Seeing off the Local Threat to Irreducible Knowledge by Testimony

I commit myself to the 'common-sense restraint' that testimony is in fact an important source of knowledge. This has the consequence of making the dispute between reductionists and anti-reductionists a question of the possibility of reducing the epistemic status of testimony to that of other epistemic resources such as perception, memory and inference. I accept arguments (from Coady and Stevenson) against the possibility of global reductionism, but little importance can be attached to their victory if the local reductionist threat (from Fricker) is not met. The strength of the local reductionist case rests on the plausibility of a distinction between developmental and mature epistemic phases, and on a reductionist stipulation of default settings. I claim that the distinction is either irrelevant or detrimental to the local case, and that default settings are more perspicuously thought of as due to the irreducible reliability of testimony.