Sceptical worries have been at the centre of most of our discussions about knowledge so far. A different but perhaps more common approach to epistemology is to seek out an analysis of what knowledge is. If we can say what knowledge is, then we should be able to identify in particular cases if we know or not. The subject of this chapter, reliabilism, grew out of this tradition of philosophy.

Justified true belief

Our story begins this time with Plato rather than Descartes. Knowledge, Plato tell us in the Theaetetus, is justified true belief. That knowledge requires truth seems straightforward enough. If my belief is not true, it is mistaken and mistakes cannot possibly be knowledge. It also seems very reasonable that knowledge requires justification. For example, I might be asked to guess how many grains of sand there are in a bucket. Let us say I guess there are 3,127 and I form a belief that there are 3,127 grains of sand in the bucket. It just so happens to turn out that my belief is correct, so I have a true belief. Nobody would describe this as knowledge. I am just lucky that my belief corresponds to how things are. The difference between mere true belief and genuine knowledge appears intuitively to be that in cases of genuine knowledge I am justified in asserting my belief. I can offer some reason in virtue of which I am entitled to say what I believe.