which is necessary in order that we can make inferences about the logical sums is ordinarily information that has been collected by statistical means, and is probable. Our inferences, therefore, about the logical sums of different collections are ordinarily probably probable.
The statistical probability which is the consequence of the principle of induction should, in the future, be correlated and compared with the various other theories of probability. In particular its connection with the “probability” which involves the calculation of chances should be determined.
The principle of induction is a consequence of certain properties of collections. As a consequence of these properties there are also certain rules which must be followed in applying the principle. As another, it appears that the only meaning which can be attached intelligently to the word probable is this—that the probable is that which would be true if we knew that our will to choose fairly among all possible samples was successful. The principle of induction derives its validity from the metaphysical assumption that there exists a collection of potential experiences each member of which is what it is absolutely. From this assumption the propriety of many of the methods of reasoning of the natural sciences may be demonstrated.
Somerville, Mass.
A Clue to Holt’s Treatment of the Freudian Wish
I find myself in close agreement with Professor Watson’s recent estimate[1] of Edwin Holt’s brilliant little book on The Freudian Wish. “That Holt’s wish is not Freud’s” (p. 862) and that there is from my point of view, very fortunately—more of Holt than of Freud in the book (p. 902); that we none the less are indebted to Holt for his emphasis on Freud’s genuine contribution to general psychology; that Holt is justified in his effort to replace the sensation as unit of psychology (p. 861) ; that his effort is not wholly successful and that, in particular, he does not present us with a satisfactory substitute for “Meynert’s justly criticized scheme”—to all these conclusions I subscribe very cordially. I also agree with Mr. Watson that the book has a thoroughly “behavioristic tendency” (p. 904), while yet “in many places” it has not rid itself from “subjectivism.” But Watson, I think, has net found the clue to Holt’s apparent see-saw between “hehaviorism” and “subjectivism.” Yet it lies ready
- ↑ This Journal, Vol. XIV., pp. 85–92.
