when he began his story-book. The thought that seems to dominate the whole was at best an after-thought, and what we have is an attempt to bring a mass of heterogeneous material into some kind of connection with the history of the second Persian war, the only part of the work that has any true coherence. By processes familiar to the student of Homeric criticism, evidence is adduced to show the imperfect welding of the mass and the incompleteness of the work. References are made to stories that are never told, to persons that never recur, and not only so, but, according to these critics, the narrative fails of a proper close. It should have ended earlier or later. Even the marvellous style has not escaped criticism, and attempts have been made to prove that the histories lack the supreme touch of the master's hand.

But we must beware of over-analysis. If Herodotus did not set out with a fixed purpose and a definite plan, if he began as a curious traveller, and not as a systematic investigator, still there must have been in his mind a general conception of the universe, which grew clearer and sharper as he matured, so that long before he came to write his preface the work of his life was revealed to himself. Revision, elaboration, there must have been, but the unity of the work lies in its moral purport.

At all events it is this unity, this grasp of a vast and varied material, that is one of the special claims set up for Herodotus as the originator of a new department of literature, and before proceeding to consider the other points in which the historian differentiates himself from his predecessors it is necessary to give some brief account of the earlier writers of Greek prose.

The extent of Herodotus's obligations to his predecessors is variously estimated, and there are those who have made Herodotus out to be an unscrupulous conveyer of other men's labours. But the charge of plagiarism, to which modern writers are extremely sensitive, glanced harmless from the armour of the Greek of the best period. Later Greeklings have much to say about stealing one from another. The true masters helped themselves to what they needed. A good thing was

common property. Was Sophocles in debt to Herodotus, or Herodotus to Sophocles, or both to a third? To whom did Herodotus and Euripides owe the common estimate of life as an accident? Yes, Herodotus had predecessors. He may have learned something of Miletus from an obscure writer, Cadmus of Miletus, of Lydia from Xanthus, called the Lydian, and there are those who think that he owes much to Hecatæus of Miletus, an earlier and haply not so far inferior Herodotus. True, he cites Hecatæus but four times-once to magnify the importance of the historian's calling, once to approve his judgment, once to criticise, and once to show that his authority is not overwhelming. But German professors are always ready to attribute to others the processes of their own craft, and there is scant evidence that the work of Herodotus is in any sense a compilation. From whatever sources he drew his material, he was the originator of a new line of literary work. Before him the epopee had served as history, and the eldest Muse of the nine was commissioned to tell of the glories of the past. But poetry is not history, for history means investigation; and poetry is inspiration, not investigation, and the beginnings of prose presaged the death of the epos. The earliest writers of prose were called logographers (λoyoypápo or Moyoπolol), “writers of accounts." When Logos comes in, Mythos retreats. There is no λóyos in Homer, and the first prose writers were the first critics. The logographer was something else than the translator of poetry into the language of every-day life. In the early period the poetical form is imperative, and the art of verse is earlier and easier than the art of prose. For mere narrative the epos would have sufficed. Chronicles and genealogies—and, for that matter, travels-might have been composed in hexameters to the end of time. But the Logos is that which reckons, that which calculates, that which takes stock, and this must be borne in mind when we consider the predecessors of Herodotus. The opening words of the Genealogies of Hecatæus show that he approached tradition in no reverent spirit, and that he scouted the false and foolish stories which he found current. Herodo

calendar. He is a critic of far greater acumen than was once supposed, and the assaults on his honesty have not demolished the importance of his evidence. Of course, we must carefully distinguish between the facts for which he vouches and the reports that he simply repeats, but in the vast number of details it is not always clear when he is speaking of his own knowledge and when he is giving mere hearsay. The trouble we have found in establishing his itinerary recurs when we follow the historian along the track of events. Those who study Herodotus critically must be content to put every narrative to the test, to tap every wheel on which the train runs so smoothly. But no historian can escape this scrutiny, and the good faith of Thucydides himself, who claims so much greater accuracy than does Herodotus, has been mercilessly impugned. Even in antiquity the untrustworthiness of Herodotus was a jest, and the frivolous scepticism of a later age selected for ridicule the very points on which the testimony of Herodotus has been sustained by recent exploration. Travellers' tales are proverbial, and Herodotus may have contributed to the currency of the proverb. Indeed, it must be confessed that, being no linguist, he was at the mercy of guides and interpreters, and yet he has not seldom fared better at the hands of modern critics than those who had access to native documents. The native documents require as much interpretation as do the stories of the native dragoman, and he who takes a romantic epic for a sober chronicle is no better off than a man who had no command of Persian. And then much, very much, depends on the tradition of the text, and in all matters involving numbers we must be careful to suspend our judgment. The account of a flood in northern China, which took place in October, 1887, showed a vast difference in the estimates of the loss of life, which ranged from one million to seven millions; and to the intrinsic difficulty of exactness in figures we must add the trouble of the Greek double notation, which lent itself readily to all manner of

errors.

But an introduction like this can not go into the various

points which have been raised and met in regard to the trustworthiness of Herodotus, and some space must be given to the author's conception of the world, for, after all, every history reflects the historian more or less.

In order to do justice to Herodotus's conception of his task we must remember that he belonged to a sceptical generation. The Persian war had brought about what might be called a revival of religion, but the old leaven was working. The protest that Xenophanes had made early in the century against the anthropomorphism of Greek religion had not lost its effect. Pindar is orthodox, but refuses to believe the myths that he deems dishonouring to the gods. Eschylus shows by his attempts to justify the ways of God to men that there is a profound dissidence to be bridged. Sophocles wears his faith with a difference. Euripides is counted by some an outand-out rationalist. The leading agnostics were close contemporaries of Herodotus. Sophistic influences are traceable in his style. The famous debate of the third book reminds one of Protagoras, and the jingles that bear the name of Gorgias are heard now and then. Under these conditions Herodotus is best understood as a conservative, not to say reactionist. He holds fast to the old doctrine. He does not deny the existence of the gods outright. He goes so far as to point out the influence of poetic fancy on the Hellenic notion of the several gods, and extols the purer ideas of the Persians; but he reveres tradition, he is a Church of Greece man, and he has a strong faith in the divine power that regulates the universe. To him, as to Anaxagoras, another friend of Pericles, there is a wise Providence that keeps everything in balance. This balance is now called Nemesis, now vengeance, now envy of the gods. Mind not high things," for high things are brought low. The dread genealogy of Surfeit (Kópos), Outrage (Tẞpis), Disaster ("Arŋ), of which the poets tell, is true of nations as of individuals. The innocent suffer with the guilty. Nor do the instruments of divine vengeance escape if not modest in their office. There must be no excess, no overstepping the bounds. The Divine gives satisfaction to the

66

injured, champions the weaker, and restores the balance. All this is ingrained in the Greek temper. All this is Hellenic, as it is Herodotean.) The historian keeps strictly to the sphere of national thought, and proof texts might be drawn from Solon and Theognis and Pindar, as well as from Herodotus. We must go farther down before we come to the "Chance central of circumstance," to Fortune, such as Thucydides conceives her and as Polybius conceives her. This poetical justice simplifies the scheme of the universe-makes it perhaps suspiciously simple. God in history becomes too much a deus ex machina, a too convenient untier of knots; but even our century demands an increasing purpose that runs through the ages, and even those who have rid themselves of religious formulæ are not always superior to spiritual manifestations, to the study of psychical research. There are those who have gone so far as to call Herodotus a sceptic lined with a spiritualist.

Superior to his predecessors in his conception of his task, superior, after all, in his critical method, Herodotus's greatest, it is fair to say his unapproachable excellence lies in his style. True, the style of Herodotus loses much of its charm-the Greek scholar is tempted to say all its charm-in the transfer to another idiom. Not the least of its attractiveness lies in the dialect, the leisurely Ionic, with its soft vowels, its deliberate utterance, its quavering rhythm, its old-fashioned vocabulary. Biblical English, biblical German, which has been tried, does not reproduce the tone of Herodotus. The employment of consecrated phrases in a secular narrative grates. We have a jargon, not a dialect. The chief thing gained is the simple structure of the sentence, which is common to the Hebrew original of the Old Testament and to the Greek of Herodotus. But the dialect is not everything, and as a famous Greek critic has undertaken to prove by actual experiment that the spell of Herodotus is not broken by the translation from Ionic into Attic, so a transfer from Greek into English does not efface all the Herodotean charm. The dialect is gone, it is true, a cunningly wrought robe with gleams of epic gold. The varied

B