ABSTRACT
To approach the question in a different way: if, a la early Freud, man and mouse only hallucinate cheese on the basis of a previous ‘reduction of primary needs and tensions’, and slowly learn which hallucinations pay off, then, as we have already argued, man and mouse must be able to recognize success and failure and discriminate and recognize experiences associated with one or the other. In some cases, indeed, we must allow one or more investigators the much more dangerous privilege of knowing, with a high degree of probability, whether a public piece of cheese is, or is not, present at the time, in order to say that the subject seems to be hallucinating. Such
knowing is a complex process; we do not understand either psychologically or philosophically much of what it implies. But it seems fairly clear that processes of discrimination (and recognition) in an unblind, shareable and sometimes veridical sense are integral to it, and their presupposition underpins any scientific inquiry. Hence to treat the basic observational roots of science as ‘blind responses’, for fear of God, mind or the devil, is a species of paranoic intellectual suicide.
