Page:Philosophical Review Volume 1.djvu/130

114  condition than that these premises have meaning. If we regard syllogism as employing a universal premise, it presupposes the axiom of induction, under which a universal truth is asserted and applied on the round of a particular observation. Taking next deduction to be the combination of several universals in a chain of reasoning, we here assert a relation between two universals on the ground of the relation of each to one or more intermediate universals, i.e. out of several relations we construct a resultant relation, or a whole in which the resultant relation appears as a part. This reasoning depends in the end on generalization, for it implies some such axiom as: If, when two terms are related to a third, a relation between the two is observed, then other terms similarly related to a third will have a similar relation to each other. All cases, then, tend to show that deduction involves generalization from an observed case on a single axiom; the axioms, therefore, of induction are the axioms of reasoning, and the generalization of particulars is the work of reason.

When one thinks, reads, or hears a general term, what arises in consciousness immediately and without reflection, above the sign? R. said to many men and women of diverse characters and occupations: "I am going to pronounce several words; I want you to tell me immediately and without reflection, if the word calls up nothing in your mind, or if it calls up something, and what." An answer which took more than five or six seconds was considered nil or doubtful. Words used were fourteen, in increasing degrees of abstractness: dog, animal, color, form, justice, goodness, virtue, law, number, force, time, connexion, cause, infinite. Some answered to all or some of these; others to a few; 900 answers were written out. These were of three types, the concrete type, consisting in the immediate and spontaneous substitution of a particular case (fact or individual) for the general term; the typographical type, the linking of the concept to a clear vision of a printed word, without anything more; the auditory type or auditory image (here was R.'s most noteworthy case, a literary physician who found every word "sounding in his ears"; he talked aloud in writing his books). R. tried the words in the simplest possible prepositional form (such as to suggest nothing), and got the same results: the psychological and the logical unit have the same effect. But 53 per 100 of the answers were "nothing"; e.g. to cause the answer was, "I can think of nothing."