[857] Grundlegung, p. 342.
[858] Cf. Ingram, History of Political Economy, and Denis, Histoire des Systèmes.
[859] A. Comte, Cours, vol. iv, p. 198.
[860] Cours, vol. iv, p. 328.
[861] It is interesting to learn the views of historians on this point. Meyer thinks that the object of history is not to discover the general laws of development, but to describe and explain particular concrete events as they succeed one another. Such descriptions can only be made in accordance with the rules of historical criticism, but explanation is only possible with the aid of analogy. “It is only by the use of analogy that the historian can explain past events, especially where there are psychological motives that require analysis. The explanation thus given will necessarily be of a subjective character, and from its very nature somewhat problematic.” Cf. Ed. Meyer, Geschichte des Alterthums, Introduction, 2nd ed. §§ 112 et seq. There does not seem to be any connection between this method and that of Aug. Comte. One becomes still more convinced of this after reading Langlois and Seignobos’s Introduction aux Études historiques or G. Monod’s study in historical method in De la Méthode dans les Sciences (Paris, 1909), or, finally, the numerous articles dealing with this question of method which have appeared in the Revue de Synthèse historique.
[862] Theory of Political Economy, preface to the second edition, 1879.
[863] Schmoller’s Jahrbuch contains descriptive studies of present-day commercial and industrial undertakings which are veritable models.
[864] The Present Position of Political Economy, in the Economic Journal, 1907, p. 481.
[865] We have not the necessary space in this volume to refer to the history of statistics. This science, though independent of political economy, is, however, such a powerful auxiliary that its progress has to some extent been parallel with the growth of economics. During the last twenty years the methods of interpreting statistics (we are speaking merely of observation) have been very considerably improved. The logical problems involved have been studied with much care, and the application of mathematics to these problems has proved very fruitful. No student of the social sciences can afford to neglect such mathematical theories as those of combination, correlation, degree of error, etc. The history of statistics, which contains many eminent names, from Quetelet to Karl Pearson, would certainly deserve a chapter in a book dealing with method, although there would be some risk of giving it too statistical a bias. We must rest content with referring the reader to Udny Yule’s Introduction to the Theory of Statistics, which constitutes what is perhaps the best recent introduction to the discussion concerning the method to be employed in this social science, and forms an indispensable complement to the study of the problems examined in this chapter.
[866] Dupont-White makes the remark somewhere that the State, strictly speaking, has only existed since 1789. It appears, then, that a State which is not constitutional, democratic, and liberal has none of the virtues of the true State. Such exclusion, although permissible in the publicist, is indefensible in the theorist or historian.