[SAŽETAK PREDAVANJA]

I present a strategy by which moral knowledge can be derived from non-moral knowledge, using insights from reliabilist epistemology. The strategy begins by discovering which cognitive processes generate which moral and non-moral beliefs. We can then assess the reliability of these cognitive processes for moral belief formation by considering to what extent they produce true belief on non-moral issues, and by checking whether they produce contradictory moral beliefs in different people. By retaining reliably caused beliefs and abandoning unreliably caused ones, we can move closer to moral truth. No normative ethical assumptions are required.