I have already had occasion to rise the expression “scientific psychology,” or psychology as a positive science, and the meaning of this expression must now be more carefully considered.  As a positive science, psychology is limited to the function of analyzing mental states, and of tracing their origin in previous and more simple mental states.  It has, strictly speaking, nothing to do with the question of the legitimacy or validity of any mental act.

Take a percept, for example.  Psychology can trace its parentage in sensation, the mode in which it has come by its contents in the laws of association.  But by common consent, a percept implies a presentative apprehension of an object now present to sense.  Is this valid or illusory?  This question psychology, as science, does not attempt to answer.  It would not, I conceive, answer it even if it were able to make out that the whole mental content in the percept can be traced back to elementary sensations and their combinations.  For the fact that in the chemistry of mind elements may combine in perfectly new forms does not disprove that the forms thus arising, whether sentiments or quasi-cognitions, are invalid.  Much less can psychology dispute the validity of a percept if it cannot be sure that the mind adds nothing to sensation and its grouping; that in the genesis of the perceptive state, with its intuition of something external and now present as object, nothing like a form of intelligence is superimposed on the elements of sensation, giving to the result of their coalescence the particular unity which we find.  Whether psychology as a positive science can ever be sure of this:  whether, that is to say, it can answer the question, “How do we come by the idea of object?” without assuming some particular philosophic or extra-scientific theory respecting the ultimate nature of mind, is a point which I purposely leave open.

I would contend, then, that the psychologist, in tracing the genesis of the percept out of previous mental experiences, no more settles the question, What is the object of perception? than the physicist settles it in referring the sense-impression (and so the percept) to a present material agent as its condition.

The same applies to our idea of self.  I may discover the concrete experiences which supply the filling in of the idea, and yet not settle the question, Does intelligence add anything in the construction of the form of this idea? and still less settle the question whether there is any real unity answering to the idea.

If this is a correct distinction, if psychology, as science, does not determine questions of validity or objective meaning but only of genesis, if it looks at mental states in relation only to their temporal and causal concomitants and not to their objects, it must follow that our preceding analysis of illusion involves no particular philosophic theory as to the nature of intelligence, but, so far as accurate, consists of scientific facts which all philosophic theories of intelligence must alike be prepared to accept.  And I have little doubt that each of the two great opposed doctrines, the intuitive and the associational, would claim to be in a position to take up these facts into its particular theory, and to view them in its own way.