groups the settlement of numerous vexed questions as to whether a thing happened or not. But after agreement is reached regarding the actual occurrence, differences regarding the motive will arise. Here is a realm from which science is shut out, for standards of right and wrong vary according to time and circumstance and individual belief. History is less a teacher of morals than a magazine from which moralists draw their illustrations, and until there is a science of morals there can hardly be a science of history. Or, putting the matter rather more generally, even though we accept the same set of facts, our interpretation of them will depend on whether we are by temper conservative or radical, peaceful or warlike, romantic or prosaic.

In the present case, at any rate, we shall not consider the study of the sources as a means of getting at "truth," but as a means of getting at facts. The difficulties which these afford before they are run to earth will suffice the energy of most beginners. Every few years an accepted episode is attacked on the score of inaccuracy, and after being violently defended, for a greater or less period according to its importance, is at last yielded to its foes. Thus Protestants have given up the belief that a female Pope, Joan, reigned at Rome in the ninth century; the Swiss, except in Canton Uri where the story fixed his birth, have lost confidence in William Tell; and the Welsh no longer accuse Edward I. of murdering their bards. Supported by proved errors of this kind, shallow observers have been emboldened to deny that historical reading can serve any useful purpose. Without accepting such an extreme and sceptical view, one should from the outset confront the liability to error, and regard its nature. If we begin with first principles we shall see that the vagaries of human consciousness must always occasion blunders and unwitting falsehoods. Honesty will not protect a writer against these. He may be a poor observer, his memory may fail him when he is engaged in recalling the scene, his original mental picture may be overlaid by later and incorrect impressions. An unemotional person, such as the professed critic, is likely to be the victim of tricks played upon him

by the senses, and the immense majority of historical witnesses are both emotional and uncritical. Next add to inevitable human weakness that common stumbling-block, imperfection of the record. Besides the total loss of many eminent works and the mutilated state of others--for instance the histories of Livy and Tacitus our knowledge of a whole period may depend on a single writer. It follows that matters which should have been reported from different standpoints are reported from only one. Browning showed a sound instinct in the Ring and the Book, where ten dramatic monologues turn about the crime of Count Guido Franceschini, each one portraying a special aspect of it. And then, worst of all, comes bad faith. Monks forge charter deeds, politicians defend their party, autobiographers defend themselves, and the simple desire of telling a good story peoples that limbo which cannot be called the home of liars, but in which no truth is spoken. Flagrant falsehood does little damage, because it is either self-contradictory or is quickly pilloried. But masters in the art of deceit can mingle fact and calumny with a skill which often gains for them the credence they seek. So misleading is partial correctness that a shameless cynic of the Italian Renascence, Pietro Aretino, advised those who are bent on damaging an enemy's character to tell the truth about him. By grouping the inevitable mistakes of our authorities, their deliberate falsehoods and their fragmentary condition a damaging list is formed.

And yet M. Taine, who was not credulous either by disposition or training, said that there is no such thing as a bad document -il n'y a pas de mauvais documents. He meant that every text will, if properly studied, surrender its quota of genuine information. The great fault of those who deny the truthfulness of history is that they speak sweepingly and without discrimination. Push scepticism far enough and it becomes plain that no one living can predict how his dearest friend would act in an intricate crisis, or, for that matter, how he would act himself. Still we are continually forming estimates of character which prove serviceable. Millions of facts about the past are impregnably established. In order to deny the assassina

tion of Julius Cæsar, or the coronation of Charlemagne at Rome, or the storming of the Bastille, one must be willing to question the very existence of matter. If lying and carelessness are raised in relief upon the historical page, the abundance of authentic evidence is equally prominent. Fact and fiction are often joined, not as an ounce of metal is embedded in a hundredweight of rock, but as oxygen and hydrogen are combined in water. Yet here, although the gases are apparently united past hope of separation, the chemist can easily release the atoms from their intimate connection. A glance at the twelfth extract, page 29, will illustrate the possibility of getting historical fact from a somewhat unpromising source. This passage from the Heimskringla Saga is in certain respects a mere fancy sketch of the battle at Stamford Bridge. A legendary element, picturesque and unmistakable, pervades it. The spirited speeches of the leaders in parley before the combat are wholly imaginary, or have been coloured by a much later hand. The topography is inaccurate. One cannot feel sure that the fluctuations of the fight are properly described. On the other hand, observe how much actual framework it displays. A battle was fought in eastern England, near the Humber, between Harold, King of England, and his rebellious brother, Tostig, aided by the Norwegians under their heroic king, Harold Hardrada. All three commanders were present in person and shared the danger along with their troops. Harold Godwinson won and survived. Tostig and the King of Norway were both slain. The remnant of their army retreated, and the immediate purpose of the expedition failed. Furthermore, this saga brings out distinctly the fighting mood of the Middle Ages, and through its literary quality provides a means of judging the culture of that time. Because the story that Harold Godwinson offered Harold Sigurdson "seven feet of English ground, or as much more as he may be taller than other men," is altogether too good to be true, we must not call the entire tale a lie and throw it light-heartedly away.

Instead, then, of adopting the negative position that history is fallible to an extent which renders it unprofitable, we should

accept the principle that every record has its value. Advancing a little further, we reach the collection and classification of documents. Information must be sought wherever there is a chance of finding it, and after the authorities are ascertained they must be assorted and appraised. However firmly we may believe that each piece has some merit, it is no less clear that the degrees of value vary infinitely. Criticism has an endless task in deciding what should be kept and what rejected, how much weight certain statements should carry, and whether certain authors could have had precise knowledge or only repeated current gossip and scandal. Let us now consider the most salient features of difference which historical materials present, then pass to their accumulation, and, thirdly, examine our own sources as they group themselves into species.

66

Jordanes relates that when the Visigothic chieftain Athanaric stood in the market-place of Constantinople he was awe-struck by the unwonted scene, and exclaimed, 'Without doubt the Emperor is a God upon earth, and he who attacks him is guilty of his own blood". This speech-for present purposes its verbal authenticity matters little-attests the way in which Rome's ancient and imposing civilisation affected the barbarian mind. It also brings before us through one example that whole world of historical evidence wherein stories are told without words by objects mute, inanimate and impersonal. Athanaric was impressed not by what he heard but by what he

saw.

The density of population, the din of commerce, the lofty solid buildings about him meant wealth and power and wisdom. Had he understood every stage of imperial growth the result would have been the same in arousing his admiration. He saw what proved the Romans a wonderful people, and he could therefore dispense with details of their rise and of their conquests. Now quite outside literature former generations have left memorials in their buildings, their sculptures, their paintings, their music, their tools, their weapons, their household articles. The case of Athanaric suggests itself here from its association with buildings. He and his folk were not dwellers in towns, and their simple valour was unavailing against the brick walls

of a Roman fortress. Doubtless the shops, churches and palaces would catch his attention first and detain it longest. Nor are barbarians the only men who judge a people's might and majesty by architectural remains. Plutarch, writing 500 years afterwards, declares that the construction by Pericles of the public and sacred buildings at Athens "is now Greece's only evidence that the power she boasts of and her ancient wealth are no romance or idle story". And what more convincing proofs of Roman grandeur can be named than the still extant Colosseum, Pont du Gard, and aqueduct of Tarragona ?

All the works wrought by human hands which have escaped destruction may be fairly termed historical documents; though most require translation by specialists, just as, in the case of dead tongues, Champollion unlocked the meaning of Egyptian inscriptions and Rawlinson those of Assyria. We can even be more emphatic and state that where the hands have left no traces the bodily frame will bear cross-examination. In tracing the dim history of primitive man the shape of a skull and the dimensions of a skeleton attain immense importance. The earth abounds with "documents" other than those of language— whether the words be taken separately or arranged in sentences and books. But the sources which printing has diffused surpass every other kind if tested by popular interest and their power to reach the intelligence readily. Books are the universal means of conveying knowledge, and though history does not altogether depend on the written word, this species of record is its mainstay for the ages which have elapsed since letters were invented. We reach the heart of the present discussion in asking, "By what steps does our modern historical curiosity attempt to gain a faithful idea of distant eras or movements from the literature which they have left?"

At the head of the subject stands the bold rubric, Bibliography. One must first learn the character of his sources and their amount. Gibbon's essay on the Roman law-chapter 44 of The Decline and Fall-has been praised as a marvellous performance, considering that it owes nothing to the Institutes of Gaius, which were for the time lost. Gibbon's ignorance is