Hume, Empiricism, and Miracles

David Hume, Scottish empiricist philosopher, demonstrates in his Enquiry Concerning Human Understanding that he holds belief in miracles to be irrational. He devotes an entire section (Section X) of the Enquiry to the topic. He ostensibly wishes to defend Christianity from those ``dangerous friends or disguised enemies'' (402) who would try to prove its truth through reason alone; though not mentioned by name, it is a safe assumption that he had René Descartes in mind. In actuality, he may be attempting to defend atheism (or deism, or ``agnosticism'') as a rational system of belief. Whatever his true motivation, he considers the question of miracles to be of utmost importance.

Hume's empiricism and his views on miracles

Empiricism generally holds that knowledge comes to us only through our senses. In Hume's case, this idea is taken to its logical conclusion. In particular, since there is no innate knowledge of matters of fact, we can never establish an a priori relation of matters of fact. This restriction is applied especially to the concept of causality (Sections IV, V, VII). Rather than having knowledge of an apodictic connexion between two events, we become accustomed to constant conjunction, and postulate a universal law, the universality of which which has no ground in reality (Sections IV, V). What, then, is human reason? In relation to matters of fact, it is of precisely the same type as that of animals, though more refined (Section IX). ``Reason'', according to Hume, is simply the ability to learn from experience.

Given this, an entire section of the Enquiry (Section X) is devoted to the results of applying empiricism to reports or appearances of miracles. Hume is of the opinion that we should not blindly accept reports of miracles; rather, ``a wise man . . . proportions his beliefs to the evidence'' (388). For events which fall outside of the realm of ordinary experience, there is ``a contest of two opposite experiences'' (390). On the one hand, our past experience may simply lack this new case, and the new experience be true. On the other hand, we may be deluded about the reality of the new experience. We must therefore balance the evidence for and against the veracity of the experience, and hold the belief having more evidence (though the certainty of the belief is diminished by the evidence against it).

Miracles are certainly events that fall outside the realm of ordinary occurrences; and they in fact go much further. By definition, ``a miracle is a violation of the laws of nature'' (391). That is, a miracle violates all past experience. According to Hume, ``the proof against a miracle . . . is as entire as any argument from experience can possibly be imagined''. Experience tells us, with as complete certainty as possible, that a miracle simply cannot occur; thus, experience cannot provide evidence for a miracle which would be strong enough to overcome the evidence against. The only way we can prove a miracle through experience is if ``the testimony be of such a kind, that its falsehood would be more miraculous . . .'' (392).

If the evidence against a miracle is so overwhelming, how does any rational human being come to believe in the occurrence of a miracle? ``Our most holy religion'', Hume claims, ``is founded on faith, not on reason'' (403). Christianity ``even at this day cannot be believed by any reasonable person without [a miracle]'' (404). Thus the true miracle is not in the events that one believes, but rather in the act of believing itself. One is reminded of the famous statement misascribed to Tertullian: credo, quia impossibile est.

Does it follow?

According to Hume's strict empiricism, experience is the only ultimate source of knowledge of matters of fact. We may be able to establish a priori truths, but they are merely analyses of concepts (established through what Hume calls ``demonstrations''), and do not concern questions of fact. As Albert Einstein would say centuries later, ``as far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality''.

From such a strict viewpoint, Hume's argument against miracles is rather firm. Certainly, if an event is completely contradictory to all previous experience, there is a preponderance of empirical evidence against it. Such empirical evidence is, to an empiricist, the only kind of evidence we can have for a matter of fact. Thus, Hume argues, a single event cannot outweigh a law observed in every preceeding event. This is the weakest point in Hume's argument: that an event contradicting an observed law of nature cannot outweigh the law of nature. Were this strictly true, scientific advance would not be possible: an event which contradicts a law of nature would be thrown out, whereas we should instead reconstruct the rule in such a way as to explain the new event.

While this is a salient objection, it is easily dismissed. An event which would be explained by a law of nature must necessarily be repeatable; otherwise, there would not be enough information from which to construct such a law. A miracle, on the other hand, is by its very nature unique; it contradicts every possible law of nature. A miracle is, to Hume, defined as something which contradicts a law of nature---if we could construct a new scientific law to explain the event, it is no longer a miracle. If we accept such a definition, as well as accepting the tenets of Humean strict empiricism, Hume would appear to be correct: a miracle can be reasonably considered true only if its falsehood would be a greater miracle.

The relevance of Hume's argument, then, depends on whether his definition of ``miracle'' is applicable to the events we commonly call miracles. In particular, consider Hume's principal example: the resurrection of a deceased human being (whether the individual had been Christ is irrelevant here). ``It is a miracle,'' he says, ``that a dead man should come to life'' (392). Is it truly a miracle in the Humean sense? According to Hume, it is a law of nature that the dead do not return to life. Yet we have seen individuals, clinically dead, resuscitated by techniques of modern medicine. We have thus changed the definition of ``death'' to exclude these cases. Would it not possible, though, once we fully understand the workings of the human body, to somehow alter a corpse (with nanotechnology, say) in such a way that cellular processes recommence, synaptic communication resumes, and the body is again capable of life? Only if we assume the existence of some non-physical motive force (which an empiricist would be hard-pressed to find evidence for, since our senses only perceive the physical) would this be theoretically impossible.

Hume would argue that such an explanation would strip the event of its miraculous status. Most Christian theologians would probably agree. The resurrections of Christ and of Lazarus are awarded the status of ``miracle'' precisely because they are scientifically unexplainable. If Christ were somehow risen by nanotechnology, his resurrection would not be a miracle, attributable solely to divine provenance. It seems, then, that Hume's definition of ``miracle'' is a relevant one, and his arguments do indeed have ``miracles'', under its common definition, within their scope.

There are, however, ways to be an empiricist without accepting Hume's claim about miracles. In particular, one can argue that sensual experience is not the only source of knowledge. A mystic of the appropriate persuasion would be inclined to argue that divine revelation is a valid source of true knowledge. While it is certainly experience, it is not sensual experience, and may thus have a higher priority than mere sensation. Hume would, of course, argue against such a possibility, probably by calling revelation itself a ``miracle'', subject to the same disbelief. However, revelation of this sort is not necessarily contrary to ``laws of nature''; since experience is solitary and individual in nature, one cannot necessarily argue about the nature of one's own experience by referring to the experience of others. The question is, then, could this be called empiricism? Divine revelation would seem to be even less related to sensual experience than Descartes' ``light of nature''; and it is precisely this ``light of nature'' that empiricism denies. ``Empiricist'', though, is not an exact term: perhaps there is something out there, which could be called ``empiricism'', which would allow for rational belief in miracles. Whatever it may be, one thing is clear: it is not empiricism of the Humean variety.


Hume has set before himself a difficult task---showing that miracles are epistemologically unsupportable---and succeeded quite admirably. In (at least partially) liberating Christian theology from the dogmatic influence of Anselm, Descartes, and the like, Hume helped pave the way for Immanuel Kant's Critique of Pure Reason, where the existence of a deity is held to be a question not answerable by reason alone. Whether or not one accepts the tenets of empiricism necessary for the validity of Hume's arguments, one may certainly see this part of the Enquiry as a tremendous source for further philosophical discourse.