Philosophy Jennifer Marusic

“Locke on Knowledge and Probability”

Abstract:

Locke’s epistemology is primarily concerned with answering two questions:

1. What can we know?
2. When it comes to matters that we do not know, how are we to conduct ourselves and what should our attitudes be?

My aim is to argue against a widespread way of understanding Locke’s answers to these questions and the relationship between these answers. According to this widespread view, Locke holds, in response to the first question, that we can only know a very narrow class of propositions, sometimes characterized as necessary propositions and sometimes as analytic propositions. Commentators who hold this view typically take Locke to be surprisingly nonchalant about the impossibility of having knowledge of any contingent propositions. Locke’s nonchalance, however, seems to be explained by his answer to the second question: we can and should rely on probable judgments about the world in our everyday lives.

I claim that this widespread view is inconsistent with Locke’s account of probability. I argue that Locke holds that only propositions that we know (or propositions that are themselves probable in light of our knowledge) provide evidence that can rationally support our probable judgments. Locke denies that we can rationally make probable judgments that are supported by evidence that is probabilistic all the way down. The widespread interpretation saddles Locke with a deep problem right at the heart of his epistemology: if Locke does hold that only necessary truths are knowable, then, by his own lights, all of our probable judgments about the way the world contingently is are irrational.

In order to solve this problem, I propose revisiting the reasons for thinking that Locke is committed to the view that only necessary truths are knowable. It is often thought that this follows straightforwardly from Locke’s definition of knowledge. I argue that this rests on an assumption that ought to be rejected. The assumption is that knowable propositions for Locke are all what Hume calls relations of ideas. By rejecting this assumption, we can resist the conclusion that knowledge, for Locke, is only of necessary truths.

I conclude by outlining a better way of understanding Locke’s answers to the two questions and the relationship between these answers: Locke holds that our knowledge provides evidence in light of which we make probable judgments, but that it can do so only because it includes some knowledge about the way the world contingently is.