To begin, let’s understand that one can be critical of classical empiricism – Locke, Hume – and its contemporary representatives, without denying fundamental empiricist precepts, particularly materialism. To say that the brain does not know through aggregating sense impressions into representations that then produce knowledge through associations, is not to deny that the brain is the organ that experiences, interprets, and learns. There are alternative explanations, after all.
But perhaps classical empiricism can be salvaged through neuroscience? I don’t believe so, especially as the neuroscience and behavioral studies on its fundamental precepts – translation of sense impressions into representative images and extrapolated concepts – remain inconclusive. Their inconclusive character may be the result of ungrounded assumptions and poorly phrased questions. (It should be remembered that Locke’s empiricism has been chipped away at even by more sophisticated empiricists, beginning with Hobbes, then Hume, after all.)
I’m not sure that resort to neuroscience resolves such philosophical issues, anyway. The philosophical question is not really ‘how we know’ which got trashed on the Continent by Nietzsche, then Heidegger; and here in America by Pragmatism. The question remains, what ‘knowing’ means. That’s a question that may intersect with some form of behaviorism (since what we ‘know’ has something to do with what we do), but it has nothing to do with neurons firing in the brain. Mapping neurons to images will not tell us anything at all about what we think ‘knowing’ might be.
The theories of (classical) representationalism and associationism seem to me pretty well shot and anachronistic at this point.
I suggest a different avenue towards a resolution to some of these difficulties: Charles Sanders Peirce, who originated Pragmatism partially to resolve such difficulties in classical empiricism (not ‘how’, but ‘what’, as the end result of a natural and inevitable process of sign response), especially through semiotics.
The failure of the Analytic tradition to account for and fully absorb Peirce and semiotics – with it’s implicit critiques of classical empiricism, and its potential resolutions of problems of representation, intentionality, and meaning – I think a great embarrassment for the tradition of professional American philosophy. I am not suggesting that Peircean semiotics can resolve all problems; on the contrary, it is really just a beginning. But it is a beginning that has not been allowed to begin on its own terms. Consequently, it has had to find its development elsewhere, especially in cultural studies. But its original intent was as a propadeutic to logic. (Peirce wasn’t the first to think this – the idea, in rough form, was in currency in the Middle Ages, and surfaces in Hobbes.)
Reading the article on Mental Representation at SEP, I kept thinking, ‘well, but the chain of signification will produce the same result and still leave open further opportunities for concept formation as sign interconnection, without resort to base sensory dependence.’ Conceptualization is indeed a response – to signs.
The movement of the fly signs “food” to the frog. But humans do have a choice of whether they want to call the vinyl bag stuffed with foam pellets a “bean bag chair.” The pragmatic arbitrary quality of human language usage raises difficulties for a theory that insists on baseline composition of concepts from sensory input.