Page:Decline and Fall of the Roman Empire vol 1 (1897).djvu/54

xlviii may have become obsolete before the days of the last Merwings. It is instructive to compare Gibbon's description of the social and political institutions of our Teutonic forefathers with that of Bishop Stubbs. Gibbon blends together with dexterity the evidence of Cæsar and Tacitus, between whom a century had elapsed, and composes a single picture; whereas Bishop Stubbs keeps the statements of the two Romans carefully apart, and by comparing them is able to show that in certain respects the Germans had developed in the interval. Gibbon's account of the military establishment of the Empire, in the first chapter of his work, is open to a like objection. He has blended, without due criticism, the evidence of Vegetius with that of earlier writers.

In the study of sources, then, our advance has been great, while the labours of an historian have become more arduous. It leads us to another advance of the highest importance. To use historical documents with confidence, an assurance that the words of the writer have been correctly transmitted is manifestly indispensable. It generally happens that our texts have come down in several MSS., of different ages, and there are often various discrepancies. We have then to determine the relations of the MSS. to each other and their comparative values. To the pure philologist this is part of the alphabet of his profession; but the pure historian takes time to realize it, and it was not realized in the age of Gibbon as it is to-day. Nothing forces upon the historian the necessity of having a sound text so impressively as the process of comparing different documents in order to determine whether one was dependent on another,—the process of investigating sources.