The problem of induction from a metaphysical point of view

Abstract

The aim of this study is to examine the metaphysical solutions recently proposed for the problem of induction, which we inherited from the Scottish empiricist philosopher David Hume. The problem of induction concerns its justification. One way to understand the problem is as follows: If there is some time-invariant regularity in nature, can we know (or have some degree of justification for believing) it from experience? This is a conditional question; because, if there is no regularity, then there is no problem. Also, one need not assume that there is full regularity in many regards. According to Hume, our opinions regarding the unobserved parts of nature, our beliefs based on inductive inference are not justified. He argues that it is impossible to justify placing even a small degree of trust in any of our predictions regarding the unobserved parts of nature. The power of Hume’s problem lies in the strength of this argument. Recently, we see that there is a significant effort to provide a metaphysical solution to the problem of induction. The most prominent roles in this effort are played by the defenders of the views nomic necessitarianism and dispositional essentialism. In this thesis, both nomic necessitarian and essentialist solutions to the problem of induction are examined and taken issue with. This aim is pursued along two main lines. The first line is to show that the nomic necessitarian solution is incomplete since it does not help us with the epistemic issue, that is, how we can know (or have some degree of justification for believing) that there are time-invariant regularities from experience. The second line is to show that dispositional essentialist solutions, despite being more sophisticated than the nomic necessitarian solution, are highly problematic. The result of this examination is that, as they stand, none of these approaches to the problem of induction can overcome inductive skepticism as they claim to.