INTRODUCTION TO SOCIOLOGY 489
the utility, if not the possibility, of establishing an abstract sociology, and, consequently, an abstract statics also. Appeal- ing to the example of Auguste Comte, he asserts that
In order to formulate such a social metaphysics, it would be necessary to deal with data so general and so lacking in precision that one could not draw from them either any explanation of existing facts or any prevision of future phenomena, still less any practical conclusion. For example, it would be necessary to limit ourselves to showing, like Auguste Comte, that, in addi- tion to the preliminary condition of language, the bases of social order are family, property, and religion; but giving to the word "family" a sense so general that it includes government; to the word "property" a sense so extended that it embraces all possible forms of appropriation, including prop- erty in common ; finally, to the word " religion " so unaccustomed a sense that it includes polytheism and monotheism, determinism or scientific athe- ism, along with fetichism. Thus generalized, sociology would be without framework, it would have neither form nor consistency. Inoffensive as it would remain in its indefiniteness, it would have the serious disadvantage of inclining minds irresistibly disposed to reach conclusions to the belief that the fundamental institutions of all society are everywhere identical and immutable.
The author concludes :
I think, therefore, that it is necessary to limit ourselves to a semi- concrete, schematic sociology, generalizing the data furnished by selection from history, giving preference to the great civilized nations.
Indeed, since the attempt (including that of Auguste Comte) to construct, de planu, an abstract sociology without concrete, verifiable foundations, it has been the general practice to make sociology both incompletely abstract and incompletely concrete by making a selection of historical facts. The result has been a double failure, both scientific and philosophic. It is necessary to begin, on the contrary, with the study of social elements con- sidered especially from the statical point of view; then to advance to the particular historical institutions in which these elements are blended; then to study the particular societies in their ensemble ; then, finally, but only then, to look among these elementary and concrete facts for the general relations which they have in common, independently of their transitory forms in space and time. Such is the only scientific method, the only method capable of avoiding the difficulties pointed out with rea-
