How does he know?

A per

hearer to assume that they know. If they do mention their sources of information, they generally make them appear better than they really are. son may describe himself, for example, as an eye-witness of some great battle in which he took part. But a battle is much too large and complex an affair for any single individual to see the whole of it, much less one who is busy fighting in his own part of the field. Nine-tenths or more of what the 'eye-witness' probably tells about is therefore based upon conjecture or the confused tales of other combatants often heard at fourth or fifth hand.

Hearsay.

If a witness got his information through somebody else and if he is not himself a trained investigator, his statements must be taken with the greatest caution. This secondhand evidence, or "evidence which does not derive its value solely from the credit to be given to the witness himself, but rests also, in part, on the veracity and competency of some other person", is what is known to the law as 'Hearsay '; and except in a few specified classes of cases our courts refuse even to listen to it. "The law requires . . . the testimony of those who can speak from their own personal knowledge. It is not requisite that the witness should have personal knowledge of the main fact in controversy; for this may not be provable by direct testimony, but only by inference from other facts shown to exist. But it is requisite that, whatever facts the witness may speak to, he should be confined to those lying in his own knowledge, whether they be things said or done, and should not testify from information given by others, however worthy of credit they may be.”

We do not reject hearsay altogether in common life, and there is no reason why we should. The courts can afford to do so because they give power to the litigants to compel the attendance of witnesses (if they are alive and within the court's jurisdiction) who can speak from their own personal knowledge. Moreover, in the courts the only matters which have to he proved by witnesses are those which one party or

other is not prepared to admit. Thus the procedure of the courts does not imply that nothing can be proved by hearsay outside of them. It does show, however, that in the opinion of jurists hearsay is not in general a very satisfactory kind of evidence. The objections to it are the probability that the statement of the person quoted "was imperfectly heard, or was misunderstood, or is not accurately remembered, or has been perverted"; the ease with which a lying witness can shield himself when he says he was told '; the impossibility of cross-examining the person quoted, "that it may appear what were his powers of perception, his opportunities for observation, his attentiveness in observing, the strength of his recollection, and his disposition to speak the truth"; and the frequent impossibility of telling "through whom, or how many persons, the narrative has been transmitted, from the original witness of the fact. *

If we cannot tell anything about the origin and history. of a story, if we have no reason to believe that the person who tells it to us would sift it thoroughly before telling it, and if it does not bear some very strong internal evidence of truth,

* Greenleaf, "The Law of Evidence", §§ 98, 99. The rule against the admission of hearsay does not of course exclude everything that a witness claims to have heard said. Evidence about the statements of another person is not hearsay when "the very fact in controversy is whether such things were written or spoken, and not whether they were true". If a person is being tried for slander a witness would be allowed to testify that he heard him say the slanderous words; but he would not be allowed to testify that some one else had told him that he had heard him say them. Again, evidence is not hearsay when the statements, whether written or spoken, which the witness repeats are "natural or inseparable concomitants of the principal fact in controversy ". In a murder trial, for example, a witness may testify that he heard the prisoner threaten to kill the deceased, not because there was necessarily any truth in the threat, but because it indicates a state of mind in which such a deed might be done. The threat is thus a part of the res gesta; it has, or may have, a direct or indirect causal connection with the act in question; and it is thus so much circumstantial evidence, not hearsay.

we have no right to assume that it is true. This statement applies to the ordinary newspaper anecdotes about public men, to much of the gossip that we hear about our neighbors, and to almost all of the legends of ancient heroes and mediæval saints.

ful?

Now comes the question whether a witness is truthful. If we are really weighing testimony, this question must be Is he truth- answered according to some rational principle. To depend upon our impulses for an answer is not to weigh it at all, but merely to return to the primitive condition in which we were before we discovered the need for weighing it. Hence we must not assume that a witness is truthful merely because the story he tells is pleasant or flattering, or because it is interesting or pathetic; because he is polite or intelligent, his voice and accent winning, or his manner confidential; because he was hard to find, knows so much about the subject, or is perhaps the only person whom we can find that knows anything about it at all. We must not trust him either merely because he was doubted once before and vindicated, or even because he told us himself he that was telling the truth'. So, vice versa, we must not assume that a witness is untruthful or ignore what he says merely because he is uninteresting or disagreeable.

In the effort to escape the influence of our impulses where impulses are liable to mislead, and to find some fixed test Arbitrary of veracity, there is danger of choosing something tests. purely arbitrary; of assuming, for example, that there must always be truth on both sides' when witnesses disagree, and trying to split the difference between them; or of assuming that the story supported by the larger number of witnesses must be the true one, regardless of the character of the individual witnesses and their sources of information.

Because truthfulness is right and lying wrong, people have often invented arbitrary tests based upon the supposition that God would intervene in some particular manner to vindicate the right and punish the wrong. Thus a person sometimes

says, 'May God strike me dead if what I say is not true', and in a case of this sort if the calamity named does happen to occur it is often accepted by those who see it as a divine judgment. Such tests are easily systematized and adopted by a people as a whole. Consequently the Hebrews cast lots to discover a culprit, and our own ancestors in the middle ages appealed to the ordeal or to trial by combat. Such tests still survive, although the definite appeal to God may be lost; and schoolboys assume that one of their number has 'defended his word' or proved his truthfulness if he has won a fight with the boy who impeached it. In the same way when boys play the disputed part of some game over again, the winner announces that 'that proves it '.

If such simple methods as these for settling the rights of a case and the veracity of the witnesses had been found effective, the state and the church would certainly still continue to use them. Nowadays, however, we recognize that God or Nature gives no arbitrary sign by which we can detect a lie and vindicate the truth. We are not left wholly without a test-far from it; but the test which we have is hard to apply and, like the oracles of old, it often leaves us in doubt. There is no ultimate test of truthfulness but fact, and the test of fact is nothing less than consistency with the whole course of Nature.

When the fact itself is unknown and we are trying to estimate a witness's veracity in order to find it, the most important things to consider are his personality; his special relations to the question about which he speaks; the consistency and probability of the story he tells; and its relations. to the stories of other witnesses. We must speak of each.

First the witness himself. We must not judge everybody by ourselves. Whether a person has a strong disposition to speak the truth is largely a matter of his own in- Character. dividual character. This is very different in different people, and it is something about which we cannot always know. We often can know, however, about a per

son's race, the age in which he lived, his occupation, and the set to which he belonged; and all these help to determine his veracity. An Oriental is generally less likely to speak the truth than a European, a slave than a master, a promoter than a military officer. If Othello had stopped to think, he might have known that an Italian of Iago's time, even though an officer, would not hesitate to deceive a man of alien race if he had anything to gain by it.

In considering the effects of character we must not assume that the most innocent persons are necessarily the most truthful. A child may tell the most harmful lies in all innocence simply because the thought has come (or been put) into its head, and he has not critical judgment enough to distinguish clearly between truth and falsehood or to recognize the wrongness of the latter. A child once more may lie from pure nervousness. Ask him in a threatening tone whether he has done some perfectly innocent thing which you have seen him doing, and very likely he will deny it.

Circumstances.

Next the circumstances. We must not divide people once for all into sheep and goats, those who lie or prevaricate and those who do not. "Falsus in uno falsus in omnibus" does not mean that if a person lies once he will lie always, but only that he is likely to lie again on the same occasion and about the same matter. Conversely, though a person has never yet told a lie, he may tell one in some new situation where the strain is greater. There always must be first offences. Therefore, when we have estimated a witness's general character, truthfulness and characteristic motives as well as we can, we must try to find out what special circumstances are present to influence him. Much depends, for example, upon the solemnity of the occasion on which the statement is made. A person is not likely to lie when he is confessing to his priest or when he knows he is at the point of death. The law recognizes the influence of this solemnity, and therefore it demands that every witness who comes into court shall swear to tell