The following is the first of a three-part series.
The argument from miracles seeks to prove that a religious deity (such as God) exists on the premise that only God could have caused a miracle to occur. David Hume’s “Of Miracles” has proven to be the most important philosophical essay on this argument. In his essay, Hume develops a sophisticated epistemic case against the reliability of testimony in behalf of a religious miracle.[1] There is much debate, however, about what his argument actually is, and whether or not it succeeds, in the end.
The vast literature on Hume’s argument in “Of Miracles” means there are many competing reconstructions of his argument in the literature. Perhaps the most fundamental divide among commentators turns on the question: Does Hume’s objection to the argument from miracle aim to rule out the possibility of religious miracles a priori, or does Hume remain open to the possibility that a miracle might be proved on the basis of testimony?
My preferred reading is that Hume aims to argue that religious miracles are provable in principle but virtually impossible to prove in light of the historical unreliability of religious testimony. Perhaps this reading’s strongest proponent is Robert Fogelin, whose most important work on the topic argues that Hume’s critique of miracles is widely misread and misunderstood.[2]
The objection to Hume’s argument which I develop in this article (culminating in section 3) strikes a decisive blow against Hume if we go with Fogelin’s reading of Hume. There I argue that Hume’s case against religious miracles is superfluous. It is superfluous because it is impossible to prove a religious miracle on epistemic evidence, alone. To show that no miracle can be established so as to be the foundation for a system of religion, it is not necessary to show that testimony in behalf of religious miracles is unreliable, for religious miracles have an ineliminable subjective component that makes them logically impossible to prove on epistemic grounds. Epistemic considerations can establish an event and its cause, but not how one ought to react toward either of these.
This is not to say that my critique of Hume has no bite on a traditional reading which asserts that Hume’s argument intends to rule out the possibility of religious miracles.[3] Interpreters sometimes attribute to Hume an aprioristic critique of miracles that locates his argument in the clash of proof against proof, at the end of part 1. If Hume in fact thinks that miracles are impossible, as the traditionalist view maintains, then the argument I offer here could still be read as undercutting Hume’s case against miracles.
Any such line of argument is superfluous on my account, for no clash of proof against proof is possible if satisfying a subjective condition is essential to the occurrence of a religious miracle. The reason is that the clash in question presumes that it is possible for there to be a full proof of a religious miracle. I argue, however, that a proof of a “Humean religious miracle” is not a proof of a religiously significant event.
Section 1 of this paper presents Hume’s two definitions of “miracle” and argues that Hume believed there are two types of miracle, religious and non-religious. Section 2 argues that Hume’s account of religious miracle presupposes a view of (ontic and epistemological) religious significance. Being a non-religious person, it is not surprising that Hume nowhere develops an analysis of religious significance in explicit terms. His silence notwithstanding, a Humean account of religious significance can be reconstructed from what he thinks distinguishes religious and non-religious miracles. Section 3 defends a subjective condition of epistemological religious significance which the Humean account cannot accommodate without undercutting the epistemic significance of religious miracles.
Humean Religious Miracles
Hume does not present an a priori argument against miracles, nor does he provide independent arguments in parts 1 and 2 of his essay. Rather, the arguments of both parts 1 and 2 are a posteriori (grounded in experience) and are meant to be read together: they build a cumulative and comprehensive case against miracles. What follows is a crude summary of Hume’s argument in parts 1 and 2.
Part 1 articulates an epistemic framework of reliable testimony that consists of two tests, the direct and reverse tests.[4] The direct test provides criteria for evaluating the quality of the reports themselves and on the qualifications of reporters. Hume’s criteria apply to all forms of testimony and not just to testimony in behalf of miracles. These criteria include consistency (the witnesses concur with one another, rather than contradict one another), number (they are many, not few), character (they are of unimpeachable, rather than of doubtful, character), and similar conditions. To the extent that reports of a single event meet these criteria, the probability of the reported event increases, perhaps amounting to what Hume calls a “proof.”
A proof in Hume’s sense denotes testimony that is certain—as certain as anything can be, in the light of past experience. Hume uses the term “uniform testimony” to signify reports of a single event that flawlessly pass the direct test by satisfying all criteria of reliable testimony, thereby rising to the level of a proof. In practice, multiple reports of a single event are rarely flawless. Yet an event can be (highly) probable on less than uniform reports, for reliability comes in degrees.
The reverse test appraises the intrinsic probability or improbability of an event attested; that is, an event is assigned a probability prior to examining the evidence in its behalf, based on how likely or unlikely it is on past experience. When testimony attests to a common form of experience (such as running into an acquaintance by chance) the prior probability of the event is normal, and the testimony passes the reverse test.
However, when a rare event is attested—including extraordinary and marvellous events—the prior improbability of the event can be so high (before looking at the evidence) that it supplies strong support for asserting its non-occurrence. In the limit case, the reverse test yields a proof of the non-occurrence of the event.[5] Fogelin defends the reverse test by arguing that we apply it in our day-to-day lives, not just when evaluating miracle reports. He asks us to suppose that President George Bush is said to have been observed tightrope-walking over his swimming pool. Most people’s initial reaction would be disbelief. The sheer bizarreness and improbability of such an event’s taking place casts immediate doubt on the force of the testimony offered in its behalf.[6]
Fogelin argues that there can never be a proof against a marvellous event, because such events are compatible with past experience in the sense that they have some basis past experience. A proof on the reverse test is achieved when the event attested is a miracle, which Hume defines as a violation of the laws of nature. Because laws of nature are backed by a firm and unalterable experience, a miracle, say, walking on water, runs contrary to the whole of experience. We thus have something along the lines of a proof against the miracle.
Fogelin argues that Hume’s two methods for evaluating testimony set the stage for the possibility of a clash of proof against proof. This occurs when we have a proof that a miracle occurred on the direct test, and a proof that it did not occur on the reverse test. In the clash of proof against proof, we cannot determine a priori which proof to reject and which to accept. We must therefore apply the following principle: “If the falsehood of his testimony would be more miraculous, than the event which he relates; then, and not till then, can he pretend to command my belief or opinion.”[7]
Fogelin clarifies this principle thus: “A proof on a par in strength with its counterproof yields no more than a draw. A direct proof stronger than the reverse counterproof will still be diminished in strength by the counterproof it surpasses. What is needed—to put it metaphorically—is a direct proof that outdistances the reverse counterproof by the full length of a proof.”[8] This completes the first part of Hume’s case against miracles, and part 1.
In part 2, Hume argues that, as a matter of fact, no attested miracle—especially of the religious variety—has passed the direct test. He provides four considerations in support of this claim: (1) first, no testimony in behalf of a miracle has ever been very extensive and uniform so as to pass the direct test; (2) second, human are psychologically prone to believe marvellous and surprising claims, and this tendency can overpower good sense; (3) third, miracles are observed chiefly among “ignorant and barbarous nations”—i.e., miracle reports originate in societies that are susceptible to the epistemic vices of gullibility, dogmatism, and uncritical deference to authority; (4) fourth, and finally, miracles often aspire to establish the truth of one religious system over all others.
In such cases, the miracle reports of one religions are incompatible with the miracle reports of rival religions, and since only one religion can be uniquely true we have good reason to doubt all miracle reports. The rest of part 2 provides a brief assessment of the veracity of well-known miracle reports; Hume argues that all of them fail the direct test. This concludes part 2 of Hume’s essay, and with it, his case against religious miracles.
Hume’s two definitions of “miracle”
The above summary was provided for the benefit of those that are unfamiliar with Hume’s argument. I now develop an analysis and critique of Hume’s case against miracles. As we have seen, Hume’s case depends upon his definition of “miracle” as a violation of the laws of nature. I call this first definition the Humean account of miracle—or Humean miracle, for short. In this section, I argue that Hume’s case against miracles more accurately depends upon his second definition of “miracle” (a definition buried in a footnote of part 1 of his essay). My premise is that Hume’s argument against miracles requires a distinction between religious and non-religious miracles. Having established the primary target of Hume’s argument, I will proceed to unpack his account of religious significance, in section 3.
In a footnote, Hume provides a definition of “miracle” more elaborate than the first: “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.”[9] The Humean account of religious miracle—or (Humean) religious miracle,for short—posits two necessary and sufficient conditions for a religious miracle: (1) a violation of the laws of nature (2) caused by a religious deity or invisible agent (presumably, acting in behalf of a religious deity). One question we may have is whether Hume’s second definition is an unpacking of the first definition or whether Hume thinks there are two kinds of miracles (which correspond, respectively, to his two definitions)? I argue for the latter interpretation.
Hume uses the term “religious miracle” only twice in his essay—in both cases, he does so to point out that religious miracles are more improbable (hence, more difficult to prove) than non-religious miracles. For example, in the first passage, Hume argues that humankind has a strong, natural propensity to believe the “extraordinary and the marvellous”; to this he adds that such human passions (alongside others)—which incline “the greatest vehemence and assurance”—are stronger in the case of “religious miracles.”[10] In the second passage, he makes a similar observation: “As the violations of truth are more common in the testimony concerning religious miracles, than in that concerning any other matter of fact; this must diminish very much the authority of the former testimony, and make us form a general resolution, never to lend any attention to it, with whatever specious pretence it may be covered.”[11]
This textual evidence shows that Hume distinguishes religious miracles from marvellous and mundane events. However, it falls short of showing that Hume distinguishes religious miracles from non-religious miracles. It might be thought that Hume’s second definition adds nothing of substance, since it merely makes explicit what is implicit in his first definition. That is, one might think that the very essence of a Humean miracle entails supernatural agency; hence, that a natural entity cannot be the cause of a violation of the laws of nature. Unfortunately, this view runs into textual difficulties.
Philosophically, why should we think, as a purely a priori matter, that miracles are and must be effects of supernatural causation? One might attribute this position to Hume based on the premise that to bring about a miracle an agent must act in a way that violates the laws of nature and it is impossible for a natural entity to so act, because it is part of the essence of natural entities that they are governed by them.
If the behaviour of natural entities is limited to possibilities set by the very laws which describe their behaviour, then natural entities can never violate them but can only be compelled to behave in ways that are consistent with them. On this argument, a miracle is, analytically, a supernatural effect—the product of supernatural causation—and never a natural effect—the product of natural causation.
Hume, for one, seems to disagree with the conclusion of this argument, for he explicitly asserts that a violation of the laws of nature might be caused by the “interposition of some invisible agent,” and it is unclear that an invisible agent must be a supernatural entity. It might be replied that Hume’s references to “religious deities” and “invisible agents” can be plausibly interpreted as references to supernatural agents.
However, I find it just as plausible that an invisible agent might be a natural entity. Is a ghost, for example, a supernatural phenomenon? In any case, this reading is inconsistent with textual evidence supporting the possibility of non-religious miracles. A non-religious miracle is a violation of the laws of nature that has a natural cause; as such, it is naturalistically explicable, at least in principle.[12] In what follows, I provide textual evidence to show that Hume rejects the notion that all miracles are religious miracles.
Upon concluding that religious miracles cannot be established on the basis of testimony, Hume writes: “For I own, that otherwise, there may possibly be miracles, or violations of the usual course of nature, of such a kind as to admit of proof from human testimony; perhaps, it will be impossible to find any such in all the records of history.”[13] To illustrate this, he provides a hypothetical example of an eight-day eclipse:
Thus, suppose, all authors, in all languages, agree, that, from the first of January 1600, there was a total darkness over the whole earth for eight days. Suppose that the tradition of this extraordinary event is still strong and lively among the people: that all travellers, who return from foreign countries, bring us accounts of the same tradition, without the least variation or contradiction: it is evident, that our present philosophers, instead of doubting the fact, ought to receive it as certain, and ought to search for the causes whence it might be derived. The decay, corruption, and dissolution of nature, is an event rendered probable by so many analogies, that any phenomenon, which seems to have a tendency towards that catastrophe, comes within the reach of human testimony, if that testimony be very extensive and uniform.[14]
The aim of this passage is to stress the difficulty of establishing a religious miracle. For though extensive and uniform testimony might establish the eight-day miracle—which, for Hume is a non-religious miracle—no amount of extensive and uniform testimony can establish a religious miracle. If there is any doubt that this is Hume’s position, he removes it in the paragraph immediately following the above quotation. After the quoted passage, Hume articulates a hypothetical religious miracle which he says he would immediately reject because he believes that its overt religious significance is a sure sign of a cheat:
But suppose, that all the historians who treat of England, should agree, that, on the first of January 1600, Queen Elizabeth died; that both before and after her death she was seen by her physicians and the whole court, as is usual with persons of her rank; that her successor was acknowledged and proclaimed by the parliament; and that, after being interred a month, she again appeared, resumed the throne, and governed England for three years: I must confess that I should be surprised at the concurrence of so many odd circumstances, but should not have the least inclination to believe so miraculous an event. I should not doubt of her pretended death, and of those other public circumstances that followed it: I should only assert it to have been pretended, and that it neither was, nor possibly could be real. You would in vain object to me the difficulty, and almost impossibility of deceiving the world in an affair of such consequence; the wisdom and solid judgement of that renowned queen; with the little or no advantage which she could reap from so poor an artifice: All this might astonish me; but I would still reply, that the knavery and folly of men are such common phenomena, that I should rather believe the most extraordinary events to arise from their concurrence, than admit of so signal a violation of the laws of nature.[15]
Faced with uniform and extensive testimony in behalf of Queen Elizabeth’s rising from the dead, Hume says not only that he would reject this claim on grounds that religious testimony is unreliable, but that he would conclude it “neither was, nor possibly could be real.” Hume is here speaking hyperbolically, as Fogelin argues; that is, he does not literally mean the Queen Elizabeth miracle is impossible. Rather, he means that its prior improbability is so high that, even if the testimony in its behalf is so extensive and uniform as to persuade all historians of England that it occurred, this proof would be less probable than that the Queen faked her death. Fogelin offers two reasons in support of the hyperbolic reading of Hume’s strong anti-religious language.[16]
First, the hyperbolic reading is supported by Hume’s explanation of his peculiar use of the term “proof” elsewhere in his writings.[17] Second, Fogelin appeals to something like the principle of charity (which requires that whenever a passage is ambiguous the most plausible interpretation should be attributed to an author). Given this principle, if Hume’s strong language against religious miracles can be read on one of two ways—say, either as hyperbolical language or as expressing overt religious bias—we ought to be charitable in ascribing the reading that most strengthens the author’s position.
The former is the more charitable in the present case, because it renders Hume’s strong language against religious miracles consistent with the epistemic framework of testimony he develops in part 1. The possibility of religious miracles notwithstanding, Hume thinks there has never been very extensive and uniform testimony in behalf of a religious miracle.[18]
Hume’s view that religious miracles are virtually impossible to prove on testimony explains why Hume calls his second definition of “miracle” more accurate than his first definition. It is more “accurate” in that it is the definition religious apologists tacitly presuppose when they argue that the occurrence of a miracle provides a just foundation for a system of religion. That is, condition (2) of the second definition specifies a necessary condition of religious miracles.
Though Hume is sceptical of all arguments from miracles, the textual evidence suggests that religious miracles are his primary target. Perhaps the most telling passage of all to this effect is his thesis statement, which he repeats twice (in almost identical phrasing): “I beg the limitations here made may be remarked, when I say, that a miracle can never be proved, so as to be the foundation of a system of religion.”[19]
We can now conclude that, for Hume, not all miracles are religious miracles. Yet Hume also thinks that only miracles that happen to be religious miracles can confirm the truth of religious claims. Indeed, the capacity for a miracle to confirm a religious claim is essential to what Hume means by “religious miracle.” We can further conclude that Hume’s second definition—his account of religious miracles—is not a mere unpacking of his first definition. That is, Hume seems wedded to the possibility of two types of miracle: non-religious miracles (per his first definition) and religious miracles (per his second definition). Were he committed to the view that all miracles are religious miracles, he would not be able to claim, as he does, that religious miracles have a heftier evidential burden than non-religious miracles.[20]
Alberto Urquidez is a CFD Postdoctoral Fellow at Bowdoin College. He is the author of (Re-)Defining Racism: A Philosophical Perspective (Palgrave Macmillan, 2020).
______________________________________________________________________________
[1] David Hume, “Of Miracles,” in An Enquiry Concerning Human Understanding, ed. L. Beauchamp Tom (Oxford & New York: Oxford University Press, 1999).
[2] Robert J. Fogelin, A Defense of Hume on Miracles (Princeton University Press, 2003).
[3] Chapter 2 of Fogelin’s book criticizes misconceptions of Hume’s argument in David Johnson’s Hume, Holism, and Miracles and John Earman’s Hume’s Abject Failure: The Argument Against Miracles. For a critique of Fogelin’s reading of Hume, see Robert A. Larmer, “Interpreting Hume on Miracles,” Religious Studies 45 (2009),https://doi.org/doi:10.1017/S0034412509009962. Larmer defends the traditional reading of Hume’s thesis in “Of Miracles.”
[4] These labels are Fogelin’s (see A Defense, 6-7), not Hume’s. Textual support for the direct test: “We entertain a suspicion concerning any matter of fact, when the witnesses contradict each other; when they are but few, or of a doubtful character; when they have an interest in what they affirm; when they deliver their testimony with hesitation, or on the contrary, with too violent asseverations. There are many other particulars of the same kind, which may diminish or destroy the force of any argument, derived from human testimony” (Hume, “Of Miracles,” 171). Textual support for the reverse test: “Suppose, for instance, that the fact, which the testimony endeavours to establish, partakes of the extraordinary and the marvellous; in that case, the evidence, resulting from the testimony, admits of a diminution, greater or less, in proportion as the fact is more or less unusual” (ibid., 171-72).
[5] Fogelin argues that assigning prior probabilities and improbabilities is partly a contextual matter. Consider a report of someone tight-rope walking. Normally it is unusual to witness this event during a casual walk; however, this is not unusual for someone walking in a tightrope-walking community.
[6] Under normal circumstances, it seems more reasonable to treat the report as a hoax or misinterpretation. To be sure, there may be video footage and other compelling evidence available that undermines our initial reaction (i.e., refutes the event’s prior improbability). If such direct evidence of the event is sufficiently uniform, extensive and reliable, it may well refute the reverse test.
[7] Hume, “Of Miracles,” 174. Hume reiterates this point in part 2: “When, therefore, these two kinds of experience are contrary, we have nothing to do but subtract the one from the other, and embrace an opinion, either on one side or the other, with that assurance which arises from the remainder” (ibid., 184).
[8] Fogelin, A Defense, 15-16.
[9] Hume, “Of Miracles,” 173.
[10] Ibid., 176.
[11] Ibid., 185.
[12] An event that is naturalistically explicable in principle may be an event that is currently unexplained; it may be an event that is inexplicable given the current state of the evidence. It is naturalistically explicable if it can be explained in natural terms alone (i.e., if it would be possible to do so) given a thorough and complete understanding of the relevant facts.
[13] Hume, “Of Miracles,” 184.
[14] Ibid., 184.
[15] Ibid., 184-85.
[16] Fogelin describes his two reasons as “levels of scrutiny” in textual interpretation (A Defense, 17).
[17] Hume’s use is not the logician’s use. Unlike a demonstration in deductive logic which is infallible in that the conclusion of an argument is true whenever its premises are, a Humean proof is essentially fallible because proofs are always based on past experience and the future might not resemble the past. “A proof, commonly understood,” writes Fogelin, “is something that settles a matter—something that makes further investigation unnecessary, perhaps even irrational” (ibid., 16). This understanding is based on the following passage, from Hume: “Mr. LOCKE divides all arguments into demonstrative and probable. In this view, we must say, that it is only probable all men must die, or that the sun will rise to-morrow. But to conform our language more to common use, we ought to divide arguments into demonstrations, proofs, and probabilities. By proofs meaning such arguments from experience as leave no room for doubt or opposition” (Hume). The fallibility of a Humean proof is evidence that his argument is epistemological rather than ontological: it measures rationality, based on past experience, rather than truth.
[18] Fogelin provides a hypothetical case of a religious miracle that he thinks Hume’s method requires him to accept: “To alter Hume’s own example, suppose that for eight days all was dark save for an illuminated face that simultaneously appeared throughout the world, speaking in a way intelligible to all, offering many proofs of his or her magnificence, and so on. (The story could be further filled in with universal cures, resurrections, whatever.) We would then have a case that does parallel Hume’s example of a natural miracle, and it would surely be a matter of prejudice for him to reject the testimony in behalf of the religious miracle while accepting the testimony in behalf of the natural miracle. Hume’s point, however, is that the local, sect-serving testimony that has been offered in behalf of religious miracles falls hopelessly short of standards of testimony satisfied by Hume’s imagined case of a natural miracle” (A Defense, 29).
[19] Hume, “Of Miracles,” 184. The evidence against religious miracles, claims Hume, is so overwhelming that only a charlatan or self-deluded individual could believe otherwise: “human testimony, in these circumstances, loses all pretensions to authority. A religionist may be an enthusiast, and imagine he sees what has no reality: He may know his narrative to be false, and yet persevere in it, with the best intentions in the world, for the sake of promoting so holy a cause[…]” (ibid., 175).
[20] The conclusion of my argument is one Fogelin agrees with. Regarding Hume’s second definition, Fogelin writes: “[Hume] seems to be making the intervention of a divine (or at least invisible) agent an essential feature of a miracle. Yet elsewhere, indeed, even in this very footnote, Hume uses the notion of a miracle in a wider sense that includes the notion of nonreligious miracles. The discussion in part 2 relies on this contrast between religious and nonreligious miracles. It seems, then, that Hume’s intention here cannot be to narrow the notion of a miracle to religious miracles, but only to define one particular kind of miracle—those that are supposed to depend upon divine intervention” (A Defense, 14, note 6).
