Fisika & Matematika editorial
Open AccessOA2023
An Alleged Tension Between non-Classical Logics and Applied Classical Mathematics
Timothy Williamson argues that using classical mathematics in science conflicts with adopting non-classical logics. Horvat and Toader show this tension dissolves for quantum logic and other non-classical logics, drawing broader lessons about applied logic and mathematics.
Sebastian Horvat; Iulian D. Toaderยท The Philosophical Quarterlyยท 2023ยท DOI 10.1093/pq/pqad125
The core problem
Timothy Williamson has recently argued that the applicability of classical mathematics in the natural and social sciences raises a problem for the endorsement, in non-mathematical domains, of a wide range of non-classical logics. The alleged tension arises because classical mathematics is formulated using classical logic; if one employs classical mathematics to model a domain, one seems committed to classical logic at least for that mathematical theory. But if one also endorses a non-classical logic for reasoning about the domain itself, a conflict emerges: the non-classical logic may sanction inferences that are invalid in classical logic, yet the mathematical model relies on classical logic. Williamson takes this to be a serious challenge for non-classical logics. The authors aim to reconstruct Williamson's argument, focus on the case of quantum logic (QL), and demonstrate that no problematic tension exists between the applicability of classical mathematical models to quantum phenomena and the endorsement of QL in reasoning about those phenomena. They then identify the premise that fails for QL and argue that it fails for a wider variety of non-classical logics, drawing general lessons about the relationship between applied logic and applied mathematics.
Innovation
The authors proceed by first reconstructing Williamson's argument in its general form. They then restrict the argument to the specific case of quantum logic, which is a non-classical logic motivated by the peculiar features of quantum mechanics. By examining how classical mathematics is actually applied in quantum theory, they test whether the alleged tension materializes. The method is philosophical analysis: identifying the premises of Williamson's argument, checking their truth in the quantum-logical context, and showing that at least one premise is false when restricted to QL. They then generalize the point by arguing that the same premise fails for other non-classical logics as well. Finally, they draw broader conclusions about the relationship between applied logic and applied mathematics, suggesting that the applicability of classical mathematics does not force the adoption of classical logic in the domain of application.
Introduction
Timothy Williamson has recently argued that the applicability of classical mathematics in the natural and social sciences raises a problem for the endorsement, in non-mathematical domains, of a wide range of non-classical logics. The alleged tension arises because classical mathematics is formulated using classical logic; if one employs classical mathematics to model a domain, one seems committed to classical logic at least for that mathematical theory. But if one also endorses a non-classical logic for reasoning about the domain itself, a conflict emerges: the non-classical logic may sanction inferences that are invalid in classical logic, yet the mathematical model relies on classical logic. Williamson takes this to be a serious challenge for non-classical logics. The authors aim to reconstruct Williamson's argument, focus on the case of quantum logic (QL), and demonstrate that no problematic tension exists between the applicability of classical mathematical models to quantum phenomena and the endorsement of QL in reasoning about those phenomena. They then identify the premise that fails for QL and argue that it fails for a wider variety of non-classical logics, drawing general lessons about the relationship between applied logic and applied mathematics.
Why it matters
The authors analyze why the tension seems plausible: it relies on an overly monolithic view of logic, where the logic of the mathematical model must govern all reasoning about the domain. Instead, they advocate a more nuanced picture in which different logics can be appropriate at different levels of description. The failure of the premise for QL and other non-classical logics suggests that the applicability of classical mathematics does not impose classical logic on the domain. The discussion draws general lessons: applied logic and applied mathematics are distinct enterprises, and the choice of logic for a domain is not dictated by the mathematics used to model it. This has implications for debates about logical pluralism and the role of logic in science. The authors conclude that Williamson's argument does not succeed in showing a problematic tension, and that non-classical logics can be endorsed in non-mathematical domains without conflict with classical mathematics.
Who should read this
Pembaca Research Digest dan praktisi rumpun ini.
Opening member contentโฆ