If I am asked whether Political Economy, understood, for example, as Mill understood it, is to be regarded as a science, I should have to admit that I could not simply reply, Yes. To say nothing of any errors in his logic, I should say that I do not believe that it gives us sufficient guidance even in regard to economic phenomena. We could not, that is, deduce from the laws accepted by Economists the necessary working of any given measure--say, the effect of protection or free trade, or, still more, the making of a poor-law system. Such problems involve elements of which the Economist, purely as an Economist, is an incompetent judge; and the further we get from those questions in which purely economical considerations are dominant, towards those in which other factors become relevant,--from questions as to currency, for example, to questions as to the relations of capitalists and labourers,--the greater the inadequacy of our methods. But I also hold that Political Economists may rightly claim a certain scientific character for their speculations. If their ultimate aim is to frame a science of economics which shall be part of the science--not yet constituted--of sociology, then I should say that what they have really done--so far as they have reasoned accurately--has been to frame an essential part of the prolegomena to such a science. The "laws" which they have tried to formulate are not laws which, even if established, would enable us to predict the results of any given action; but they are laws which would have to be taken into account in attempting any such prediction. And this is so, I think, because the laws are descriptions--within limits accurate descriptions--of actually existing facts as to the social mechanism. They are not mere abstract hypotheses, in the sense sometimes attached to that phrase; but accounts of the plan upon which the industrial arrangements of civilised countries are, as a matter of fact, constructed. Such a classification and systematic account of facts is, as I should suggest, absolutely necessary for any sound historical method. Facts are not simply things lying about, which anybody can pick up and describe for the mere pains of collecting them. We cannot even see a fact without reflection and observation and judgment; and to arrange them in an order which shall be both systematic and fruitful, to look at them from that point of view in which we can detect the general underlying principles, is, in all cases, an essential process before we can begin to apply a truly historical method. Anything, it is said, may be proved by facts; and that is painfully true until we have the right method of what has been called "colligating" facts. The Catholic and the Protestant, the Conservative and the Radical, the Individualist and the Socialist, have equal facility in proving their own doctrines with arguments, which habitually begin, "All history shows". Printers should