Apparently Merivale means the same thing by 'abstract' political philosophy that Mr. Bryce means by 'ideal' democracy. Both refer to a conception of human nature constructed in all good faith by certain eighteenth-century philosophers, which is now no longer exactly believed in, but which, because nothing else has taken its place, still exercises a kind of shadowy authority in a hypothetical universe. The fact that this or that writer speaks of a conception of human nature in which he is ceasing to believe as 'abstract' or 'ideal' may seem to be of merely academic interest. But such half-beliefs produce immense practical effects. Because Merivale saw that the political philosophy which his teachers studied in their closets was inadequate, and because he had nothing to substitute for it, he frankly abandoned any attempt at valid thought on so difficult a question as the relation of the white colonies to the rest of the British Empire. He therefore decided in effect that it ought to be settled by the rule-of-thumb method of 'cutting the painter'; and, since he was the chief official in the Colonial Office at a critical time, his decision, whether it was right or wrong, was not unimportant. Mr. Bryce has been perhaps prevented by the presence in his mind of such a half-belief from making that constructive contribution to general political science for which he is better equipped than any other man of his time. 'I am myself,' he says in the same Introduction, 'an optimist, almost a professional optimist, as indeed politics would be intolerable were not a man grimly resolved to see between the clouds all the blue sky he can.'[37] Imagine an acknowledged leader in chemical research who, finding that experiment did not bear out some traditional formula, should speak of himself as nevertheless 'grimly resolved' to see things from the old and comfortable point of view! [37] _Loc. cit._, p. xliii. The next step in the course of political training which I am advocating would be the quantitative study of the inherited variations of individual men when compared with the 'normal' or 'average' man who had so far served for the study of the type. How is the student to approach this part of the course? Every man differs quantitatively from every other man in respect of every one of his qualities. The student obviously cannot carry in his mind or use for the purposes of thought all the variations even of a single inherited quality which are to be found among the fifteen hundred millions or so of human beings who even at any one moment are in existence. Much less can he ascertain or remember the inter-relation of thousands of inherited qualities in the past history of a race in which individuals are at every moment dying and being born.