Perception and Evidence

The question as to whether empirical knowledge has any foundations and, if it does, just what those foundations might be, has long been an important epistemological question. The problem with which I am concerned is that of taking primitive sensory experience as the ground of empirical knowledge. I consider three attempts on the part of 20th century British and American analytical philosophers to substantiate our ordinary knowledge claims about an extra-mental, empirical reality. The first of these is the sense-datum approach to the problem, in which by using the act-object analysis of simple sensations, the independent status of things sensed was thought to be established. But I point out that the sense-datum theorists do not prove the point they set out to prove, but only succeed in illuminati: 1g the fundamental assumptions of Realism in opposition to those of Idealism. In opposition to the sense-datum Realists, I advance the adverbial analysis of sensation, and in so doing open the way to a less direct but more credible Realism. The second part considers the protocol-statement theories, according to which empirical knowledge is (or corresponds to) a truth-functional syntactical system of propositions ultimately grounded on a class of atomic propositions: the protocol statements. It becomes apparent that the protocol-statement theories must choose between subjectivism and corrigibility. The relation of empirical knowledge claims to subjective sense experiences is problematic. Finally, I relate the attempt made to rework the phenomenalistic and physicalistic epistemic systems of the Logical Positivists along the lines of warranting relations rather than implicative relations. Thus sensory experience comes to have an evidence-conferring relation to empirical knowledge claims. But that relation is a complex one which must be determined in the light of all our relevant information; particularly that regarding the causal conditions prevailing in our perceptual environment