Didn't Hume beg the question when he argued that a miracle was the most unlikely explanation for any unusual event?

Hume wrote:

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

But doesn’t this tacitly assume that 'miraculous’ is synonymous with 'most unlikely’? If so, Hume seems to be begging the question.

If we don’t agree with him that miracles are, by definition, the most unlikely thing. Does his argument still work?

Posted: May 23rd 2008

bitbutter www

On at least one reading of Hume’s argument it does seem to beg the question, relying on an assumption about the truth of what it sets out to prove, that supernaturalism is false (see the links at the end of this post for writers who disagree with this assessment).

Whether or not this criticism is legitimate, Hume’s argument is useful when considered in light of the fact that we are all born implicit atheists and those of us who become theists almost always do so through accepting the claims of our parents rather than through investigation conducted from a position of intellectual independence.

Belief in biblical miracles is required to legitimise the metaphysical claims of Christianity. Despite theistic admiration for the irrationality of belief without evidence, I hope we can all agree that only a fool would believe Jesus’ claim that he was the the son of God if he never did any magic. On the other hand though, our daily experience of the laws of nature not being broken means that we’d need a pre-existing belief in theistic metaphysics to license the rational acceptance of Jesus’ miracles.

The two beliefs are co-dependent. Children’s uncritical acceptance of supernatural claims made by authority figures (usually parents) provides one way to get into this loop. But it’s possible to think of situations which would allow an intellectually independent atheist to rationally accept a miracle claim too.

Only extremely good evidence of a Yahweh-authored miracle would be sufficient to overturn the weight of daily non-miraculous experience in favour of Christianity. Bearing in mind the mundane facts of human fallibility, the rampant superstition of the ancient world, and that people lie for personal and political gain, anonymous testimony recorded in a bronze age document doesn’t even come close to the strength of evidence we’d need to rationally accept a miracle claim.

Here are a few ideas about the kind of evidence that would enable an intellectually independent atheist to rationally accept a miracle claim (a prerequisite for accepting Christianity):

  • The witnesses are still living and can be cross examined.
  • The witnesses include known skeptics and atheists who have no previous record of mental illness or dishonesty.
  • There are clear video recordings of the event from different angles which shows no signs of being digitally tampered with.

I included some other suggestions as part of this YouTube thought experiment.

Perhaps Christians feel that it’s unreasonable to demand that an event fulfill these requirements before we can rationally attribute it to a miracle-working deity. But this shouldn’t be a problem. Assemble the most skeptical naturalists you can find in your controlled environment, power-up your measuring equipment, roll the cameras, and ask for a miracle in Jesus’ name. If Christianity is true, Jesus will oblige and you’ll have helped to save hundreds of souls. If the requested miracle doesn’t happen you’ll know that Jesus wasn’t telling the truth or that the bible is not to be trusted. In either case, someone will come away wiser.

Here are some resources that respond to this criticism of Hume’s stance on miracles:

Hume on Miracles, Frequencies, and Prior Probabilities

J.L. Mackie’s Argument Against Miracles

The Miracle of Theism (a book by J.L. Mackie)

A Defense of Hume on Miracles (a book Robert J. Fogelin)

Posted: May 28th 2008

See all questions answered by bitbutter


I don’t think it’s begging the question.

“miracle” is not a specific definition, but merely a label that encompasses a class of claim.

To illustrate, if I said that I drove my truck to work today, you would probably take me at my word.

If I said that I rode my bike 50 miles to work today, then you would probably be looking for some more concrete evidence, before you’d believe.

And if I said I teleported from my garage into my office, how much evidence do you need before you would believe?

The more common formulation is “extraordinary claims require extraordinary evidence”, and I think that’s an easier one to understand.

Posted: May 23rd 2008

See all questions answered by Eric_PK

SmartLX www

We seem to be on a miracle kick here at ATA. First the incorrupt saints, then the Hiroshima church, now a more general approach. I’m enjoying it.

David Hume had more to say about miracles, much of it now common sense. I’m not fond of this particular argument though, for much the same reason as you. To a skeptic it seems self-evident, and the begging of the question only comes from stating roughly the same obvious thing (that miracles are unlikely) twice, but to a believer it’s absolutely glaring. I don’t think it’s given many people fresh doubts.

So here’s another way of putting it. What in the world is less likely than a historical miracle? Even if you postulate that the God of Abraham exists and is capable of miracles, when is it not at least plausible that rather than God suspending the natural order of things to make a special case for someone (after not doing the same for countless others, as in the Hiroshima case) one of the following has happened?

  • One of the multitudes of natural phenomena we do not fully understand, or something we misunderstand, has jumped up and bit us by leaving us with no good natural explanation for the event, when in fact there’s a perfectly good one waiting to be discovered.
  • A religious person or people, who have much to gain from a publicly recognised miracle by their particular god, have influenced the event to make it seem more like a true miracle. This encompasses anything from unconsciously exaggerating the story, to unwittingly overlooking contrary evidence due to confirmation bias, to flat out faking the whole thing.

In all but a few cases like evolution (where biblical inerrancy or some other doctrine is at stake), when there is a known natural explanation for something nobody tries to claim a supernatural cause, even though in principle a supernatural cause could be emulating a natural one. In the presence of a natural explanation, a supernatural one is unnecessary and not even considered. So why is it so different when the natural explanation isn’t yet known? Why not just wait for it, or help look for it, unless you’re actively looking for places to insert the supernatural?

All this won’t matter much if you believe that miracles happen all the time, and that the big ones like unlikely survivals are just punctuation marks in a constant stream of God’s self-expression. But if you think God is actually somewhat discerning about his interventions, you need to look at each miracle with a strong suspicion that no, He probably didn’t have anything to do with this particular one. It’s not worthy of Him.

Posted: May 23rd 2008

See all questions answered by SmartLX

brian thomson www

But doesn’t this tacitly assume that 'miraculous’ is synonymous with 'most unlikely’?

That seems like a reasonable definition of “miracle” to me, if the opposite is “commonplace” or “ordinary”. The dictionary calls it:
bq. an effect or extraordinary event in the physical world that surpasses all known human or natural powers and is ascribed to a supernatural cause.

Note the keywords in there: “extraordinary“, meaning “beyond the ordinary”; “known human or natural powers”; “ascribed to a supernatural cause”. I think we can assume that Hume was also judging “likely” as relative to “known human or natural powers”. (Those have dramatically increased, in scope and power, since his days.)

So I don’t see any contradiction there: a “miracle”, by definition, can’t be “normal” or “ordinary” to be noteworthy at all. If something is inexplicable in terms of “known human or natural powers, that alone does not make it a miracle by that definition e.g. gravity was not fully understood in Hume’s day, but was not a “miracle”, because it is “normal”.

Another way of phrasing it is that something happened “against the odds”, but you then have to question what the odds are, and what that means. To use an example, if someone is shot in the heart, it is likely that person will die: it is possible to make a statement of probability about that person’s chances of survival, based on past experience. Say, for the sake of this example, that the probability is 1% survival.

Does that mean that the 100th gunshot victim, the survivor, is the beneficiary of a “miracle”? Whether you use that term or not depends on your existing religious beliefs, but if you ask the victim, family or doctor(s), what are the odds (!) that they will call that a “miracle” if they did not already believe in “miracles”? (Outside Hollywood, that is.)

We tend to talk of odds in statistical terms, covering large “populations”. What if two out of a hundred such gunshot victims survived in e.g. a month? Is that a miracle for one of them, and if so, which one? No: statistical odds of this type describe larger populations of people, and do not specify what will happen in small populations. (What if none survive? Is that an anti-miracle?) So, what if it is seen, over a long period, that two out of every hundred people with bullets in the heart do survive? Then the odds have improved, for understandable reasons, or may have been inaccurate in the first place.

The media can be incredibly sloppy when it comes to describing probability: you hear phrases such as “one-in-a-million chance of survival” which sounds great, but no-one thinks critically about what that means, or how accurate it is. If that “one-in-a-million chance of survival” comes up… it’s a miraculous escape! Unless it was really a one-in-ten chance, that is..?

A quick coda, now I’ve had more time to think about this question. it seems to me that the original questioner is concerned that Hume is defining “miracle”, in a way that “stacks the deck” against a “God” explanation. My interpretation of Hume is that he was merely reporting the way “miracle” was already defined in common usage. I don’t think Hume had an “agenda” in the way we typically find today.

Posted: May 23rd 2008

See all questions answered by brian thomson


Is your atheism a problem in your religious family or school?
Talk about it at the atheist nexus forum