I wonder if we could make an argument from the number and nature of Christian miracle claims (and perhaps even miracle claims which are not peculiarly ‘Christian’), to the implausibility of all such miracle claims being false. The naive notion that in order to believe an extraordinary proposition one needs extraordinary evidence is answered in probability theory by saying that one is justified in believing in something unlikely (even incredibly unlikely) so long as it is even more unlikely that we would have the positive evidence we do have if it were not the case. Thus, for example, one can be justified in believing that they have the winning ticket for the lottery if and when their numbers turn up, so long as they have good reason to believe they aren’t being punked. Similarly with the apologetic for the resurrection, we are justified in believing that Jesus rose again from the dead, however unlikely it would seem, precisely because it is even more unlikely that we would have the evidence we do have if he had not been raised. So, what of a probabilistic argument from miracles?
Such a suggestion perhaps wouldn’t be easily accepted. We often hear people say “there is not a single verified miracle” or “there has never been a single verified miracle”, which is of course just rhetoric. For all the examples of miracles that we do have, the antagonist will simply say that it wasn’t ‘verified’ to their satisfaction. If we say, as we can, that it was verified by scientists and/or medical doctors, the antagonist will simply say that these scientists or doctors weren’t doing their job right, or they didn’t have enough time to be thorough, or else she (the antagonist) will accuse those scientists or doctors of being charlatans, or worse -persons of religious faith. However, her criteria for determining whether such instances of ‘verification’ qualify as verification by antagonistic standards are ideologically determined: since miracles do not happen, if it was observed once then it was likely the fault of the observer, and if it can be repeated, then it simply becomes a new datum which our scientific theories have to account for. For the Naturalist, no miracle can in principle be verified to satisfaction – the instance they begin to put forward relatively realistic criteria for verification they are met with innumerable accounts of ‘miracles’ which have been verified in just those ways. From Fatima, to the resurrection of Christ, contemporary miracles along with miracles in the past, we have the best empirical evidence one could have hoped for.
Now, how could one go about actually putting numbers in to quantify over all miracle claims, rather than just one? When one uses Bayes theorem with just one miracle (the Resurrection), there are already problems or controversies when it comes to plugging in the background information. However, supposing one were to use Bayes’ theorum on a number of different miracle claims (including Fatima) – I anticipate some problems. Of course, the selection of miracle claims would have to be blind (one cannot simply take the most convincing cases) in order for it to be made an argument for miracles in general, one would need a sampling of miracle-claims considerably large, and then we would have to ask what the probability is that we would have the evidence we do for each if it were not truly miraculous. Following this we would have to quantify over all our sample cases and come out to some probability that we would have all the evidence we do if not a single miracle had occurred. The cumulative probability from a random sampling could work, but of course it is difficult to know what criteria we should use for determining which cases we can select from (how much background information do we have, etc). Technically, we wouldn’t really be checking for miracles directly, but rather paranormal events which have theological significance (thus, we would not take into account UFO sightings or such things, and instead would stick to those events which are recognized by the religious to be miraculous).
In any case, it would be interesting to see if this could be done, and if it could it would be interesting to actually do it. We would need a way to select from a pool of miracle claims, select a large enough sample, plug in the probability of each being authentic, and then quantify over all of them to find the probability that not a single one would be authentic.
In the meantime, I think any honest person of goodwill who exposes themselves to the very many very convincing claims of miracles will find two curious things about them: First, their nature and frequency seem to lead to a tacit recognition that there is something different about Christianity (and especially Catholicism). Second, miracle claims are too many, too sincere, and too surprising to be the result of religious imagination. For instance, I recommend that all interested persons take a listen to this discussion on miracle stories, on the podcast/radio-show Unbelievable?