Niklas Dahl
Affiliation: Lund University
Category: Philosophy
Keywords: Modality
Date: Wednesday 3rd of September
Time: 18:00
Location: Gen. Henryk Dąbrowski Hall (006)
View the full session: Modal Cognition & Problem Solving
My goal in this talk is to sketch an account of how alethic modal claims can be understood in terms of belief revision; the process of changing our beliefs as a result of encountering new information. This process is also one which we can simulate. When we consider hypothetical cases, such as supposing that P were the case, then we move to a belief-state where we have come to accept P and consider what else we accept there.
Essentially, the motivating idea of the framework I propose is this: if we look to how we normally evaluate an alethic modal claim, then we can see how such modal notions relate to hypothetical belief revision. What we normally do when we consider what's possible is to see if there is some change in belief which would lead us to accept the claim under consideration. And the changes in belief which underwrite such evaluations are simulations of exactly the same process by which we change our beliefs in response to new information. Building on ideas from Frank Ramsey (1929), and explored for counterfactual conditionals by Peter Gärdenfors (1988), I propose and explore Ramsey Tests for modal claims.
(□I) You ought to accept □φ if you accept φ under revision by ψ for all sentences ψ. (□E) If you accept □φ, then you ought to accept φ under revision by ψ for all sentences ψ. (♢I) You ought to accept ♢φ if there is some sentence ψ such that revision by ψ is consistent and leads you to accept φ. (♢E) If you accept ♢φ, then there ought to be some sentence ψ such that revision by ψ is consistent and leads you to accept φ.
These provide schemas for how to reason with alethic modal claims; characterising their norms of use. Further, drawing on an idea from Greg Restall (2012) - that alethic modality concerns subjunctive supposition, whilst epistemic modality relates to indicative supposition - I also argue that the resulting modal notions can be distinguished from epistemic modals, despite their epistemic basis.
One of the main up-shots of this approach is that it supports a formal semantics for modal logic. As I will briefly explain in the talk, we can use the Modal Ramsey Tests together with a characterisation of the norms governing belief revision to construct a formal semantics for well-known modal logics. This extends earlier work by Niklas Dahl (2023) on how we can adapt the AGM-framework for belief-change (Alchourrón, Gärdenfors, and Makinson, 1985) into a system of semantics which is sound and complete for propositional logic. This allows us to study how axioms of modal logic correspond to specific norms on belief revision. Further, it shows how we can explain different types of alethic modality in terms of what restrictions are placed on belief revision. All in all, it provides a replacement for possible worlds suitable to inferentialists and logical expressivists alike.
Essentially, this way of explaining the information conveyed by alethic modal claims is in terms of their inferential role. Belief revision, in turn, is meant to model the kind of epistemic change which we undergo all the time as we encounter new information. As such, it is part of our normatively governed epistemic practice. Our epistemic norms provide bounds for which ways of revising beliefs in the light of new information are allowed and disallowed. What alethic modal vocabulary does for us, then, is allow us to convey what beliefs about the world are within the bounds of our epistemic practice. Claims about possibility convey information about what one may believe within those bounds. Claims about necessity convey information about what one is bound by our epistemic norms to believe. Indeed, given that one is already a competent participant in the epistemic practice of revising one's beliefs, the Modal Ramsey Tests are sufficient to introduce the notions of necessity and possibility. In this way, they make explicit, in the sense discussed by Robert Brandom (1994, 2008), the limits for how our beliefs can change which were imposed by our epistemic norms even in the absence of that modal vocabulary.
Given this connection between which modal claims are acceptable and which norms one takes oneself to be bound by, the framework I'm proposing here is congenial to the view which Amie Thomasson (2020, 2023, forthcoming) has advocated for in recent books and papers: modal normativism. Drawing on empirical work within systemic functional linguistics (Halliday, 1973, 2009; Eggins, 2004), the core idea of this view is that the best way to understand metaphysical modal vocabulary is as a tool for conveying and negotiating semantic norms. Apart from providing a congenial way to explain modal logic, the belief revision framework extends the normativist view by providing a general account of the information conveyed by alethic modal claims. This allows us to generalise the approach beyond beyond metaphysical modality to also explain alethic modalities in general - such as the logical, physical, and nomological - and how such modalities can be ordered in terms of increasing strength. Further, as I will argue in the talk, the account presented in Thomasson (2022) of the semantics for specifically metaphysical modals, follows as a special case of the Modal Ramsey Tests.
Finally, time permitting, I will briefly sketch an answer some questions of modal epistemology. Specifically, I will show how the account is able to meet the well-known objections to modal knowledge put forward by Christopher Peacocke (1997, 1998) and Peter Van Inwagen (1998) against accounts built on possible worlds. Doing so allows me to illustrate the methods by which we can obtain modal knowledge within the present account, which makes every question about alethic necessity or possibility either a conceptual or an empirical matter.