we all possess some natural (as opposed to revealed) way of knowing that and where the unassisted potentialities of nature (as opposed to a postulated supernature) are more restricted than the potentialities that, in fact, we find to be realized or realizable in the universe around us.

This is a very old and apparently very easy and tempting assumption. It can be found, for instance, in Cicero's De Natura Deorum, and hence presumably much earlier, in Cicero's Greek sources. Nevertheless, the assumption is entirely unwarranted. We simply do not have, and could not have, any natural (as opposed to revealed) criterion that enables us to say, when faced with something that is found to have actually happened, that here we have an achievement that nature, left to her own unaided devices, could never encompass. The natural scientist, confronted with some occurrence inconsistent with a proposition previously believed to express a law of nature, can find in this disturbing inconsistency no ground whatever for proclaiming that the particular law of nature has been supernaturally overridden. On the contrary, the new discovery is simply a reason for his conceding that he had previously been wrong in thinking that the proposition, thus confuted, did indeed express a true law; it is also a reason for his resolving to search again for the law that really does obtain. We certainly cannot say, on any natural (as opposed to revealed) grounds, that anything that actually happens is beyond the powers of unaided nature, any more than we can say that anything that any man has ever succeeded in doing transcends all merely human powers. For our evidence about the powers of nature in general, and of men in particular, is precisely and only everything that things and people do. For a scientist to insist that some recalcitrant fact constitutes an overriding of a still inviolably true law of nature is—to borrow Rudolf Carnap's mischievous analogy—as if a geographer were to maintain that the discrepancies between his maps and their objects show that there is something wrong with the territories concerned.

The insistence of the scientist, insofar as he is simply a scientist, on always seeking strictly universal laws is itself rooted in the fundamental object of the whole scientific quest: if scientists are to find comprehensive explanations, they must discover universal laws. A scientist's refusal to accept the idea that in any single case nature has been overridden by supernatural intervention is grounded partly on precisely the above-mentioned lack of any natural (as opposed to revealed) criterion for distinguishing natural from supernatural events, and partly on his commitment—which is chiefly what makes him a scientist—to continue always in the search for completely universal laws, and for more and more comprehensive theories. In view of this, it need be neither arbitrary nor irrational to insist on a definition of a "law of nature" such that the idea of a miracle as an exception to a law of nature is ruled out as self-contradictory.

The seductive but erroneous idea that we do possess some natural means for the identification of the supernatural is one that, in some respects, parallels the notion that it is logically possible to derive prescriptive norms from knowledge of what is, in some purely descriptive sense, natural. In each case there are adherents for whom the division between natural and supernatural, or between natural and unnatural, is nothing but an incoherent muddle. Likewise, in each case there are others who, in support of their choice, are prepared to deploy some more or less elaborate structure of theoretical justification.

problem of evidence

All of this argumentation, although both relevant and (in spirit at least) thoroughly Humean, has little in common with the line of argument Hume chose to develop in the section "Of Miracles." Although this line of argument is equally methodological, it treats the question of miracles as it arises in the field of history rather than as it might impinge upon natural science. Hume was primarily concerned not with the question of fact but with that of evidence. The problem was how the occurrence of a miracle could be proved, rather than whether any such events ever had occurred. Consequently, even if Hume was successful, the way would still remain clear for people to believe in miracles simply on faith. In his own mordant way, Hume himself was happy to allow for this, but he always insisted that "a wise man proportions his belief to the evidence."

This concentration on the evidential issue means that Hume's thesis, however offensively expressed, is nevertheless at bottom defensive. Hume hoped that he had discovered "a decisive argument … which must at least silence the most arrogant bigotry and superstition, and free us from their impertinent solicitations … an argument which … will … with the wise and learned, be an everlasting check to all kinds of superstitious delusion…." These words were very carefully chosen. The whole argument was directed to the wise—to those, that is, who insist on proportioning their belief to the evidence. It did not show that the substantive claims of the bigoted and superstitious are in fact false. It was intended to serve as a decisive check on any attempt to solicit the assent of rational men by producing proof of the occurrence of the miraculous. In particular, the object was to interdict the second movement of the standard apologetic attack as outlined above.

If for present purposes a certain amount of misguided psychologizing is ignored, the following would appear to be the gist of Hume's "everlasting check." There is, he remarked, "no species of reasoning more common, more useful, and even necessary to human life than that derived from the testimony of men and the reports of eye-witnesses and spectators." Yet all testimony must ultimately be subject to assessment by the supreme court of experience. Certainly there are, as Hume observed, "a number of circumstances to be taken into consideration in all judgments of this kind." Yet "the ultimate standard by which we determine all disputes … is always derived from experience and observation." (Of all people, Hume, as the author of that most famous paragraph in the Treatise of Human Nature, should have said not "is, " but "ought " always to be so derived.)

The weight of the testimony required must depend on the apparent credibility of the events reported. If the events are in some way marvelous and rare, then the testimony for them has to be treated with more circumspection than the witness to everyday occurrences. But supposing that the testimony is for events that, had they occurred, would have been genuinely miraculous: we are then confronted with a paradoxical dilemma, proof balanced against proof. However overwhelming the testimony might have appeared were it not being considered as evidence for a miracle, in this peculiar case the testimony must always be offset against a counterproof. In Hume's own words, "A miracle is a violation of the laws of nature; and as a firm and unalterable experience has established these laws, the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined."

In the first part of section X, Hume argued generally from the concept of the miraculous—from, as he put it, "the very nature of the fact." In the second he deployed several more particular assertions about the corruptions to which testimony is liable, urging that such corruptions are exceptionally virulent where any religious issue is involved. He also added a further consideration relevant to any attempt "to prove a miracle and make it a just foundation for any … system of religion."

This consideration was expressed badly and was entangled in one or two inessential errors and confusions. But a letter makes clear Hume's intent. The point is that if the occurrence of some sort of miracle is to serve as a guarantee of the truth of a system of religion, then there must not have been any similar miracle under the auspices of a rival system, the truth of which would be incompatible with the truth of the first. Consequently, insofar as we are considering a miracle not as a putative bald fact but as a possible endorsement of the authenticity of a revelation, we have to throw into the balance against the testimony for the miracles of any one candidate revelation all the available testimony for all the miracle stories presented by all the rival systems that are inconsistent with the first. In its appeal to a necessary conflict of evidence, this argument resembles the paradoxical dilemma expounded above.

Miracles and the Philosophy of History

Enough already has been said to suggest that there is more to Hume's check than a trite insistence that since the occurrence of a miracle must be very improbable, it would have to be exceptionally well evidenced in order to be believed. C. S. Peirce was in possession of the vital clue (which he seems never to have exploited fully) when he remarked, "The whole of modern 'higher criticism' of ancient history in general, and of Biblical history in particular, is based upon the same logic that is used by Hume" (Values in a Universe of Chance, edited by P. P. Wiener, New York, 1958, pp. 292–293). When we follow this clue, it becomes obvious that Hume himself saw "the accounts of miracles and prodigies to be found in all history, sacred and profane" as presenting a methodological problem. This section on miracles constitutes the outer ring of Hume's defenses against the orthodox religious apologetic. But at the same time it is also part of his contribution to an understanding of the presuppositions and the limitations of critical history.

This fact seems not to have been appreciated as it should have been. There is, for instance, no reference to Hume's section "Of Miracles" in R. G. Collingwood's The Idea of History (Oxford, 1946); and neither Collingwood nor F. H. Bradley seems to have had any idea of the extent to which Bradley's own essay, "The Presuppositions of Critical History" (Collected Papers, Vol. I., Oxford, 1935), echoed arguments first developed by Hume. It is worthwhile to consider possible causes of this neglect. In part it is to be attributed to the insistence (at one time universal) on treating section X, "Of Miracles," as though it were a separate and disingenuous essay, irrelevantly inserted into the first Enquiry simply to cause scandal and thereby push up sales. This perverse and gratuitously offensive notion has misled interpreters to overlook some extremely relevant remarks in Part I of section VIII which concern the inescapably uniformitarian presuppositions of both the natural and the social sciences. Even those who have succeeded in appreciating section X as a very considerable piece of argumentation have been inclined to pigeonhole it as being a contribution to the philosophy of religion only. Certainly Hume's argument does, in the first instance, belong to the philosophy of religion; and this, of course, is how Hume presented it. Yet, as we have already seen, it also has a place in the philosophy of science. The fact that Hume appreciated this is perhaps suggested by his proposal that if, against all reasonable expectation, there were to be sufficient historical evidence to establish that the "miracle" of a universal eight-day eclipse had occurred in January 1600, "then our present philosophers [scientists], instead of doubting the fact, ought to receive it as certain; and ought to search for the causes whence it might be derived." It is surely significant that in this one context, and inconsistently with his own official definition of miracle, he spoke not of "a violation of the laws of nature," but rather, and more weakly, of "violations of the usual course of nature."

The same nodal argument which thus has a place in both the philosophy of religion and the philosophy of science belongs equally in the philosophy of history. For what Hume was contending (with certain lapses and hesitations) is that the criteria by which we must assess historical testimony, and the general presumptions which alone make it possible for us to interpret the detritus of the past as historical evidence, must inevitably rule out any possibility of establishing, upon purely historical grounds, that some genuinely miraculous event has indeed occurred. Hume concentrated on testimonial evidence because his conception of the historian, later illustrated in his own famous History of England, was of a judge assessing with judicious impartiality the testimony set before him. But the same Humean principles can be applied more widely to all forms of historical evidence.

The fundamental propositions are first, that the present detritus of the past cannot be interpreted as historical evidence at all, unless we presume that the same basic regularities obtained then as today; and second, that in trying his best to determine what actually happened, the historian must employ as criteria all his present knowledge, or presumed knowledge, of what is probable or improbable, possible or impossible. In his first work, the Treatise of Human Nature (II, iii, i), Hume had argued that it is only on such presumptions that we can justify the conclusion that ink marks on old pieces of paper constitute testimonial evidence. Early in the first Enquiry, in the first part of section VIII, he urged the inescapable importance of having such criteria. In a footnote to section X, he quoted with approval the reasoning of the famous physician De Sylva in the case of a Mlle. Thibaut: "It was impossible she could have been so ill as was proved by witnesses, because it was impossible she could, in so short a time, have recovered so perfectly as he found her."

flaws in hume's account

Two very serious faults in Hume's presentation of his argument may obscure the force and soundness of De Sylva's reasoning, as well as the fact that this sort of application of canons to evidence is absolutely essential to the very possibility of critical history.

The first fault is a rather wooden dogmatism of disbelief. For against all his own high, skeptical principles, Hume tended to take it for granted that what in his own day he and all his fellow men of sense firmly believed about the order of nature constituted not just humanly fallible opinion, but the incorrigible last word. He was thus betrayed into categorically dismissing as downright impossible certain reported phenomena that the later progress in the study of abnormal psychology and of psychosomatic medicine has since shown to have been perfectly possible. But the moral to be drawn from these lapses into dogmatism is not that Hume was mistaken in insisting that the critical historian must apply canons of possibility and probability to his evidence, but that he failed to appreciate that all such canons are themselves subject to criticism and correction.

The second major fault in Hume's treatment is both more serious and more excusable. He was unable to provide an adequate account of the logical character of a law of nature. Hence, he could not offer any sufficiently persuasive rationale for employing, as canons of exclusion in historical inquiry, propositions that express, or that are believed to express, such natural laws. The way may thus seem to be open for a historian who holds different presuppositions, yet still remains truly a historian, to endorse as veridical stories of events that, had they occurred, would have been truly miraculous. (For a sustained study of such attempts to have it both ways, see T. A. Roberts, History and Christian Apologetic, London, 1960.)

This problem of the logical nature of natural laws has, of course, many more aspects than those that immediately concern us here. But it is important first to emphasize that it is at least as much a problem for Hume's immediate opponents as for Hume. For it is his opponents who need a strong sense of "miracle," in which the miraculous can be distinguished from the merely marvelous. It is tempting, but entirely wrong, for the spokesman for the miraculous to think that he can afford to triumph over Hume's difficulties without being himself committed in any way to producing his own account of the character of laws of nature—an account that shall be more satisfactory as an analysis and yet, at the same time, consistent with the things the spokesman himself wants to say about the miraculous. His dilemma, to repeat, is that he needs to be able to accommodate simultaneously both the strong laws and the spectacular transgressions.

nomological propositions

Casting back to the reasoning of De Sylva, it can now be seen that (and how) it constitutes a paradigm of critical history. For it is only and precisely by presuming that the laws that hold today held in the past and by employing as canons all our knowledge—or presumed knowledge—of what is probable or improbable, possible or impossible, that we can rationally interpret the detritus of the past as evidence and from it construct our account of what actually happened. But in this context, what is impossible is what is physically, as opposed to logically, impossible. And "physical impossibility" is, and surely has to be, defined in terms of inconsistency with a true law of nature. Or rather, since this sense of "impossible" is prior to the development of science proper, it might be said that what is physically impossible is whatever is inconsistent with a true nomological proposition.

Both causal propositions and those expressing laws of nature fall under the genus nomological. Although Hume himself concentrated on the causal species, what he said can easily be extended. In his view, when we say that A is the cause of B, the main thing we are saying is that B 's are constantly conjoined with A 's—never as a matter of fact A and not B, or, in modern terminology, A materially implies B. Of course, he went on, people think they are asserting not a mere constant conjunction, but some real connection, and in a way this is right. The fact is, according to Hume, that there is a connection, but that it is a psychological one: we have formed a habit of associating the idea of an A with the idea of a B.

Yet this account of causal propositions cannot be adequate. All causal propositions entail subjunctive conditionals. (A subjunctive conditional, appropriately enough, is a proposition of the form, "If it were … it would.") Thus, "A 's are the only things which cause B 's" entails "If A were not to occur (or to have occurred) B would not occur (or have occurred)." But no variation on the material implication theme, with or without benefit of associationist psychological speculation, can be made to entail any such subjunctive conditional. Furthermore, the same essential inadequacy afflicts any extension of a Humean analysis to cover nomologicals in general. For a nomological is, by the above definition, a contingent proposition that entails some contingent subjunctive conditional.

The essential difference between the contingent "All X are ϕ " and the equally contingent "Any X must be ⥽ " is that the former can be expressed as a material implication, "Not both X and not ⥽, " whereas the latter cannot be so expressed, because it is a nomological, entailing such subjunctive conditionals as "If there were to have been an X (which in fact there was not) it would have been a ⥽. " The nomological goes far beyond the statement of a mere conjunction of X and ⥽ as a matter of fact. It asserts also a (contingent) connection between X and ⥽. For although the nomological is no more logically necessary than the corresponding material implication, it says not merely that, as it happens, a constant conjunction has been, is being, and will be maintained, but also that it would be and would have been maintained regardless of what anyone did or might have done. To assert the nomological is to assert that the conjunction is one that can be relied upon. It is for this reason that experimental evidence is so essential to our knowledge of nomologicals: the obvious and ultimately the only satisfactory test of the reliability of a law is to subject it to strains. It is for the same reason that a knowledge of nomologicals provides, at least in principle, a guarantee of repeatability. To say that the conjunction of B 's with A 's is reliable is to say that any time anyone likes to produce an A he will thereby bring about a B.

the historian's approach

In the light of the above discussion, we can again consider the question of historical evidence for the miraculous. The critical historian, confronted with some story of a miracle, will usually dismiss it out of hand, asking first only whether it can be used as evidence, not for the occurrence reported, but for something else. To justify his procedure he will have to appeal to precisely the principle Hume advanced: the "absolute impossibility or miraculous nature" of the events attested must, "in the eyes of all reasonable people … alone be regarded as a sufficient refutation." Our sole ground for characterizing the reported occurrence as miraculous is at the same time a sufficient reason for calling it physically impossible. Contrariwise, if ever we became able to say that some account of the ostensibly miraculous was indeed veridical, we can say it only because we now know that the occurrences reported were not miraculous at all.

objections to the historian's approach

To this representation of the procedure of the critical historian there are two main objections. First, it will be argued that such an approach to what purports to be historical evidence for the miraculous is irrationally dogmatic, for in this instance the historian seems to be represented as dismissing all evidence that conflicts with his own fundamental prejudices and as defending a closed system in which his professional predilections are guaranteed against falsification by a "Heads-I-win: Tails-you-lose" argument. This is a very understandable objection. It is made more plausible by the regrettable fact that there have been, and still are, many historical writers whose actual procedures correspond rather too closely to this suggested representation. Also it is, of course, true that the dilemmas generated by the tension implicit in the concept of the miraculous must necessarily seem to their victims to have a "Heads-you-win: Tails-I-lose" aspect. Nevertheless, the critical historian is not committed to the sort of bigoted dogmatism the present objection attributes to him.

Nomological laws and reports of miracles

As Hume was insisting from first to last, the possibility of miracles is a matter of evidence and not of dogmatism. For, to proceed beyond Hume, the nomological proposition that provides the historian's canon of exclusion will be open and general and of the form "Any X must be ⥽. " The proposition reporting the (alleged) occurrence of the miracle will be singular, particular, and in the past tense; it will have the form "This X on that particular occasion was not ⥽. " Propositions of the first sort can in principle be tested at any time and in any place. Propositions of the second sort cannot any longer be tested directly. It is this that gives propositions of the first sort the vastly greater logical strength that justifies their use as criteria of rejection against the latter. It will indeed be only and precisely insofar as we have evidence sufficient to warrant our assertion of the general nomological that excludes the particular historical proposition that we shall have sufficient reason to claim that the event it reports would have been genuinely miraculous.

The logic of evidence

Suppose that in some particular case the evidence for a miracle appears extremely strong. Then perhaps the historian may ask himself whether the nomological proposition that precludes this event is after all true. It could, in principle at any rate, be further tested. If, as is possible, it were shown to be false after all, then perhaps the event so strongly evidenced did indeed occur. But by the same token, that event could now no longer be described as truly miraculous. This, surely, is what has happened in the case of so many of the reports of astonishing psychosomatic cures, which Hume himself, in his capacity as a historiographer, too rashly dismissed. (Consider, for example, his contemptuous rejection of the stories of faith healings by the Emperor Vespasian and of the many cures associated with the tomb of the Jansenist Abbé Paris, all in section X of his first Enquiry. ) Alternatively, the nomological proposition might survive even our further tests. Hume should be the last one to deny that it must remain always conceivable—logically, that is, as opposed to physically possible—that the event in question did in fact occur. Yet in this case, no matter how impressive the testimony might appear, the most favorable verdict that history could ever return must be the agnostic, and appropriately Scottish, "not proven."

Need for canons of evidence

The second objection to the above representation of the procedure of the critical historian suggests that there is something arbitrary or at least optional about the appeal to canons provided by some of our knowledge, or presumed knowledge, of what is probable or improbable, possible or impossible. Once again there is some ground for this objection. Certainly we can choose whether or not we will try to act as critical historians. But once that fundamental choice is made, there is nothing arbitrary and nothing optional about insisting on the employment of these canons. For the essential aim of the historian is to get as near as he can to a full knowledge of what actually happened, and why. To do this he must find and interpret evidence, for belief unsupported by evidence may be true, but it cannot constitute knowledge. Yet to interpret the detritus of the past as evidence, and to assess its value and bearing as such, we must have canons. And for a rational man, these canons can only be derived from the sum of his available knowledge, or presumed knowledge. It is not the insistence on the systematic employment of these always corrigible canons that is arbitrary; what is arbitrary is to pick and choose in the interests of your ideological predilections among the available mass of miracle stories, or to urge that it is (psychologically) impossible that these particular witnesses were lying or misinformed and hence that we must accept the fact that on this occasion the (biologically) impossible occurred. If one once departs in such arbitrary ways from these canons of critical history, then anything and everything goes. (For examples of precisely this sort of arbitrariness, see M. C. Perry, The Easter Enigma, London and New York, 1959.)

Possible justifications for belief in miracles

Nothing that has been said in this article decisively closes the door on faith. We have been concerned only with questions about the possibilities of having good reasons for belief in the miraculous. Again, nothing has been said to preclude the production of nonhistorical and nonscientific considerations that might, either by themselves or with the aid of historical or scientific evidence, justify our belief that certain miracles did indeed occur. Perhaps one might develop some defensible system of rational theology that would provide criteria both for identifying particular occurrences as miraculous and for separating the true miracle stories from the false. Hume tried to rule this out also, of course, in section XI of his Enquiry, and elsewhere. But it has been no part of our present task to examine arguments against natural theology. Finally, it is perfectly possible to develop a new concept and to apply to it the term "miracle." There is never anything to keep anyone from simply changing the subject.

Bibliography

historical studies

Grant, R. M. Miracle and Natural Law in Graeco-Roman and Early Christian Thought. Amsterdam, 1952.

Lecky, W. E. H. History of the Rise and Influence of Rationalism in Europe. London, 1865. Chs. 1–3. These chapters carry on approximately from where Grant leaves off.

Stephen, Leslie. English Thought in the Eighteenth Century, 3rd ed. London, 1902. Vol. I. Contains the classic account of the entire controversy of which Hume's "Of Miracles" formed a part.

logical and methodological studies

Augustine. The City of God, XXI, 8.

Hume, D. An Enquiry concerning Human Understanding, VIII–XI.

King-Farlow, J. "Miracles: Nowell-Smith's Analysis and Tillich's Phenomenology." International Philosophical Quarterly 2 (1962). Although this article appeared in a Jesuit journal, it seems to accept most of the sort of methodological criticism presented in the present article. The bibliography of recent Roman Catholic literature on the subject is especially valuable.

Lewis, C. S. Miracles. London and New York, 1947. Mozley (see below), Tennant (see below), and Lewis all give Anglican views, of which Tennant's is perhaps the most liberal and Lewis's the most conservative. The latter employs a theoretical structure of the type referred to in the discussion of the problem of identifying the supernatural in order to uphold the possibility of miracles. For a discussion and development of the ideas touched on in that section, see the exchanges between H. H. Dubs, A. Lunn, and P. Nowell-Smith in the Hibbert Journal (1950–1952).

Mill, J. S. A System of Logic, III, 4 and 25.

Mozley, J. B. On Miracles. London, 1865.

Roberts, T. A. History and Christian Apologetic. London: S.P.C.K., 1960.

Smith, G. D., ed. The Teaching of the Catholic Church, 2nd ed. London, 1952.

Spinoza, B. Tractatus Theologico-Politicus, VI.

Tennant, F. R. Miracle and Its Philosophical Presuppositions. Cambridge, U.K.: Cambridge University Press, 1925.

Thomas Aquinas. Summa contra Gentiles, III, 98–107.

illustrations of main methodological points

Huxley, T. H. "Agnosticism and Christianity." In Lectures and Essays. London, 1902. An attack on Newman (see below).

Middleton, Conyers. A Free Inquiry into the Miraculous Powers which are supposed to have subsisted in the Christian Church from the earliest ages through several successive centuries. London: R. Manby and H.S. Cox, 1749. A pivotal work in the great eighteenth-century controversy described by Stephen.

Morrison, F. Who Moved the Stone? New York: Century, 1930. A gripping, if very unsophisticated, study of New Testament evidence.

Newman, J. H. "Essay on the Miracles Recorded in Ecclesiastical History." In The Ecclesiastical History of M. L'Abbé Fleury. Oxford, 1842. This essay by the future cardinal deserves to be read alongside Middleton's Free Inquiry (see above).

Perry, M. C. The Easter Enigma. London: Faber and Faber, 1959. An attempt to apply the findings of psychical research to the interpretation of the New Testament documents. It illustrates both an admirably undogmatic flexibility about what in fact is and is not possible and a certain inconsistency in applying the canons of critical history.

Schweitzer, Albert. The Quest for the Historical Jesus, from Reimarus to Wrede. London: Black, 1910. A classic history of a historiographical subject. It is interesting to note that the work contains no reference to Hume.

Thomson, J. M. Miracles in the New Testament. London: Edward Arnold, 1912. An important but neglected work by a historian who later achieved distinction in the field of secular history.

West, D. J. Eleven Lourdes Miracles. London: Duckworth, 1957. A study of the best evidence by a psychiatrist and sometime research officer of the Society for Psychical Research.

Antony Flew (1967)

Encyclopedia of Philosophy