Can Miracles Happen? Answering Hume and Spinoza

Imagine a courtroom in which the verdict is fixed before the first witness is sworn. However many people testify, however sober and consistent and unmotivated to lie, the judge has ruled in advance that their testimony cannot be heard, because it concerns a kind of event that—he has decided beforehand—simply does not happen. That is precisely the posture many take toward the central claim of the Christian faith. When the resurrection of Jesus is raised, a great many thoughtful people do not weigh the evidence and find it wanting; they decline to weigh it at all, because they hold that a miracle is either impossible or, whatever the testimony, never something a reasonable person could believe.

If that verdict is sound, the rest of the case is moot. There is no point examining an empty tomb or an early creed if resurrection has been ruled out at the door. So before any historical question can be honestly opened, one prior question must be faced: is there a good in-principle argument that miracles are impossible, or never rationally believable? This essay is about that question, and only that question. Its thesis is deliberately modest: the two classic attempts to shut the door—Baruch Spinoza’s argument that miracles are impossible, and David Hume’s that they are never credible—both fail. If they fail, the door is open. Whether any particular miracle actually walked through it is a separate, historical inquiry, taken up elsewhere.

Stating the Question Carefully

Two questions must be kept rigorously apart. The first is a question of possibility: could a miracle happen at all? The second is a question of actuality: did this or that specific miracle happen? These are answered by different means. Possibility is settled by philosophy—by asking whether the very concept is coherent and whether the universe is the kind of place open to such an event. Actuality is settled by evidence—by testimony, documents, and the ordinary tools of history. Spinoza and Hume both attempt to settle the actuality question a priori, in advance of any evidence, by reasoning from the nature of law itself. That is the move this essay contests.

It is worth stating plainly what will not be argued here. Nothing below shows that any miracle has occurred; nothing below establishes the resurrection, or that miracles are common, or that every wonder reported in Scripture happened as told. The claim is narrower and, precisely because it is narrow, more secure: the a priori bar is illegitimate, so miracle claims must be examined case by case on their evidence rather than dismissed at the outset. Removing the roadblock is not the same as completing the journey—but you cannot begin the journey while the roadblock stands.

What a Miracle Is—and Is Not

Careless definitions have done more to confuse this debate than any argument. A miracle, in the classical sense, is not simply a very strange or very rare event; a stunning coincidence is not a miracle, and neither is a phenomenon we cannot yet explain. Nor is it best described as nature “breaking down.” The most defensible definition is that a miracle is a special act of God who transcends the natural order—an event that would not have occurred had nature been left to its own resources. Strikingly, the atheist philosopher Antony Flew, no friend of theism for most of his career, framed a miracle in just such terms: as something that would never have happened if nature had been left to its own devices.¹ The concept, then, is not incoherent even to those who reject it; the dispute is over whether such an act is possible and whether it ever occurs.

One further point of care, kept deliberately light and shared across the Christian traditions: a miracle in the biblical sense is not a bare display of power but a sign—an act tied to a message and a messenger, meant to authenticate a word from God. This essay does not adjudicate which miracles are genuine, nor when or whether they occur today; those are questions on which sincere Christians differ. The point here is only conceptual: a miracle is God acting purposively within a world he made, not a random rupture in the fabric of things.

Descriptions, Not Prescriptions

The decisive move in the whole debate is a distinction about what a law of nature is. (The metaphysics of laws is treated at length in a companion essay;² here only one facet matters.) A law of nature, on the most defensible account, is a description of how things regularly behave, not a prescription that dictates what must happen—and even those who hold that laws genuinely govern take them to govern only how things go when nothing interferes. The law of gravity does not compel a stone; it tells us, accurately, how unsupported bodies do in fact move when nothing interferes. That last clause is everything. The regularities hold when nothing interferes—and agents interfere constantly.

Consider an outfielder catching a falling ball. He does not repeal the law of gravitation; the law describes exactly what the ball would do left to itself, and it goes on describing that even as his glove intervenes. A finite agent, by introducing a new causal factor, produces an outcome the regularity alone would not have produced, without abolishing the regularity in the slightest. If a finite mind and hand can do this within nature, an infinite Mind who authored nature is hardly barred from doing the same on rare occasion and for a purpose. A miracle, so understood, is not the violation of a law but the introduction of a cause the law never claimed to exclude. This is why the loaded word “violation”—which both Spinoza and Hume lean on—quietly smuggles in the conclusion it is meant to prove.

The Closed Box and the Open One

Underneath Spinoza’s argument—and the popular “science has disproved miracles” slogan—lies an assumption rarely stated: that the physical universe is a causally closed system—a sealed box in which nothing acts except other things inside the box. (Hume’s subtler, purely epistemic version does not depend on this; it is taken up in the next section.) On that assumption, of course, no miracle can occur, because there is no one outside to act. But whether the box is closed is not a finding of science; it is a metaphysical commitment. Naturalism assumes closure; theism holds that the universe is open to the God who made it. And here is the pivot: whether the universe is a closed system is exactly the question in dispute between the theist and the naturalist. To assume closure and conclude that miracles are impossible is to argue in a circle—to help oneself to the very premise the theist denies. If a God capable of creating a universe exists at all, the possibility of his acting within it follows almost trivially; the one who made the box is not trapped inside it.

Spinoza: Miracles as an Absurdity

The bolder of the two arguments is the older. Benedict de Spinoza, writing in his Tractatus Theologico-Politicus of 1670, held that miracles are not merely unbelievable but impossible. His reasoning, reconstructed fairly, runs like this: the laws of nature are the eternal, necessary decrees of God; whatever happens follows necessarily from those laws; therefore an event contrary to the laws of nature would be an event contrary to God’s own nature, which is absurd. Spinoza states it without flinching: “Nothing, then, comes to pass in nature in contravention to her universal laws, nay, everything agrees with them and follows from them, for whatsoever comes to pass, comes to pass by the will and eternal decree of God.”³ He concludes that “a miracle, whether in contravention to, or beyond, nature, is a mere absurdity.”⁴

Stated at full strength, the argument has real force: if the laws of nature genuinely are immutable and inviolable, then by definition nothing can violate them, and a miracle so defined is a contradiction in terms—like a married bachelor. One cannot simply wave this away.

But it does not survive inspection, for two reasons. First, and fatally, it begs the question by definition. Spinoza builds inviolability into the meaning of “law of nature” and then discovers, unsurprisingly, that laws cannot be violated. But whether the regularities of nature are absolutely immutable—impervious even to their Author—is precisely what is in dispute. Define laws as descriptions of nature’s ordinary course, and the argument evaporates: a description of what usually happens is not breached by an unusual act, any more than the outfielder breaches gravity. Second, Spinoza’s conclusion was already contained in his metaphysics. He was a pantheist who famously identified God with Nature itself—Deus sive Natura, “God, or Nature”—so that there simply is no God who transcends the natural order and could act upon it from without.⁵ Given that God, miracles are indeed impossible; but that is not a discovery about miracles, it is a restatement of pantheism. Against a theist who affirms a transcendent Creator, Spinoza’s argument has no purchase; it refutes a position no theist holds.

Hume: Miracles as Never Credible

The argument that actually shapes the modern mind is subtler, and more formidable. David Hume, in Section X of his Enquiry Concerning Human Understanding (1748), conceded the ground Spinoza had over-claimed. He did not try to prove miracles impossible. He argued instead that no testimony could ever make belief in one reasonable—an epistemological argument, not a metaphysical one, and all the stronger for its restraint.

Hume begins with a principle few would reject: “A wise man, therefore, proportions his belief to the evidence.”⁶ He then defines a miracle as “a violation of the laws of nature” and presses the consequence: “and as a firm and unalterable experience has established these laws, the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined.”⁷ The whole weight of human experience of nature’s regularity stands, he says, against any report of its breach: “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.”⁸ From this he draws his celebrated maxim, offered in Part II as “a general maxim worthy of our attention”: “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.”⁹ In practice, Hume thought, such testimony never arrives; the witnesses to miracles are too few, too credulous, too eager, and too apt to found new religions, for their falsehood ever to be more astonishing than the wonder they report.

This is a genuinely powerful argument, and it deserves to be felt before it is answered. It appeals to the same evidential caution that makes us doubt a tabloid headline; it does not deny God or the possibility of the supernatural outright; and it seems only to ask the believer to be consistent about weighing evidence. It has persuaded serious philosophers ever since, and any reply that treats it as foolish has not understood it.

Where Hume’s Argument Breaks Down

Yet for all its polish, the argument fails—and it fails in three distinct places.

First, the premise that the evidence for the regular is always greater than the evidence for the rare is simply false. A single well-attested, unrepeatable event can be rationally believed on adequate testimony; the frequency of an event and the strength of the evidence for it are two different things. We accept, on good testimony, that a particular golfer once made a hole-in-one, without demanding that he repeat it, and without weighing against his witnesses the countless times balls have not gone in. More tellingly, the naturalist’s own worldview affirms events believed to have happened exactly once—the origin of the universe and the origin of life among them. These are, on the standard scientific picture, unrepeated and in some sense unrepeatable; yet no one thinks it irrational to believe they occurred. If unique events can be rationally believed elsewhere, Hume’s blanket rule cannot be right. The Humean will reply that a miracle is not merely rare but contrary to the whole uniform run of experience—a stronger charge than mere infrequency. But that objection was met above: a miracle introduces a new cause into the ordinary course of things; it does not breach the regularity that holds when nothing interferes, and so it does not carry the maximally low prior the objection assumes.

Second, on its strongest reading the argument argues in a circle. Hume’s claim that “a firm and unalterable experience” stands against every miracle assumes what it needs to prove. How could one know that experience is uniformly against miracles unless one already knew that every miracle report, past and present, is false? But that is the very conclusion in question. C. S. Lewis pressed exactly this point: if we assume in advance that all reports of miracles are false, then of course experience is uniform against them—but we are then reasoning in a circle, having smuggled the conclusion into the premise.¹⁰ To label experience “uniform” is not to report a finding; it is to prejudge every disputed case at once.

Third, the argument weighs the evidence the wrong way. Hume, in effect, piles the entire mass of ordinary experience against the specific testimony for a specific event, and concludes the testimony must lose. But that is not how rational assessment of a particular claim works. The right method is not to add up how often the extraordinary fails to happen, but to weigh the specific evidence for this claim—the number, independence, and reliability of the witnesses, their access to the facts, their motives. Here the most striking witness is not an apologist at all. John Earman, a distinguished philosopher of science and no friend of theism, subjected Hume’s essay to a rigorous probabilistic (Bayesian) analysis in a study pointedly titled Hume’s Abject Failure. Earman concludes that Hume’s argument, where it is original, is largely without merit, and that Hume overlooked how multiple independent witnesses can, in principle, make even a highly improbable event credible.¹¹ That verdict, from a secular philosopher of the first rank, is worth more than a shelf of sympathetic defenders.

The State of the Scholarship

It would be triumphalist, and false, to say the debate is over. Hume’s essay remains one of the most discussed pieces in the philosophy of religion, and it retains able defenders. J. L. Mackie, in The Miracle of Theism, offered a careful reconstruction of Hume’s reasoning and judged that the balance of probability will almost always lie against accepting a miracle report—a serious case, honestly made, that any responsible treatment must reckon with.¹² The live dispute is not whether miracles are worth taking seriously as a concept—that much is now widely granted—but how to weigh testimony for extraordinary events, and how high the prior improbability of a miracle should be set.

What has genuinely shifted is the collapse of the confident a priori dismissal. Richard Swinburne’s work argued decades ago that there is no good reason to rule out miracles in advance and that historical evidence for them can carry real weight.¹³ And the weight of recent scholarship, reflected in reference surveys of the field, is that Hume’s argument does not establish its sweeping conclusion; even critics of miracles increasingly rest their case on the assessment of particular evidence rather than on Hume’s general maxim.¹⁴ The center of gravity has moved from “miracles are impossible or never believable” to “let us look at the evidence in each case”—which is exactly where the honest inquirer wanted the conversation to be.

Conclusion: What This Does and Does Not Establish

The argument of this essay is a door-opener, not a proof. It does not show that God exists—that case is made elsewhere, from the beginning of the universe, its fine-tuning, and the moral law. It does not show that any miracle has occurred. It certainly does not show that the resurrection happened; that is a historical claim, to be tested by the ordinary evidence of the empty tomb, the early creed, and the transformed witnesses.¹⁵ What it establishes is this and only this: neither Spinoza nor Hume has given a sound in-principle reason to refuse the evidence a hearing. Spinoza’s argument begs the question and presupposes a pantheism no theist accepts; Hume’s rests on a false premise, a hidden circle, and a mistaken method of weighing.

If God exists, miracles are possible; if miracles are possible, then whether one occurred is a question for the evidence, not for a rule laid down in advance. The believer need not claim more than that—and need not claim less. The courtroom that fixes its verdict before hearing the witnesses is not being rigorous; it is being closed. The reasonable thing, and the honest thing, is to unbar the door and let the testimony be heard. Those wishing to press on to the historical case will find it taken up in the companion essays on the minimal facts of the resurrection and the empty tomb.