1. Witnesses provide the most clear and connected proof, and the least liable to misunderstanding; but yet a proof which is entirely dependent on veracity, on intelligence, on absence of prejudice, and on clear memory, and is hence the least dependable kind of evidence in some cases.
2. Material facts, which may be very conclusive; such as A’s footprint in B’s garden, or A’s chisel left in B’s house, at a burglary. If the fact is certain, the conclusion is proved; but the danger lies in misunderstanding the fact.
3. Exhaustion, which may prove A guilty because no one else could have done the deed; as when A and B are seen in a railway carriage at one station, and at the next stoppage B is found murdered and A leaves the carriage. There may be not a trace of other evidence, but this is enough.
4. Probability, as when A is last seen with B, and B proceeds to deal with the property of murdered A. This kind of evidence is enough to hang a man, solely from presumption.
Now let us look at these kinds of evidence about the past of man.
Witnesses.
1. Witnesses, the documents, which give a clear and connected statement. They may be either primary, as a stone inscription or an autograph letter; or secondary, as compiled histories or subsequent copies. No other kind of evidence is so easy to follow; yet this is a proof in which we are entirely at the mercy of the prejudices, the ill-will, the frauds, and the blunders of others, and it is hence the least dependable kind of evidence in some cases. The speeches of Thucydides, the bias of Suetonius, the wonders of Livy, the romances of William of Malmesbury, and the forgery called Richard of Cirencester, each plunge us deeper and deeper into the doubtfulness of written documents; to say nothing of the casket letters or Ossian.
Material facts.
2. Material facts, when rightly understood, are the most conclusive evidence. They may be in a single object, as a palaeolithic flint rechipped over and over in later ages; or a foreign ornament used on an object of dated style, as a Maori tatued head in a daguerreotype would prove the tatuing to be known between 1840 and 1860; or a restruck coin with one type over another, as Barchocheb over Hadrian; or an added inscription, so familiar on Egyptian statues. Or the evidence may consist in a collocation of objects, such as a group of things found together in a tomb; or the superposition of strata of ruins in a town. In the case of a single object there are few possibilities of misunderstanding the evidence; but in strata or tomb-groups there is a chance of older things being reused. Such chances of error are, however, extinguished by the recurrence of instances; and the finding of certain things together in several cases under different circumstances is one of the strongest kinds of evidence, such, for instance, as the name of Amenhotep III often found with the Mykenaean pottery, both in Greece and in Egypt.
Exhaustion.