Miracles

The equation further on in the essay didn't translate very well from word to this blog.

What are the most significant philosophical objections to religious belief in miracles, and how best might one respond from the standpoint of Christian belief?

Traditionally, Christian theism has held that belief in miracles is fundamental to its worldview.  The apostle Paul writes, “And if Christ has not been raised, your faith is futile...” (1 Cor. 15.17).  However, since the enlightenment, religious belief in miracles has been thoroughly tested.  Following on from the philosophical objections of the deists (Craig, 2008, 248), arguments against belief in miracles came thick and fast, the most prominent of these being Hume’s argument in section X of his ‘Enquiries concerning the human understanding’ (Hume, 1963).  In this current essay the best of these arguments and some of their modern day expansions will be discussed in order to determine their significance.  Following on from this the various Christian responses to these arguments will be weighed in order to find out which response is to be judged the best from the standpoint of Christian belief.  For this essay a miracle will be defined as “a violation of a law of nature by a God” (Swinburne, 1970, 11) unless otherwise stated.

There are two main lines of argumentation when it comes to philosophical objections to religious belief in miracles. One line focuses on the a priori impossibility of miracles (cf. Craig, 2008, 248; Palmer, 2001, 184) and the other, made famous by Hume, argues that miracles are so intrinsically improbable “That no testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous...” (Hume, 1963, 116).  The task, then, is to measure which objection (or family of objections) is the most significant.  

            On inspection, the first argument seems the weakest.  Indeed, prominent scholars like that of Mackie (1992, 23) and Palmer (2001, 184), Hume and Flew (Palmer, 2001, 186) all reject the idea that miracles can be ruled out a priori.  This a priori argument rests on the assumption that the natural law is a closed system, or as Palmer (2001, 184) writes, echoing Mckinnon, “laws of nature state what happens, anything that happens must conform to those laws.”  Naturally, if this was the case, the concept of ‘miracle’ would be logically impossible as the laws of nature could not be violated, by definition miracles could not occur. However, as has been rightly pointed out by Swinburne (1970, 26) natural laws, “describe what happens in a regular and predictable way.” Our understanding of natural law is reached by inductive reasoning and is thus based on probabilities, not necessities and therefore miracles cannot be ruled out a priori. 

            The second argument however, is much more challenging and not as easy to resolve. It is necessary to look at this in more detail to understand it more fully.  Hume’s argument from the improbability of miracles is based on two separate forms of evidence (Earman, 2000, 20), those being experience and testimony.  According to Hume there is a conflict between the two in the case of a miracle.  We have a), the testimony given by a person for the miracle and b), uniform experience which he says counts against the occurrence of miracle.  Hume argues that uniform experience always outweighs testimonial evidence for miracles, and that testimonial evidence is more likely explained in naturalistic terms (Hume, 1963, 113-114) so that a miracle can never be verified. 

Antony Flew went on to enhance Hume’s argument by focussing on historical studies with reference to miracle claims.  Flew writes, “...whether or not anything did happen in the past inconsistent with what we at present believe to be a law of nature, one cannot possibly know on historical evidence that it did so happen.” (Palmer, 2001, 217).  Flew says that by its very nature historical enquiry cannot tell us whether a miracle has occurred.  Historical evidence, then, according to Flew is subordinate to the “evidence still available to us” (Palmer, 2001, 186) and according to this evidence, which is “directly verifiable”, miracles are “physically impossible” (Palmer, 2001, 186). Flew’s argument, and Hume’s argument, poses a significant problem for religious belief in miracles.

Following on from his main argument in ‘Of Miracles’ Hume puts forward another four arguments in order to back up his claim that testimonial evidence is unreliable in assessing the miraculous (Palmer, 2001, 180).  His final argument of these four is of particular interest.  Hume (1963, 121-122) writes that 1) the aim of any miracle “is to establish the particular system to which it is attributed...”, 2) that this happens in all religions (or at least a good deal of them), and 3) therefore the miracle claims of each religion cancel each other out.  This argument seems quite strong prima facie and will be returned to later.

It is recognized, then, that Hume puts forward some fairly well reasoned arguments, and, stemming from this, modern scholars like Flew have strengthened this argument with particular reference to historical study.  Of course Hume’s arguments and those of the deists did not go unquestioned by the Christian community of the period.  Indeed, there were a plethora of responses from the religious and philosophical community in defence of religious belief in miracles by the likes of Jean Le Clerc, Jacob Vernet, Thomas Sherlock and William Paley to name only a few (Craig, 2008, 252-256).  Despite this, it was the philosophical objections to religious belief in miracles “that posterity gave an eye to...” (Craig, 2008, 259). 

It seems that there are at least two main ways in which Christians might respond to these arguments.  The first way in which the Christian could respond is by accepting the arguments against religious belief in miracles.  For example, theologians like Bultmann (Bultmann, 1953, 5) and Tillich (Palmer, 2001, 199) redefine the concept of miracle as ‘myth’ or as ‘symbolism’, thus dismissing the traditional understanding.  However, whether this is the best response for the Christian is to be questioned.

This approach is not likely to appeal to the regular Christian in the pew, nor is it in line with traditional and orthodox Christian thought as was briefly pointed out in the introduction (cf. Geivett, 2010, 595).  Moreover some scholars imply that this approach isn’t actually Christian at all, in that it leads to “Christless and even atheistic Christianity” (Craig, 2008, 248). Thus those Christians who are not inclined to the bultmanian or tillichian approach will be glad to hear that it is not the only available Christian response to objections against belief in miracles, and nor is it a necessary one.  A second and more palatable approach can be taken which retains the traditional and more orthodox stance.  The Christian and non-Christian alike (Cf. Earmann, 2000) have employed strong counter arguments in response to Hume and Flew, some so confidently and damningly so that it is alleged that Hume’s argument from the intrinsic improbability of miracle is an “abject failure” (Earmann, 2000, 3). 

One of the lines of argument that can be taken in response to Hume is appealing to the probability calculus.  Since Hume wrote ‘Of Miracles’ numerous developments in probability theory have taken place, revealing hole’s in Hume’s reasoning (Craig, 2008, 270).  He is mistaken on a number of points. Firstly, Hume focuses almost solely on the intrinsic improbability of the miraculous, ignoring the wider factors that need to be taken into account when calculating probability (Craig, 2008, 272). This is a mistake.  In calculating the probability of a miracle ‘M’ we not only need to take into account its intrinsic probability given our background knowledge ‘B’ (The nature of ‘B’ will be expanded on later), but also its explanatory power ‘E’.  The calculation is as follows (cf. Craig, 2008, 271),

Pr(M|E&B)                Pr(M|B)             Pr(E|M&B)    

                                                                =                        x

Pr(not-M|E&B)              Pr(not-M|B)       Pr(E|not-M&B)

On the right hand side of the equation we have two ratios.  Firstly, that of the intrinsic probability of ‘M’ (Pr(M|B)) and secondly, to the right, the explanatory power of  ‘M’ (Pr (E|M&B)), the denominators being the probability of their negations.  If the explanatory power of ‘M’ is sufficiently high to be equal to or greater than the intrinsic improbability of the miracle, then ‘M’ may be more probable than not (Craig, 2008, 272).  What is shown by the probability calculus is that Hume’s argument is flawed in that he leaves out significant factors contributing to calculating the probability of ‘M’.  He gets close to understanding this, but ultimately fails to expand on or develop this sufficiently (Craig 2008, 272).

            Secondly, Hume incorrectly goes about calculating the intrinsic probability of ‘M’ by using the frequency model of probability. Craig (2008, 274) paraphrases Earman in pointing this out, “...the frequency model of probability simply will not work in this context.”  If one were to take this theory of probability on board it would undermine a good deal of scientific study as well (Craig, 2008, 274).  Therefore this mistake, in conjunction with the others, shows that Hume largely fails in his argument against religious belief in miracles.

            But what of Hume’s other argument, that the miracles of different religions cancel each other out?  Mackie seems to think that this argument can be dealt with or at least avoided by taking the stance of theological liberalism or pluralism (Mackie, 1992, 15).  Again, this is a viable option, but not preferable to those who are not in favour of a more liberal explanation.   Another way of responding is to point out, along with Craig (2008, 277) that the question is ‘empirical’ and demands investigation as to whether the evidence supports a Christian claim (the resurrection for example) as opposed to a non-Christian miracle claim. Earman (2000, 70) agrees in that Hume had to “...leave the high ground and descend into the trenches where...there were opponents...[who] were prepared to argue for...the superiority of the New Testament miracle stories over heathen miracle stories.” 

What can be said of Antony Flew’s argument against religious belief in miracles on historical evidence?  As it happens, this argument, like those of Hume has also been challenged.  Flew’s argument rests on the proposition that historical evidence is subordinate to scientific evidence on the grounds that “scientific propositions, unlike statements about the past, can be repeatedly verified.” (Palmer, 2001, 192, cf. Swinburne, 1970, 42).  But as Swinburne (1970, 42) points out, “We can go on and on testing for the truth of historical as of scientific propositions”. Both historical and scientific propositions are “established ultimately in a similar kind of way” (Swinburne, 1970, 43). Thus, Swinburne asks why, now, should we assume that scientific propositions take precedence over historical propositions?  “To make the supposition that they are to be treated differently is to introduce a complicating ad hoc procedure for assessing evidence.” (Swinburne, 1970, 43).  Therefore the argument put forward by Flew also fails.

            One can say therefore, that there could be enough evidence of particular kinds, to “outweigh the evidence of physical impossibility.”(Swinburne, 1970, 51).  Nevertheless, “There is, even on theistic assumptions, a strong presumption in favour of a natural explanation for any physical event.” (Geivett, 2010, 601)  Indeed those in favour of claiming that a miracle has occurred bear the burden of proof (Palmer, 2001, 195).  But in acknowledging this, one more factor should be noted in calculating the probability of the miraculous.  The crucial component in considering the probability of ‘M’ will be the intrinsic probability of ‘M’ given background knowledge ‘B’.  Thus the intrinsic probability of ‘M’ is relative to ‘B’.  If one assumes already that God doesn’t exist then the intrinsic probability of ‘M’ occurring will be very low.  However, if one assumes theism, “...and that this theism grounds a reasonable expectation that God would act on behalf of humanity-” (Geivett, 2010, 604, cf. Swinburne, 1970, 65, cf. Lewis, 1947, 4), then the intrinsic probability of ‘M’ will increase significantly.  “For if there exists a being able to produce V [violations of laws of nature] (as a god would be), it is natural to attribute them, if otherwise inexplicable, to his agency.” (Swinburne, 1970, 65).  Thus, a cumulative case could be made for Christian theism on this basis. The Christian could build up a case, by arguing for God’s existence and then going on to apply this in an argument for the occurrence of a miracle, say, the resurrection.  Therefore, not only can the Christian respond that miracles are indeed, verifiable, but they can include arguments from miracles in conjunction with arguments from natural theology to compose a cumulative apologetic (cf. Swinburne, 1970, 69).

It has been seen that numerous philosophical arguments have been made against religious belief in miracles.  It was noted that the ongoing philosophical dialogue concerning miracles originated in the enlightenment period and that Christian and none Christian alike endeavoured to understand miracle claims within the prevailing zeitgeist.  From this background, Hume’s most significant argument arose and was built on by many, the most prominent being Antony Flew.  It was shown that the Christian could respond in two main ways.  The first being to dismiss supernatural miracle claims, the second being to employ philosophical counter arguments to the objections of Flew and Hume in order to preserve the rationality of religious belief in the miraculous.  The second was deemed the best in that it retained the possibility of affirming traditional, orthodox Christian belief.  These counter arguments have been shown to be successful, and can be used positively by the Christian in an accumulative apologetic for his or her worldview as the probability of such claims largely depends on background information.  Therefore, there is good reason to think that this second approach is the best and most successful way for the Christian to respond to philosophical objections to religious belief in miracles.

Word Count: 2107

 Bibliography

Bultmann, R. (1953). New Testament and Mythology. In H. W. Bartsch (Ed.), Kerygma and Myth: A Theological Debate (pp 1-44). London: S.P.C.K.

Craig, W.L. (2008). Reasonable faith. (3^rd ed.). Illinois: Crossway Books.

Lewis, C.S. (1947). Miracles. London : Harper Collins Publishers.

Mackie, J.L. (1992). The Miracle of Theism. (Fifth impression.). New York: Oxford University Press.

Palmer, M. (2001). The Question of God: An introductory sourcebook. London: Routledge.

Earman, J. (2000). Hume’s Abject Failure. New York: Oxford University Press, Inc.

Swinburne, R. (1970). The Concept of Miracle. London: Macmillan and Co Ltd.

Geivett, R.D. (2010). Miracles. In C. Meister and P. Copan (Ed.), The Routledge Companion to Philosophy of Religion (pp 595-605). Abingdon: Routledge.

Hume, D. (1963). Enquiries Concerning the Human Understanding and Concerning the Principles of Morals. (2^nd ed. Impression of 1963). London, Oxford University Press.