One of the most fertile sources of error in modern political thinking consists, indeed, in the ascription to collective habit of that comparative permanence which only belongs to biological inheritance. A whole science can be based upon easy generalisations about Celts and Teutons, or about East and West, and the facts from which the generalisations are drawn may all disappear in a generation. National habits used to change slowly in the past, because new methods of life were seldom invented and only gradually introduced, and because the means of communicating ideas between man and man or nation and nation were extremely imperfect; so that a true statement about a national habit might, and probably would, remain true for centuries. But now an invention which may produce profound changes in social or industrial life is as likely to be taken up with enthusiasm in some country on the other side of the globe as in the place of its origin. A statesman who has anything important to say says it to an audience of five hundred millions next morning, and great events like the Battle of the Sea of Japan begin to produce their effects thousands of miles off within a few hours of their happening. Enough has already occurred under these new conditions to show that the unchanging East may to-morrow enter upon a period of revolution, and that English indifference to ideas or French military ambition are habits which, under a sufficiently extended stimulus, nations can shake off as completely as can individual men. CHAPTER V THE METHOD OF POLITICAL REASONING The traditional method of political reasoning has inevitably shared the defects of its subject-matter. In thinking about politics we seldom penetrate behind those simple entities which form themselves so easily in our minds, or approach in earnest the infinite complexity of the actual world. Political abstractions, such as Justice, or Liberty, or the State, stand in our minds as things having a real existence. The names of political species, 'governments,' or 'rights,' or 'Irishmen,' suggest to us the idea of single 'type specimens'; and we tend, like medieval naturalists, to assume that all the individual members of a species are in all respects identical with the type specimen and with each other.