attestation by contemporary witnesses as the essential condition for justifying belief in an alleged fact, he thus proceeds :

'The main difference between the divergent schools is as to the extent to which contemporary attestation may be presumed without direct and positive proof. Both assume the same mode of proving an historical fact: but the former refuse to infer the proof from the existence of an oral tradition; the latter consider that inference legitimate. The former deny that the existence of a popular belief with respect to the past, derived from oral reports, raises a presumption that the events narrated were at the time of their supposed occurrence observed by credible witnesses, and by them handed down to posterity. The latter, on the other hand, hold that the existence of such a popular belief (combined perhaps with some accessory circumstances) authorises the conclusion that the current story was derived from credible contemporary witnesses, and has descended from them in a substantially unfalsified state.' . The diffe

rence between the opposite opinions on this subject is therefore a difference of degree rather than of principle. Nobody asserts that all history must be taken directly from the reports of percipient witnesses. No historian applies the strict rule of judicial evidence, that all hearsay reports are to be discarded. In treating of the period which precedes contemporary history, all persons admit traditionary, secondary, or hearsay evidence, up to a certain point. The question is, where that point ought to be fixed?' (Vol. ii. p. 490.)

After a few words (p. 494.) upon the difficulties which beset 'the application of rules of evidence to the semi-historical or 'crepuscular period, a period of which some knowledge has 'been preserved, though by imperfect means, and in a dete'riorated state,'-the last result is thus given:

-

'All the historical labour bestowed upon the early centuries of Rome will, in general, be wasted. The history of this period, viewed as a series of picturesque narratives, will be read to the greatest advantage in the original writers, and will be deteriorated by reproduction in a modern dress. If we regard a historical painting merely as a work of art, the accounts of the ancients can only suffer from being retouched by the pencil of the modern restorer. On the other hand, all attempts to reduce them to a purely historical form, by conjectural omissions, additions, alterations, and transpositions, must be nugatory.'

Those who are disposed to labour in the field of Roman history will find a worthier reward for their toils, if they employ themselves upon the time subsequent to the Italian expedition of Pyrrhus.' . . . . . ‘In this history, much must remain incomplete, uncertain, and unknown but the great outlines are as firmly marked as in a modern history, composed with brighter lights and from ampler materials and the historical inquirer will meet with a richer return for his labours, than if he bewildered himself with vain attempts to distinguish between fact and fiction, in the accounts of the foundation

of Rome, the constitution of Servius, the expulsion of Tarquin, the war with Porsena, the creation of the dictatorship and tribunate, the decemviral legislation, the siege of Veii and the capture of Rome by the Gauls; or even the Licinian rogations, and the Samnite Wars.' (Vol. ii. p. 556.)

We subscribe to these conclusions fully, so far as regards Roman history under the Kings and prior to the Republic. As to the period of the Republic, we cannot adopt them without some qualification. Sir George Lewis fairly states the question: It being admitted, that there is a certain point antecedent to the beginning of contemporary history, up to which point historical research is legitimate and reasonable-where is this terminus to be fixed in regard to Rome? He would fix it at the landing of Pyrrhus in Italy. But we submit that this is within the actual limits of contemporary Grecian authorship, and, in a certain sense, even of Roman authorship through the speech of Appius in the Senate, which was preserved to later times. This terminus is therefore too low to correspond with the principles laid down. The Licinian laws can hardly be thrown into category of the unsearchable—along with the foundation of Rome.

the

In fixing the upward terminus, we perceive no index so appropriate as the beginning of contemporary notation; which is, in truth, contemporary history in fragment and rudiment. Wherever matters of fact and of public import were recorded, even though detached and without coherence, historical research becomes admissible.

Now in Rome, the substratum of notation' can be traced up to the commencement of the Republic, but not higher. No similar notation belongs to the regal period: at least, if any such existed, it never crops out, but is irrevocably submerged and undiscernible. Accordingly, the suitable upward terminus for historical research is, in our view, the commencement of the Republic. We consider the kingly period as lying above the limit of historical research, and as 'a series of picturesque narratives' in which no matter of fact had ever any recorded existence apart from fiction. Comparing Roman with Grecian history, we regard (conformably to Sir George Lewis's view) all that precedes the Roman Republic as corresponding to heroic or legendary Greece; we consider the two first centuries of the Republic as corresponding to Greece between the first recorded Olympiad (776 B. C.) and the year 500 B. C. To the first of the two, the microscope of the historian is inapplicable. Respecting the second, we cannot say the same; for there are, or

were, some recorded realities which an attentive contemplation may hope to magnify and bring into fuller day-light, both in themselves and in the consequences deducible from them.

This is the only line of demarcation which we see any theoretical reason for drawing. Whether the researches into the history of the early Republic will turn out very fruitful, or will yield much of new certainty and new probability, is a different question. We are not sanguine in hoping that they will: but neither are we sanguine respecting those investigations, recommended by Sir George Lewis as preferable, into the later history of the Roman Republic; where there was once much contemporary information, now entirely lost, and represented by little except the Epitomes of Livy. What we expect from farther study of the early Republic, is, not so much a corrected version of the facts of detail, as better and clearer views of the institutional practice and development, gathered by combination, inference, and cautious hypothesis, from a variety of distinct sources. Books on Roman antiquities (especially the excellent work of Becker and Marquardt) already teach us much respecting the magistracies and constitutional growth of the Republic: but we acquire no knowledge (beyond the literal statements as they stand) respecting the period of the Kings. And though there is much fanciful conjecture in Niebuhr, it is indisputable that many portions of Roman republican antiquity (the Agrarian Laws especially) are far better understood than they were before his writings.

Discountenancing as Sir George Lewis does all historical inquiry into Roman history anterior to Pyrrhus, it is natural that he should pronounce, as to that period, All attempts to 'reduce the accounts of the ancients to an historical form by 'conjectural omissions, additions, alterations, and transpositions, 'must be nugatory.' This is perfectly true respecting the period of the Kings, but we are not prepared to pronounce the like peremptory verdict (must) about the two first centuries of the Republic. The former (as we have above remarked) contains none of the genuine materials of history; the latter contains some, in greater or less proportion. In our view, wherever the genuine materials of history exist, all the processes above indicated are frequently indispensable, to bring out of them either continuous narrative or determinate results. It is by going through such elaboration that history is distinguished from a mere collection of depositions.

The manner in which Sir George Lewis sets forth the discrepancies between Livy and Dionysius, and the tone of his criticisms on Niebuhr, tend to suggest two impressions, which

we are by no means sure that he would sanction, but from which we certainly dissent. 1. That discrepancies, as many and as great, are not to be found between contemporary wit2. That the Niebuhrian spirit of hypothesis and recombination is illegitimate in principle,-not simply objectionable from abusive excess in Niebuhr's hands.

nesses.

Now we think that contemporary witnesses often form a multitude with every variety of dissonance and contradiction*: and that if, out of such perplexities, an historian is to construct & narrative setting forth the true or the probable, he cannot proceed without a large latitude of preference and hypothesis. Even with the most unexceptionable historians with Gibbon or Mr. Hallam-the narrative supplied to the reader is a result put together in their own minds, founded upon an attentive study of all the evidences, yet not without many inferences, comparisons, and eliminations of their own. Neither of these authors could have performed their task, if conjectural omission, 'addition, alteration, and transposition,' had been forbidden. We know that these liberties are liable to much abuse, and that they have been abused by Niebuhr. But in commending salutary vigilance of criticism on this eminent man, in so many instances of his arbitrary dealing with evidence, we must at the

6

*As a parallel to the discrepancies between Livy and Dionysius, we transcribe the following account of the original authorities respecting the wars in La Vendée, from the beginning of 1793 downwards. We have here contemporary witnesses, under the full publicity of modern times, described by M. Michelet, eminent both as an historian and as a laborious examiner of original archives. (Histoire de la Révolution Française, vol. vii. p. 78.)

'Le livre le plus instructif sur l'histoire de la Vendée (j'allais dire, le seul) est celui de Savary, père du membre de l'Académie des Sciences: Guerres des Vendéens, par un officier, 1824. Dans les autres, il y a peu à prendre. Ce sont des romans, qui ne soutiennent pas l'examen les noms, les dates, les faits, presque tout y est inexact, faux, impudemment surchargé de fictions. Je le sais maintenant à mes dépens, après avoir perdu des années dans la critique inutile de ces déplorables livres. Savary donne les vraies dates, et un nombre immense de pièces : les notes de Canclaux, de Kléber, d'Oppenheim, y ajoutent un prix inestimable.'

We know the work of Savary, and can certify that it fully merits the encomiums bestowed upon it by M. Michelet. But to compose such a work, requires a combination of ability, diligence, and opportunity, such as are rarely brought together in the same person. How many periods are there of human affairs, in which there are contemporary authors approximating to the dark side of Michelet's picture, without any such witness to control them as Savary!

same time guard against what appears to us an opposite extreme. We cannot disallow the constructive imagination of the historian, nor lighten his responsibility by tying him down to a literal

sequence.

While claiming for historians this freedom of judgment, in their laborious task of eliciting probability out of conflicting statements and analogies, we should be glad if it could always be exercised subject to such a censorship as that of Sir George Lewis. No man interested either in ancient history or in the general theory of historical study, can read his book without profit; but none will profit by it so much as those who, adopting his conclusions only in part, account the two first centuries of the Roman Republic a subject still open to historical research and philosophical explanation.

ART. II. Tagebuch des Generals Patrick Gordon, während seiner Kriegsdienste unter den Schweden und Polen vom Jahre 1655 bis 1661, und seines Aufenthalt in Russland vom Jahre 1661 bis 1699. Zum ersten Male vollständig veröffentlicht durch Furst M. A. OBOLENSKI und Dr. Phil. M. C. PosSELT. Moskau, 1849-1851. 2 band.

(The Diary of General Patrick Gordon, during his Military Service with the Swedes and Poles from the Year 1655 to 1661, and his Residence in Russia from the Year 1661 to 1699. Published completely for the first time by Prince M. A. OBOLENSKI and M. C. POSSELT, Ph. D. Moscow, 1849-1851. 2 vols.)

IT

T is one of those singular coincidences which history sometimes presents to us, that the policy of Peter the Great towards Turkey, so steadily and successfully carried on by the Russian Government for two centuries, was originally planned and conducted by a cadet of Lord Aberdeen's family, General Patrick Gordon, the familiar friend and adviser of the Czar Peter the Great, and the conqueror of Asof. The higher,' say the editors of this Diary in their preface, the gigantic scheme of Peter the Great is estimated, and the more it is considered as a plan which his successors have, to the present day, kept ' in view and pursued, the more is the attention drawn to the ' events, whether obstructing or promoting its prosecution, which 'have occurred, and to the persons who were called to be the instruments of its accomplishment. Among those persons, if 'one can be named as the leader of the ideas of the young Czar, it was certainly Patrick Gordon.'

6