Epistemic relativism

3. Is epistemic relativism problematic?

Many philosophers consider epistemic relativism a dangerous or troubling doctrine. It is, however, not easy to find plausible arguments justifying this negative attitude. This section briefly sketches two lines of argument that might motivate opposition to relativism, although I do not think either argument very persuasive.

The first charge against relativism is that it is nihilistic because it simply gives up on the project of distinguishing good reasoning from bad, and embraces a sort of epistemic anarchy. From our previous discussion, however, it should be clear that the ‘anything goes’ slogan is a singularly inappropriate one for many relativistic accounts of cognitive assessment. Many versions of reliabilism are relativistic. But reliabilists are certainly not epistemic anarchists – quite the contrary. Reliabilism offers an extremely demanding account of cognitive evaluation. For a given cognitive agent in a given historical setting, it will typically be the case that a reliabilist evaluation will rank one system of reasoning higher than another. Rarely will it be the case that reliabilism ranks all contenders on a par.

A second complaint against relativism is that it threatens the connection between cognitive inquiry and truth. For if the epistemic relativist is right, then there may be a pair of people whose systems of reasoning are very different from one another, though each system is optimal for the person using it. We can expect that on being exposed to essentially the same data these people will sometimes end up with very different sets of beliefs. When this happens it is unlikely to be the case that both sets are true; at least one set of beliefs will be substantially mistaken. Since at least one person will end up with false beliefs, and since ex hypothesi they are both using optimally good cognitive systems, it can not be the case that good cognition always leads to true beliefs.

What this argument shows is that if the epistemic relativist is right, then good reasoning does not guarantee truth. But it does not show that good reasoning and truth are unconnected. If, for example, we adopt a reliabilist account of cognitive evaluation, then people who reason well will do the best job possible at producing truths and avoiding falsehoods. To expect more than this seems unreasonable.