form of argument is never very weighty. At best, it can but establish a strong probability; it can never be conclusive.

We might, for example, be able to show that the conditions on the planet Mars were almost identical with those obtaining on the earth; but that would not be proving that Mars was inhabited by living beings, much less by human beings like ourselves.

Reasoning from analogy is, however, by no means to be discredited. It can never, indeed, demonstrate to a certainty, and at times it may even be misleading; yet on the whole it has been found to be a method of reasoning having great practical value. It is peculiarly adapted to furnish hints or starting-points for new lines of investigation. In the field of scientific research, analogy has often pointed the way to new discoveries or new applications of familiar laws and principles. It was reasoning from analogy that enabled Darwin, for example, to hit upon his famous theory of natural selection. In studying the methods of breeders of plants and animals, selection, he observed, was the clue to their success. To improve a variety, they uniformly selected the best, that is, the fittest for the particular purpose in view, and allowed those only to survive and propagate their kind. Why might not, he reasoned, the same principle hold true in nature? Why might there not be an improvement of races or varieties by natural, as well as by artificial selection? The hint thus obtained led ultimately to the theory, now accepted as a truth by virtually all scientists, that species owe their origin to a process of natural selection.

6. REFUTATION

Argumentation may be destructive as well as constructive; that is, it may equally as well aim to prove a proposition false as aim to prove it true. Most argumentative compositions, indeed, are a mixture of constructive and destructive reasoning; for it is often necessary, in order to establish the truth of a given proposition, to destroy belief in its opposite. In any case, the successful argumentation is the one which not only advances positive proof in support of its proposition, but meets all weighty objections that either are, or may possibly be, urged against it. Destructive argumentation, or argumentation which consists in showing that an opponent's conclusions are wrong or wrongly arrived at, is usually called refutation.

To be able to manage refutation well, the disputant must not only know what constitutes proof, but must be able to detect errors in reasoning. Such errors are usually known as fallacies. Fallacies may arise either from lack of definition of the terms used, from misinterpretation of evidence, or from improper methods of making inferences. To point out a fallacy in an opponent's argument, is, of course, to invalidate any conclusion that may rest on that argument.

Fallacies may take a great variety of forms. It is scarcely worth while to enumerate them all here, however; it will suffice if a few of the more common are distinguished.

The ambiguous term. This fallacy consists in using, in an argument, a word or term in two or more

senses, though ostensibly in one only. Sometimes the fallacy is quite transparent; again, it is so subtle as almost to escape detection. No one, probably, is deceived when the stump speaker triumphantly declares that X ought to vote the Republican ticket because he is a Republican and believes in a republican form of government. But a fallacy like that in the following argument is very apt to pass unchallenged: "He is the Representative in Congress of our district; therefore he should really represent us, do as we should do, were we ourselves there to act in our own behalf.' The best way to expose fallacies of this kind is to insist rigorously on a careful definition of all the important terms used in the argument.

Begging the question. The fallacy here consists in taking for granted something that has to be proved. The most common form in which it occurs is in the use of what are called question-begging epithets. Thus if, in attacking the acts or policies of a political opponent, we begin by calling them " nefarious" or "unstatesmanlike," and then proceed to condemn them as nefarious or unstatesmanlike, we obviously assume what it should be our business to prove.

Arguing beside the point. It sometimes happens that a disputant finding it hard to prove the proposition he began with, proves some other proposition very much like it, and assumes that it is virtually the same as that he wished to prove. In such a case, he is said to argue beside the point, or to ignore the point at issue. Dr. Johnson's refutation of Berkeley's idealism by kicking a stone is an example of a variety of this fallacy. The appeal to the special interests,

passions, or prejudices of a particular individual (argumentum ad hominem, as it is called) is also a variety of this fallacy.

Assuming that to be true of the whole which is true only of the part, and the converse. Fallacies in which assumptions of this kind are made are technically called the fallacies of composition and of division, respectively. Thus if we argue that because participation in collegiate athletic contests benefits the particular individuals who take part in them, therefore such contests have a beneficial effect on the student body as a whole, we commit the fallacy of composition. Contrariwise, if we argue that because the Republican party deserves well of the country, therefore X, the Republican candidate for district Y, ought to be reelected, we commit the fallacy of division.

The false cause. The most common form of this fallacy, perhaps, is that usually known as post hoc ergo propter hoc (after this, therefore because of this), in which the assumption is made that because one thing follows another, there is therefore a necessary connection between the two things. X, for instance, takes a certain quack remedy for the rheumatism, and finds, after a few days, that his rheumatism is all gone; he leaps to the conclusion, accordingly, that the remedy has cured his malady, whereas it may have had no effect one way or the other.

Hasty generalization. The fallacy here consists in assuming that to be true generally which happens to be true in one or two particular cases,—that is to say, in inferring the existence of a law or general truth from a too narrow basis of observation. Because we

happen to have been unlucky in certain ventures which we began on a Friday, it does not follow that Friday is an unlucky day on which to begin anything. Yet a great many people argue in precisely this fashion.

The disputant must remember that even if a given argument used by an opponent is shown to rest on a fallacy, it does not necessarily follow that his conclusions are false. They may be true for other reasons that he urges, or for reasons that he fails to urge. It is only when the fallacious argument is an essential part of his proof, that the exposing of the fallacy means the overthrow of his conclusions. Effective refutation lays bare the cardinal points of a chain of reasoning and shows that they are incapable of supporting the conclusions that rest on them.

A particularly effective method of refutation, wherever it can be employed, is that known as reductio ad absurdum,-that is, showing that an opponent's arguments lead to manifest absurdities when carried to their logical conclusion. A passage from Webster's Reply to Hayne well illustrates the method. Senator Hayne, with others of his party, contended that in case of a plain, palpable violation of the Constitution by the general government, a State may interpose; and that this interposition is constitutional." In the course of his reply, Webster cited the tariff of 1828 and observed as follows:

The tariff is a usurpation; it is a dangerous usurpation; it is a palpable usurpation; it is a deliberate usurpation. It is such a usurpation, therefore, as calls upon the