David Hume's Case Against Miracles

A miracle is an event that is not naturally possible. It is supernatural. The supernatural makes it possible. The alternative is to think that only natural things happen.

David Hume contended that regardless of whether miracles happen or not it is irrational to believe in them. They are not believable.

The Christians look at the wording used by the David Hume. Then they resort to manipulation to try and make it look ridiculous. Their rejection is not based on what Hume meant but on their distortions.

Let us spell it all out.

Hume's Wording

the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined.

Wilful Christian Distortion

He is saying that past experience tells us what natural law is. But this is stupid. Just because nobody ran at x speed before in a race in the past does not mean nobody can do it now.

The True Meaning

He is not talking about unique natural events - every event is unique but about events that are magical such as somebody running a hundred miles in two minutes.

Hume's Wording

Why is it more than probable, that all men must die; that lead cannot, of itself, remain suspended in the air; that fire consumes wood, and is extinguished by water; unless it be, that these events are found agreeable to the laws of nature, and there is required a violation of these laws, or in other words, a miracle to prevent them?

Wilful Christian Distortion

He is being biased - he is saying that a natural explanation is always more likely than a miracle. He does not consider the fact that there might be good evidence for at least some miracles.

The True Meaning

We all know that testimony is not always reliable which is why testimony that the sun will go black tomorrow magically is not as convincing as somebody saying the sun will act as normal.  The sun acting normal happens more than testimony being right.  Hume was using this principle to show that it is irrational to believe in miracles on testimony.

For Hume it is very unlikely that a testimony to a miracle is true.  Even if you think that is too strong, it is still the case that a miracle testimony is unlikely to be true.  And we have to remember that in some cases, testimony need not be believed or disbelieved but just ignored.

Hume is discussing the assumptions we must make before we start considering evidence so that we can believe what the evidence points to.  There is nothing biased about that.  In fact it is what makes him unbiased.  Evidence requires that you make some assumptions about what evidence is.  To recognise that x stole from your purse you have to assume that no alien did it invisibly with some super-science and that no demon made you think x did it.  Evidence by its nature is anti-supernatural.  Evidence is what you get when you assume there is no supernatural interference.  The Christians say Hume is biased and should believe in miracles if the evidence is good enough.  But that is contradictory when evidence is presuming miracles don't happen!The Christians accuse Hume of saying evidence for miracles does not make them credible just because he doesn't want to believe. They lie about his motive.

The Christians do not really believe their objection to Hume. When a criminal says that a witch planted the evidence pointing to him as the bad guy and that he is innocent of any crime, the Christian does not start saying, "Let us see what evidence there is that he is telling the truth." The Christian simply dismisses his claim. The Christian holds that under natural law it must be assumed that he is lying or mistaken.

It is far more important to investigate a miracle claim made by a criminal like that than it is to investigate a miracle claim by some girl that the Virgin Mary miraculously appeared to her. Why? Because an innocent man may be found guilty and condemned for something a miracle was to blame for. Somebody gets hurt.

The Christian is more concerned about wanting to believe that God puts his stamp on Christian doctrine by doing miracles to show the doctrine is true and more concerned about getting people impressed by the religion and amazed than goodness and honesty. The people that run to apparition sites do not go to help the sick - even if they do help the sick. There are sick people in other places and they don't bother with them. It is really the magic and the experience of magic they are after.

The Christian only takes miracle reports seriously when he wants to. The harm done by not investigating the criminal's claim is irrelevant to him. The Christian cares about winning the religious argument more than anyone or anything.

If you investigate a claim that a girl saw the Virgin, then you should investigate the criminal's claim. The latter should be prioritised. Imagine what that would do to the world? Every criminal would be saying that demons or aliens or witches set him up. The implication is sufficient to show why miracles should not be believed or countenanced and should not be sought. They are a licence to complete madness. If the believer does not go that far, then he becomes a hypocrite. He loses any right to criticise the mad miracle beliefs of others that he disagrees with.

But once you permit belief in miracles or endorse it, you have no choice but to take the consequences. You have no right to say that considering the criminal's claim is not an option because it will be disastrous. If the criminal has a point then it is not disastrous but rather very difficult. You cannot argue that it is wrong to give the criminal a sincere hearing regarding the supernatural claims just because it is difficult or dangerous. Claims being difficult or dangerous do not make them untrue.

It is fair to say that even if the evidence for a miracle is convincing it is actually not enough. But surely it's contradictory to say a miracle is both convincing and not? Surely that is biased? Not as long as we see that there can be evidence for things that are not true.

We can admit that the evidence is good for a prince having been turned into a toad but we will still not take that as justification for thinking it really happened. Even religion agrees with us sceptics on that.

There is evidence for miracles that are rejected by the Church that is good as (and often better than) the evidence for the miracles that are accepted. If the miracle believer is honest, then he admits that there is evidence for one alleged miracle that contradicts that for another. For example, two apparitions of the Virgin Mary may give opposite messages. If miracles are signs from God about what religion or version of faith is true then he does not agree with himself!

Wilful Christian Distortion

Hume rejected miracles on the a priori basis as impossible

The True Meaning

The claim that Hume a priori dismissed miracles is totally false.  A priori means on principle.  It accuses him of assuming miracles don't happen any more than cheese is made only of lead.  It suits the Christians to lie about his argument as he is so influential.  All he says is that testimony for miracles is not a good enough reason to believe in them.  And the reason is that people lie and make mistakes and miracles have been reported and believed despite being clearly false.  Testimony is certainly not a reason to take miracles seriously.  He might actually accept some of today's miracles that claim to have the backing of science as genuine!

By the way, Hume's writing on miracles is assumed to have two parts - an a priori part first which dismisses miracles on a point of principle and the second which casts doubt on any evidence for them being good enough.  This is not true.  His entire argument is about how evidence and observation themselves show us that testimony has limited value and is not enough to establish a miracle as credible.  However he does lay out some conditions that make it reasonable to accept that a miracle has happened.  His problem is that these conditions have never been met and the miracles that religion considers essential for belief happened in the distant past.

Wilful Christian Distortion

Hume is biased against miracles

The True Meaning

If he has a bias it is against human nature.  The argument is not about miracles strictly speaking but about how and when human testimony cannot be accepted.  The lie is told to put people off his argument for Christians know fine well that human testimony and how valid it is is a complex and controversial thing and all of us know there are problems.   People reject human testimony even from supposedly reliable people all the time so the person who follows Hume just goes a step further and calls miracle testimony dubious.

The absurd consequences of belief in miracles warns us to beware them. Believe in them only as a last resort. Thankfully we never reach that last resort.  Hume was trying to help us so if people put their bitterness aside and let him speak for himself they might learn something. 


Babbage says you always get independent witnesses to everything even if indirectly. He says that Hume is only thinking of the witness to a miracle as an individual and is not thinking of the witnesses as a group.  If so, then Hume in that case refutes the believability of miracles such as Lourdes which had only one witness, Fatima where only one witness talked to the apparition and it talked back, Banneux, the apparition of Jesus to St Paul, the apparition of Jesus raised from the dead to Mary Magdalene, the apparitions of Gabriel to Muhammad and the apparition of God and Jesus to Joseph Smith. As a sceptic, I'd be happy with that.  It refutes enough even if not all.  There were definitely no independent witnesses of any kind at all to any of this.  Now most accounts of miracles involve one witness.  The majority of miracle claims lacking credibility makes us wonder about the ones that had more than one witness.  The problem is how human nature seems to want miracles to be true too much.  It reduces the multiple witness claims to maybe not probably or certainly.  The lower the number of witnesses and the more evidence or reason to believe they are influencing each other the more that maybe shrinks.

Consistency in a witness or among witnesses is a negative test not a positive one.  It does not mean that what they say is probably true.  It only says, "No problems with consistency so we must now look to see if the reason for the consistency is that the tale is true.  That involves looking for new data."


Bayes Theorem is put up as a challenge to those like Hume who say that miracles may happen but are too improbable for believability.  For Bayes, good evidence shows that an extremely unlikely and unbelievable event is now to be considered probably correct and therefore believable.  It is obvious that Bayes even if some miracles pass the test religion should be scrapping the ones that don't.  Don't hold your breath. Miracles such as Jesus allegedly walking on water are definitely out.  Religion cares neither about Bayes or Hume or if they have a point.

Bayes tells us that we should not consider even a kiss turning a frog into a prince to be impossible. But it is so close to impossible that we may as well call it that. It says science needs to leave the tiniest chink, it is barely there, in case the kiss will do it.

Bayes does fit Hume for Hume like it said that the chink allowing belief in miracles is hardly there at all so it can be ignored. Hume argued that a miracle is indeed so improbable we should not believe but if a testimony to a miracle would need a miracle to be wrong then we should. Like Bayes he saw there are cases where the extremely improbable can be true.

Those who say that Bayes overthrows Hume never apply it to their pet miracles such as the resurrection of Jesus.  Plus if Jesus rose from the dead we need another miracle to keep him alive until now.  Nobody talks about that.  All they do is say Bayes overthrew Hume.  We get a bare statement instead of a case study for each one of their favourite miracles.

The Miracle Loophole backfires

Religion looks for a chink in science to allow for miracles. 

Karl Popper informed us that though you can learn something and confirm it by observation and experience over and over again there might be at least one rare exception. Maybe somewhere a cat has had a kitten fathered by a dog. For that reason you need to admit what it would take to prove your beliefs wrong. Thus all statements are to be testable. To be useful scientifically, a statement has in principle to be open to being shown to being wrong. No matter how sure you are, leave some room for refutation. Anything that cannot be proven wrong is going too far and is about dogma not learning and is about faith not science. Until you are proven wrong then make your generalisations from observations.

So if a man is dead for days and comes back glowing like a sun that would have to be regarded as natural and not a miracle. Science then that has not been discovered by us would be held to be the cause.

It would mean that science then is wrong to say dead men stay dead and that would have to be scrapped.  This is not a chink then after all!

Either science can generalise or it is useless.

The case against magical beliefs is watertight and you must choose science.


Let Hume have the last word, "I have discovered an argument [...] which, if just, will, with the wise and learned, be an everlasting check to all kinds of superstitious delusion". He is not saying all miracles will necessarily be such delusion but seeks to protect us from ones that are.  So far, as far as the evidence goes all miracle claims are in that category.  That is not bias but just what methods for assessing evidence say. 

No Copyright