I did but slightly and summarily. I avail myself, therefore, gladly of this second opportunity you offer me. Goethe remarks how convenient it would be if in human life we were allowed to do everything twice over. He means, no doubt, that we might correct our mistakes, as it were, in a second edition. I, too, find it convenient that I am allowed to deliver the presidential address twice over; but not because I wish to unsay or to qualify anything ; it is because in going over the ground again I may put myself at a different point of view, and bring into the front of the scene what was left before in the background. My subject, then, to-day, will be the practical effect of the two rival methods of treating history -the literary and the scientific method.

6

Though I treated this question but summarily last year, yet I imagine I laid down a proposition which could not easily be misunderstood, and which, if true, was weighty enough. I affirmed that all politics depend upon history. Let me start then again from this principle. It is a paradox, no doubt, to those who repeat the word "history" without defining it. What! history tells us about Greece and Rome, and other curious places which were important two thousand years ago, or about William the Conqueror and Richard Cour-de-Lion, about wars and tyrants, and the strange things that happened when the world was half barbarous. What can all this have to do with politics? And what better plan could a man adopt for confusing his judgment and disqualifying himself for sober practical affairs than to fill his head with such old-world stories?' Such are the notions of history which prevail naturally under the literary treatment of it. It is supposed to be romantic, and concerned only with remote times, because literary historians for the success of their books choose romantic subjects, and dress them in poetical diction, and affect remote periods in which they can escape from political controversy. All such objections fall to the ground at once if we lay it down that history is simply the mass of facts that can be collected concerning the actual existence and development of the organism called the state, that accordingly it deals with the recent and the present just as much as with the

past, and that it has no predilection whatever for what is romantic or unusual-this being a perversion which has been introduced through the literary treatment of it-but seeks those facts from which important inferences concerning the life of states can be drawn. This being laid down, the connexion between history and politics becomes immediately apparent. Once conceive that states may be studied in this comprehensive and inductive manner, you at once see that such study may be the proper basis of all statesmanship and politics. You see that the divorce which now separates history from politics may be the effect of a particular perverse mode of treating history.

History then, so considered, is no mere agreeable recreation, but of all studies the most practical and the most important. All that we call in the largest sense politics, all great affairs affecting large numbers of people, depend upon it. When we attempt to deny this, when we say that the kind of knowledge needed in public affairs is practical rather than bookish, when we point to successful statesmen who have despised history, all we prove is that history has been so badly studied, so badly organised, that the substitute for history a shrewd man may provide for himself was practically better than the history taught in schools. In like manner, so long as medicine is in its infancy, the empiric may be a better doctor than the regular practitioner. What passed for history until a very recent time, could be of no practical use. Its facts were untrustworthy, its generalisations rhetorical and not serious. In these circumstances the practical man might wisely neglect it. But even then he only disregarded what was known technically as history; the real thing he did not and could not dispense with. Out of Blue Books and such statistics as he could come at, out of conversation with older statesmen, he made up for himself a fund of knowledge which, though perhaps he did not. know it, was history-history not reaching back far, fragmentary and unsatisfactory, but yet considerably better than what in those days passed for history in the schools.

Perhaps few of us realise how great is the change which has

taken place of late years in history, or how great are the results which must shortly follow from that change. We hear, indeed, of great activity in historical research, new documents consulted, new views of men and periods coming into fashion; but does it occur to us that all this activity, all this progress, must some day give history, as such, a position and influence in practical affairs that it has never yet enjoyed? Theoretically, I suppose, we shall all admit that history ought to be the guide and oracle of public affairs. If we had before us a clear record of the past, if all the tendencies had been accurately traced, all the forces measured, we should have the best possible indicator to guide us onward into the future. Hitherto all this has been purely theoretical. We have had no such clear record, but a record so confused and false that scarcely any use could be made of it. We have therefore accustomed ourselves to other guidance, we have fallen back upon mere custom or else upon à priori principles, such as in all other subjects we have long ago recognised to be untrustworthy. But in the meanwhile historical method has been improved and reformed. We have now before us a mass of historical facts on which it is worth while to found generalisations. How different was it when Locke or Montesquieu tried to speculate upon political science! They could scarcely adduce a fact which does not now appear utterly worthless, legendary, or else misunderstood. It is quite otherwise now; and so now it is worth while to call to mind again that history after all, if only history can be made to speak clearly, has the secret we want to learn. After all in politics as in other departments, principles ought to be grounded upon observation abundant and accurate; they can in fact be safely grounded upon nothing else; so that the politician who tells me he cares nothing for history, merely means he cares nothing for remote history, or that he distrusts the history he finds in books, and prefers the history he can discover for himself.

And now, then, if history is a study so momentous, if such great issues hang upon it, I ask you, ought history to be studied in the literary or scientific manner? Hitherto the historian has held

an uncertain position on the boundary line between literature and science. In which province shall he take up his definite place?

This is as much as to ask whether the political truths which we hope to discover by means of history, are truths which lie at all out of the way, are at all difficult to be discovered, or such as will at all surprise us when they are discovered. There is no use in making a parade of scientific exactness if nothing important is to come of it. The literary method will be quite sufficient if there is really nothing beyond the lists of kings, the wars and treaties, the parliaments and legislation, which strike the eye first in history, if beyond all this we do not look for any generalisations, any discoveries analogous to those who have been made by the students of plants and animals. If this is all, then let us of course be exact, let us verify our dates with care, let us consult original documents faithfully, and weigh evidence scrupulously. But in that case it would be absurd to lay much stress upon accuracy; it would be mere pedantry to insist upon minute details. In that case, perhaps, the best course would be to class the subject under the head of literature, and the best history would be the history written in the most glowing, imaginative, and popular style.

In that case history would come quite close in the classification of subjects to a subject with which it is often compared, and sometimes confounded, biography. Biographies too ought to be accurate, but after all not much depends upon their minute accuracy, as we do not mean to found generalisations upon them. Accordingly, though we blame a biography if it is inaccurate, we do not greatly praise it for being simply accurate; we praise it for warmth, vividness, insight into character. Just so the literary school of historians regard history, which they conceive as "the essence of innumerable biographies." They collect facts, not as inductive philosophers, not that they may found general propositions upon them, but as artists, that they may produce an effect by means of them. They do so because they do not believe that any system of new important truth is to be discovered through history. According to them there is no doubt a kind of wisdom to be gained from

history, a certain knowledge of human nature, but it is a wisdom which can never develop into science, but begins and ends in weighty aphorisms, useful maxims like those of Bacon or Macchiavelli. Tacitus is the idol of this school.

They set no

bounds to their admiration for his artful turns of expression, his pregnant sentences, his graphic pictures, and the solemn effective pose of the historian himself. And yet if there are really problems to be solved by history, and you go to Tacitus for a solution of the momentous problems which his age presents, you may find him perhaps not only a great genius, but even as satisfactory a historian as could be expected in the time and place, but by no means a model such as ought to be imitated now. You find that you can

not trust his facts, that he has related what is effective rather than what is true; as to his famous reflections, you find that they would be more instructive if they were less elaborately pregnant, and that they ought to have taken the form of full and carefully supported explanations, instead of being conveyed in oracular hints.

Thus the literary school of history starts from a postulate that the historian has nothing particular to discover, and may, therefore, devote himself to elegant narration. It postulates, therefore, the exact opposite of that fundamental proposition which I laid down above. It does not for a moment imagine that politics depend upon history, that what we believe in politics would be found, if examined, to rest upon evidence which, good or bad, is historical in its nature. On the contrary, it habitually assumes that we come to history with a political system ready made, a system established independently, and not to be modified by anything which history may reveal. I may take Tacitus as an illustration here too. There runs through his writings a tacit assumption that the senatorian view of the imperial system of Rome is of course just. This view is not advocated, facts are not expressly adduced to prove it, but it is assumed throughout as if it were established by some evidence higher and stronger than that of history. day in the majority of historical works.

So it is at the present They assume a political

system which is supposed to have been proved elsewhere, and not