Abstract
This thesis argues that contemporary epistemology faces a challenge that has seldom been explicitlyrecognized, and even less often seriously confronted. I call this the problem of possibilities. It is the
problem that epistemic evaluations and norms, as well as the arguments or principles that define
them, are possibility sensitive. By this I mean that arbitrary decisions in how possibility spaces are
modelled affect the outcomes of these arguments and principles. Sometimes it is the epistemic
agent whose model of possibility space has this effect, and sometimes it is that of the epistemic
evaluator (or, epistemologist).
The thesis demonstrates that possibility sensitivity shows up in many aspects of contemporary
epistemology, attempts to unite its differing incarnations into a single theme, and argues that there
is no easy escape from the problem. I also argue that the solution to these problems lies in finding
formal methods to epistemically evaluate possibility spaces.
The thesis also explores two ways of incorporating the epistemic value of conceiving the possible
into our evaluations and normative assessments of epistemic agents. The first way is to directly
measure the epistemic value of a set of concepts or a possibility space with a utility scoring rule. The
second way is to adapt accuracy-based scoring rules for credences to enable comparisons between
credence functions on distinct domains. Relatively little previous work has been done on either
approach, and I conclude find that existing methods have serious flaws. I also argue that the inability
of accuracy scores to handle comparisons between credence functions defined on distinct domains,
may amount to a refutation of veritism.
| Date of Award | 20 Jun 2023 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Jason P Konek (Supervisor) & Richard G Pettigrew (Supervisor) |