Has David Hume shown that it can never be reasonable to believe in miracles? Part 1

Scepticism about miracles is hardly a modern innovation. But what the eighteenth-century Scottish philosopher David Hume brought to the table was an argument that purported to show that in principle it can never be reasonable to believe a report of a miracle, no matter how credible the testimony. Commenting on the significance of his own argument, which can be found in Section 10 of his book Enquiry Concerning Human Understanding, Hume wrote that,

I flatter myself, that I have discovered an argument…which, if just, will, with the wise and learned, be an everlasting check to all kinds of superstitious delusion, and consequently, will be useful as long as the world endures.[1]

Even if Hume displayed more than a touch of hubris with these remarks, it would be hard to overstate how much impact his argument has had. A great deal of historical scholarship on the New Testament, for example, has proceeded on the assumption that miracles are ruled out from the get-go, and this mindset is due in no small part to the influence of Hume.[2]

Before we go any further, especially for the benefit of any philosophers who might be reading this, I want to be up front about the fact that the interpretation of Hume’s discussion is disputed. By this I mean there are different views of what Hume actually intended to say. So what I’m calling “Hume’s argument” is really an interpretation of Part I of Hume’s discussion of miracles – albeit an influential and longstanding interpretation of it – according to which Hume is trying to show that there can never be enough testimonial evidence to outweigh the improbability of miracle, because of the kind of thing a miracle would be.[3] I’m not going to be talking in this article about Part II of Hume’s discussion, where he contends that miracle reports in practice tend not to be reliable.

Here is a particularly crucial passage from Hume’s discussion of miracles:

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…[4]

What exactly is Hume saying here? Let’s look at this by way of an example: suppose that every time we’ve ever observed any object being released in mid-air (without obstruction) it falls to the ground. In light of this, a scenario in which I let go of a ball in mid-air and it just hovers would be a dramatic departure from what we have always observed in the past. According to Hume, it is therefore extremely improbable that a scenario like that would occur. What Hume seems to be saying, then, is that the more that an alleged event would be a departure from everything we’ve observed previously, the more improbable it is that such an event really took place. A miracle would be an event that — by its very definition — would be a striking departure from all that we’ve observed up until that point. And, given how Hume thinks the probability of an event is determined (the more of a departure from past experience: the more improbable), he concludes that a miracle is maximally improbable: as improbable as anything could ever be.

The other key aspect of Hume’s argument is captured in the following passage:

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. [5]

Philosophers call this Hume’s balancing principle: what this principle says is that it is reasonable to accept testimony to a miracle only if it’s even more unlikely the testimony is mistaken than that the alleged miracle occurred. The commonsense principle that “extraordinary claims require extraordinary evidence” is basically saying the same thing. In a moment I’ll give some examples to illustrate the principle. Hume certainly wasn’t the first person to notice this principle, but it plays a key role in his argument. [6]

In general, when we consider a report of an alleged event, there are basically two factors that we’re weighing against each other. On the one hand, there’s what we can call the intrinsic likelihood of the event: in other words, before we’ve encountered any direct evidence for this event, how likely should we think it is that this sort of event would occur? On the other hand, there’s the direct evidence that the event occurred — for example, someone’s testimony claiming that it occurred, or a photograph that seems to depict the event, and so on.

When we consider whether an alleged event occurred, what we’re doing (at least implicitly) is weighing the direct evidence for the event against the intrinsic unlikelihood of the event. Philosophers sometimes formalize this weighing up procedure mathematically using Bayes Theorem, but I won’t get into that here — suffice it to say that Bayes Theorem is really just a way of trying to make precise and explicit something we already try to do intuitively.[7]  In a nutshell, the idea is that the more intrinsically unlikely an alleged event is, the stronger the direct evidence for it will need to be in order for it to end up being reasonable to believe that the event really occurred.

Let’s look at an example of how this works. Suppose I tell you that my wife had a cup of tea with breakfast this morning. How intrinsically likely or unlikely is that? Well, presumably it’s not at all intrinsically unlikely — English people are well-known for enjoying a cup of tea with breakfast. Accordingly, the amount of direct evidence you need in order to make it reasonable to believe my claim is not very much at all. Even though you know very little about me, my testimony is enough to make it perfectly reasonable for you to believe that my wife had a cup of tea with breakfast this morning.

Let’s consider a second example. Suppose my friend John claims to have seen a man levitating: sitting cross-legged and hovering three feet above the ground. Let’s suppose, with Hume, that it’s intrinsically very, very unlikely that there really was a man levitating (again, because of how much an event of this sort would depart from previous experience of the world). In order for it to be reasonable for me to believe that there really was a man levitating, the evidence supplied by John’s testimony is going to have to be really, really strong: and that’s to say, it needs to be really, really unlikely that John is mistaken (i.e. that he’s lying to me or is sincere but deceived). Or to put it another way, since it’s so intrinsically unlikely that a man was really levitating, the direct evidence is going to have to be overwhelmingly strong in order to outweigh the intrinsic improbability.

So, we can put this altogether and summarise Hume’s in-principle argument against believing in miracles like so:

First premise: Testimony to an alleged event should be believed only if the direct evidence (the testimony) is strong enough to outweigh the intrinsic unlikelihood of the event, i.e., only if it’s more improbable that the testimony is mistaken than that the alleged event has really occurred.

Second premise: Testimony to a miraculous event is never going to be strong enough: it’s always going to be more likely that the testimony is mistaken than that the miracle really occurred, because a miracle would be a deviation from all that has happened before and so would be maximally improbable.

Conclusion: Testimony to a miraculous event should never be believed.

In sum, Hume’s argument against believing testimony of miracles is built on the idea that miracles, by definition, are events that deviate from an otherwise uniform experience of nature. Because such deviations are maximally improbable, no amount of testimony could ever be strong enough to outweigh their intrinsic unlikelihood. This leads to his famous conclusion: testimony to a miraculous event should never be believed.

Not surprisingly, philosophers have had a lot to say about Hume’s argument since its publication, and that’s what we’ll look at in the next article. [8]

References

1. David Hume, Enquiry Concerning Human Understanding (1777), Section X, “Of Miracles”
2. See Section 3.3 of Timothy J. McGrew, “Miracles,” The Stanford Encyclopedia of Philosophy, (2019) Edward N. Zalta (ed.), URL = https://plato.stanford.edu/entries/miracles/
3. I am particularly following the interpretation offered by Alan Hajek, “Are miracles chimerical?”, Oxford Studies in Philosophy of Religion, Volume 1. (Oxford: Oxford University Press, 2008), pp. 82-104. Hajek’s interpretation is affirmed by McGrew, “Miracles,” and is similar to the interpretation given by John Earman, Hume’s Abject Failure (Oxford: Oxford University Press, 2000).
4. Hume, Enquiry, Section X
5. Hume, Enquiry, Section X
6. For more on Hume’s balancing principle, see Earman, Hume’s Abject Failure, pp. 38-43
7. For an example of the use of Bayes’ Theorem in weighing up the evidence for the resurrection, see Timothy McGrew and Lydia McGrew, “The Argument from Miracles: A Cumulative Case for the Resurrection of Jesus of Nazareth,” in The Blackwell Companion to Natural Theology, edited by William Lane Craig and J. P. Moreland, (Oxford: Blackwell, 2009): pp. 593-662.
8. For a comprehensive overview, see McGrew, “Miracles”