"Hume Defines Miracles Out of Existence” and “It’s a Circular Argument”

Tapping My Hume Sign #1

For the first part of this series, I’m lumping these two criticisms of Hume together because they’re basically arguing the same thing: Hume is committing a logical fallacy, begging the question, and therefore his argument is invalid. Specifically, Hume either defines miracles as impossible (and therefore they cannot happen) or that he presupposes (somehow) that they cannot happen.

C.S. Lewis is probably the one who popularized this argument. I will take shots at Lewis in a future post.¹ Here, I hope to demonstrate here that this is a bad argument against Hume because it misunderstands him on a (very) basic level.

A Priori vs A Posteriori

Hume elaborates in his Enquiry Concerning Human Understanding (the book within Of Miracles is published) that a matter of fact (a posteriori) was different from a relation of ideas (a priori). Relations of ideas were true by definition (like logical proofs), while matters of fact were subject to falsification (that the sun rose this morning and will rise tomorrow).

Hume’s argument on miracles is not a priori, about definitions. Rather, it’s an a posteriori argument and assessment of miracles and miracle testimony, given patterns about them.

In Of Miracles, Hume himself says that he’s speaking of matters of fact four times. Just to name two of them, he says that experience is “our only guide in reasoning concerning matters of fact,” and when he describes ranges of assurance “from the highest certainty to the lowest species of moral evidence,” he is talking about matters of fact.²

Since When Does Hume Rely Just On Definitions?

Now, a couple quotes without full context may not seem convincing to someone who hasn’t done a tremendous amount of research. That’s fine! As we’ll see in this post, however, any honest look at the structure of his essay reveals that it’s obviously not a priori.

But before going to that degree of detail, I must make a small observation. We could debate whether Hume is, strictly speaking, a logician (I’m not sure?), but he’s certainly not a philosopher of language. He does not make arguments based on the definition of words. Most of his work is informed by his understanding of matters of fact. Indeed, a recurring theme his philosophy is the inadequacy of logic by itself to explain human reasoning (induction) and behavior (the is/ought gap).

I think viewing Hume’s argument or his maxim as a language trick reveals a lack of basic understanding of the kind of philosophy Hume practiced. That by itself should be a red flag for anyone assessing whether the person raising this objection to Hume is familiar with his argument.

Look At the Structure

Even still, the structure and length of Hume’s essay Of Miracles casts doubt on the “definition” and “circularity” rebuttal. If the argument were that simple, why did Hume go on for thousands more words? The answer is that it’s obviously not that simple!

Robert Fogelin put forward the most succinct summary of the structure of Hume’s argument:

Stated broadly, the task of part 1 is to establish the appropriate standards for evaluating testimony in behalf of a miracle of any kind; the task of part 2 is to show that reports of religious miracles have not in the past met these standards. Taking experience as his guide, Hume further concludes that there is no likelihood that they will ever do so. (Fogelin 9-10)³

In this way, the foundation of Hume’s argument is factual. There is no question begging or defining miracles out of existence. Whether or not the argument is compelling to you depends on:

1. How you evaluate facts about probability,

2. Facts about the probability of miracles (i.e that they are unlikely),

3. And how you evaluate facts about miracle testimony.

As far as I can tell, most people agree with the premise that miracles are unlikely and thus the evidence to establish one is high. We also agree that human testimony is fallible. Where Hume gets controversial is how he formulates his probability and simultaneously concludes/infers that all miracle testimony is bad or that miracles are thus impossible given-what-we-know. These are interesting nuances I have written about at the link below.

How To Understand David Hume's Argument On Miracles

Perhaps you disagree with Hume’s legalistic formulation of probability or his language of impossibility. Perhaps you think his use of words like “impossible” or “proof” or “demonstration” are anachronistic and confusing. That’s fine!

But that’s not the same thing as Hume begging the question or defining miracles out of existence!

You Can Prove Hume Wrong

In this way, Hume’s argument isn’t circular because you can change matters of fact that inform it. If matters of fact that inform his conclusion change, they can render his conclusion false.

Most interestingly, it’s somewhat easy to point out the facts that need to be proven wrong:

• If it’s the case that Hume’s standard of establishing a miracle is factually wrong, the argument technically fails. (This is the main way philosophers engage with Hume’s argument, as far as I can tell)

• If it is the case that religious miracles in the past have been attested by consistently credible by witnesses, the argument also fails. (Religious people like to argue this; it’s worth mentioning that, for Hume, it’s not enough to demonstrate a handful of credible witnesses for the miracles you personally like, but a great mass of all miracle witnesses)

• There are other implicit factual assumptions made in this argument (like the reliability of memory and experience) that can also be proven false. (Skeptics may do this, but they don’t really care to rebut Hume’s argument; also Hume may agree with them, but as a mitigated skeptic would say we have no choice but to assume some of these implicit assumptions to function)

The problem is that many of those arguments are hard to make! It’s hard to have a discussion about Bayesianism vs pre-Pascalian probability.⁴ That takes work, and weeks if not months if not years of study. It’s much easier to spend an afternoon or week falsely asserting his argument is circular or a word game.

And so, many people don’t rise to the occasion of arguing against Hume on the factual matters, they instead retreat to defeating a straw man.

As for me, so long as it is the case that “questionably credible witnesses observing singular events and attesting to an event that violated the laws of nature” is less compelling to the neutral observer than “the uniformly observed and inferred laws of nature remained intact and witnesses were mistaken,” I will side with Hume.

1. As his chapter on Hume in his book on miracles is fortunately short, but unfortunately awful.

2. Shout out to John Loftus’s write up, I basically used his passage instead of searching through the essay directly

3. Yes, he says “in behalf” it’s a direct quote. I was going to make a joke about Fogelin being British (may he rest in peace) but he was American!

4. Now, if I’m putting my cards on the table: I think the more interesting discussion is on the merits of Hume’s theory of probability. I think if you presume that Bayesian probability is the best formulation of epistemic probability or that there are factual flaws in Hume’s formulation of the probability calculus, that’s sufficient reason to think Hume’s argument is wrong. I will try to talk about this in a future post, but I think if you grant Hume his assumptions about probability, his argument is fine. What’s more, I think even if it’s a flawed/wrong argument, it can nevertheless be salvaged, as Peter Millican has done.