David Hume's definition of a miracle (refining, paraphrasing): X is a miracle if (and only if) not X is an even greater miracle.

Is this definition problematic in any way?

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.... 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. ~David Hume

  • can you post a quote? i got started on hume's on mriacles, but got distracted by my stupidity.
    – user67675
    Oct 18, 2023 at 5:04
  • 1
    Done! @prof_post. Post edited Oct 18, 2023 at 5:21
  • ok thanks. difficult to answer without linking it to the rest of the text, but feels, ha
    – user67675
    Oct 18, 2023 at 5:39
  • if you want to keep going with it, you might want to consider the 2nd sense of miraculous and wonder if Hume is being cute, given his phrasing of the whole thing "remarkable and bringing very welcome consequences"
    – user67675
    Oct 18, 2023 at 5:42
  • 1
    Hume's passage is not cryptic and does not define miracles, as can be seen from his use of "miraculous" in the supposed defining. As a definition, it would be circular. He instead gives a condition on when a testimony of a miracle (however defined) is to be believed. The condition is that the falsehood of such testimony "would be more miraculous" than what it testifies to. It is also clear from context what Hume's definition of a miracle is. It is a highly improbable event given past experience, perhaps, deemed impossible under normal circumstances. He explicitly refers to probability.
    – Conifold
    Oct 18, 2023 at 10:48

3 Answers 3


Your definition is a contradiction. Given X is an ordered-set element with ordinal number n iff not-X is an ordered-set element with ordinal number m>n, therefore if not-X is an ordered-set element with ordinal number m, then not-not-X = X must be an ordered-set element with ordinal number n>m, therefore n>m>n, therefore n != n.

It's also a mis-paraphrasation of Hume, who describes a sound, although rough, probabilistic heuristic for guessing whether an alleged event transpired, not a definition of any kind.

See the relevant SEP article.

Addendum edit:

Hume does define a miracle earlier in Enquiry: "a violation of the laws of nature." By "laws of nature" he seems to mean the collected knowledge of events which experience has told us are common. I have some problems with that apparent meaning, but that's not relevant to this post.

  • Above me paygrade, your explanation is. Anyway, you see a contradiction which is interesting. Can you dumb it down, your analysis, or perhaps utilizing a different set of ideas? Oct 18, 2023 at 5:29
  • @AgentSmith If something happens, it's the same as it not not happening. If something happening is a miracle only if something not happening is a greater miracle, then something not happening is only a miracle only if something not not happening is a greater miracle, but that doesn't make any sense because something happening can't be a greater miracle than itself.
    – g s
    Oct 18, 2023 at 5:45
  • Yes, thereabouts. not possible that X > X. Anything else? Oct 18, 2023 at 5:51
  • Yes: you can't construct a useful definition or test for being a particular kind of thing if the only test for whether something is a that thing requires knowing whether something else is that thing.
    – g s
    Oct 18, 2023 at 6:16
  • Indeed @g s, indeed! Oct 18, 2023 at 8:16

He's saying that he's only prepared to believe in something out of the ordinary if it seems more believable than the alternative explanations. It reminds me of a program on TV last night which showed four people around a ouija board supposedly communing with a spirit. I was inclined to doubt it, as I thought the alternative explanation- that they were deliberately pushing the planchette around- was much more likely. Hats off to GS, by the way, for his very clever answer.

  • Moat interesting. Oct 19, 2023 at 2:22

It is problematic, but for much more subtle reasons than the one you propose. It is NOT a circular definition--Hume is too cautious and clever to be caught out that easily. It's a heuristic for judgment, similar to Ockham's Razor.

However, taken at face value, it doesn't allow for any novel beliefs. If you reject anything that doesn't fit into your existing framework of beliefs, that framework of beliefs will not be capable of any evolution.

Ultimately Hume's beliefs are founded on an axiomatic empiricism, which, due to his other commitments, he has to treat as self-evident. If you share his core beliefs, his arguments are persuasive. But if, for instance, you have a different judgement of what counts as "more probable" then his heuristic will fail to secure an unambiguous result.

  • What makes Hume's definition problematic? Oct 19, 2023 at 2:21
  • It's not a definition, it's a heuristic, and I've done my best, above, to explain why I think that heuristic is problematic. Oct 19, 2023 at 13:38
  • Ok. Apologies, my bad. I didn't read carefully enough. Oct 20, 2023 at 2:30
  • So miracles are novel events that are incongruent with existing, accepted belief systems. Is this your definition of a miracle? Oct 20, 2023 at 2:39
  • MIne? No, not in the slightest. I'm not personally in sympathy with any of Hume's views. But I do have a lot of respect for him. He doesn't tend to make any easy mistakes. In terms of impact I consider him one of the worst philosophers ever. But in terms of technical skill, he's one of the absolute best. Oct 20, 2023 at 12:09

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .