1. Historical grounds.—Here the facts are believed on testimony; the testimony of men who had means of knowing them. That such witnesses should have lived at the time when the facts in question took place, is the great and essential condition of their credibility.
2. The belief ex necessitate.—A fact which, at the time of its first announcement could only have been known from having been witnessed by a cotemporary, but which at some later period is shown from other facts to have been real, is to be admitted unreservedly; the evidence in its favour being of the highest kind. Of this sort are such astronomical facts as, in the present state of our knowledge, can be ascertained independently of experience, but which, when first notified, could only have been ascertained by experience.
3. Traditional Grounds.—Here the immediate authority to the person who is informed of a real or supposed fact, is some one who had not the possibility of knowing the facts in question from being contemporary with them; but who heard it from some one who was so contemporary—or else heard it from some one who heard it from some one, &c., ad infinitum. Here the statements are possible or impossible, probable or improbable. If possible, they may be true; if probable, they are likely to be so. In neither case, however, are they historical facts; that is, there is no testimony founded upon a knowledge of the event.
4. The true elements in unreasonable traditions.—A series of necessary and connected antecedents to a given effect, inductively obtained is an ethnological ground of belief, or an ethnological fact; and it is based on inductive reasoning. A series of unnecessary and unconnected antecedents, derived from the imagination, is a false ground of belief, and in most cases this takes the form of mythological tradition. It does not, however, necessarily follow, that, because a body of tradition may, on the whole, be unreasonable, or even impossible, it is therefore wholly deficient in grounds of belief. The doctrine ex nihilo nihil may here apply. It may fairly be argued, that, absolute invention is so difficult, that in all error there is some truth. Granted. It may, then, be argued, that a criticism analytical enough to evolve the residuum is a scientific (or literary) possibility. Granted. But who is the critic? I fear that his appearance is optandum magis quam sperandum.
5. The inductive method consists in the assumption of certain causes as the necessary antecedents of a known event; and they are good or bad according to their scientific or unscientific character. To take as the first fact in the history of Greece, the existence of a poem like the Iliad in the ninth century B.C., to ascertain the state of society that it implies, and to appreciate the civilization involved therein, is an ethnological argument; whilst, to assume a certain amount of time for such to have grown up in, is an argument from effect to cause, and is good or bad, according as it assumes no more than is absolutely necessary.
Now, if we ask upon which of these five principles we believe in the antiquity of the Chinese civilization, it will certainly not be the first.
I am not prepared to wholly exclude the second; indeed, I have not the means of forming an independent one on the subject. At the same time I know that, in respect to the Chinese astronomical calculations many good judges are incredulous, and many of those who are not so are at variance in their opinions.
The third is essentially admissible for a limited period only.
The fifth remains open for consideration.
In the application of what may be called the doctrine of necessary antecedents, I believe, for my own part, that we must take the China as described by Marco Polo in the fifteenth century; and if we put the development there exhibited on a level with that of the China of the present century, we are giving to the advocate of antiquity full as much, perhaps more, than he can fairly demand. I submit that the time necessary for the growth of such a phenomenon need not exceed a few centuries.