Of Miracles is a philosophical essay by David Hume. It was written as part of his book "An Enquiry Concerning Human Understanding" but was omitted in the original edition to avoid religious offence. The essay was published in 1748 and forms part of modern editions of "An Enquiry". It argues that no testimony has yet established the occurrence of a miracle. The essay attacks the argument from miracle testimony.
While Hume provides an interesting foundation for the discussion of miracles and proposes a useful standard of evidence, the essay fails to draw any valid conclusion against the occurrence of miracles, or the validity of miracle testimony in general. The standard of evidence proposed is similar to that used by modern historians: Ernst Troeltsch argued that there must be some analogy between historical events and current events.
Summary of Arguments
Part 1 of the essay proposes a standard that testimony must meet to establish the occurrence of a miracle. Part 2 argues that no current testimony of a miracle meets the standard established in Part 1.
Personal experience is the best standard of evidence, although it is not infallible. Testimony has a lesser evidential weight than direct personal experience. Some cases of cause and effect are probabilistic rather than certain (such as a week in summer is usually, but not always, warmer than the week in winter). Therefore, we proportion our belief to the available evidence. [This is Evidentialism.] An infallible experience would provide certainty. Since this is not available, we weigh the evidence and incline to the stronger conclusion, but with some caution. [Hume argues for Fallibilism.] A belief is a particular conclusion with a certain probability of it being true, depending on the strength of the evidence.
We may evaluate the validity of testimony by comparing it to facts derived from personal experience. Generally, testimony has some value because of the memory of witnesses, their inclination to truth and their avoidance of shame in being caught lying. We disregard testimony of the witness if they are delirious or have a reputation for falsehood.
"There are a number of circumstances to be taken into consideration in all judgements of this kind; and the ultimate standard, by which we determine all disputes, that may arise concerning them, is always derived from experience and observation."
- — David Hume
If there is contradictory evidence, we consider which conclusion has the greater evidence and are inclined toward the stronger but with our assurance diminuted by contrary evidence.
Testimony for events that are contrary to our experience are therefore considered doubtful. Either our expectations based on personal experience are incorrect or the testimony is incorrect. However, our personal experience has given some assurance that the testimony is valid and at this same time given us an expectation that implies this testimony is false. As before, we proportion our belief to the evidence.
A story may be so contrary to our expectations that no argument from authority could support it. Also, some events are outside our experience but are not miraculous.
"A miracle is a violation of the laws of nature [...] a firm and unalterable experience has established these laws"
- — David Hume
[R. F. Holland argued the definition of a miracle being contrary to natural laws does not have to be accepted and miracles may be defined to include fortunate coincidences.  Hume's essay rules out fortunate coincidences as miracles. However, Hume's argument holds within his own definition. Fortunate coincidences may have been excluded by Hume because they are not particularly convincing when used to argue for the existence of God, which is Hume's ultimate target.
Earman pointed out that Hume's argument doesn't flow from the definition he established.  This lack of clarity in terminology is a valid criticism. It is good practice in an essay to properly define what is being discussed. However, the specific definition is actually of less importance if a property of miracles is that they are contrary to our expectations based on personal experience. The essay makes most sense if this is taken as axiomatic. Not all people use the term miracles in this way or agree with Hume that they are uncommon but assuming this axiom is sufficient for the argument to proceed. If Hume had said he was discussing "fantastical myths" rather than "miracles", this may have avoided some confusion.
Although a stretch, Hume might mean "the laws of nature" in the sense of "expectations based on personal experience" rather than "the principles governing the universe". This would make this otherwise jarring definition consistent with the rest of the essay.]
"Nothing is esteemed a miracle, if it ever happen in the common course of nature. [...] There must, therefore, be a uniform experience against every miraculous event, otherwise the event would not merit that appellation."
- — David Hume
[Readers might consider disregard Hume's various definitions of miracles and consider this as a starting axiom.]
"And as a uniform experience amounts to a proof, there is here a direct and full proof, from the nature of the fact, against the existence of any miracle; nor can such a proof be destroyed, or the miracle rendered credible, but by an opposite proof, which is superior."
- — David Hume
[This is perhaps the most controversial section of the essay. Hume claims we have "uniform experience" of the non-occurrence of miracles. George Campbell  and C. S. Lewis  argued that our experience is not uniform because testimony exists for the occurrence of miracles. However, Hume is referring to the uniformity of personal experience, which he defined as to exclude miracles.
Hume claims we have uniform experience and appears to claim that this is "proof" against the existence of miracles. Because of this, Hume has been accused of a hasty generalization by John Earman, in that he apparently rules out miracles based on a limited set of personal experiences (what Earman refers to as the "straight rule of induction").  However, when Hume refers to a "proof", he does so in a fallibilist context and could be understood as "strong but potentially fallible evidence" (which is the maximum weight he would admit to any evidence). In this context, a "superior proof" may trump a lesser proof. This quote above is therefore a strong but not definitive argument against the existence of miracles, until such time as the appearance of a "superior proof" of their occurrence. This places the burden of proof on the person claiming a miracle occurred. The possibility of a superior proof is the main topic of Part 2 of the essay.
If Hume had employed the straight rule generalisation, as suggested by Earman, Hume would have treated the probability of a miracle as zero and he would not have needed to consider the validity of testimony. Since Hume repeatedly balances direct experience with testimony, this is clearly not Hume's intent.
Levine write that Hume "appears to argue" that we can rule out testimony of miracles a-priori.  Fogelin said this would be a "gross" misreading of Hume.  Hume's intent is for evidence to be considered before coming to a conclusion (particularly the a-posteriori fact that human testimony is relatively unreliable). Even if Hume's argues for the improbability of miracles and we reject his argument because it begs the question, this does not affect the essay's overall argument regarding the validity of testimony regarding miraculous occurrences. In summary, personal experience largely suggests the non-occurrence of miracles but this is obviously not definitive.]
"A miracle may be accurately defined, a transgression of a law of nature by a particular volition of the Deity, or by the interposition of some invisible agent.'"
- — David Hume
[It is curious that Hume includes multiple definitions of a miracle, however the cause of a miracle is of no consequence to Hume's argument. ]
"That 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"
- — David Hume
We can believe a miracle more than likely occurred IF the probability of the testimony being incorrect is less than the probability of the miracle occurring. In the case of contradictory evidence, I base my belief on the balance of the evidence.
[The conclusion of the chapter is that miracles are uncommon occurrences and that Evidentialism may be applied to both human testimony and personal experience. A critic of Hume might point out these are also his axioms and his argument has not made much headway!
Based on the standard established in Part 1, no testimony has yet been established a miracle as a likely possibility. This is because:
- No one has had enough integrity, good-sense, education and ability to establish their testimony beyond all doubt.
- We are fascinated by stories of miracles and gossip, regardless of truthfulness. People also spread stories of false miracles from zeal and good intent. The witnesses of such miracles may have their rationality compromised by rhetoric and their curiosity. There are many instances of false miracles motivated by vanity.
- Miracles are more commonly recorded in "ignorant and barbarous" circumstances but are less commonly asserted when people are educated and have effective communication. Sceptics of ancient times did not have the resources to refute miracles (except for occasional instances such as the case of Alexander of Abonoteichus) but we have better means to detect frauds in modern times.
- Many religions claim supported from miraculous occurrences. It is not possible for them all to be correct since the proof of a certain God disproves the others. Therefore the argument from miracles for the existence of deities results in an inconsistent conclusion.
[The argument from miracles also does not establish which God was responsible or his attributes. In principle, a single religion could be vindicated by reliable testimony of miracles but this has not yet occurred. The evidence for the miracles of each religion are largely comparable and if we lower the standard of evidence, we allow many contradictory examples to be admitted.]
There are several miracles recorded by generally reliable historians including:
[Hume assumes that the reader is willing to reject the alleged miracles he cites and does not make a great effort in rebutting them. His point is that the testimony for these cases are relatively strong but we are still willing to reject them.]
- Roman emperor Vespasian cured a blind man and a lame man in AD 70. The event is recorded by Tacitus  and Suetonius. The circumstances of the event and its recording ought to lend it some weight.
"To which if we add the public nature of the facts, as related, it will appear, that no evidence can well be supposed stronger for so gross and so palpable a falsehood."
- — David Hume
- Jean François Paul de Gondi's account in his Mémoires of the door keeper of Saragossa Cathedral having regained a leg. 
"He [Gondi] considered justly, that it was not requisite, in order to reject a fact of this nature, to be able accurately to disprove the testimony, and to trace its falsehood, through all the circumstances of knavery and credulity which produced it."
- — David Hume
- Many miracles witnessed by the tomb of François de Pâris around 1731. Although they were witnessed by many reliable people, "the absolute impossibility or miraculous nature of the events" is a sufficient refutation. There has even been ridiculous comparisons between these miracles and the miracles recorded in the Bible.
[It might have been clearer for Hume to say the events are contrary to our experience, rather than appear to discount miracles a-priori. His hyperbole plays into the hands of his critics. The distinction drawn between biblical and recent miracles may have been a tactic to avoid offence or possibly Hume being ironic, rather than Hume avoiding supposedly strong evidence of miracles.]
" It was at last observed [of the miracles of the tomb of François de Pâris], that where men are heated by zeal and enthusiasm, there is no degree of human testimony so strong as may not be procured for the greatest absurdity"
- — David Hume
The Molinist authorities, not wanting to undermine human testimony for miracles, claimed the miracles were the work of the devil.
- In Racine's history of the convent of Port-Royal, he records that the niece of the famous Pascal was healed of an fistula of the eye when touched by a sacred relic.  This example is little known but far more appealing than more famous miracles because of its connection to celebrity.
[Hume recommends this last miracle as far more persuasive propaganda, probably in irony.]
Some historical accounts are credible but not all testimony is as reliable. Suppose two enemies at war had both claimed victory in a number of battles, it would be difficult to establish the truth. This is similar to the case of miracle testimony by Greek or Christian writers. [Both cases cannot be true.]
The bias of testimony should be considered and caution exercised if it is motivated by:
- their natural inclinations or propensities or
- self delusion.
False testimony is often encouraged by the general population who are eager for stories of superstition and wonder. It is often difficult to verify testimony of the miraculous directly, particularly if it occurred in a remote location. Often, wise people do not see the need to falsify a claim of a miracle and the evidence against it is not preserved. No means of detection remains apart from their testimony (and our experiences).
"Upon the whole, then, it appears, that no testimony for any kind of miracle has ever amounted to a probability, much less to a proof; and that, even supposing it amounted to a proof, it would be opposed by another proof; derived from the very nature of the fact, which it would endeavour to establish. It is experience only, which gives authority to human testimony; and it is the same experience, which assures us of the laws of nature. When, therefore, these two kinds of experience are contrary, we have nothing to do but substract the one from the other, and embrace an opinion, either on one side or the other, with that assurance which arises from the remainder. But according to the principle here explained, this substraction, with regard to all popular religions, amounts to an entire annihilation; and therefore we may establish it as a maxim, that no human testimony can have such force as to prove a miracle, and make it a just foundation for any such system of religion."
- — David Hume
[Hume attempts to establish a maxim to disregard miracle testimony:
- The most reliable testimony in existence has only amounted to false miracles,
- from (1), no testimony that meets the standard of evidence specified in part 1,
- and we will not have better testimony in future,
- from (2) and (3), we can generally discount testimony of miracles as insufficient.
Point (3) has the least justification in the essay and may be a hasty generalization of point (1). Although modern times have better ways to record events (e.g. using video, photographic and forensic methods), we also have advanced methods for faking evidence (such as computer generated graphics).
A more limited argument would be:
- Testimony must meet the standard in part 1 to be considered as probable.
- Examples of miracle testimony given earlier are recorded on relatively good authority but from (1) are clearly improbable.
- From (1) and (2), we can discount any further testimony unless it has greater evidential weight than the testimony already considered.
This essay uses probability intuitively and uses the term "substract" in an analogical sense. It does not mention Bayes' theorem, which is fundamental to modern probability theory. A mathematical approach to miracle testimony was taken by Condorcet and Babbage, however it is difficult to mathematically account for the many factors that affect the strength of evidence. ]
There is insufficient testimony of miracles to argue for the existence of God on this basis. Historical accounts of extraordinary events are acceptable (such as a hypothetical, unexplained darkness for 6 days) if the testimony is very consistent, extensive and from multiple independent sources.
[Hume is often criticised and satirised for allegedly ruling out historical events such as the existence of Napoleon.  However, Hume is not a sceptical extremism and clearly states the standard testimony must meet. The existence of Napoleon meets this standard and Hume argues that miracles do not. Hume discusses scepticism in more depth in Dialogues Concerning Natural Religion, particularly in Part 1.
Some critics claim that Hume rules out all belief in unlikely events, such as the result of a lottery.  This is a false analogy because any particular lottery result is relatively likely compared to a miracle such as rising from the dead. Also, the reporting of lottery results is relatively reliable and lacking in bias. In this case, the evidence on each side is comparable and justifies a suspected winner contacting the lottery to attempt to claim the prize.]
If historians state that queen Elizabeth I had died and come back to life after one month, I would find it more likely the story originated in the knavery and folly of men, rather than an actual occurrence that is so contrary to common experience. If this miracle was used as an argument for religion, it would still be rejected and perhaps even more so because false testimony of religious miracles is very common. Francis Bacon adopted a similar suspicion of testimony from religion, magic or alchemy.
The reasoning in this essay protects Christianity from its dangerous friends and disguised enemies. Religion is better founded on faith rather than evidence. If we examine the Pentateuch [i.e. the Torah or first five books of the Old Testament] which was written in a barbarous time, long after the events, with no corroborating evidence, we see it gives an account that is very different from the present world. Therefore, it is more likely these stories are false than they actually occurred.
[No wonder Hume is considered an atheist. Biblical literalism is ruled out by Hume because of the stories of fantastic miracles.]
The argument in this essay also applies to testimony of prophecies.
"[The] Christian Religion not only was at first attended with miracles, but even at this day cannot be believed by any reasonable person without one. Mere reason is insufficient to convince us of its veracity: And whoever is moved by Faith to assent to it, is conscious of a continued miracle in his own person, which subverts all the principles of his understanding, and gives him a determination to believe what is most contrary to custom and experience."
- — David Hume
[Hume defends belief in Christian miracles but continues to attack any rational basis for this belief. This contradicts his earlier evidentialism and is a form of special pleading. It is likely the last paragraph was an attempt to defuse Christian hostility to this essay.]
- Michael R. Licona, Jan G. Van der Watt, Historians and miracles: The principle of analogy and antecedent probability reconsidered, HTS Teologiese Studies / Theological Studies; Vol 65, No 1 (2009), 6 pages. doi: 10.4102/hts.v65i1.129 
- John Earman, Hume's Abject Failure: The Argument against Miracles (Oxford: Oxford University Press, 2000) 
- R.F. Holland "The Miraculous". In American Philosophical Quarterly 2, 1965: pp 43–51 (reprinted in Swinburne)
- George Campbell, A dissertation on miracles, p. 31-32, London: T. Tegg, 1824 
- C.S. Lewis, Miracles, (New York Macmillan Publishing Company, 1960), 102.
- Robert J. Fogelin. A Defense of Hume on Miracles, Review by Richard Otte 
- Hume and the Problem of Miracles: A Solution By Michael Levine, p. 13
- Marcello Truzzi, On the Extraordinary: An Attempt at Clarification, Zetetic Scholar, Vol. 1, No. 1, p. 11, 1978
- The Complete Works of Tacitus, tr. Church and Brodribb, 1942, Part 4, para. 81.
- Jean François Paul de Gondi, Mémoires [Memoirs], in Œuvres, ed. Marie-Therese Hipp and Michel Pernot (Paris: Gallimard, 1984). EHU Sect. 10.26.
- Mary Shelly, Lives of the Most Eminent Literary and Scientific Men of France
- Timothy McGrew and Lydia McGrew, The Reliability of Witnesses and Testimony to the Miraculous, in Probability in the Philosophy of Religion, 2012, eds. Jake Chandler and Victoria S. Harrison
- Historic Doubts Relative To Napoleon Buonaparte by Richard Whately 
- Bruce Langtry, "Hume, Probability, Lotteries and Miracles", Hume Studies Volume XVI, Number 1 (April, 1990) 67-74