Hume's Abject Failure: The Argument Against Miracles (review)

As his uncompromising title announces, John Earman considers Hume's famous account of miracles in the Enquiry Concerning Human Understanding to be an "abject failure." More than this, the author judges Hume's well-known arguments to have failed in multiple ways. As he states in the opening paragraph: "It is not simply that Hume's essay does not achieve its goals, but that his goals are ambiguous and confused." Hume's deliberations on this topic, Earman further contends, are unoriginal, and where original, insubstantial. Further, the logic of the position expressed in the essay "reveals the weakness and poverty of Hume's own account of induction and probabilistic reasoning." Finally, "the essay represents the kind of overreaching that gives philosophy a bad name." While there are occasional hints in the book that the author is himself not completely immune to this latter tendency, for the most part these bluntly stated objections to the famous essay are well supported. This book is a comprehensive and persuasive demolition of Hume's much vaunted claims about the credibility of miracle reports, and it issues a serious challenge to standard readings.

Hume's Abject Failure is divided into two parts, the first of which takes the form of an analysis and critique of Hume's arguments. This eighty-page section is followed by another hundred pages of primary materials, mostly eighteenth-century, which address issues related to Hume's essay. It is clear from the format of the book that Earman intends to take seriously the intellectual context in which Hume's arguments first appeared. This a most welcome development. Many treatments of Hume's "Of Miracles" tend to be profoundly ahistorical. What becomes immediately apparent from a consideration of this context is the startling unoriginality of Hume's chief argument. The principle that uniform experience overrides testimony to the occurrence of a miracle was aired long before Hume claimed it as his own, although it can be conceded that in Hume's hands it took on a particularly persuasive form.

More importantly, however, this is an argument which does not work. Earman claims that Hume would have been aware of this had he availed himself of the developing science of probability, and in particular Thomas Bayes's pioneering attempts to quantify degrees of belief and credibility. Hume's probabilistic arguments, even on the most charitable interpretation, simply fail the test of a Bayesian analysis, and a number of Earman's chapters forcefully demonstrate this. Having exposed the dubious nature of the argument based on competing testimony, Earman goes on to makes good his claim about the inadequacy of Hume's view of induction, showing how it would rule out not merely miracle testimony, but many empirical claims of the kind typically made in the sciences.

It is perhaps a little unfair to charge Hume with having articulated a flawed argument the limitations of which only become obvious with the application of a sophisticated calculus unavailable to him. There is no evidence that Hume ever read Bayes. Earman points out that Richard Price, in his 1767 critique of the argument, drew Hume's attention to the possibility of a Bayesian approach, but Hume was apparently reluctant to incorporate quantification into subsequent editions of the Essay. Whether this amounts to a major failing on Hume's part is debatable. It is also worth noting that at times Earman seems to blur the contemporary distinction between "internal" and "external evidences" (miracles were instances of the latter). This distinction is important for understanding why Hume did not adopt the subjective notions of "miracle" of Locke and the Newtonians, and also why in section two of the Essay he deploys the 'contrary miracles' argument. There is also the puzzle as to how so acute a philosophical mind as Hume's could in this instance have...