Imposing Risk

A Normative Framework

Bibliographic Data

ID10691578
AuthorsTony Reeves (0000-0003-1733-8414, Binghamton University, corresponding author)
Year2018
Publication date2018-05-17
Open AccessYes
TypeARTICLE
VenueThe Philosophical Quarterly (JOURNAL)
Journal identifiersISSN: 0031-8094 • E-ISSN: 1467-9213
PublisherOxford University Press (PUBLISHER • GB)
DOI10.1093/pq/pqy020
OpenAlexW2804051032
LanguageEN
Citations received17

John Oberdiek's Imposing Risk develops a novel contractualist approach to risk that valuably contributes to the now emerging subfield of risk ethics. Oberdiek defends positions on three fundamental issues: how to define the risk of an act, why risking is morally significant, and what standards govern the permissibility of risk taking. The treatment of the first two matters is more sustained than the third, so I focus on these. Understanding risk as the probability of a harm, Oberdiek begins by elaborating the challenge of saying which probability assignment is morally salient. Subjective theories of probability (Bayesian approaches) assign probability based on a given person's actual level of credence in an outcome. Since subjective accounts rely on actual estimation, they are perspective-dependent, as different persons can have different levels of credence even with the same evidence, and perspective-indifferent, as they cannot privilege any particular perspective. This leaves subjective approaches insufficiently normative from the viewpoint of moral theory, as we have no authoritative basis for adjudicating differences in risk estimates, and thus cannot say which estimates are relevant to justifying activity (p. 24). Relative frequency, as an objective account, founders also. The relative frequency approach assigns risk to an act by classifying it and determining which proportion of the reference class produces the harm in question. This offers the prospect of an objective standard, but faces the well-known reference class problem: relative frequency theory itself cannot privilege any particular classification, and different classifications will have different frequencies. Thus, as with subjective theories, there will be no the risk to assign to an act for moral assessment (p. 31).

Actuarial science · Class (philosophy) · Credence · Economics · Epistemology · Harm · Normative · Perspective (graphical) · Political science · Positive economics · Privilege (computing) · Statistics · Artificial Intelligence · Computer Science · Law · Law, Economics, and Judicial Systems · Mathematics · Philosophy · Psychology · Social Psychology

Tools

Open DOISci-Hub