296 THE AMERICAN JOURNAL OF SOCIOLOGY
likeness, has played in human association; but observation, arrange- ment and generalization of facts.
But the book fails to reconcile a more radical and fatal antith- esis than either of the preceding. It would be useless to make con- jectures about the causal relations of these anomalies. It is evident, however, that different parts of the book were written under widely different subjective conditions. Their incoherence is too marked to escape even a superficial critic. After Professor Giddings has had time to see how the different parts of his copy look together in print I am sure he will perceive that the argument is a picturesque yoking together of the scientific ox and the speculative ass. The alternative title points to an analytic examination of reality. The preface/^ contra proposes an a priori process. Book I, Chapter III, goes off on the other tack and outlines a positive method, and finally the main conten- tion of the volume returns to the programme of the preface, with inter- pretation by assertion and deduction in the place of demonstration. I do not apply the statement to the book in detail, but of the structure of the main argument, in which the details have their setting, I do not hesitate to say that its spirit throughout is that of pre-Cartesian specula- tion, rather than of post-Darwinian science. This is all the more notice- able because nobody is more acute and punctilious than Professor Giddings in judging others by scientific canons. In various parts of Book I he has formulated principles of positive procedure as justly as any scientist could desire. x Yet he deliberately chooses to cast his main argument and to mass his material in the mould of speculation and deduction, instead of organizing the material at his command so as to show its precise inductive value. He uses formulas of scientific reasoning with admirable precision, but there is no evidence in this book that he has "experienced" science.
The key to Giddings' own explanation of his reasoning is in his account of the " objective " and the " subjective " method in sociology. If space permitted I should undertake to show that these terms have at least two sets of connotations in the book, with consequent con- fusion. In brief Giddings sometimes means by the " objective " method " the explanation of society in terms of physical law," and by the "subjective" method " the explanation of society in terms of voli- tion or motive." 5 At other times, and particularly when there is a dearth of facts in correspondence with his hypothesis, he means by the
1 Pp. 39, 40, 49, 55, 56, 70, etc. Pp. 1 1 and 36.
