Julian Vincent Laggner
University of Vienna
Regarding the question of what a propensity’s degree is and how it fixes a probability, there are broadly speaking three ways of answering it. (1) The degree is a probability: the view of nearly every historical propensity theorist, including Popper, Fetzer, Miller, and Gillies, which commits them to what Suárez calls the identity thesis. (2) The degree is real, non-probabilistic, and unbounded, as Anjum and Mumford hold in What Tends to Be. (3) There is no degree, as I argue is implicit in Suárez’s account.
Position (1) was long standard, but it collapses once both halves of the identity thesis are rejected: Salmon’s counterexamples (1979) show that identity1 “cannot generally hold” (Suárez 2014, p. 221), and Humphreys’ paradox (1985) removes identity2. A further reason is explanatory circularity: propensities cannot ground the facts of probability, in Anjum and Mumford’s own words, “on pain of circularity, if propensities were themselves already probabilistic in nature” (Anjum & Mumford 2018, p. 56).
Anjum and Mumford use overdisposing, cases in which a power has “a stronger magnitude than what is minimally needed to bring about a particular effect” (Anjum & Mumford 2018, p. 50), to distance themselves from every other propensity theorist. They accept the conclusion of Humphreys’ paradox but hold that overdisposing gives an independent argument for non-identity. An unbounded scale then raises the problem of compressing unbounded quantities into the bounds of the probability calculus.
Following Psillos and Ioannidis (2019), and adding arguments of my own, I argue that this account faces questions it cannot answer. These problems range from the absence of a clear link between tendencies and probabilities, through an unexplained and overly weak account of causal production, to an epistemological problem.
I conclude that these difficulties are generated by positing a degree at all, and so cannot arise for Suárez’s Complex Nexus of Chance (2020), on which propensity, chance, and frequency are distinct and mutually irreducible. That account further offers a worked epistemology of propensity ascription (Ruyant & Suárez 2025) and a close connection with statistical modelling practice. Suárez’s indexed probability account (2018, 2025) resolves Humphreys’ paradox without abandoning the Kolmogorov axioms, and does so while remaining compatible with a range of accounts of the nature of chance.
Anjum, Rani Lill, & Mumford, Stephen (2018). What Tends to Be: The Philosophy of Dispositional Modality. New York: Routledge.
Humphreys, Paul (1985). Why propensities cannot be probabilities. Philosophical Review 94 (4):557-570.
Psillos, Stathis, & Ioannidis, Stavros (2019). [Review of the book What tends to be: The philosophy of dispositional modality, by R. L. Anjum & S. Mumford]. Notre Dame Philosophical Reviews, 2019.03.35.
Ruyant, Quentin & Suárez, Mauricio (2025). Probabilistic empiricism. European Journal for Philosophy of Science 15 (2):1-19.
Salmon, Wesley C. (1979). Propensities: A discussion review. Erkenntnis 14 (2):183-216.
Suárez, Mauricio (2014). A critique of empiricist propensity theories. European Journal for Philosophy of Science, 4 (2): 215-231.
Suárez, Mauricio (2018). The chances of propensities. British Journal for the Philosophy of Science, 69 (4): 1155-1177.
Suárez, Mauricio (2020). Philosophy of Probability and Statistical Modelling. Cambridge University Press.
Suárez, Mauricio (2025). The possibilities in propensities: Emergence and modality. In T. Knuuttila, T. Grüne-Yanoff, Y. Sjölin Wirling, & R. Koskinen (Eds.), Modeling the possible: Perspectives from philosophy of science. Routledge.
Chair: Aphrodite Simalas
Time: 14:40-15:10
Location: SR 1.006
SOPhiA 2026 