22 THE AMERICAN JOURNAL OF SOCIOLOGY
and scholastic methods, and champions the claims of empi- ricism. Schaeffle may be said to avoid the formal discussion of methodology ; but so far as his own method of treatment is con- cerned, it may be characterized as combining philosophy with psychology, the propositions being illuminated with observations from the legal and economic points of view. Induction and deduction are combined with history.
Three of these authors express in so many words the fact that there is a limit to sociological knowledge. Comte is least definite ; while Spencer's well-known division of all reality into the knowable and the unknowable requires only passing men- tion. Lilienfeld recognizes the field of the unknown when he asserts that Uranfang and Endziel cannot be scientifically founded. Schaeffle maintains that we know only force ; that such matters as the problem of evil must be taken as existing facts, and that sociology does not go back of these.
2. The use of analogies. In spite of all that has been said on the uses and abuses of biological analogies in the study of sociology, Professor Ward probably voiced the truth in the matter, and at the same time gave evidence of his open-minded- ness and toleration, when he said, in effect, that so long as the biological analogy can be used as a cord upon which the beads of useful knowledge can be strung, we must welcome the use of analogies and encourage the men who are working so faithfully to elaborate "biological sociology." The present writer is inclined to believe that from Comte to Worms the so-called biological sociologists have in their own minds been inclined not to make the analogy the important thing; although it must be admitted that some of them have literally caused their analogies " to walk on fours." Comte goes less into detailed analogy than the other three writers here under consideration, although he compares faithfully individual and race development. Spencer speaks of analogies and of parallelisms, and he almost touches the ludi- crous when he asserts that these analogies become increasingly clear when we recognize that every considerable organism is a society; however, he formally asserts that his analogies serve only as a scaffolding by means of which the structure of
