top of page

Episode 34 - Hume, Defeater of Miracles

One of the main reasons people disregard the resurrection of Jesus is because they don't believe miracles are possible. Whether they know it or not, their anti-supernatural worldview has been heavily impacted by a philosopher named David Hume, who gave a few arguments against miracles. Even today, many philosophers will say that Hume did such a good job defeating miracles that they can't possibly consider the possibility that supernatural events occur. For this episode, we'll look at Hume's arguments, and see how they stand up.

Music from

Audio of the Ultimate Questions Podcast is found anywhere podcasts are available.


Hello and welcome to the Ultimate Questions podcast.

This podcast is brought to you in association with “Culture at a Crossroads”, which is a podcast hosted by David Mann from Life 100.3. His goal is to help navigate different cultural challenges, and how we can engage with those around us on these pertinent issues. In his latest episode he had on the leader of Village Church, pastor Mark Clark, who has written the books The Problem of God, and The Problem of Jesus. You can check out his podcast on all platforms or by going to

Recently on the Ultimate Questions podcast we’ve been evaluating the arguments against miracles, since that’s one of the biggest reasons skeptics won’t accept that Jesus actually rose from the dead. For today, I want to take a look at David Hume, who probably had the largest impact on Western culture in the movement away from believing in the miraculous. While he definitely had an impact in other areas, one of his biggest claims to fame is that many consider him to have defeated miracles entirely. Hume wasn’t really the originator of a lot of his thoughts against miracles (for example he took a good bit from John Locke). Additionally, his arguments are, quite frankly, pretty bad. Many philosophers today recognize that he committed fallacy after fallacy, and that good sound reasoning wasn’t really his strong point. For the most part, I would argue that the culture was ripe for his kind of skepticism. During this time, science was beginning to really boom, and because of the Protestant Reformation, many people were thinking about religious and philosophical matters in a much more open minded and critical way. Due to the scientific mindset, and the willingness to challenge religion, many within the higher echelons of academia were falling back into what’s called “deism”. While deists do believe in some kind of ultimate God that created the universe, they would believe that this God doesn’t really interact with His creation at all. This makes for a much more palatable religion, at least for those during the beginnings of the scientific revolution. In their mind, they don’t have to bother themselves with all the silliness of miracles and angels and old stories from the Bible. While many of them would still call themselves Christians, they stopped believing in the core fundamentals of the faith. This is the sort of cultural backdrop that Hume entered into.

Hume was a Scottish philosopher born in 1711, which was about the beginning of the Western cultural moves away from orthodox Christianity. The people were already falling into deism, and Hume gave them an excuse to fully leave behind the supernatural, and feel philosophically justified in doing so. That said, the reason Hume eventually got popular was because he was arguing for something that many “wanted” to be true, and he did so with incredibly powerful, and even aggressive prose. It was not because his arguments were any good, however, since even people within his own time were able to show how fallacious his reasoning was, as we’ll see in this podcast. Two good examples of people during Hume’s time debunking him are George Campbell and Richard Price, who wrote showing many of the different ways Hume’s arguments failed. However, despite being proven fallacious, Hume’s arguments still caught on, to the point that many academics, even today, won’t even consider the possibility of miracles, because of Hume. As an example of this, during the early 1800s, a German liberal theologian named David Strauss said that, because of Hume, we can no longer believer in miracles, simply because Hume was so brilliant, and so correct in his reasoning against the miraculous. In an outrageous example, during the 1960s there was a lecture given by James Norman D. Anderson at Harvard University, where he made an incredibly strong argument for the resurrection of Jesus Christ. One in attendance was Harvey Cox, who is an incredibly accomplished theologian, and who held one of the most prestigious positions at Harvard Divinity School. Harvey Cox does not believe in the supernatural or miraculous side of Christianity. When he heard this lecture by Anderson arguing for the resurrection of Jesus, he acknowledged how persuasive it was. However, he then admits that people like himself have a “Humean mark” on their brows, that while they try to affirm Jesus is “in some way” resurrected, deep down they have doubts that resurrections are even possible. In other words, Harvey Cox heard a strong argument for the resurrection, had no real counter argument to speak of, and instead, simply deferred to the wisdom of Hume, because Hume did such a good job defeating miracles. For another case of the impact of Hume, while I was doing research for this episode, I found a writer named Julian Baggini, who loves Hume. He seems to appreciate certain aspects of Christianity, while at the same time dismissing anything supernatural about it, which is very much the sort of attitude we would expect in someone who enjoys Hume’s work. Baggini wrote that Hume is “the amiable, modest, generous philosopher we need now”. His article then went on to praise Hume for all his contributions, including the arguments against the miraculous, without really recognizing that these arguments failed miserably. In the article, he references a study done among professional philosophers. In this study, one of the questions asked which non-living philosopher do they most identify with. The winner was Hume. So, even though Hume’s arguments failed, as we’ll look at soon, he has had a massive influence on the world, and is still incredibly relevant, even today, because people “want” to disbelieve in miracles, and they use Hume’s arguments to feel justified in their disbelief.

As I’ve said many times now, Hume’s arguments fail, miserably. This is recognized in many academics, to the point that there’s even a book titled “Hume’s Abject Failure”, which goes into great detail about all the problems with his arguments. The book is written by John Earman, who is accused of writing it for the sake of Christian apologetic purposes, but he flat out says that he does not appreciate Christian theology, which I think is safe to assume means that he’s not even a Christian. While some people do still love Hume, as we just saw, his arguments against miracles all commit fallacies, with the fallacy of circular reasoning being his main way of trying to argue his point. We’ll look at four different ways he tries to prove that miracles don’t happen.

One of the main arguments Hume gave against the miraculous is an argument by definition. This argument explains what a miracle is, and shows that it’s literally impossible, by definition. Many skeptics grab hold of this argument, firstly because it’s easy to remember, and also because it feels like it’s a homerun. However, as I’ve said, it commits the fallacy of circular reasoning, also called begging the question. The argument defines a miracle as a violation of the laws of nature. Now, at first glance, that seems like a reasonable definition of what a miracle would be. The problem enters in after that, when Hume says that the laws of nature are not capable of being broken. So, if a miracle is breaking a law that cannot be broken, then a miracle is by definition impossible. From this perspective, trying to say a miracle happened would be like trying to say there’s a married bachelor.

The big problem with this sort of argument is that it’s assuming the conclusion before even giving the argument. The two statements are the sorts of things a person would only agree with if they already believed that miracles are impossible, which is by definition what circular reasoning is. If someone who believes in miracles were to agree with both these statements, it would just show they didn’t fully comprehend what was happening. If we agree that a miracle is the breaking of a law of nature, then a religious person would merely admit that they believe God can break the laws of nature if He wishes. After all, if God created the laws of nature, then it would seem entirely possible He could work outside of them. Even if we consider the act of creation, that breaks the laws of nature, since matter and energy were being created out of nothing. So, if we agree with the first statement about miracles breaking the laws of nature, then we would not agree with the second statement that the laws of nature cannot be broken.

That said, I don’t think we even need to grant Hume’s definition of what a miracle is. It seems to me that the definition of a miracle would be something more like a supernatural being interfering with the natural world. For example, suppose a person was about to get run over by a truck, and an angel pushes them out of the way. Did a law of nature get broken? I wouldn’t think so. An angel is a causal agent, meaning it’s the sort of being that can be in a cause-and-effect relationship, making things happen. If an angel pushes someone, the angel is just doing the sort of thing that angels are capable of. It’s a supernatural entity interfering with the natural world. So we don’t need to consider a miracle as breaking the laws of nature, and even if we did, God is capable of breaking the laws of nature, and regardless of these points, the argument is structured in a way that it’s circular. The argument against miracles by definition fails, miserably.

Our second argument from Hume is that miracles go against our experiences. Basically, Hume wrote that nobody actually experiences miracles, and all the miracle claims we’re dealing with are merely stories in ancient mythologies. So even if we read a story from two thousand years ago that someone healed people, we don’t see that today, so we’re justified in thinking these stories are merely stories, and not based in reality. The problem here is that this statement is just false; many people do claim to have witnessed miracles. While Hume might not have personally experienced anything miraculous, that doesn’t mean this is the case for everyone. I can picture Hume sitting in a room, smoking cigars with his friends, with him asking them if they’ve ever seen a miracle. They all respond negatively, and then laugh at the silly religious people for believing such nonsense. That doesn’t count as evidence. Merely asking a handful of people doesn’t work as actual research. If we actually do the research by polling people, we find that there are many people claiming to have witnessed miracles. George Mavrodos wrote a critique on Hume for the International Journal for Philosophy of Religion, and in it he argued this point. He showed that, if you do an actual poll, very quickly you do get miracle stories. For me personally, each time I teach my apologetics class I ask for a raise of hands from those people who have seen a miracle, and each time, nearly every hand goes up. I recognize it’s a situation stacked in favor of miracles, since I teach at a religious school, but the point is, if you actually go around asking people, doing a real poll, you do get miracle stories. So we have two options, either Hume used an incredibly small sample size, or he ignored the miracle stories he did hear, which would be confirmation bias. Either Hume is claiming to have done research without actually doing it, or he’s committing a fallacy. So after looking at Hume’s argument against miracles by experience we see, once again, that Hume’s arguments fail, miserably. If you’re interested in looking at examples of miracles that work to show the miraculous is part of many people’s experiences, we’ll look at a few present day miracle stories in the next episode of the podcast.

Our third argument from Hume against miracles is that it is always more rational to believe there has been some mistake or deception, than to believe a miracle has occurred. Now, even just with that brief statement, I hope you can see how it’s problematic. Imagine getting into an argument with someone about some issue, and after you present your case, they reply with, “you just need to presume that your wrong, then you’ll see how right I am.” It’s obviously a bad way to argue, and is completely unreasonable. Right off the bat, we can tell that this argument, just like our first argument, commits the fallacy of circular reasoning. The only people who would ever agree with Hume here would be people who already disbelieve in miracles. If someone does believe in miracles, there’s no way they would be swayed merely by telling them they would be more rational to believe they were mistaken, or had been deceived. So why did Hume hold to such an obviously terrible argument? Hume believed that it is impossible to ever identify something as being a miracle, and in that point, I can somewhat understand where he’s coming from. For example, a few episodes ago I mentioned that my mother had been challenged by a friend to seek God for financial help. My mom then prayed to God for $333 dollars, which is what she needed in the moment, and a minute later while opening the mail, she found a cheque for $333 dollars. In this case, how do we know it was actually God providing for her? How do we know it wasn’t just a lucky coincidence? Yes, it’s incredibly bad odds to receive exactly what you needed, exactly when you asked for it, but still, how can we be “certain” that this was a miracle? Or, if there’s some sort of supernatural healing, how do we “know” that it was God healing the person, and not just some random act that science doesn’t understand yet? Hume was incredibly skeptical, and we’ll go into more detail about the lengths of his skepticism in a few minutes. But for now, if we grant that it’s impossible to be certain an event is a miracle, does that then imply that miracles are “impossible”? Let me repeat that, if we grant that it’s impossible to be certain an event is a miracle, does that then imply that miracles are “impossible”? I don’t see how. I can understand why Hume would argue against the concept of certainty, since he was quite skeptical in general, but it doesn’t follow logically that, since we can’t know something for certain, that we are then justified, or even obligated, to doubt it completely. For example, suppose for a moment my wife makes me dinner, and I sit down to eat. Can I be “certain” that this food is safe? Can I be 100% sure that it isn’t poisoned? In fact, my wife actually has previously given me food poisoning (although it was accidental), so how can I be guaranteed that it won’t happen with this meal? You will likely admit, I can’t be certain. However, that does not mean I am justified in believing that this meal is in fact poison. I definitely am not “obligated” to believe that this food is dangerous. For Hume, he was such a skeptic, that he thought that nothing should be believed, unless you can have absolute certainty in it. So I suppose every time he ate food he was breaking his own logic. The real point here is that we can’t have absolute certainty for basically anything, and yet, it’s perfectly acceptable to believe these things anyways, and in fact, we do it all the time, and it would be impossible to live life without believing things even when we lack certainty. Even though we cannot know for certain that some event is a miracle, that does not then mean we must believe there has been some sort of mistake or deception. The correct way to look at each miracle claim, or anything at all really, would be to evaluate the evidence, and determine what the best explanation of the data is. So, once again, when we look at Hume’s argument against miracles by assuming every miracle claim has a mistake behind it, we see that Hume’s arguments fail, miserably.

Our last argument from Hume against miracles is that no amount of evidence could ever possibly establish a miracle, because miracles are so improbable. And once again, hopefully you can see how problematic a statement like that is, in the fact that, again, it’s circular reasoning in that the only people who would agree with that statement are going to be people who already disbelieve in miracles. You’ll notice that this argument looks quite similar to last episode’s argument of the “Sagan Standard”, which says that “extraordinary claims require extraordinary evidence”, except that Hume is just being blunt about it, and showing the fallacious nature of this right out of the gate. Last episode I pointed out that this sort of requirement is essentially saying that, no matter how much evidence you give, it will never be enough, which is fallacious. It’s circular reasoning, confirmation bias, and moving the goal posts, all at the same time. However, Hume goes one step further, and flat out declares the fallacious aspect of this argument in his writing.

Just to show you that I’m not misrepresenting Hume here, earlier in this episode I mentioned an article by Julian Baggini that sang the praises of Hume, so let’s look at how he expressed this argument Hume gave. He wrote that Hume “argued that it would never be rational to accept the claim of a miracle, since the evidence that one had occurred would always be weaker than the evidence that such things never happen.” Now with this kind of wording, we can start to see why someone might think this is a good argument. After all, we have countless examples of times where the events have played out perfectly naturally, without miracles, and very few where there has been a supernatural claim. So in this sense, we will always have more examples of natural phenomenon, which will always outweigh supernatural phenomenon. Therefore, the evidence for the natural will always outweigh a miracle claim. One problem here is that this method seems to be trying to employ scientific methodology to miracles, which as we looked at last time, doesn’t work. This is a supernatural event, and so we can’t apply natural methods to it. A miracle is an exception to the general rule, so we should expect that it won’t happen often. This method is treating all the cases where there wasn’t a miracle as though they work as evidence against the times where there was a miracle. It’s a bit confusing, but basically, it’s trying to say that, since people aren’t “always” healed, they must “never” be healed. This is because science works by examples, to find the natural cause of things. For example, let’s look at the case of an angel pushing someone out of the way of a truck. We have many counter examples where a truck is coming at a person, and they don’t get pushed out of the way by an angel. So, if we get, let’s say, a thousand cases where the person gets run over, and then we have this one case where someone says an angel pushed them out of the way, then, by Hume’s reasoning, we have a thousand times as much evidence against the miracle than we have for it. I think this is quite problematic, because at no point were we trying to argue that people are “always” saved from trucks. Someone getting hit by a truck does not work as evidence against someone else being saved from a truck. Now again, this is confusing, and I think it’s largely because Hume is incredibly mistaken. Let’s see if we can go deeper into this point, to see exactly what’s happening, and why it fails.

As I mentioned earlier, Hume was incredibly skeptical. He was even skeptical of induction in general, which is the way that science works. Hume believed that examples cannot work as evidence, since they can never get us to the point of certainty. As an example, Hume literally made the point that, just because the sun has always come up every day, does not guarantee that it will come up tomorrow. That’s how skeptical he was. Hume was so skeptical that it was to the point of irrationality, so we shouldn’t be surprised to find irrational thinking throughout his arguments. Now that said, Hume recognized that we cannot live like this, as I’ve mentioned earlier with the food poisoning example, so he said that we need to act based on imperfect beliefs. In other words, even if we lack certainty, we still need to believe things that we aren’t certain about. Now with that in mind, why not apply that same reasoning to miracles? When we have evidence that a miracle has occurred, why ignore that? What we should do, is we should weigh the evidence and believe what is the “most likely” explanation of the data. In this case, Hume would say that we have more evidence against any miracle claim, than we do in favor of that miracle. However, this seems very problematic. For example, if you have the winning lottery ticket, the odds against you having the winning ticket far outweigh the odds of you having the winning ticket. By Hume’s reasoning, you shouldn’t believe you have the winning ticket, no matter what, because the amount of tickets that lost will always outweigh the winning ticket. In Hume’s mind, even though you really do have the winning lottery ticket, you would be more reasonable to believe there has been a mistake. And I don’t just mean that you should be skeptical; you should actively disbelieve that you have the winning lotto ticket. I recognize that this is quite a confusing point, but I really believe that’s because Hume’s point is just so bad. As another example of this, just to see how ridiculous Hume’s reasoning is, we can look at another analogy that was going around during his time. Suppose for a moment you have a person living in the tropics, who has never seen snow. Even if many people come to this person, telling them about snow and ice, the person from the tropics would be entirely justified in continuing to disbelieve in the idea of water becoming a solid. However, for Hume, he wouldn’t stop at merely saying that the person is justified in still being skeptical. Instead, Hume would say this person ought to outright reject the idea of snow and ice, based on his own observations. After all, this person from the tropics has seen countless examples of water staying as a liquid, and thus, he has far more reason to think water always remains a liquid, and therefore he is obligated to reject any notion of snow or ice. The obvious problem here is, snow and ice really do exist! Personally, I find this example very similar to the point skeptics make here against the resurrection, that all of our observations have been that dead men stay dead. Simply because you have more examples of it not happening, does not then imply that it’s literally impossible. As we can see with the person from the tropics example, he would be wrong to disbelieve that ice exists. While Hume bit the bullet here, hopefully we can recognize the terrible kind of thinking that this sort of skepticism brings about.

To tackle this argument even more, lets look at this concept of probability that Hume seems to rest his argument upon. For Hume, if you have many more examples of a miracle not happening, then that evidence will outweigh any evidence for the miracle. Basically, the probability of an event being a miracle goes way down, because of the evidence of cases where miracles didn’t happen. Ignoring the fact that this doesn’t actually logically follow, one of the big factors here involves your presuppositions. If you presume atheism, then miracles will be incredibly improbable, or even impossible. However, if you presume theism, then miracles become possible. To go further, if you presume the God of Christianity, then miracles become incredibly probable. No one begins this sort of investigation in a vacuum, without taking anything else into consideration. Additionally, we can flip this whole discussion on its head, and ask the opposite question: What are the odds of this event being non-miraculous? To rephrase this, what are the odds that we would have this event, if it weren’t miraculous? In many cases, it seems like the odds would be incredibly bad. If we take the example of my mom getting the money she prayed for, in the exact moment she prayed for it, what are the odds of that happening, if we assume atheism, and assume it wasn’t a miracle? Obviously, it would be very, very low. Then, if we grant the God of Christianity, what are the odds of God doing a miracle like this? What kind of explanatory power does the idea of a miracle grant this sort of event? Basically, we need to compare the odds of the event occurring if it’s not a miracle, and the odds of the event occurring if God actually does exist and performed a miracle here. In other words, if we presume for a moment that it’s a miracle, does that cause the situation to make more sense? If we deny that it’s a miracle, it would seem that it would actually be even more unbelievable, given the odds of receiving the amount of money needed, in exactly the moment it was prayed for. It being a miracle would actually seem to be not nearly as improbable. Therefore, shouldn’t we go with the more probable explanation?

I recognize that this talk of probability might be difficult, so let’s look at this again. The odds of a miracle are quite low, since most of the time, miracles don’t happen. However, even though the odds of a miracle might be low, the odds of this sort of event happening without it being a miracle seem to be even lower than the odds of a miracle. With the evidence in hand, that my mom actually did receive the amount of money needed in the exact moment it was prayed for, we then ask the question, “what’s the best explanation of the data?” It being a miracle would actually have more explanatory power than the lack of a miracle. The real goal here is to find out the truth of what happened, and this isn’t done by probability; it’s done by weighing the evidence. This is exactly how history works. Historians do not evaluate the likelihood of events; they take all the evidence, and try to find the best explanation of the data. If we only evaluate the likelihood of the event on its own, then there would be all sorts of events throughout history that we couldn’t believe, simply because they’re so improbable. History is filled with crazy stories that are incredibly unlikely, and yet, the evidence shows us that they very likely occurred. For example, there’s a huge list of coincidences between president Lincoln and president Kennedy. The coincidences are incredibly unlikely, and yet, given our historical data, we still believe in the historical facts. It would be silly for us to stop believing the data, simply because of the likelihood of it. Instead, historians follow the evidence where it leads, and believe the best explanation of the data.

If we then apply this kind of reasoning to the resurrection of Jesus, we can definitely grant that every other time we witness death, it’s permanent. However, many examples of the permanence of death doesn’t then make it absolutely impossible for a resurrection to occur. If we look at the probability aspect, yes, a resurrection is highly improbable. However, how probable is it for us to have the wealth of data we’ve gone over in earlier episodes of this podcast, if Jesus wasn’t raised? That seems even more unlikely. If we grant the God of Christianity, and we follow the evidence where it leads, the best explanation is that Jesus actually did rise from the dead.

As we’ve seen today, Hume has a few ways that, on the surface, might seem like acceptable arguments against miracles. However, by critically evaluating them, we can see that they are very much fallacious, usually being cases of circular reasoning. We can’t simply define miracles out of existence. We actually do have many cases of miracle claims. Also, if someone always assumes there’s been a mistake or deception when there is a miracle claim, then that person is begging the question, and presuming naturalism before even beginning the conversation. And lastly, to say no amount of evidence could establish a miracle is ignoring all evidence, and just reinforcing your own bias, rather than actually evaluating what the best explanation of the data is. Hume’s arguments do not work. Our secular society’s bias against miracles is actually built upon a foundation of Hume’s arguments. Because of this, people won’t even consider the possibility of miracles. However, since Hume’s arguments fail, everyone should start really considering the evidence on both sides, rather than fallaciously discounting any and all miracle claims.

Over the past few episodes we’ve looked at all the reasons and arguments why people are against believing that miracles are possible. As we’ve seen, these arguments don’t work. However, even if all the arguments against miracles don’t work, do we have any reasons to think miracles actually do occur? To finish up this section of the podcast on miracles, I thought it would be a good idea to look at some cases of miracles, by letting a few of my friends give their accounts of moments where God has clearly been at work. So I hope you’ll join me next time, as we listen to a few stories of present-day miracles, here on the Ultimate Questions Podcast.


bottom of page