Here as elsewhere he is not originating any new idea.  Thucydides had pointed out the difference between the real and the alleged cause, and the Aristotelian dictum about revolutions, [Greek], draws the distinction between cause and occasion with the brilliancy of an epigram.  But the explicit and rational investigation of the difference between [Greek] and [Greek] was reserved for Polybius.  No canon of historical criticism can be said to be of more real value than that involved in this distinction, and the overlooking of it has filled our histories with the contemptible accounts of the intrigues of courtiers and of kings and the petty plottings of backstairs influence—­particulars interesting, no doubt, to those who would ascribe the Reformation to Anne Boleyn’s pretty face, the Persian war to the influence of a doctor or a curtain-lecture from Atossa, or the French Revolution to Madame de Maintenon, but without any value for those who aim at any scientific treatment of history.

But the question of method, to which I am compelled always to return, is not yet exhausted.  There is another aspect in which it may be regarded, and I shall now proceed to treat of it.

One of the greatest difficulties with which the modern historian has to contend is the enormous complexity of the facts which come under his notice:  D’Alembert’s suggestion that at the end of every century a selection of facts should be made and the rest burned (if it was really intended seriously) could not, of course, be entertained for a moment.  A problem loses all its value when it becomes simplified, and the world would be all the poorer if the Sybil of History burned her volumes.  Besides, as Gibbon pointed out, ’a Montesquieu will detect in the most insignificant fact relations which the vulgar overlook.’

Nor can the scientific investigator of history isolate the particular elements, which he desires to examine, from disturbing and extraneous causes, as the experimental chemist can do (though sometimes, as in the case of lunatic asylums and prisons, he is enabled to observe phenomena in a certain degree of isolation).  So he is compelled either to use the deductive mode of arguing from general laws or to employ the method of abstraction which gives a fictitious isolation to phenomena never so isolated in actual existence.  And this is exactly what Polybius has done as well as Thucydides.  For, as has been well remarked, there is in the works of these two writers a certain plastic unity of type and motive; whatever they write is penetrated through and through with a specific quality, a singleness and concentration of purpose, which we may contrast with the more comprehensive width as manifested not merely in the modern mind, but also in Herodotus.  Thucydides, regarding society as influenced entirely by political motives, took no account of forces of a different nature, and consequently his results, like those of most modern political economists, have to be modified largely