Should I believe in miracles?

29 Aug 2022

Could one ever be justified in believing that a miracle has taken place? To answer this, we must first examine the nature of justified belief, which I define as having true reasons that support the belief in question. These reasons must consist of evidence. I will consider testimonial, scientific, or philosophical evidence to be a sufficient requirement for attaining justified belief in miracles. The aim of this essay is to demonstrate that it can be reasonable to believe in miracles. In order to establish my argument, this analysis will critique the objections of David Hume.

There are many different definitions of a miracle [1]. For the sake of this essay, we will use David Hume’s ‘transgressive miracle’ definition. Hume writes ‘a miracle is a violation of the law of nature’, where a ‘law of nature’ constitutes that which is commonly observed by human experience. [2] A miracle is by definition a rare event. It defies what humans recognise as unalterable patterns within the universe. For example, someone rising from the dead would be counted as a miracle, since resurrection is not commonly observed. A transgressive miracle is an event that acts contrary to common observation.

By analysing Hume’s objections, I will present a defence of miracles by demonstrating the following propositions to be true. These propositions independently do not prove miracles, however, using each proposition to create an accumalative case, one could hold a rational and reasanable belief in the existence of miracles.

It is because these propositions are true, one can be justified in believing that a miracle has taken place.


Before I establish my three premises, in part two of his paper, Hume makes four preliminary arguments that may undermine my case [3]:

a) No miracle has been reported by a sufficient number of educated persons.

b) Men are inclined to believe in the miraculous.

c) Miracles are only found amidst barbaric and uncivilised people.

d) No number of witnesses are sufficient to testify that a miracle has taken place.

Hume uses these claims as accumulative evidence for his conclusion, ‘the christian religion… cannot be believed by any reasonable person’. [4] Hume’s propositions cannot only be proven false, but they do not support his conclusion. I will briefly demonstrate the insufficiency of claims (a)-(c) with assistance from contemporary christian philosophers, leaving my refutation of preposition (d) for my main argument. [5]

Firstly, proposition (a) states that the people who report miracles are uneducated, and few in number. This is not true, especially in terms of the christian religion. For instance, the Acts of the Apostles and the Gospel of Luke contain a large number of biblical miracles. Both books were written by a man called Luke who was a practicing physician and doctor; an educated man who did not grow up with a religious upbringing. Luke is simply one example of an educated witness. He is not the only man who wrote in detail about miracles. There are many educated first century writers. Matthew was a Tax Collector highly literate in multiple languages and likely formally educated. Paul was likely the most educated I have mentioned so far: a scholar on the Hebrew bible, whose handwriting could be mistaken for a proffesional scribe. All these writers likely had formal education. While one may doubt the content of these testimonies, it is vain to suggest the reason for doing so is because the writers do not hold to Hume’s unexplained criteria of being educated enough. Furthermore, the reason Christianity expanded so rapidly was due to the eye-witness testimony of people from Galilee, Judea, Jerusalem, Idumea, Tyre, Sidon, Samaria, and many other places [6]. Many miracles were witnessed by large crowds exceeding 5000 people. [7]. To claim that the thousands of people who came to witness miracles were all uneducated is simply unjustifiable. Therefore, proposition (a) does not apply to the christian religion.

Secondly, while proposition (b) is somewhat truthful, it is needlessly pointed at this discussion. The opposite is also true; miracle sceptics are disposed toward rejecting miracles. Either way, people are inclined to believe what they want to believe, even if it goes against the evidence. This is in no way a particular argument against miracles, rather an argument against human beings. While it is helpful to understand when confirmation bias may exist, pointing out the possibility of confirmation bias does not prove that all people operate with it.

Thirdly, like (a), proposition (c) is false in the context of the christian religion. Christianity, a religion full of miracles, began in Judea; a civilised society that was part of the Roman Empire. No historian would classify Judea, a roman province, as barbaric [8].

Conclusively, these preliminary arguments bear little weight when considering whether we are justified in believing miracles. Hume’s arguments do not support his conclusion, that we cannot maintain justified belief in christian miracles.


One can conceive the concept of a miracle, why might a person be hesitant to accept there possibility? The conditions needed for a miracle to be possible highly depend on whether there is an omnipotent God. Setting this matter aside, we must not pre-conclude the impossibility of a God who is responsible for miracles, nor the impossibility of miracles. If one believes that miracles are not possible, prior to investigating the evidence, then all evidence in favour of miracles will be discounted. In order to have an intelligible and objective discussion of evidence, we must begin with an a priori belief in the possibility of miracles.

David Hume’s first error is that he a priori rejects the possibility of miracles. Preliminary argument (d) indicates that there is no sufficient testimony that can prove a miracle. Hume admits this when he states that a sufficient way to refute a miracle is by referring to ‘the absolute impossibility… of the events’ [9]. Hume clearly presupposes miracles are impossible, using that to tell us that he will never be convinced by testimony. Hume inadequately rejects testimonial evidence that otherwise could justify one’s belief in miracles.

Scientific evidence is that which can be observed, tested, and repeated [10]. Miracles by nature are at odds with scientific evidence as they cannot be repeated, nor can they be tested. If a miracle can be predictably repeated, then it would be a natural and common observation in the universe. A repeatable miracle, by definition, ceases to be a miracle; a miracle must be rare. It is pointless to test miracles by tools of the natural world, as miracles by definition defy the natural world. If I could miraculously walk on water, one can test the density of water and measure the air pressure. However, if I could explain how the miracle is done by referring to these factors (commonly observed relationships), then I am no longer dealing with a miracle. Another problem we may encounter is that the christian religion is established by historical miracles. No historical fact can be exactly repeated or tested under the scientific method [11]. This remains true for miracles. Thus, the only way we could possibly use the scientific method to establish belief in miracles is through observation.

Hume’s a priori commitments, exclude him from being able to observe miracles. Even if Hume witnessed a miracle, he would conclude that he was deceived, much like any witness that supports miracles. Alternatively, one could claim that if God demonstrated a miracle to them now, they would believe in the biblical miracles. Both approaches demand conditions for belief that can never be attained. Presume I could maintain a justified belief in miracle A, under the condition that God provides miracle B, to tell me that I am not being deceived. When I encounter miracle B, how do I know I am not being deceived further? I would need a further miracle C, to validate that miracle B is not a deception. I would then need a miracle D to tell me that miracle C was not a deception, and so on. Here, we encounter a problem of infinite regression. In light of this, Hume cannot discern whether observable evidence is deception. Therefore, he rejects scientific evidence for miracles.

Hume has rejected both testimonial and observable evidence, yet his paper does not give sufficient reasons for doing so, nor does it provide a philosophical case for the logical impossibility of miracles (that is the exact factor in the emperical world which prevents miracles from being true). Rejecting all evidence before an investigation is not only anti-academic, but demonstrates that Hume is the person who is employing bias (see preliminary argument (b)). Here, Hume can be compared to the churchman whom Galileo beckons, ‘come and see for yourself the evidence, just take a look through my telescope.’ Hume’s reply is ‘there is no need, for what you say is impossible’. In order to be objective, one must look through the telescope and examine the evidence, and only then, can he explain why what he sees is true or false.

While miracles are ontologically possible, are they logically possible? In other words, if we were to set out a series of philosopical premises that do not depend on the natural world, could we discount the possibility of miracles? Baruch Spinoza argued that since the laws of nature are dependent on God’s nature [12], any transgression of the laws of nature, by God, is a transgression of His own nature [13]. This presents a logical problem as to whether God could actually perform a miracle. This problem arises out of a false relationship between God and the laws of nature, which is not nessasary to maintain a rational belief in miracles. Alternatively we can appeal to the famous Christian document, the Westminster Confession, which states that God by ‘His own will, freely, and unchangeably, orders whatsoever comes to pass.’ [14]. In other words, the future, and hence the laws of nature, are not caused by God’s eternal nature, but caused by his will. The reason that he orders the laws of nature to be constant is purely the free choice of God. This means that God can at any time, if he so pleased, suspend or change the laws of nature for a moment, in order to demonstrate a miracle. [15] Hence, if God performed a miracle, there is no contradiction, as God never transgresses His own nature. As long as this conception of God is adhered to, Spinoza’s argument against the logical impossibility of miracles fails.

In conclusion, premise (1) of my argument is established: miracles are logically possible [there is no philosphical reason which prevents it from being possible] and ontologically possible [there is no physical contradiction in reality which prevents it from being possible].


Why might a high or low probability of miracles work for or against the existence of miracles? A skeptic could simply accept the possibility of miracles (in theory), but limit the probability to such a small figure that one cannot ever be justified in believing a miracle has taken place. Under this strand of sceptisism Hume suggested that miracles can be believed or rejected according to probability. The evidence for miracles is human testimony, while the evidence against miracles is human experience. Therefore, by weighing the evidence, for and against, it is more far more probable that miracles do not occur.

According to Hume, the experience of the whole of humankind testifies against miracles. My first objection is that there are two problems with this belief. First, there is no uniform opinion of whether miracles are possible, nor is there a unified experience of the human race. Hume writes, ‘it is a miracle, that a dead man should come to life; because that has never been observed in any age or country.’ [16]. This, however, is the very claim that we are trying to prove. [17]. This is Hume’s conclusion and is being offered up as a premise in his argument. Not only that, but his claim is also highly dependent on who you ask. Speak to a Christian and they will say that miracles have been observed. If there is at least one group of people who claim that miracles have been observed, Hume’s claim that all human experience testifies against miracles is false. Second, I would call into question Hume’s epistemology. Should we affirm such a criterion that suggests that since everyone believes one thing it is rational for me to believe it too? Of course not. If everyone believed that miracles were impossible, it has no bearing on whether I should believe that miracles are impossible. Truth and belief should be founded on primary evidence. If Hume’s argument is sound, in Galileo’s time, I must rationally believe earth is the centre of the universe, despite the evidence. If truth is not dependent upon the opinions of others, neither should the methods for which I come to hold justified belief.

It seems far more reasonable to believe that the evidence against miracles, is not human experience, but rather, Hume’s experience. If Hume, however, rejects miracles a priori, even if he experienced one, he would either not recognise it or consider himself deceived. Hume’s epistemology does not allow him to know whether he has actually witnessed a miracle. His experience is dominated by an a priori belief in the impossibility of miracles, not evidence. Therefore, his personal experience can hardly be used as evidence against miracles. Hume’s claim can be reduced to; ‘I do not believe in miracles since I have pre-concluded that I have not personally experienced one’.

My second objection is that Hume does not have a problem with miracles, but rather improbable events. If someone playing golf said that they got a hole-in-one, and Hume has never gotten a hole-in-one, then Hume would be unable to believe the testimony.[18]. If we gathered all the lottery winners that have ever existed and got them to personally testify that they won the lottery, Hume could not rationally believe them, as he has never personally won the lottery. Hume might reply, ‘others claim to have experienced these events, so I can believe them’, to which I would reply, ‘others have claimed to have experienced miracles so I too can believe them’. Further, Hume cannot account for events that are so extremely improbable that they have seldom occurred. Hume’s personal experience dictates that he cannot rationally believe in: the beginning of the universe, the beginning of life, any diagnosis of a rare disease and the event of humans walking on the moon. This is because these events are not within the realm of his experience, neither are they commonly observed by other people. These examples are proof that Hume’s argument from probability, which seeks to invalidate testimony, is far too strict.

Just because I have demonstrated that Hume’s probability argument is insufficient does not suggest that miracles are probable. Whether you believe it is probable that at least one miracle has occurred in world history, is highly dependent on your a priori commitments. If I assume that God does not exist, then I can argue that it is improbable that a miracle has occurred in world history. On the other hand, as William Paley rightly argues, if an omnipotent and personal God exists then it is highly probable that he would interact with his creatures on earth, and highly likely that his interaction on earth will involve miraculous events. [19]. Whether miracles are probable to you, may well depend on your belief regarding God’s existence. Stating that it is unlikely that miracles have never occurred in world history is not an argument, but a confession that you do not believe in the existence of a personal God. Therefore, it is safe to presume premise two: if God exists, miracles are probable.


For my third premise, there is valid evidence that supports miracles, I will set forth no new information, but simply summarise my discussion of ‘evidence’. In order to have justified belief in miracles, I must not a priori reject the possibility of miracles, as Hume does, but examine the evidence. Evidence that could support miracles, as I have stated, could be categorised as testimonial, scientific, or philosophical.

Hume rejects all testimonial evidence, including every historical source which might contain legitimate evidence. Hume justifies this rejection for two reasons: miracles are impossible, and miracles go against the experience of humans. I have shown these two reasons to be insufficient. First, to claim that miracles are impossible is what Hume is trying to prove and reflects his a priori rejection of them. Second, miracles do not go against the experience of all humans, unless you first reject the testimony of human beings that testify that miracles exist, which is a rejection of undesirable evidence. Further, just because Hume has not personally experienced a miracle should not be sufficient to establish that miracles do not occur. Since Hume’s reasons for rejecting testimonial evidence fail, testimonial evidence stands as valid evidence for a miracle.

Hume rejects all scientific evidence for miracles. The only scientific evidence for miracles that is valid, by definition, is observation. Hume, however, argues that the human senses can be deceived, and hence, would conclude that he is being deceived if he ever observed a miracle. Therefore, he rejects observation as valid evidence for a miracle. Consequently, Hume denies evidence prior to its investigation since miracles are impossible. Once again, we see Hume’s conclusion attempting to establish itself as a premise in his argument. Since Hume’s rejection of scientific evidence fails, if I believe in the possibility of miracles, then saw a miracle, It would be rational for me to believe a miracle has taken place. [20].

Finally, I have rejected one of Hume’s philosophical arguments against miracles, supplying philosophical evidence (the probability argument) for miracles in its place. Hume argues that it is improbable that miracles occur. I instead state that if we accept the preconditions for a miracle to take place are true, namely that God exists, then it is highly probable that a miracle has or will take place in world history.

Conclusively, Hume’s rejection of evidence is ungrounded. This means that the ordinary forms of evidence are valid, and we can use testimonial, scientific, or philosophical evidence to obtain justified belief in miracles (premise three).


I have proved that miracles are logically and ontologically possible, miracles are logically probable under the right assumptions, and there is valid evidence that supports miracles. My analysis criticised objections from Hume and Spinoza, as well as many anticipated objections. Furthermore, I have considered many preliminary arguments from Hume that might bear weight in this discussion. I argue that instead of rejecting miracles a priori, they should be considered in light of the evidence. It is not my aim to present evidence in all three categories, however, there is abundant material in favour of miracles. I am not arguing that people are not often deceived, nor that we should believe evidence without scrutiny. Instead, I argue, if we properly analyse the evidence for miracles, case by case, considering the possibility that miracles can occur, it is possible, if the evidence permits, to obtain justified belief in miracles. I advocate that we should, as rational beings, look through Galileo’s telescope and take what we see seriously.


[1] I personally favour the definition: ‘A miracle is that which cannot be explained by natural causes.’ I do not have time to contest Hume’s definition here, thus, I will address Hume on his own terms.

[2] David Hume, ‘An Enquiry Concerning Human Understanding, Section X: Of Miracles’, Hume on Religion (Fontana, 1963), p. 210.

[3] David Hume, ‘An Enquiry Concerning Human Understanding, Section X: Of Miracles’, Hume on Religion (Fontana, 1963), pp. 212-226.

[4]Ibid. p. 226.

[5] William Lane Craig, The Problem of Miracles: A Historical and Philosophical Perspective (2020), p. 15.  John Lennox, ‘Violating Nature: The Legacy of Hume’, God’s Undertaker: Has Science Buried God? (Oxford, 2009), p.199.

[6] Mark 3:38, John 4: 4-26.

[7] Matthew 14:13-21.

[8]T. Editors of Encyclopaedia Britannica, ‘Jewish Palestine at the time of Jesus’, Encyclopedia Britannica (2021), [,]( accessed on 28 September 2021.

[9] David Hume, ‘An Enquiry Concerning Human Understanding, Section X: Of Miracles’, Hume on Religion (Fontana, 1963), p. 220.

[10] T. Editors of Encyclopaedia Britannica, ‘Scientific Method’,  Encyclopedia Britannica (2020),  [,]( accessed 28 September 2021.

[11] Carey B. Joynt, and Nicholas Rescher. ‘The Problem of Uniqueness in History’ History and Theory 1, no. 2 (1961), [,]( pp.150–62.

Carol E. Cleland, ‘Methodological and Epistemic Differences between Historical Science and Experimental Science’ Philosophy of Science 69, no. 3 (2002), [,]( pp.447–51.

[12] This view holds to the dependency thesis of God. All universal laws are expressions of God’s nature.

[13] G. H. R, Parkinson, ‘Spinoza On Miracles And Natural Law’ Revue Internationale de Philosophie 31, no. 119/120 (1/2) (1977), []( p.151.

[14] Westminster Confession, Chapter 3.

[15] J. Alph Turretin, Traité de la vérité de la religion chrétienne, 2nd ed., 7 vols., trans. J. Vernet (Genéve, 174555), p. 5, 272.

[16] David Hume, ‘An Enquiry Concerning Human Understanding, Section X: Of Miracles’, Hume on Religion (Fontana, 1963), p. 211.

[17] John Lennox, ‘Violating Nature: The Legacy of Hume’, God’s Undertaker: Has Science Buried God? (Oxford, 2009), p.202.

[18] Frank Turek and Norman Geisler, ‘Miracles: Signs of God or Gullibility?’, I Don’t Have Enough Faith to Be an Atheist (2004), p. 207.

[19] William Paley, A View of the Evidences of Christianity, 2 vols., 5th ed. (London, 1796), pp.3-15.

[20] Charles Leslie, A Short and Easy Method with the Deists (London, 1815). p.13.