David Hume's Anti-Miracle Argument Is Poor

David Hume, as you know, wrote on miracles. He didn’t like them. Which is to say, he didn’t believe them. But then Hume didn’t believe we could know any cause. It was he who is greatly responsible for foisting inductive skepticism upon us, a poorly state from which we are only now recovering. He said things like, “We have no reason to believe any proposition about the unobserved even after experience!” His italics, his exclamation mark. That kind of thing did enormous wreckage to science, though science is only now coming to see it.

Given his arch skepticism, it’s no surprise Hume didn’t hold with miracles. But he didn’t have his anti-miracle argument published in his usual manner, preferring to slide it in quietly. It’s said he held back because he was frightened of Christian reaction. There’s some truth in that, but I don’t believe it was the whole reason. He might have pointed to some vague fear as an excuse, but I think he didn’t push it because it wasn’t ready for prime time. It is poorly written, fractured, and not well thought out. Hume was far too intelligent not to have seen this. So I think he sat on the text. See for yourself.

Let’s do our best to steelman his position. No easy task. This, for instance, is a typical sentence: “A miracle is a violation of the laws of nature; and as a firm and unalterable experience has established these laws, the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined.”

Briefer, this is, “Miracles are impossible because miracles don’t happen”. Well, that’s only a small joke, but it is also a fair summary of his belief. We can gather that Hume’s definition of a miracle is a violation of the “law of nature”.

I’m not sure Hume took “laws” in the same literal sense most scientists today do. For them, “laws” are fixed ubiquitous constant unbreakable forces. You must not disobey the “laws”! Or, rather, you cannot, even if you want to. If you think you can, or have, it’s because there was another, deeper “law” that superseded the ones you thought were in force. For modern scientists, “Miracles are impossible because miracles don’t happen because laws” is their argument. The circularity does not bother them. Just don’t ask them where their “laws” come from.

Hume might have taken “laws” the modern way, or maybe in an older way which merely meant any natural, and no supernatural, cause. In any case, he, like modern scientists, did not think claims in which natural causes were “violated” could be believed. And so a miracle, for him, is some effect (change) with a supernatural cause, where the natural cause has been set aside.

That might sound right to you, but it’s only one possible definition. Here’s another. Since Aristotle we have the first mover argument, which shows that any change that is happening, right now, has to be caused, at base and in the chain of causes that comprise the Event, by a First Cause, right now. Not at the start of the universe, which then wends it way forward operating by laws. But every single change, as they happen—right now—has a first mover or cause. This First Cause is necessary, and it allows the secondary causes that science may investigate.

Thus every change in the world is a miracle, of a sort. The world existing moment-by-moment is a miracle. Yet that doesn’t capture the spirit—no charge for these jokes, friends—of miracles. Which are things like a man rising from the dead. Hume: “But it is a miracle, that a dead man should come to life; because that has never been observed in any age or country.” What he should have written is that “because I don’t believe any of the reports of those raised from the dead.” Because raisings from the dead have indeed been observed, and reported in ages and countries. I don’t need to remind you of the Most Famous Case.

Supernatural cause of a change in the world is, for now, a good enough definition of a miracle. Now, besides First Cause, there may be any number of miracles happening here, there, everywhere, which are not observed or recorded. But there is no direct proof of that one way or the other. Hume is only interested in reported miracles. Hume’s real argument is whether or not to believe reports of miracles, and is not a direct argument against miracles. It is epistemological and not ontological (metaphysical).

The plain consequence is (and it is a general maxim worthy of our attention), ‘That 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; and even in that case there is a mutual destruction of arguments, and the superior only gives us an assurance suitable to that degree of force, which remains, after deducting the inferior.’ When anyone tells me, that he saw a dead man restored to life, I immediately consider with myself, whether it be more probable, that this person should either deceive or be deceived, or that the fact, which he relates, should really have happened. I weigh the one miracle against the other; and according to the superiority, which I discover, I pronounce my decision, and always reject the greater miracle. If the falsehood of his testimony would be more miraculous, than the event which he relates; then, and not till then, can he pretend to command my belief or opinion.

You now see what I mean about the writing. Hume was far from careful here, overloading words like miraculous where he meant improbable.

Those following the Class (which should be all of you) know that mathematical notation can help or hinder. Here I think it helps. Followers also know that all probability, with no exceptions, is conditional on the evidence assumed for the proposition of interest.

For Hume, one proposition of interest might be: “The person is lying or mistaken about a claimed miraculous event”. And he wants to calculate the probability of it based on certain assumptions “X”. I.e.

Pr(Person is lying or mistaken | X) .

“X” will be something like: This man reported an event which he claims is supernaturally caused; many previous similar reports have proved to be natural and not supernatural; I myself have only seen a long string of natural events; indeed, the laws of nature forbid supernatural action.

With that evidence X, the probability the report is false is 1. Which means the probability of a miracle is 0. No reported miracle can ever overcome that last clause on “laws”. If you start by assuming miracles are impossible, which is what the clause insists, then all reports of miracles must be mistaken or frauds. Yet if you accepted it, you would be blind to the Lord himself giving you a sign. Nothing can ever bring you to belief. Remove the clause, though, and the probability is not 1, though it might still be close to it because most reports have in fact been false, and most things operate, or appear to operate, by natural secondary causes.

Indeed, that is the position the Church takes. It downplays mundane reports of miraculous events, for the good reason most reports of the miraculous are mistaken (Jesus’ image in burrito wrappers, for instance). It takes great supporting evidence for the Church to officially confirm extraordinary miracles. Ordinary ones, in the statistical sense, like transubstantiation, it confirms daily.

Probability is not act. Decisions are acts. You might decide, like most, the probability a certain event is miraculous is low. That does not necessarily mean you do nothing about it, or do not believe it. That depends on what the miraculous report was, how you view it, and what you might do about it, and a host of other things. An act is decision rule which is a function of the probability (even informally). That is beyond our purpose here. We only want to know whether miracles are possible. A miracle you judge to have non-zero probability is possible. Hume seems to argue the chance of any miracle is zero or as close to zero as you like.

Hume’s proposition of interest is not quite the same as our example. It’s hard to see what it is, though. He said to believe a report of the miraculous the testimony’s falsehood “should be more miraculous” than the event “which it endeavours to establish”. But the event the testimony wishes to establish is the miracle. That is two propositions of interest, not one, with a decision rule lathered on top. And with “miraculous” not meaning miraculous but uncertain. The decision rule is: Believe the event is a miracle if

Pr(Testimony false|X) < Pr(Event a miracle | X).

In other words, if the probability the testimony is false is less than the probability the event is judged a miracle, believe the miracle; otherwise disbelieve. The totality of evidence used to judge both statements is the same, and must be. Which is odd, as we’ll see. About that evidence, Hume said “there is a mutual destruction of arguments, and the superior only gives us an assurance suitable to that degree of force, which remains, after deducting the inferior.” That’s as clear as a bureaucrat’s memo.

He could have made it simple by saying

Pr(Miracle| X & Testimony false) = 0,

which is the probability the miracle is true is 0 given X (whatever that might be) and assuming the witness erred. Which no one can disagree with. And then he could have said

Pr(Testimony false|X) ~ 1 (or even = 1) ,

which says the probability the testimony is false given X is 1 or close to it. Hume spends much acreage in his essay arguing for human frailty in reporting miraculous events, which few will disagree with, in general. Which means, for the sort of X Hume entertained, no miracle could be established: “Upon the whole, then, it appears, that no testimony for any kind of miracle has ever amounted to a probability [i.e. a chance], much less to a proof”. He judged all previous testimonies false, or nearly so. This becomes his X. So any new miracle testimony (he does not witness himself) will only have low probability, or zero.

As with all probability, the real interest is, or should be, the discussion of X. Hume’s X says, at least, that attested miracles were false or illogical and whatever. Even though he has not, as none of us could, investigated them all. He thus judged all, if not false, then of so low a probability to not be worth investigating. Which is incredibly “dogmatic”. Because some claimed miracles are so profound in consequence, that even if their likelihood is judged low, it does not follow that they should be dismissed.

An analogy is psychic abilities. Briefly, I take it all the reports I have seen or read before have been false or flawed. I have investigated many, and found these to be mistaken. I have conducted my own experiments and found them to be false. But I haven't investigated all (though I show how), there are many intriguing reports I have never got to, and I have no proof why such things should be ontologically impossible. When a new claim arises, I therefore judge its probability as low.

Most claims are small and my decision rule is not to investigate. But some are more important, like Rupert Sheldrake's staring experiments, and my rule is that these are worthy of investigation. Even though, like with Hume, I give a low chance it is real, I cannot dismiss it.

As I cannot dismiss all claims out of hand until such a time I have a proof, one way or the other, that such things are ontologically impossible or possible. It would be crankish and stiff-necked to exclude from investigation something ubiquitous in human experience, like psychic phenomena or miracles.

Hume doesn’t have any ontological proof against miracles. He skirts around one with his talk of “laws”. It stands that

Pr(Miracle| Laws) = 0,

for any miracle. That is indeed an ontological claim, but then the burden in on Hume to prove when God cannot intervene, or that he could but doesn’t exist, and that the “laws” can explain themselves. He did none of these things.

Subscribe or donate to support this site and its wholly independent host using credit card click here. Or use the paid subscription at Substack. Cash App: $WilliamMBriggs. For Zelle, use my email: matt@wmbriggs.com, and please include yours so I know who to thank. BUY ME A COFFEE.