Mackie and Hume on Miracles

The first chapter of The Miracle of Theism, is entitled Miracles and Testimony. It is a discussion of David Hume's argument against miracles. Hume made the argument that no miracles could be believed by any rational person, unless the denial of the miracle's occurrence would take an even greater miracle, and even then the miracle would be believed in very tentatively. After much discussion Mackie states Hume's primary argument like this:

• "Where there is some plausible testimony about the occurrence of what would appear to be a miracle those who accept this miracle have the double burden of showing that this event took place and that it violated the laws of nature" (pg 26)
• A: If it the miracle violates the law of nature then it is immensely improbable that the miracle occurred.
• B: If the event took place, but did not violate the laws of nature then it can be explained in natural terms.

Of course there are some assumptions that Hume makes; that one should proportion one's belief to the evidence; and that a miracle is a violation of the laws of nature. So if a miracle is to be defined in other ways one can probably avoid this argument. I take it that the former assumption is uncontroversial.

Now it doesn't seem to me as though B is at all problematic. If a miracle is to be used as evidence for something supernatural, pointing to something that could very well be natural does not constitute as good evidence for the supernatural. However A appears to be a bit more tricky, at least it needs some reinforcement as to why a miracle would be as immensely improbable that they are. To determine how unlikely any event x that violates a law of nature is it will be important to define what a law of nature actually is. Mackie says that "the laws of nature...describe the ways in which the world...works when left to itself, when not interfered with. A miracle occurs when the world is not left to itself" (pg 20). Now a miracle that violates the laws of nature is extraordinarily improbable because we have empirical evidence that what the laws of nature describe what nearly always, if not always, obtains. Laws of nature describe not only the way the world works, but the way it is. That means that in order to go about witnessing the falsifying of the laws of nature one would have to climb quite a tall mountain of probabilities. However laws of nature could be indeterminate, in other words they may not describe what always obtains but what the chances are of something obtaining. And this highlights the dirty fork of the argument, if it is the case that laws of nature do not always obtain then something which may appear to be a violation of a law of nature may actually not contradict nature at all. I have the feeling that what this leads to may be to strong of a conclusion, however for the most part this argument appears quite sound (though Hume has been challenged and supported by Bayes Theorem recently, this is probably the realm where the real arguing about this argument will take place.)

Here is a helpful discussion on this topic: http://philosophicaldisquisitions.blogspot.com/2011/02/miracles-series-index.html