specific statement in the sources. Yet a point is not proved by two or three bits of favorable evidence if the available sources are such that ten or twelve confirmations might be expected. Or the available source material may be so scanty that it does not offer sufficient basis for any positive conclusions. One should try to determine not only how much evidence there is on the point in question and in what degree it bears upon that point, but also how much there ought to be. The quantity of evidence pro and con must be measured not only absolutely, but also in proportion to the available source material, and further with regard to the source material which is missing.
After adopting and working out in full detail through practise the method of research which has here been but suggested, historians will still have the duty of showing others, not only their results but the solid foundations on which these rest and the process by which they were attained. This obligation is recognized to-day to the extent that a historical work without bibliography, footnotes or references to the sources is not considered scholarly. But this is not enough. A list of the books that the historian has used is far from fulfilling the requirement which science makes of a complete analysis of the source material. It gives little idea of the character, scope, applicability and reliability of the source material, even though some word of comment be attached to each title. Fault may be found, moreover, with the footnote as a means of proof. And here is criticized the footnote at its best, rather than when, with pretentiously long quotations in foreign tongues and with superfluity of scholarly digression and learned small talk, it degenerates from a pillar by which science supports its results into a pedestal upon which erudition poses; or when, instead of bearing upon the main point, it gives specific references merely for some accompanying illustrative detail which proves nothing. At its best the footnote less actually proves a point than furnishes indications how one can set about proving it for oneself. Then, owing to the erroneous notion about "facts" of history, each footnote attaches itself to some one point in the text. Thus all the notes taken together supply means for verifying particular statements, but do not substantiate general truths nor lead us to scientific propositions. That must be done in the text, if at all. But here is a third disadvantage that the proof contained in the notes, besides being subdivided there and so losing the strength of union, is further detached from the proof in the text. Yet the very existence of the footnote bears witness that the literary method of presentation at present employed in the text is unsatisfactory for scientific purposes, testifying that the whole would become even clumsier and more confused were the attempt made to embody the notes in the text.
The remedy is a presentation primarily scientific, a unified and uninterrupted presentation of the complete historical process from raw material to finished product. The historian will reveal his investigations as well as their results. After stating and defining his problem
