Bayesian reasoning provides a rigorous framework for evaluating extraordinary claims, yet it is only as sound as the assumptions encoded in priors, likelihoods, and model structure. One often overlooked principle is the necessity of an "Unknowns Reserve"-a probability allocation for mechanisms not yet conceived or understood. History is replete with examples where premature attribution of phenomena to divine agency collapsed under the weight of later discoveries: lightning once attributed to Zeu…
Read moreBayesian reasoning provides a rigorous framework for evaluating extraordinary claims, yet it is only as sound as the assumptions encoded in priors, likelihoods, and model structure. One often overlooked principle is the necessity of an "Unknowns Reserve"-a probability allocation for mechanisms not yet conceived or understood. History is replete with examples where premature attribution of phenomena to divine agency collapsed under the weight of later discoveries: lightning once attributed to Zeus yielded to atmospheric electricity, diseases once thought to be demonic aictions yielded to microbial theory, and eclipses once feared as divine portents yielded to orbital mechanics. This paper argues that properly applied Bayesian reasoning must institutionalize such a reserve as an epistemic norm, particularly in assessing historical miracle claims that can no longer be tested under scientific scrutiny. Without this reserve, Bayesian models risk premature closure, inflating the plausibility of the miraculous and underestimating the vast space of possible natural mechanisms. By situating the Unknowns Reserve within the broader philosophy of science (Jaynes, 2003; Dawid, 2005), cognitive psychology (Loftus & Palmer, 1974; Gigerenzer, 2002), and historical epistemology (Hume, 1748/2007; Earman, 2000), the paper shows why apologetic deployments of Bayes that omit it misrepresent the mathematics. The conclusion is that the Unknowns Reserve is not skepticism for its own sake but a structural safeguard for rational inquiry in contexts where evidence is fragmentary, testimony is fallible, and Bayes' theorem is often invoked as a tool for adjudicating extraordinary claims, such as the resurrection of Jesus or reports of miraculous healings (Carrier, 2012, 2014). The formalism promises neutrality: evidence is weighed against competing hypotheses, with posterior probabilities reflecting the relative strength of each. Yet the neutrality of the framework depends on whether all live alternatives are represented fairly. When unimagined or unknown mechanisms are excluded from consideration, the evidential weight is artificially funneled toward favored hypotheses. This omission is.
