Dissertation, St Andrews (
2017)
Copy
BIBTEX
Abstract
The present work explores various ways in which contingent evidence can impact
metaphysics, while advocating that, just as a scientific realist allows for ampliative inferences
to the unobservable, ampliative inferences from possible evidence can warrant possibility
claims that lie beyond the reach of sensorial imagination. In slogan form: possible evidence
is a guide to possibility. Drawing on Shoemaker’s (1969) argument for the possibility of time
without change, I advocate the following principle: If there is a possible world at which the
observable facts make it objectively reasonable to conclude that p, then we should conclude
that p is possibly true. This provides a route to contingentism in metaphysics, for, if one
considers that there are worlds in which the observable facts make it objectively reasonable to
conclude that p, and worlds in which the observable facts make it objectively reasonable to
conclude that not-p, then my principle tells us that we should conclude that possibly-p and
possibly not-p, i.e. that p is contingent. This contingency in what is reasonable to conclude, I
suggest, occurs most saliently in debates where evidence of phenomenal experience and
empirical science are marshalled to support one theory over another. I also explore some
consequences of taking possible evidence to be a guide to possibility in this way, among them
being an interesting modal analogue of the lottery paradox.