the hardiness of the survivors, which is the cause, not the effect, of their having lived through such a training. So, loading a gunbarrel to the muzzle and firing it does not give it strength; though it proves, if it escape, that it was strong1"

§ 8. (3) Of the Ignoratio Elenchi.

This fallacy occurs whenever in any debate or discussion the conclusion arrived at, or the argument advanced, is not to the point: you wish to disprove or establish a certain proposition, and for this purpose you advance arguments which lead to a conclusion which is quite irrelevant to the subject at hand. For example, you wish to prove that a certain doctrine is false; and instead of adducing facts or principles or both, which really disprove it, you dilate upon its consequences, upon the small number of its adherents, upon the moral qualities of its promulgators, and so forth. The way in which the Theory of Evolution is at the present day attacked by some, and defended by others, will furnish us with very apt illustrations of this fallacy. On the one hand, many popular speakers and writers attempt to refute it by arguments which have reference only to its consequences, to its appearance of absurdity, and to the prejudices and sentiments of the people; and, on the other, many of its defenders attempt to prove it by arguments which are no better than the former, having reference only to the high authority of the scientific men who believe in it, to their numerical strength, to the grandeur and beauty of the Theory, to the impossibility of the popular doctrine being true, and so forth. Both the opponents and the defenders of the Theory are equally guilty of the fallacy of Ignoratio Elenchi, inasmuch as they do not address themselves to the facts and principles really bearing upon the question.

Whately describes and illustrates the more important forms of this fallacy as follows:

“It is evident,” says Whately, "that Ignoratio Elenchi may be employed as well for the apparent refutation of your op

1 Whately's Elements, 9th Edition, p. 135.

ponents' proposition, as for the apparent establishment of your own; for it is substantially the same thing to prove what was not denied, or to disprove what was not asserted. The latter practice is not less common; and it is more offensive, because it frequently amounts to a personal affront, in attributing to a person opinions, &c., which he perhaps holds in abhorrence. Thus, when in a discussion one party vindicates, on the ground of expediency, a particular instance of resistance to Government in a case of intolerable oppression, the opponent may gravely maintain, that 'we ought not to do evil that good may come—a proposition which of course had never been denied; the point in dispute being 'whether resistance in this particular case were doing evil or not.' Or again, by way of disproving the assertion of the 'right of private judgment in religion,' one may hear a grave argument to prove that 'it is impossible that every one could be right in his judgment.' In these examples, it is to be remarked that the fallacy of Petitio Principii is combined with that of Ignoratio Elenchi; which is a very common and often successful practice,―viz., the Sophist proves, or disproves, not the proposition which is really in question, but one which is so dependent on it as to proceed on the supposition that it is already decided, and can admit of no doubt; by this means his ‘assumption of the point in question' is so indirect and oblique, that it may easily escape notice; and he thus establishes, practically, his conclusion, at the very moment he is withdrawing your attention from it to another question. For example, an advocate will prove, and dwell on the high criminality of a certain act, and the propriety of severely punishing it; assuming (instead of proving) the commission.

"There are certain kinds of arguments recounted and named by logical writers which we should by no means universally call Fallacies; but which when unfairly used, and so far as they are fallacious, may very well be referred to the present head; such as the Argumentum ad hominem (or personal argument), Argumentum ad verecundiam, Argumentum ad populum, &c., all of them regarded as contradistinguished from Argumentum ad rem

or ad judicium. These have all been described in the lax and popular language before alluded to, but not scientifically: the Argumentum ad hominem, they say, 'is addressed to the peculiar circumstances, character, avowed opinions, or past conduct of the individual, and therefore has a reference to him only, and does not bear directly and absolutely on the real question, as the Argumentum ad rem does'; in like manner, the Argumentum ad verecundiam is described as an appeal to our reverence for some respected authority, some memorable institution, &c., and the Argumentum ad populum as an appeal to the prejudices, passions, &c., of the multitude; and so of the rest1."

"The fallacy of Irrelevant Conclusion (Ignoratio Elenchi) is nowhere more common than in protracted controversy, when one of the parties having attempted in vain to maintain his position, shifts his ground as covertly as possible to another, instead of honestly giving up the point. An instance occurs in an attack made in the system pursued at one of our universities. The objectors finding themselves unable to maintain their charge of the present neglect (viz., in the year 1810) of Mathematics in that place (to which neglect they attributed the late general decline in those studies), shifted their ground, and contended that that University 'was never famous for mathematicians'; which not only does not establish, but absolutely overthrows, their own original assertion; for if it never succeeded in these pursuits, it would not have caused their late decline2"

§ 9. Besides the fallacies we have mentioned above, two more, namely, the Non sequitur and the Fallacy of many questions, are also given under the class of material fallacies. The first occurs when the conclusion does not in any way follow from the premisses, when, in fact, there is no logical connection between the two, anything being inferred from anything else. The second occurs when, by way of asking questions, certain assumptions are made in regard to certain things or persons: "In what subjects did you fail?" This question assumes

1 Whately's Elements, pp. 141-—142.

2 Ibid. pp. 143—4.

(1) that you appeared at an examination, and (2) that you failed in more than one subject; while the real fact might be just the

reverse.

All these fallacies, as I have already said, do not properly belong to Deductive Logic. It is no part of Deductive Logic to ascertain whether a certain premiss is true or false, or whether a conclusion or an argument advanced by a party is irrelevant to the subject in debate. The Petitio Principii, indeed, would seem to be a fallacy of Deduction, inasmuch as the rules of Deductive Inference imply that a proposition can not be proved by means of itself, that a proposition, when inferred, must be inferred from others which are severally different from itself.

§ 10. Exercises.

1. In testing an argument consisting of a single categorical syllogism, the following method should be followed —

(i) Find the conclusion and note its subject and predicate which are, respectively, the minor and the major term of the syllogism.

(ii) Find the term which is not in the conclusion. It must be the middle term. (a) See if there be any other; if there is, then the argument involves the fallacy of four terms. (b) See if the middle term be ambiguous; if it is, then there is the fallacy of ambiguous middle. (c) See whether the middle term be distributed; if it is not, then there is the fallacy of undistributed middle.

(iii) Find the premiss which contains the minor term, and the premiss which contains the major term; and these two premisses are, respectively, the minor and the major premiss.

(iv) See if there be any term which is undistributed in either premiss, but distributed in the conclusion. If there is, then there is an illicit process.

If there be none of the above fallacies, then the argument is valid. To confirm this,

(v) Find the figure and mood of the syllogism, and see if the mood is a valid one in that figure.

2. In many cases the invalidity of an argument may be detected on mere inspection. For instance, when it contains two particular or two negative premisses, or when the middle term is not distributed, or when one of the premisses is negative and the conclusion affirmative, or, lastly, when one of the premisses is particular and the conclusion universal.

3. The method described above seems, on the whole, to be the best. But there are of course other methods, which may also be applied to verify the result obtained by it or to test the argument independently. For example, the figure and the mood of the syllogism may be at once found; if the mood be a valid one in the particular figure, the syllogism will be valid. Or the figure being found, the syllogism may be tested by the canon or the special rules of that figure; if it conform to the canon or to the rules, it will be valid. Or the syllogism may be tested by the method of the comparison of the diagrams: if the conclusion follow in every case, it will be valid; if it do not follow in a single case, it will be invalid1.

4. If an argument consists of more than one syllogism, that is, of a train of reasoning, it should be analysed into the constituent syllogisms; and each of them should be tested as described above. If any of the premisses be understood or suppressed, they should be supplied, and the constituent syllogisms fully expressed. In the case of Enthymemes, the suppressed premiss, whether true or false, should be supplied. In the case of Dilemmatic and other mixed arguments, they should be tested by their rules, and reduced to the categorical form. In the case of Extra-logical or Material fallacies, the student should be able to refer them to their respective classes and show where the fallacy lies.

Examples.

Test the following arguments:

1. Every metal conducts heat; every metal conducts electricity: therefore every substance that conducts heat conducts electricity. 2. No minerals are plants; no plants are animals: therefore no minerals are animals.

3. All plants are organized; no crystals are plants: therefore no crystals are organized.

شا

1 Read also the directions given in Part III. Chap. v.