nate an indeterminate amount of probability or improbability, e. g., the presumption that snow will fall in New York in winter, but not in New Orleans. In general there is always a presumption against what is not. Thus, until 1866, there was a presumption against the successful laying of an Atlantic cable. Much later there was a like presumption against talking by electricity (telephone). There is still a presumption against flying machines.

One legal assumption, or presumption, calls for especial study-viz. the law of Testimony.

TESTIMONY AND AUTHORITY.

70. Testimony is usually classified as an argument from a "sign" (see end of § 67). This view is not philosophical, it fails to discriminate between a statement and our acceptance of it. If we believe what a certain witness says, we accept his statement as a fact, it is to us a fact, as if we ourselves had seen with our own eyes and heard with our own ears. And this fact is usually a sign, i. e., an individual fact from which and similar facts we infer a cause or motive. But why do we accept the statement as true? No theory of the "sign" will give an answer. To find one we must assume that men in general tell the truth, and that this individual witness is peculiarly qualified to tell the truth. Without such assumption there can be no testimony. In fact, society itself, we may boldly say, could not subsist without truth-telling; a community of liars would go to pieces.

But as testimony rests upon assumption, we are always free to reject it for valid reasons, i. e., we may show that the witness was imperfectly informed, or inaccurate, or even untruthful.

The special qualifications of a witness may be summed up under two heads:

a. His ability to observe accurately.

b. His disposition to tell truly.

These two qualifications are tested:

a. By cross-examining him.

b. By comparing his statements with those of other wit

nesses.

Cross-examination proper is possible only in a law-trial. After the witness has testified, in direct examination, in answer to questions put to him by counsel for the side that has summoned him, the counsel for the other side crossexamines him. That is, the adverse counsel plies him with questions likely to bring out every defect or inaccuracy in his statements, or to lead him to contradict himself.

The following is from the trial of Lord Gordon for treason, 1781. The principal witness against the accused was William Hay, who professed to give a full and true account of the rioting in the streets of London. When cross-examined by Mr. Kenyon, Hay involved himself in numerous contradictions and absurdities. Thus:

Q. How do you know it was the same person you saw in the fields? Were you acquainted with him?

A. No.

Q. How, then?

A. Why, he looked like a brewer's servant.

Q. What, were they not all in their Sunday clothes? [A previous statement by the witness.]

A. Oh yes, they were all in their Sunday clothes.

Q. Was this man with the flag then alone in the dress of his trade? A. No.

Q. Then how do you know he was a brewer's servant ?—After a long pause, in which the witness seemed to be on the point of running out

of court-

A. Because there was something particular in the cut of his coat, the cut of his breeches, and the cut of his stockings.-BAKER: Argumentation, p. 115.

In addressing the jury Lord Erskine quoted this remarkable break-down, with the caustic comment: "Gentlemen, you will not, I am sure, forget, whenever you see a man about whose apparel there is anything particular, to set him down for a brewer's servant."

In a debated question of history, which may turn upon the accuracy of a statement written by some person long since dead, cross-examination is impossible. The student can only examine the statement minutely, to see if it is consistent with itself at all points, and compare it with the generally accepted data of history. We must admit, once for all, that the sifting of historical evidence is extremely difficult, and that many of its problems will remain unsolved. The present tendency is to reject all tradition and to accept only the recorded statements of persons contemporary with the events.

The general credibility of a witness is enhanced by various circumstances, e. g.:

a. In a law-suit, by the gravity of the occasion, and by the taking of a solemn oath, with heavy penalties attached.

b. By reputation in the community at large for intelligence and honesty.

c. By the concurrent testimony of other witnesses.

d. By stating something contrary to his own interests. Authority. Usually a witness testifies only to matters of fact, i. e., to what he has seen or heard. But occasionally, even in law-suits, he is called upon for his opinion, e. g., in suits for infringement of patent-rights, in trials for murder by poisoning, etc. Especially in the identification of signatures to legal documents. When the alleged writer of the signature, the proper person to identify it, is dead, some other person familiar with his writing, e. g., a bankcashier, may be called upon to certify that the signature is in his judgment genuine.

Testimony of this kind is technically called expert testimony. Authority is more general in its application than expert testimony. It is not restricted to law proceedings. All men are in the habit of resorting to authority. Thus we consult a dictionary for the spelling or meaning of a word, an encyclopædia for the general facts of a man's

life, a text-book for the rules and principles of a science. Our acceptance of the authority rests on the assumption that a scholar who has made special studies in the subject is trustworthy. We ourselves have not the time for such special study; besides, we have no reason for doubting the book. ANALOGY.

71. Most of the reasoning in the affairs of human life is by way of Analogy. We may regard analogy as a nonscientific practical induction-deduction, or inferring from resemblances. A simple example is this: A. sells to B. on credit; the ground of his confidence is, perhaps, the circumstance that B. has already paid him several times for similar articles. From this he infers a permanent willingness and ability in B., i. e., a general principle applicable to this particular instance. Or, if A. and B. have not had as yet any dealings, A. may know that B. belongs to a class in the community which is in the habit of paying debts, and apply this general principle.

In analogical reasoning certainty is never attainable, but only probability. This should be constantly borne in mind in following political and historical discussions, which are nearly always maintained by arguments based upon analogy.

Thus, if we analyze the arguments by which a statesman advocates a given measure, we shall usually find them to turn upon the observation that similar measures have been profitably adopted in a similar condition of affairs in the previous history of his nation, or of some other nation constituted like his. But similarity is not identity. Were the present condition and the previous condition identical (which they never are), the inference would be scientifically conclusive. But as the conditions are only similar, the inference is only one of probability, and this probability becomes higher the closer the similarity approximates to identity.

The value of analogy, therefore, depends upon the degree of its approximation to identity. To ensure a reasonable degree, two practical rules should be observed:

a. Discover as many points of resemblance as possible.

b. Examine closely the points of difference and show that they are accidental, i. e., not essential to the present question.

These rules are easy to formulate, but they are not easy to apply. History and politics are full of examples of false analogy, due to the violation of a or b; sometimes, of both. That is, the reasoner does not get together enough points of resemblance, or he overlooks points of serious difference. E. g., it is poor reasoning to infer that, because the body suffers when the heart is enlarged, therefore a nation is enfeebled when the capital city increases rapidly in population. The resemblance between the heart and the city is too slight, too fanciful. It is also false analogy to argue that an absolute government is the best, by comparing it to the control that a father exercises over his children. The weakness lies in overlooking two points of difference: the one, that when we think of a father we really mean a good and wise father, whereas an absolute ruler may be neither good nor wise; the second, that men are not children.

Another weak argument, but much less obvious, is this. To increase the quantity of wheat, coal, iron, etc. in a given community, is to increase the productive capacity of the community; therefore, to double the amount of gold and silver in circulation is to increase the productive capacity. The argument overlooks at least one fundamental difference between coal and gold. Coal is of service only as it is consumed; the more of it consumed, the greater good to the consumer and to the community. But gold is not consumed; it merely circulates from hand to hand. Increasing the amount circulated does not increase