Dawkins v. Fatima

In my previous post, I quoted from Dawkins' Unweaving the Rainbow: Science, Delusion and the Appetite for Wonder, as follows.
It is the 70,000 witnesses that impress. Could 70,000 people simultaneously be victims of the same hallucination? Could 70,000 people collude in the same lie? Or if there never were 70,000 witnesses, could the reporter of the event get away with inventing so many?
Let's apply Hume's criterion. On the other hand, we are asked to believe in a mass hallucination, a trick of the light, or a mass lie involving 70,000 people. This is admittedly improbable. But it is less improbable than the alternative: that the sun really did move. … If the sun moved in truth… an even greater miracle would have to have been perpetrated: an illusion of non-movement had to be staged for all the millions of witnesses not in Fatima. And… if the sun had really moved at the speed reported, the solar system would have broken up. We have no alternative but to follow Hume… and conclude… that the miracle of Fatima never happened. Moreover, it is not at all clear that the onus is on us to explain how those 70,000 witnesses were misled. [My bold emphasis.]
 I mentioned the logical flaws in such arguments, so I had better explain myself.

Let us conduct a thought experiment. In order to determine the way a large crowd of people experience the sun, we arrange to conduct tests at a number of venues at which crowds of from 50,000 to 120,000 people gather. Candidates for such gatherings spring readily to mind. With the co-operation of the organisers of these gatherings, we ask the attendees to glance towards the sun, while reminding them of the dangers of doing so. We further ask them to report to one of the many data collectors present, whether they noticed any peculiar behaviour of the sun. What expectations would we have? Apart from being sued by a large number of people because of the damage they sustained to their eyes in following our instructions, we can confidently expect that at none of the venues would any untoward phenomena be reported of the sun.

If, however, at the Portugese national soccer championship, 70,000 people suddenly cried out in terror as the sun seemed to tear itself from the heavens and hurtle towards the earth, how would we respond? Could we repeat what Dawkins has said above, and conclude that it never happened? Let's be clear about it. In this case, as in Fatima, it is that 70,000 people saw the sun behave in inexplicable ways, ways contrary to the "laws of nature." We might argue, with Dawkins, that the failure of others away from the venue to observe these phenomena means the the sun did not actually change it's behaviour at that time. However, would Dawkins himself dare to argue that it is not at all clear that the onus is on us to explain the phenomena? I very much doubt it. Some hypotheses would have to be offered. But how could they possibly be tested, especially if, across all of experimental venues, this turned out to be a unique event?

Unique events are always problematical for science. Scientific methodology is most comfortable with the observation of interchangeable subjects and controlled conditions. As either of these reduces, the problems of drawing generally valid conclusions multiply. Unique events in uncontrolled circumstances—the stuff of most of our lives—most of the time defy any but serendipitous observation. Why, for example, do we hold no expectation at all that the above thought experiment would yield such a result?

Our lack of expectation tells us nothing about the likelihood of the events at Fatima, because of the extra dimension, inapplicable to our experiment. That dimension is is the supernatural intervention of the Virgin Mary. My point in proposing the experiment is to demonstrate the different criteria that are applied to events thought to be naturalistic, and to those thought to be supernaturalistic. Why two standards for the judgement of the evidence?

The miracle of the sun at Fatima was witnessed by so many because it had been predicted by the children.  Mary, in the preceding apparition, had promised that some such event would occur on that date, and at that time. A falsifiable prediction was made, based on the hypothesis that the Blessed Virgin Mary was the one who was appearing, and that she, through her Son, had the power to initiate such events. Dawkins, as a scientist, should be very pleased with such a set of conditions, even if he is incensed by the success of the prediction.

Hume's criterion, which to Dawkins seems...unassailable, he quotes as follows:
…no testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous than the fact which it endeavours to establish. 'Of Miracles' (1748)
Hume is talking about testimony, because his primary aim is to destroy confidence in the testimony of the Bible, and particularly, though not exclusively, the New Testament. Whereas the miraculous events of the OT occur in the Law and the more ancient histories, the NT is the witness, the testimony, of the contemporaries of Jesus, and Christianity was founded on this testimony. What he is demanding is that the testimony to, say, the Resurrection and subsequent appearances of Jesus, be of such a kind that it would require a miraculous revelation even more astounding than the Resurrection itself, to engender disbelief.

Is this feasible? Of course not. All human testimony carries a burden of uncertainty. Furthermore, supernatural events, by definition, are not repeatable, so no testing can be carried out ex post facto to reproduce the reported event or events. In that, it's like the unrepeatable moments of our lives. It is the observed stability of many classes of events in the natural environment that gives us the confidence to repeat an experiment, on the understanding that the unrepeatable components of the new event do not detract from the common element that we abstract from the experimental reality.

Fatima, though, is a event that comes perilously close, for sceptics, to fulfilling Hume's criterion. Surely the falsehood of the testimony of 70,000 witnesses would be miraculous? Dawkins accepts that, but fails the witnesses on the test of comparative miraculousness. But how many witnesses would it take to convince Dawkins? Had the whole of the daylit human population witnessed the event, it would not have been enough, because "the solar system would have broken up."

That is to deny the nature of the demonstration that was Fatima. What drew 70,000 to Fatima was the promise that the Trinity would act in support of Mary's appearances and messages to the children. They who created the natural universe, space and time, would act on their creation so as to demonstrate its dependence upon them.

Another word for testimony in this context is evidence, and when it suits his purposes, as it frequently does, Dawkins attacks everyone who does not share his worldview, as having no evidence to support their contentions. Here, though, Dawkins reveals that he simply disregards evidence which is contrary to his philosophical presupposition of exclusive materialism. For this criterion is nothing but a statement of a deeply held a priori philosophical commitment.

Hume's test is prima facie anti-scientific. Nonetheless, it is the way most scientists proceed. They draw conclusions based on their own judgements of what is feasible and infeasible, reasonable and unreasonable, possible and impossible. Consequently, observations which wildly contradict prevailing scientific theories are dismissed as freakish outliers. Only if such observations persist might they begin to trouble the scientific conscience. Scientific consensus, however, can be a very unscientific thing.

UPDATE 2nd March 2012
UPDATE 21st March 2012
UPDATE 16th November 2012
UPDATE 16th October 2014