Is an Increase in Probability Always an Increase in Evidential Support?

Abstract

Peter Achinstein has argued at length and on many occasions that the view according to which evidential support is defined in terms of probability-raising faces serious counterexamples and, hence, should be abandoned. Proponents of the positive probabilistic relevance view have remained unconvinced. The debate seems to be in a deadlock. This paper is an attempt to move the debate forward and revisit some of the central claims within this debate. My conclusion here will be that while Achinstein may be right that his counterexamples undermine probabilistic relevance views of what it is for e to be evidence that h, there is still room for a defence of a related probabilistic view about an increase in being supported, according to which, if p(h|e) > p(h), then h is more supported given e than it is without e. My argument relies crucially on an insight from recent work on the linguistics of gradable adjectives.and partially wrong. If we focus on the positive form of 'being evidentially supported' or 'being evidence that h' in its ordinary sense, then Achinstein seems to be right in claiming that e being evidence that h cannot be fully captured by mere appeal to an increase in probability. Probabilist views that claim otherwise are problematic. However, if we focus on comparative forms, e.g. 'h is more supported given e, than without e', then there is still a room for probabilists to escape Achinstein's argument, or so I will argue. More specifically, I will suggest that Achinstein hasn't established that comparative evidential support cannot be understood in terms of probabilistic relevance, since a central premise in what appears to be Achinstein's main argument towards that more specific conclusion is problematic. The premise in question relies on a mistaken assumption.I will not question here what appears to be Achinstein's central claim, namely, that e being evidence that h cannot be fully understood in terms of an increase in probability. This article aims to show that the view that an increase in probabilities entails an increase in comparative support escapes Achinstein's counterexamples. In a sense, then, my proposal amounts to a new conciliatory suggestion. The central intuition from Achinstein's approach might be right. Our common-sense judgements about cases of increase in probability not always amounting to something being evidence that the hypothesis is true might be correct. However, this is not to say that the probabilistic accounts are not interesting and useful tools for thinking about science and some aspects of confirmation and comparative support in particular. What our discussion will show, I hope, is that as long as evidential support in science has anything to do with the ordinary, common-sense understanding of evidential support, being evidentially supported cannot be merely reduced to an increase in probability, even though Achinstein's arguments don't establish that comparative evidential support cannot be understood in terms of probabilistic relevance.In what follows, I first introduce Achin...

Graded epistemic justification

“… 16 For more on the dual life of ‘confident’ see Logins 2019 . For an exploration of another epistemically significant gradable adjective that seems to exhibit a maximal only interpretation, namely ‘supported’, see Logins 2020 . …”

Is Evidential Support the Same as Increase-in-Probability?

“…For Replies to Achinstein, seeBar-Hillel and Margalit (1979),Maher (1996), andRoush (2004). SeeLogins (2020) for a detailed discussion and analysis of the debate between Achinstein and his critics.5 Achinstein has responded to this response. Briefly, he denies that there are two notions of support: static and dynamic.…”

“…Logins (2020), in response to Achinstein's lottery example, grants that Support-IP is false. Still, he suggests a more restrictive version of Support-IP, which is solely concerned with comparative claims of the following type: "H is more supported given E than it is without E".…”