On the pragmatic and epistemic virtues of inference to the best explanation

Abstract

In a series of papers over the past twenty years, and in a new book, Igor Douven (sometimes in collaboration with Sylvia Wenmackers) has argued that Bayesians are too quick to reject versions of inference to the best explanation that cannot be accommodated within their framework. In this paper, I survey their worries and attempt to answer them using a series of pragmatic and purely epistemic arguments that I take to show that Bayes’ Rule really is the only rational way to respond to your evidence.

On the Ecological and Internal Rationality of Bayesian Conditionalization and Other Belief Updating Strategies

“…Of course, Greaves and Wallace's (2006) argument is not the only one that has been advanced on behalf of conditionalization. A more recent argument shows that Bayesian conditionalization "dominates" all other updating rules, where a strategy is said to dominate another one if and only if it is better (or at least not worse) in all possible worlds according to some metric (Pettigrew, 2021). For reasons of space, I will not go into a detailed discussion of the argument.…”

“…First, the way we "score" the extent to which a strategy is rational should not be rigged against Bayesian conditionalization in the way that the above-mentioned example (where I pay you to have non-probabilistic degrees of belief) is rigged against probabilism. Indeed, Pettigrew (2021) argues that some of Douven's examples are flawed in precisely this way. Second, any alternative updating rule that is purported to be ecologically more rational than Bayesian conditionalization should have some independent theoretical motivation.…”

Abductive Reasoning in Science

“…also, e.g.,Weisberg, 2009;Henderson, 2014;Pettigrew, 2021). By contrast, I use "Inference to the Best Explanation" to refer more narrowly to the inferential account of abductive reasoning developed byHarman and Lipton, among others.…”

“…A "fair" bet is one that has an expected value of zero given the agent's personal probabilities, that is, roughly such that the agent can expect to break even in the long run if she repeatedly made the same bet.45 See alsoPettigrew (2021) for a version of this argument that doesn't appeal to rational betting behavior, but instead argues that using Bayesian Conditionalization (rather than Abductive Conditionalization) maximizes the expected accuracy of one's credences.https://doi.org/10.1017/9781009353199 Published online by…”