What then are theresources upon which sociology can draw when it sets itself the task ofexplaining the laws of progress and of social order? They are, first ofall, the same that have been used so successfully in the naturalsciences: observation, experimentation, and comparison.
Observation does not mean the unguided quest for miscellaneousfacts. "But for the guidance of a preparatory theory," the observerwould not know what facts to look at." "No social fact can have anyscientific meaning till it is connected with some other social fact" bya preliminary theory. Hence, observation can come into its own onlywhen it is subordinated to the statical and dynamic laws of phenomena. But within these limits it remains indispensable.
The secondscientific method of investigation, experimentation, is onlypartly applicable in the social sciences. Direct experimentation is notfeasible in the human world. But "experimentation takes place wheneverthe regular course of the phenomenon is interfered with in anydeterminate manner. . . . Pathological cases are the true scientificequivalent of pure experimentation." Disturbances in the social body are"analogous to diseases in the individual organism," and so the study ofthe pathological gives, as it were, privileged access to anunderstanding of the normal.
The scientific method of inquiry ofcentral importance to the sociologist is comparison, above all,because it "performs the great service of casting out the . . . spirit[of absolutism]." Comparisons of human with animal societies will giveup precious clues to "the first germs of the social relations" and tothe borderlines between the human and the animal. Yet comparisonswithin the human species are even more central to sociology. The chiefmethod here "consists in a comparison of the different co-existingstates of human society on the various parts of the earth'ssurface--these states being completely independent of each other. Bythis method, the different stages of evolution may all be observed atonce." Though the human race as a whole has progressed in a single anduniform manner, various populations "have attained extremely unequaldegrees of development" from causes still little understood. Hence,certain phases of development "of which the history of [Western]civilization leaves no perceptible traces, can be known only by thiscomparative method," that is, by the comparative study of primitivesocieties. Moreover, the comparative method is of the essence when wewish to study the influence of race or climate on human affairs. It isindispensable, for example, to combat fallacious doctrines, "as whensocial differences have been ascribed to the political influence ofclimate, instead of that inequality of evolution which is the realcause."
Although all three conventional methods of science mustbe used in sociology, it relies above all on a fourth one, the historicalmethod. "The historical comparison of the consecutive states ofhumanity is not only the chief scientific device of the new politicalphilosophy. . . it constitutes the substratum of the science, inwhatever is essential to it." Historical comparisons throughout thetime in which humanity has evolved are at the very core of sociologicalinquiry. Sociology is nothing if it is not informed by a sense ofhistorical evolution.
From Coser, 1977:5-6.