THE PERCEPTION OF SPACE. (IV.) 545

the four or five greatest monuments of human genius in the scientific line ? If truth impels I must fain try, and take the risks. It seems to me that Helmholtz's genius moves most securely when it keeps close to particular facts. At any rate it shows least strong in purely speculative passages, which in the Optics, in spite of many beauties, seem to me fundamen- tally vacillating and obscure'. The " empiristic " view which Helmholtz defends is, that the space-determinations we per- ceive are in every case products of a process of unconscious inference. 1 The inference is similar to one from induction or analogy. 2 We always see that form before us which habitually would have caused the sensation we now have. 3 But the latter sensation can never be intrinsically spatial, or its intrinsic space-determinations would never be overcome as they are so often by the "illusory" space-determinations it so often suggests. 4 Since the illusory determination can be traced to a suggestion of Experience, the " real " one must also be such a suggestion : so that all space intuitions are due solely to Experience. 5 The only psychic activity re- quired for this is the association of ideas. 6 But how, it may be asked, can association produce a space- quality not in the things associated ? How can we by in- duction or analogy infer what we do not already generically know ? Can " suggestions of experience " reproduce elements which no particular experience originally contained ? This is the point by which Helmholtz's "empiristic" theory, as a theory, must be judged. No theory is worthy of the name which leaves such a point obscure. Well, Helmholtz does so leave it. At one time he seems to fall back on inscrutable powers of the soul, and to range himself with the ' psychical stimulists '. He speaks of Kant as having made the essential step in the matter in distinguish- ing the content of experience from that form space, of course which is given it by the peculiar faculties of the mind. 7 But elsewhere again, 8 speaking of sensationalistic theories which would connect spatially determinate feelings directly with certain neural events, he says it is better to assume only such simple psychic activities as we know to exist, and gives the association of ideas as an instance of what he means. Later, 9 he reinforces this remark by con- fessing that he does not see how any neural process can give rise without antecedent experience to a ready-made (fertige) perception of space. And, finally, in a single 1 P. 430. - Pp. 430, 449. a p. 42 8. 4 P. 442. 5 Pp. 442, 818. 6 P. 798. 7 P. 456 ; see also 428, 441. 8 P. 797. 9 P. 812.