Science in Christian Perspective



The Problem of Miracle in the Apologetic from History
Department of Philosophy
University of Tulsa
Tulsa, Oklahoma 74104

From: JASA 30 (December 1978): 154-163.

From at least the seventeenth century until the twentieth, a cornerstone of Christian apologetics has been what can be called "the apologetic from history." Its strategy was to argue, first, that there is sufficient historical evidence to warrant belief that certain historical events-crucially the resurrection of Jesus-have occurred; and then, that the "miraculousness" of these events supplies rational justification for believing the religious teachings of the person through whom the events took place, namely Jesus.1 Robert Boyle, John Locke, Joseph Priestley, William Paley, Joseph Butler, and many others endorsed this as the strongest bulwark for the claim that Cod, through Jesus, has made available to man "revealed truth" about Himself.2

This mode of argument has, of course, become theologically unfashionable in the twentieth century. Karl Barth proposes that "Belief cannot argue with unbelief; it can only preach to it"; H. Richard Niebuhr urges that Revelation is "confessional" and that Protestant theology is essentially "subjective"; Bultmann and Tillich reinterpret the Christian proclamation as "existential": virtually all of the distinctively twentieth-century theological traditions converge in an antipathy toward giving arguments, especially historical arguments, for the claims or commitments of the Christian venture. Thus, in his Easter sermon for the New York Times, we find Martin Marty advising "otherbelievers, nonbelievers, or antibelievers": 

Yawn, please, whenever a preacher tries to "prove" the resurrection. Your boredom will help us face the issue of faith. Silly putty proofs and reasonings insult you and thoughtful Christians. They convince only the convinced. Nervous apologists have to use logic and history to prove that a tomb was empty. But Easter rises from the experience of faith-then and now.

Marty's advice, one hardly needs to document, reflects the reigning theological consensus: objective historical enquiry is irrelevant to the question of the "validity" of Christianity.

There are, however, two exceptions to this rule. Among evangelical intellectuals, there is a strong remnant still maintaining that objective historical evidence does provide strong reasons for believing the theological claims of Christianity. The late C. S. Lewis endorsed this defense; and especially under the auspices of the Inter-Varsity Christian Fellowship, proponents of this position exert considerable influence on Christian college students.4

Second, in recent years some distinguished analytic philosophers have brought the issues concerning the relations between "faith" and "history" under critical scrutiny. They have argued that the twentieth-century attempts to insulate Christian commitment from the results of historical enquiry are, philosophically, highly objectionable. But while thus arguing that historical evidence is relevant to assessing Christian belief, they have gone on to claim that this relevance is negative in import: it provides reasons only for rejecting Christianity.5

These two exceptions to the current theological consensus supply the rationale for this paper. For it is my distinct impression that virtually all the debate among those under the evangelical umbrella has concerned itself with the theological challenges to the apologetic from history-with, that is, questions about whether this apologetic is theologically appropriate. Consequently, those apologists who argue that it is appropriate have ignored, or dealt most superficially with, the very different challenges posed by analytic counter-apologists.

My aim in this paper is to bring to a focus one of the spotlights of contemporary analytic criticism, and, having done this, to use it to highlight some problems in the historical argument of a leading contemporary evangelical apologist, John Warwick Montgomery. I do this, not with the intention of debunking the historical approach Montgomery employs, but rather in the hope that, by promoting further discussion, whatever is of value in this approach might prove its mettle.

Historicity and Miracle

There are two basic first-order questions at stake in the apologetic from history. The first is whether spe cific alleged events, such as the resurrection of Jesus, actually occurred. The second is whether such an event, if it occurred, would constitute a "miracle."

Each of these questions elicits a "prior question" of a methodological sort. Before we can properly ask the first question, we must answer the prior question: How should we go about determining whether such-and-such an alleged event actually occurred as alleged? That is; By what criteria are we to assess reports alleging that this event occurred?

And before we can properly ask the second question, we must ask the corresponding prior question: How can we defensibly determine the "miraculousness" of an event? That is: By what criteria, if any, can we defensibly determine whether an event is a "miracle"?

I shall be concerned primarily with this second "prior question."

The term "miracle" is, however, a "multiguous" one:6 in different contexts it receives very different meanings. For our purposes here, we can sufficiently reduce this multiguity by reminding ourselves that the term must be given a particularly strong definition when it is employed in an apologetic from history. For the apologist from history, having purportedly established the occurrence of certain events, needs then to argue that the "miraculous" character of these same events gives us justification for believing that the miracle-doer is a trustworthy teacher of religious truths. The "miracle," that is, must function as "Divine attestation," a stamp of approval from God upon the teacher through whom the miracle occurs. Accordingly, apologists from history have sensibly tried to define "miracle" as-minimally-an event which is "contra natura," or "beyond the powers of created things." After the rise of science in the seventeenth century, this was further articulated in terms of "the laws of nature": a "miracle" was usually defined-again minimally-as "a transgression of the laws of nature."7

This is the definition David flume invokes in his infamous critique of the apologetic from history.8 It should be noted that this definition does not-countless critics of Hume notwithstanding-rule out the attempt to theologically explicate "nature's laws" as themselves the result of the continuous activity of God in nature. The bite of the definition is simply that a "miracle" is an event which rationally compels a man to admit (if he is rational): "Only God could do this thing; nature alone could not!" If the order of "nature alone" is itself explicated in terms of the continuous activity of God, of His "general concourse" with Creation, then a miracle must be defined as an event which could occur only by a special voluntary act of extraordinary power. For only so can it hear the weight of the apologetic from history. As Antony Flew puts it:

It is only and precisely in so far as it [miracle] must involve an overriding from outside and above-an event which, so to speak, Nature by herself must be unable to contrive-that such an event would force the conclusion that a transcendent Power is revealing itself.
This being so, it will get the apologist nowhere fast to urge that such a notion of the miraculous [as Hume invokes] is somehow quite unsound. He is the one who needs it, if, that is, the occurrence of a miracle is to serve as the credentials of his candidate revelation.9

Granting, then, the apologetic necessity10 of defining

Among evangelical intellectuals there is a strong remnant still maintaining that objective historical evidence does provide strong reasons for believing the theological claims of Christianity.

"miracle" in this way, the crucial question is: By what criteria are we to judge whether or not an event is a miracle, in this hall-blooded revelation-certifying sense of the term? Events do not, after all, come with attached tags telling us whether or not they are produced by special acts of Divine agency.

It seems to be rarely recognized, especially by those who still espouse the apologetic from history, that developments in science and in the philosophy of science have greatly increased both the necessity and the difficulty of answering this question.

The necessity: because the last few centuries of science have repeatedly turned up events which were strikingly contrary to what the theories of the time implied nature is capable of contriving. At the time they are first observed, such "anomalies" may be unique, and practically speaking unrepeatable. One thinks, to cite one instance of massy, of the startling observation of a supernova in the sixteenth century. No one had seen such a thing before, no one knew whether it would he seen again, and it was contrary to the then-established theory that the celestial region is "incorruptible"-comprising entities which, by their nature, can suffer neither generation nor destruction.11

At least by the wisdom of hindsight, we know that it would be apologetically and scientifically disastrous to deem such anomalies as "miracles." It would he apologetically undesirable, both because it would lead to a baffling proliferation of "miracles," and because such "miracles" would be uncomfortably ephemeral as those less prudent prophets who deemed the sixteenth-century nova a "miracle" died too soon to learn. For like this nova, the most startling anomalies have regularly led to the development of new scientific theories which adequately explain the supposed "miracle" in terms of strictly natural processes. This historical reality also shows why it would be scientifically' disastrous to regard such anomalies as miracles: for it is only by so much as they are treated as the effects of not-yet-understood natural processes, that they prod the search for new and more adequate scientific theories.

These same historical realities which make it necessary for the apologist to supply the criteria in question also make it difficult for him to do this-more difficult now than it was, say, for Robert Boyle in the seventeenth century or William Paley in the eighteenth. In these earlier centuries, especially after the astounding successes of Newtonian dynamics, it could he maintained with some plausibility that Newton's "inductive method" yields knowably true and complete theories of natural processes-theories which would never have to be revised or abandoned in their proper domains. This confidence in the "absolute truth" of inductively-certified scientific theories may well be why the apologetic from history thrived as it did: for by so much as one scientifically "knows" (or thinks one knows) what natural processes are capable of contriving, one knows also what they are incapable of. That is, an event which is contrary to what is entailed by an "infallibly known" scientific theory could cogently be argued to be a "miracle."12

But this epistemological confidence in "inductive method" was, we have since learned, much too optimistic. The revolutionary overthrow of Newtonian dynamical theory in the twentieth century brought forcibly home what the more perceptive methodologists had long suspected: even our best scientific theories are fallible, and may have to be radically revised in the light of new experimental findings. One can thus no longer appeal to "inductively established" scientific theories as providers of criteria for demarcating that of which nature is capable, from that which, because it cannot possibly be produced by natural processes, is necessarily miraculous.

In short: the apologist from history must provide anew some set of defensible criteria for determining which "anomalies" are properly to be regarded as "miracles," and which are to he regarded instead as indices of the inadequacy of our current theories of natural processes. And it is clear that the onus of providing such criteria is on the apologist, not upon his opponent: for it is the apologist who must show that it would be unreasonable to regard his putative miracles merely as persistent-hut-temporary natural anomalies, akin to the nova in the sixteenth century. Indeed, the onus is even more stringent than this. The apologist must show that it would he unreasonable, in certain eases, to remain "agnostic" about the matter, i.e. to leave it as an open question.13

The apologist from history need not, of course, provide infallible criteria for determining "miraculousness"; but he at least needs to supply criteria that enable us to judge when it is more reasonable to regard an event as a miracle than to regard it as a persistent natural anomaly. He may choose to attempt this by speaking of the "degree of probability" that such-and-such an event is a miracle. But to justify this way of speaking he must do more than wave his arms in' the direction of Butler's aphorism:14 he must supply defensible quantitative (degree-yielding) criteria by which can assign such "probabilities" to events. Unless such criteria are provided, Flew can rightly contend that the apologist is being subjective and arbitrary in his selection of certain anomalies, and not others, as instances of the miraculous. To my knowledge no such criteria have yet been provided.15

Natural Law and Miracle

In light of what has been argued, I do not see how one could sustain the very different moral that John Montgomery tries to extract from the history of science. Montgomery writes:

But can modern man accept a "miracle" such as the resurrection? The answer is a surprising one: The resurrection has to be accepted [given the historical evidence] just because we are modern men, men living in the Einstein relativistic age. For us, unlike people of the Newtonian epoch, the universe is no longer a tight, safe, predictable playing field in which we know all the rules. Since Einstein no modern has had the right to rule out the possibility of events because of a prior knowledge of "natural law."16

Contrary to the historical generalization here implied, there simply was no "Newtonian epoch" which confidently "ruled out" miraculous events. A great many flags were flown under the Newtonian banner in the eighteenth and nineteenth centuries, and it would be foolhardy to try to summarize here the diverse and often conflicting ways that "Newtonians" have viewed natural laws and miracles.17 But this much can he said without qualification: it is clear that for Newton himself, and for the circle of disciples who first defined what it was to be a "Newtonian," Newton's scientific proposals led to a great stress on the extent to which the "laws of nature" themselves are necessarily sustained by-rather than autonomous from-the continuous active power of God.18 Furthermore, the confidence of these Newtonians that we could empirically know the true laws of nature never led them to skepticism about whether miracles occur. Their confidence was simply that, insofar as God is working by his ordinary "laws," certain sorts of events (miracles) can not occur. This did not at all "rule out" miracles, except of course for those who made the further Deistic assumption that God must always act in accordance with these regular, inductively discoverable "rules," which are thus deemed irrecusable even for the Ruler. For two centuries, Newtonians consistently could-and persistently did-reject this Deistic assumption. Twentieth-century "Einsteinians" can, and do, continue to accept it-as did Einstein himself.19

The shift from Newtonian to Einsteinian physics has, in fact, very little philosophical relevance to the question of whether we have the right to rule out the possibility of miraculous events. This question hinges primarily upon whether one chooses to believe that the natural order is "closed" or "open," and Einstein's revolution is philosophically irrelevant to this question. It is relevant, not to the question of whether it is possible that miraculous events can occur, but rather to the question of whether, if they occur, we could ever know them to be "miraculous." And its relevance here derives not from the physical content of Einstein's theories, but simply because by overturning Newtonian dynamics, these theories brought home the epistemological fact that the best of theories are fallible, and can be asserted only provisionally, "until further notice."

The correct epistemological moral to draw from the Einsteinian revolution is thus not: "Aha, now we see that miracles are possible after all!"; rather it is: "If we can no longer claim to know what natural processes in themselves are capable of producing, how then can we know whether any startling anomaly is a 'miracle'?" The crucial question is thus underscored: If miracles do occur, by what criteria can we distinguish them, qua miracles, from those natural events that are startling only because our theories of nature (and the expectations these theories give us) are defective?

Suppose, as a "Gedanken experiment" to make the issue vivid, that it were established that Uri Geller does bend metal bars across a room, by some extraordinary power. Would this be something producible only by God's special agency, and thus count as a "miracle"? If not, by what criteria are we entitled to claim that walking on water, for example, would fall in the category of the miraculous, though Geller's telekinesis would not?

Montgomery, it seems, is caught by his own argument. If we do not have "the right," by "prior knowledge of natural law," to say that certain sorts of events are beyond the capabilities of natural processes, then by what right can we say, when confronted with a resurrection, that this is an event that could be produced only by a special act of Divine power?

A Circular Argument and Existential Escape

One needs only to read a bit between the lines to see how Montgomery disposes of this problem. Writing of the resurrection, which in his view provided "the final proof of the truth of Jesus' claim to deity," Montgomery says:
...we must go to the One who rose to find the explanation [of his rising], and His explanation, though we may not like it, is that only God Himself, the Lord of life, could conquer the powers of death.20

This tactic reveals how much the logic of the apologetic from history can get twisted under the (implicit) pressure of the need for criteria for determining miraculousness. If Montgomery intends to he offering the traditional argument-and he gives no hint of an alternative to it-he has rendered it completely circular. Originally, the apologist argued that we are justified in believing Jesus' teachings because his authority has received Divine attestation via miracles-events which clearly could only come from God. But now, the reason offered for believing that the crucial event is indeed a miracle is that Jesus teaches that it is, i.e. that "only God himself" could produce it. The circle is closed.

The crucial question is thus sidestepped. For the fact that a person has a certain extraordinary power neither entails that he knows, nor, if he knows, that he is truthful about, the true explanation for this power. If it were factually established that Jeanne Dixon could prophesy, or that Uri Geller could bend spoons across a room, would we be rationally obliged to accept an explanation of their powers they proffer, simply because other humans cannot do what they can do? By the only criterion Montgomery provides, Un 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." Once this is rejected as a specious criterion, one need only add that the fact that "we may not like" Jesus' explanation is, even for the most Calvinistic of us, not a good reason for believing the explanation to he true.

But an extensive caveat is in order here. It might be felt that my criticism of this passage is unfair, because Montgomery does not intend it as a detailed solution to the problem of miraculousness. My reply to this is twofold. First and simplest, the passage under scrutiny is the only argument Montgomery provides in History and Christianity for what is surely its most crucial premise: that establishing the truth of a historical claim (about Jesus' resurrection) enables one to infer the truth of a theological claim (about Jesus' divinity). For this reason alone, any circularity introduced into the argument by this passage is, I think, fair game.

The second and more disturbing reason is that the chasm thus created in the argument of History and Christianity is, to my knowledge, not bridged anywhere in Montgomery's writings. The caveat, of course, is that my knowledge of the Montgomery corpus is not exhaustive: since I have not read everything he has written, it is possible that in some essay I've overlooked, my misgivings have already been remedied. I do, however, know of several places where he touches on the problem I have raised. That these are not remedies is the burden of what follows.

Consider, first, a passage in his essay on "Biblical Inspiration" in which he argues that the apologetic from the resurrection has as its model Jesus' own mode of arguing from an empirically verifiable claim (that he has the power to heal the man with palsy) to a theological claim (that he has the power to forgive sins) which Montgomery admits is not, per se, empirically verifiable. Commenting on the Marcan account, Montgomery writes:

Does He [our Lord] leave his forgiveness claim in the realm of the unverifiable, as have numerous religious leaders through the ages? By no means; he connects the theological claim with an empirical claim whose verifiability is not only possible but inevitable. The argument thus runs: "You do not believe that I can forgive sins. Very well; I cannot show you that directly. But if I show you that I can, by my Divine power, remedy the empirical sickness that connects with the sin problem, will you have any reason left for denying my power
to work in the theological sphere?" The empirical, objective healing of the palsied man was performed that men might "know that the Sun of man bath power on earth to forgive sins"-a fact that, had our Lord not coupled it with an objective test, could have been dismissed as meaningless and irrelevant to those who had doubtless heard such claims many times before. In precisely the same way does the New Testament present Christ's resurrection as the objective ground of belief in the theological significance of his death on tile Cross.21

For the sake of argument, let us grant the exegetical part of Montgomery's thesis here, and assume that Jesus and the New Testament writers did argue in this way. The critical apologetic question the!) is: if this mode of argument is cogent, what makes it cogent?

The crux is that if Jesus' theological claim does receive attestation from the healing of the palsied man, it does so not simply because the healing-claim is an empirically verifiable claim about an observable event. The attestation requires also that the healing-event be, knowably, a very special sort of observable event: that it have a special and peculiar property by virtue of which, if one can heal palsy, one's theological claims ought to he believed. Like the man in John 9 whose blindness He healed, one must argue that Jesus' healing deeds could he done only by a man "of God," invested with the special power of God. This, of course, is what the classical apologists from history meant by arguing that attestational events are knowably miraculous.

Obvious as this seems, it also seems that Montgomery overlooks it; for he sheds no light on how this could he plausibly argued. True, he does take the premise of Jesus' argument to assert that Jesus "can, by [His] divine power, remedy the empirical sickness that connects with the sin problem" (italics mine), but this is to assert precisely what has not been shown: for though the healing is an observable fact, that it is done by virtue of Divine power is not. One could observe whether the event occurred, but one could not observe whether it was a miraculous occurrence. (Nor, given this, are we helped much by Montgomery's invocation of the idea that empirical sickness is in some way connected with the "sin problem": I should hate to think that a medical degree is any index of one's authority to forgive sins!) So apologetically, we are left quite as much in the dark about the nature of the biblical "inference" from the healing-claim to the forgiveness-claim, as we are about how Montgomery gets from the truth of a historical claim about the resurrection to the truth of a theological claim about Jesus' deity.22

About the only thing I can find in print that sheds any light on Montgomery's views on this, is a comment
made by Montgomery's colleague Paul Feinberg, in his defense of Montgomery's philosophy of history against a critique written by Ronald Nash. Nash took Montgomery to task for claiming that historical events "carry their interpretation with them," citing as an illustration Montgomery's statement that "when the historical facts of Christ's life, death, and resurrection are allowed to speak for themselves, they lead to belief in his Deity and acceptance of his account [of the supernatural character of the resurrection]." Feinberg's reply-submitted with Montgomery's endorsement-gives a plausible defense of the view that historical facts generally provide the means for assessing rival interpretations of the events. But one wants to know how this can work when peculiarly "miraculous" interpretations are in question: how is it that "the facts themselves" justify interpreting a resurrection as a revelation-certifying miracle that rationally warrants belief in Jesus' claim to deity? On this Feinherg simply
says, "significance arises from the nature of the event. Death, for instance, is significant because it is an ultimate human existential concern." He then adds in a footnote: "This is significant in light of Nash's discussion of Montgomery and the resurrection ."23

If the footnote has any relevance to Nash's discussion at all, it must, I think, be read as suggesting the following: It is because death is "an ultimate human existential concern" that Jesus' death and resurrection, "when allowed to speak for themselves, lead to belief in his deity and acceptance of his account," Now, as a possible psychoanalytic description of the processes by which Christians come to hold their beliefs I will not quibble with this, since I am not a psychoanalyst. But Feinherg clearly also intends to be endorsing this process as exemplifying a reasonable kind of inference: for Nash was-as I am-asking not for a genetic explanation, but rather for a normative rationale. And as an answer to the normative question, I find Feinberg's proposal not a little odd: for the claim is then that we are somehow justified in our belief that the resurrection is a revelation-certifying miracle, because of the fact that we humans have a basic existential need to transcend death. I confess that the logic of this escapes me. But it surely will not do to argue that the process is one in which we "allow the facts to speak for themselves": to the contrary, Feinberg's proposal implies that what leads us from the historical facts to our theological interpretation are not the facts themselves, but rather the existential hang-ups (to put it with less epistemic charity) that we bring to the facts.

Besides his general endorsement of Feioberg's article, I know of only one other place where Montgomery has publicly expressed this existential element in his apologetic. In a dialogue first published in Christianity Today, Montgomery asserts that historical enquiry can tell us that the resurrection occurred, but that it cannot tell us what the explanation of its occurrence is. When asked what good the historical information then is, Montgomery replies, "Plenty, if you have a death problem-because you are obviously going to wonder why in thunderation this happened."24

Like Feinherg's comments, this still leaves us pretty unclear about what Montgomery thinks the episteioic relevance of our "death problem" is. Though it is risky to read too much between the lines, his other comments indicate that Montgomery is offering us something like the following: "Though historical enquiry, in itself, cannot tells us what the true explanation of the resurrection is, it can tell us what Jesus taught its true explanation to he. And given the relevance of the event to our existential needs, we are being most reasonable when we go to the resurrected one for our explanation of it."

Its relevance to our "death problem" thus seems to have become the apologetic surrogate for the traditional claim that the resurrection is knowably miraculous, in an objective, revelationcertifying sense. In view of Montgomery's avowed empiricism, this existential turn is both surprising and-to me-dubious: for since when have our human needs-however existentially fundamental-become a defensible substitute for empirical

The apologetic from history must provide anew defensible criteria for determining which "anomalies" are properly to be regarded as miracles.

evidence? That such needs guide the questions we find it important to ask, is reasonable. That they genetically explain why Christians come to hold the beliefs they hold, is not entirely implausible. But that they provide reasonable warrant for those beliefs seems, from an empiricist's point of view, indefensible.

Not Simply a Theoretical Problem

Some might be tempted to dismiss the problem of supplying the criteria in question as "purely academic," even "pedantic." "After all," it might he claimed, "on a practical level there are surely few people who would, if convinced on historical grounds that the resurrection did occur, just shrug it off as another anomaly to be put on the scientific agenda of outstanding research problems. Even Antony Flew (it might be ventured), if convinced of the historicity of the resurrection, would irresistibly respond as did a centurion to a lesser wonder: 'Surely this was the Son of God!' So the problem of formulating criteria is purely 'theoretical.'

For those who take the apologetic from history seriously, there are two reasons why this had better be avoided. First, because by it apologetics degenerates from a concern with what is rationally believable, into a policy based on practical psychology-and by biblical standards, dubious practical psychology at that.25 It might be true that even Antony Flew's psychological makeup is such that, if actually confronted with a resurrection, he could not help but believe it to be a revelation-certifying miracle. This would be an interesting fact about Antony Flew. But would it at all justify the claim that Flew has gotten closer to the truth? For the "could not help but" is in itself only a psychological necessity; and if it is not guided by reasons, it is irrational (even if Antony Flew himself couldn't resist his psyche in the crunch).

There is, secondly, an even more far-reaching issue at stake. Flew has argued, to my mind formidably, that the question of whether we are justified in asserting that the resurrection occurred, depends upon whether we can justify asserting that this event, if it occurred, would be genuinely miraculous. He builds a case that if we have no defensible criteria by which to identify such events as miraculous, then on the available evidence we cannot even justify claiming that our putative miracle occurred.26 If Flew is correct, the apologist who allows "psychological makeup" to replace rationality on the issue of miraculousness is not even going to be able to establish the historicity of his alleged miracle. It is beyond the scope of this paper to give Flew's position an analysis as extended as it deserves. The crux of it, however, shall he unpacked by critically considering one more assertion from Montgomery.

Determining Improbabilities

Montgomery writes,

Of course, attempts have been made to "explain" the resurrection accounts naturalistically. The German rationalist Venturini suggested that Jesus only fainted on the cross, and subsequently revived in the cool tomb. This "swoon theory" is typical of all such arguments: they are infinitely more improbable than the resurrection itself, and they fly squarely in the face of the documentary evidence. Jesus surely died on the cross, for the Roman crucifixion teams knew their business (they had enough practice). He could not possibly have rolled the heavy boulder from the door of the tomb after the crucifixion expcricnce.27

What Flew's discussions force one to probe here is Montgomery's confident assertion that certain possibilities, like that envisioned by the swoon hypothesis, are "infinitely more improbable than the resurrection itself." If one reflects a bit on the empirical procedures by which we normally estimate "improbabilities," this confidence seems, at the least, to need some explanation. For surely the basis of our judgment that it is improbable that a Roman crucifixion team could err in their grisly business is this: in all other eases of which we have knowledge, they did not err. If the amount of "practice" they had is invoked, the inferential procedure remains the same: the probability that for a crucifixion team "practice makes perfect" can be estimated only on the basis of how often, in other eases of this kind, practice does indeed make perfect.

But what is the outcome when, using the same procedure, we compare this with the probability of a resurrection? One need not even resort here to the per tinent observation that trained medical doctors have, on documented occasions, mistakenly pronounced a person in a deep "swoon" to be officially dead. For setting such instances to the side, the following consideration alone is decisive: 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. Bracketing the instance in question (lest the question be begged) the normal procedure therefore yields the following verdict: granting that it is to some degree improbable that a crucifixion team would mistake a man in a swoon to he dead, it is yet more improbable, "infinitely more improbable," that a dead man would not stay dead .28 The verdict is parallel if one examines, in the light of our normal procedures for estimating what is and is not "possible," Montgomery's categorical assertion that "Jesus could not possibly have rolled the heavy boulder from the door of the tomb after a crucifixion experience."

There is a parenthetical point whose outcome should be noted here. In an oft-quoted chapter on Hume, C. S. Lewis argues29 (and his argument seems to have gained universal currency among evangelical apologists) that such estimates of "antecedent probabilities" are relevant only to predictive judgments about whether a future event will occur under specified conditions: that when the event in question is in the past and we are appraising testimony alleging to have witnessed it, this kind of probability is totally irrelevant. This "reply to Hume," which seems to have originated with Joseph Butler, should have been laid to rest long ago. It rests on a knot of confusions that were adequately untangled and criticized by John Venn over a century ago, in his classic The Logic of Chance.30

Sensitized by Flew's analysis, we thus face the following dilemma. It is only by employing normal probability-estimating procedures that the apologist can assert that certain possibilities (such as that entertained by Venturini) are to some degree improbable. But if we consistently apply these same procedures to the possibility envisioned by the resurrection hypothesis, it is rendered yet more-staggeringly more-improbable than the others. This is, I believe, packed into Flew's concise summary of the Humean critique:

The heart of the matter is that the criteria by which we most assess historical testimony, and the general presumptions which alone make it possible for us to construe the detritus of the past as historical evidence, most inevitably role out any possibility of establishing, upon purely historical grounds, that some genuinely miraculous event has indeed occurred.31

The Verifiability Principle

The contemporary philosophical challenge concerning the historicity of Christian miracles is thus not the old naturalistic ontological prejudice to the effect that "the resurrection didn't happen because such an event would violate the laws of nature, and the laws of nature, we moderns know, are inviolable:" in his frequent attacks on this, Montgomery is heating a horse that has, for analytic counter-apologists, long been buried, The Flewian challenge is rather a methodological-epistemological one to the effect that "given the presuppositions that alone make possible the historical method (and given our lack of criteria for determining 'miraculousness'), one of the limitations of the method is that it is, by its nature, impotent to tell us whether alleged miracles have occurred-even if in fact they have."

Now, Montgomery has tried to rebut a position very similar to this (allegedly developed by Karl Barth), again by invoking the Verifiability Criterion. His argument is that the Verifiability Principle shows clearly that if historical method cannot provide access to the resurrection, then it is meaningless to say that the resurrection occurred in space-time history (i.e. as an observable event, in the past; in Historic, as distinct from Geschichte). He writes, "If Christ's resurrection really occurred in history, then historical investigation will [in principle he able to] indicate it -for to deny this is to make meaningless the sentence that the resurrection occurred.32

This argument rests on a misapplication of the Verifiability Principle: even if one grants the Principle (which I do not), the conclusion does not follow. For the application overlooks the relevance of a distinction that is crucial to responsible application of the principle, which its exponents always insisted on. In the verifiability literature, the distinction is usually referred to as the difference between "verifiability in principle" and "verifiability in practice." Properly understood, the Verifiability Criterion stipulates that for a sentence to be meaningful it need only be verifiable in principle; whether or not it is verifiable in practice is totally irrelevant to its meaningfulness (though it is, of course, very relevant to whether we could claim it to be knuwably true). It is only by virtue of this crucial distinction that, for example, it was meaningful to say in 1800 that "there are craters on the other side of the moon," even though there was not then (and might never have been) any method actually available for testing the statement. For all that is required is that some method of testing the statement (e.g. space travel) be imaginable, regardless of how unlikely it may be that we shall ever be able, in practice, to actually carry out the method.33

Once this elementary distinction is understood, it is quite obvious that the assertion "the resurrection occurred in history" can be meaningful even if historical method is by its nature impotent to verify this claim. All that is required is that some method other than the one historians use be imaginable, that could in principle verify the statement. And it takes only a little imagination to conceive of such a method. It is surely conceivable, for example, that someday a time-machine will be invented that would enable us to go hack to Jerusalem and check things out first hand. This conceivability is all that the Verifiability Criterion re(lures: that time-travel is not (and might never be) actually available to us is as irrelevant to the meaning fulness of saying "the resurrection occurred in history (though present "historical method" cannot show it) as the fact that in 1801) space travel was not (and might never be) available to men was irrelevant to the meaningfulness of them saying "the other side of the moon has craters (though present methods cannot show it)

Montgomery's appeal to the Verifiability Criterion

Evangelical apologists from history have not yet awakened to the contemporary analytic challenge.

is thus misguided. Even granting the Principle as our criterion of meaningfulness, Barth could consistently both deny that historical method has access to the resurrection, and yet assert that it is meaningful to say that the resurrection occurred as a real space-time event in the past. (Indeed, the verifiability criterion, properly applied, shows us precisely why this is consistent!) By the same token, Flew's willingness to say (and, ipso facto, think that it is meaningful to say) that the resurrection might have actually occurred in history, is entirely consistent with his argument that historical method, by its nature, could never give us sufficient reason to rationally believe that it has in fact occurred. The Verifiability Criterion, contra Montgomery, provides no escape from the Flewian dilemma stated above.

The Challenge of the Contemporary Analytical Approach

The only readily apparent way for the apologist from history to avoid the horns of this dilemma, then, is to adopt a policy of systematic inconsistency with respect to the probability-estimating procedures he employs. He must employ the normal procedures when appraising the possibilities envisioned by naturalistic alternatives to the resurrection hypothesis, and abstain from these procedures when gauging the probability of the resurrection. Such a policy might not be as indefensible as it at first blush appears to be. After all, if and when Cod does intervene in the normal course of events, one would not expect the normal probability estimating procedures to be appropriate.

But to actually make such a policy defensible, one would have to be able to specify in advance when it is appropriate to abstain from applying the normal procedures. To pull this off, one would have to have some means of determining what sorts of events are genuinely miraculous, and one would have to have ways of "retrodicting" those conditions under which such miracles could reasonably be expected to occur.

This, of course, brings us back to the need for criteria for determining "miraculousness" that I began with. To surrender such criteria into the hands of "practical psychology" (perhaps baptizing our own psychological responses as "recognitions of what is selfevident") would not merely make the issue of whether an event is a miracle a matter of subjectivity-although this alone should drive any serious apologist from history to existential despair. It would also threaten to preclude the possibility of establishing on tough-minded historical grounds that our putative miracle even occurred. This threat, which we are obligated to Flew for presenting so lucidly, might he dissolved by a more searching analysis: I am not claiming that Flew's arguments are irrefragable. The purpose of this paper is fulfilled if by probing Montgomery's position with some Flcvian questions, I have to some extent vindicated my suggestion that evangelical apologists from history have not yet awakened to the contemporary
analytic challenge.


1Here and elsewhere I intend "rational" to mean "reasonable": it is not to he identified with "logically provable" as the so-called "rationalists" (who might more perspicuously be labeled "logicalists") in their poorer moments thought. To ask whether a belief is rational is to ask whether, all things considered, there are good reasons for holding it: but "good reasons" need not, and generally do not, purport to provide demonstrative proof. Cf. S. E. Toulmin, The Uses of Argument (Cambridge: Cambridge University Press, 1958).
2I do not know to what extent this apologetic was employed before the seventeenth century. Augustine seems to invoke it ("Freedom of the Will," ch. 2, sec. 5), and the clearest prototype in the New Testament seems to be John 9:3034. On the dangers of anachronistically reading the modern apologetic intent into the various ways the biblical writers appealed to "history," see D. Ivan Dykstra, "Historicity," re-run with corrections in The Reformed Review 27 (Autumn, 1974): 60-68. But Dykstra's argument suggests that the apologetic from history is a "modern phenomenon" in the sense of being peculiar to mid-twentieth-century evangelicals. If this is his intent, it is mistaken: at most this apologetic strategy is modern in the sense in which science is modern-like "modern science," it may not have taken hold until the seventeenth century.
3The New York Times Magazine, March 30, 1975, p. 87.
4See for example C. S. Lewis, Miracles, A Preliminary Study (New York: Macmillan, 1947), esp. p. 113. Other evangelical proponents include John Montgomery, Clark Pinnock, John Cerstner, Michael Green, F. F. Bruce, John Stott, B. C. Sproul, and Daniel Fuller.
5See especially Antony Flew's discussions in Home's Philosophy of Relief (New York: Humanities Press, 1961), eh, 8; God and Philosophy (New York: Dell, 1967), ch. 7; and "Miracles," in P. Edwards, ed., The Encyclopedia of Philosophy (New York: Macmillan, 1967). See also R. Hepburn, Christianity and Paradox (Suffolk: C. A. Watts, 1958), ch. 6 and 7. Flew's most recent discussion is "Parapsychology Revisited: Laws, Miracles and Repeatability," in The Humanist xxxvi (May/June, 1976), pp. 27-30.
6I borrow this useful word from F. R. Tennant who coined it in his Miracle and Its Philosophical Presuppositions (Cambridge: Cambridge University Press, 1925),
7TIse argument assumes, of course, that God would not give such miraculous attestation to a teacher teaching theological falsehoods; to justify this assumption would require a particularly strong "natural theology" which few contemporary apologists from history (unlike their predecessors in earlier centuries) even attempt to provide.
Two important early-modern discussions of the meaning
of "miracle" are found in H. C. Alexander, ed., The Leib niz-Clorke Correspondence (New York: Barnes and Noble, 1965), from Clarke's fourth reply on, passim; and in John Locke The Reasonableness of Christianity, with a Dis course on Miracle ed. I. T. Ramsey (Stanford: Stan
ford University Press, 1958).
8David Hume, An Enquiry Concerning Human Understanding (New York: Bobbs-Merrill, 1955), p. 122. Specious characterizations of Home's epochal critique of the apologetic from history are multitudinous: the outstanding corrective is Flew's Home's Philosophy of Relief, ch. 8. Flew effectively criticizes the popular reply to Hunse that C. S. Lewis gives in Miracles, eh. 13.
9Flew, God and Philosophy, p. 148.
10This is a "conditional necessity": if one employs the apologetic from history, one must invoke this definition (or a closely similar one) to sustain the argument.
11See T. S. Kuhn, The Copernican Revolution (Cambridge: Harvard University Press, 1957), pp. 206-207.
12For a discussion of this image of Newtonian inductivism, as it was propagated by the Scottish common-sense realists, especially Thomas Reid, see Laurent L. Laudan, "Thomas Reid and the Newtonian Turn of British Methodological Thought," in B. E. Butts and J. IV. Davis, eds., The Methodological Heritage of Newton (Oxford: Blackwell and Toronto University Press, 1969). I believe, but cannot argue here, that conservative theological seminaries continued to use Reid's texts as authoritative on "scientific method" long after virtually everyone else recognized their superficialty; and that the Princeton apologists Charles Hodgc and B. B. Warfield-presuppose Reid's image of scientific method. For a clear illustration of the way Reid's inductisism earmarked popular evangelical apologetics in the nineteenth century see G. P. M'Ilvaine, The Evidences of Christianity (Philadelphia: Smith. English and Company, 1861 ), pp. 375-391. For a pithy discussion of the development of the rival "fallibilist" trend in scientific methodology -which contrary to Montgomery's suggestions began long before the twentieth century-see L. L. Laudan, "Peirce and the Trivialization of the Self-Correcting Thesis" in R. Westfall and B. (here, eds., Foundations of Scientific Method: The Nineteenth Century (Bloomington: Indiana University Press, 1973), The unexcelled analysis of the impact of images of scientific method on the history of thanmatology is Tennant's Miracle and Its Philosophical Presuppositions.
13Adolf Criinbaum has posed this challenge very succinctly in "Science and Ideology," The Scientific Monthly 79 (July 1954): 15-16.
14That "Probability is the very guide of life," from the Analogy... I have the disconcerting feeling that twentieth-century evangelical apologists are often as unreflective about invoking the concept of "probability" as their predecessors were in invoking the formulas of "demonstrative inductions" and "self-evident truths" that Reid told them were the core of empirical method. The last thirty years of epistemological research have led epistemologists to an increasingly widespread apprehension that very little light is thrown on the nature of most empirical inferences by the theory of probability; and many philosophers of science (for example Karl Popper and his followers) urge that it is indefensible to speak of scientific theories, for example, as having some estimable "probability" of being true.
15To my knowledge, the only sustained attempt to supply such criteria is R. C. Swinburne's The Concept of Miracle (New York: Macmillan, 1970). Swinburne's approach, taking a major cue from Ninian Smart's Philosophers and Religions Truth (SCM Press, 1962), is in my judgment inadequate; but this will have to be saved for another paper.
16John Montgomery, History and Christianity, available in either reprint or book form from, respectively, His reprints or Intervarsity Press) Downers Grove, Illinois, 1964, 1965). (Also reprinted with minor changes in Montgomery's Where Is History Going? [Grand Rapids: Zondervan, 1969].) Page references are to hook form. P. 75.
17An appropriate entrance into contemporary historiography of "the Newtonian epoch," for those who want to move past Montgomery's simplification, is provided by P. M. Heimann, "Newtonian Natural Philosophy and the Scientific Revolution," History of Science 11, pp. 1-7. See also M. C. Jacob, "Early Newtoniauism," History of Science 12, p. 142-146.
18The definitive work remains Helene Metzger's Attraction unieerselle et religion siatnrelle chez qnelrtoes commentateurs anglais de Newton (Paris: Hermann, 1938). See also E. A. Burtt, The Metaphysical Foundations of Modern Science (New York: Doubleday, 1925), c-h. 7; and Ahcxandre Koyre, Newtonian Studies (Chicago: University of Chicago Press, 1965), ch. 1. In the primary literature, two very revealing sources are Unpublished Scientific Papers of Isaac Newton, eds. A. R. Hall and M. B. Hall (Cambridge: Cambridge University Press, 1962), pp. 139 fl; and Newton's letters to Bentley, in Isaac Newton's Letters and Papers on Natural Philosophy, ed. I. B. Cohen (Cambridge: Harvard University Press), especially the famous third letter. As antidotes to the widespread misinterpretation of this third letter, which has it that Newton did not really believe in action-at-a-distance, this should be read in conjunction with Richard Bentley's "A Confutation of Atheism," in Newton's Papers and Letters, esp. pp. 240241; E. Meyerson, Identity and Reality (New York: Dover, 1930), pp. 452-456; and L. Laudan, "Comments on Buehdahl," in R. Steuwer, ed., Minnesota Studies in the Philosophy of Science, vol. 5, pp. 230-238.
19Einstein expressed his conviction: "I believe in Spinoza's God who reveals Himself in the orderly harmony of what exists, not in a God who concerns himself with the fates and actions of human beings." Sehilpp, ed., Albert Einstein: Philosopher-Scientist (La Salle: Open Court Publishing Company, 1970), pp. 659-660. The "deistic" connotation of this passage, however, is misleading: in other contexts Einstein made clear that he did not believe in a God "behind' this orderly harmony of what exists; rather, for him (as for Spinoza) God is this orderly harmony.
20Montgoniery, History and Christianity, p. 76.
21"Inspiration and Infallibility: A New Departure," in The Suicide of Christian Theology (Bethany Fellowship, 1971), pp. 344-345.
22Montgomery's more general thesis in this passage is that the Verifiability Criterion of Meaning shows how the theological statement is rendered cognitively meaningful by virtue of Jesus' connection of it with the verifiable healing-claim. At least as it stands, this thesis involves an error that the exponents of the Verifiability Principle were careful to avoid. The problem is that if any statement A has verifiable consequences, and thus is meaningful, then the logical conjunction of A-and-B will also have verifiable consequences -and thus ostensibly be meaningful-even if statement B is a blatant piece of metaphysical nonsense. For this reason, in applying the Verifiability Principle one most stipulate that statement B is not made meaningful by virtue of the verifiable consequences entailed by A-and-B, unless those consequences are in some respect different from the consequences entailed by A alone. This crucial requirement seems to me to completely undermine Montgomery's thesis that the meaningfulness of Jesus' healing-claim somehow spills over to make his forgiveness-claim meaningful: for the verifiable consequences entailed by the conjunction of these two claims are not, it seems to me, different from the verifiable consequences entailed by the healing-claim alone.
23Both articles are in Christian Scholar's Review: see Feioberg, "History: Public or Private?" (Summer, 1971), p. 329; and Nash, "The Use and Abuse of History in Christian Apologetics" (Spring, 1971), p. 221.
24"Faith, History, and the Resurrection," Christianity Today 9 (March 26, 1965): 4, reprinted as an Appendix to History and Christianity, p. 91. This existential appeal strongly resembles one strand of Wolfhart Pannenberg's position, which is summarized and criticized by Herbert Burheno in "Pannenberg's Argument for the Historicity of the Resurrection," Journal for the American Academy of Religion 40 (1972): 377.
25Cf, Luke 16:19-31. The implication of this passage is that "psychologically" men may well find a resurrection quite "religiously unconvincing." The apologist from history must claim that they would in this case be suppressing the rationally recognizable truth; but this claim is arguable only if the apologist first produces the criteria that make genuine miracles "rationally recognizable" as such. One "critic" of Flew (Ersvin Lutzer, in "Putting Christian Faith on the Line," His magazine [April, 1974], pp. 24-25) seems to think that, even without being able to provide such criteria, he is entitled to charge that Flew "has cut himself off from the possibility of discovering a revelation from God" by having "decided [in his "heart"] to live without confronting whatever god or gods there may be." Transparently, as an attempt to bypass the need for criteria, this is nothing but an ad horn inem reply.
26Cf. Flew, God and Philosophy, ch. 7; and "Miracles."
27History and Christianity, pp. 76-77.
28The justification with which we are entitled to assign a certain degree of improbability to the occurrence of event X under specified conditions depends upon (among other things) the size of our "sample," i.e. upon the number of times we have had occasion to observe whether events like X do or do not occur under these conditions.
29Lewis, Miracles, p. 104.
30These confusions often give rise to the assumption that the historian can properly criticize "non-veridical" historical hypotheses (i.e. those which "explain away" historical testisssony as the product of delusion, fraud, or mistake) by pointing out the antecedent improbability of the events posited by the hypotheses, but that "veridical" hypotheses (i.e. those taking the testimony at face value) are immune to this kind of criticism. This assumption, for example, seems to underlie Daniel Fuller's argument in "Historical Method and the Resurrection," The Journal of Bible and Religion 34 (1966): 18-24. But John Venn shows-I believe conclusively that the value of testimony of generally reliable witnesses is tremendously depreciated, when such witnesses testify to having observed an event with a low antecedent improbability. I most here simply refer the reader to The Logic of Chance (New York: Chelsea, 1962), ch. 12, 16, and 17.
31God and Philosophy, p. 145.
32"Inspiration ...," p. 330.
33This distinction is discussed at length by Arthur Pap in his An Introduction to Philosophy of Science (New York: Macmillan, 1962), pp. 18-22.