Although it is not possible to take space for a very intricate example of the skeleton of an argument, it is hardly fair to give nothing more complex than a simple syllogism; and the following may assist the formation of a more clear conception of the form in which reasoning should be put. Suppose the proposition which is to be proved to be, “The Norsemen discovered America before Columbus.”

Taking a few of the more obvious arguments which might be advanced in support of this proposition, and arranging them so as to begin with the more generally allowed and easily proved, we have:

1. The frequent appearance in European literature before Columbus of allusions to a land across the sea.

2. The story in the Icelandic Sagas.

3. Norse remains in America.

These proofs will be sufficient for purposes of illustration. Let us examine them in detail a little. Under each of these proofs—which it is convenient to call subordinate propositions—lies a syllogism, whether it is fully stated or not. The writer must be entirely clear in his own mind what this is, whether it seem to him well to state it explicitly or not. Here the syllogism of the first subordinate proposition, briefly stated, is:—

This brings us face to face with the necessity of supporting premises with facts. To support the first sub-proposition there must be citations from pre-Columbian literature. This is a mere matter of research. One reason for putting this especial proof first is that in supporting it it is possible to begin with facts which cannot be questioned. It is true that the very next step will bring us upon doubtful territory, but we start from firm ground. The moment that the passages are quoted, the possibility occurs to the mind that they may be taken to refer to lands then known, or as the expression of mythical fancies. These objections must be met. An argument can no more pass an unanswered objection than a locomotive can cross a bridge from which a span is missing. Reasons must here be given for connecting with the New World the passages cited. It will be no less important to show the reasons for supposing that the information which the ancient writers possessed of the New World came from the Norsemen. The rest may all be allowed, and yet be held to have no bearing on the thing to be proved, so that this link in the chain must be made strong and evident.

This last point illustrates the sort of questions which are likely to arise in regard to arrangement. Is it well to introduce here the proofs that this knowledge of another continent came from the Norse, or would it be better to wait until the Icelandic Sagas have been spoken of? In the latter case, the parts of the argument may be more closely bound together, and it gives an air of fairness to the whole when the writer is willing to go back a good way in his argument to take up possible objections and answer them. Against this is to be balanced the possibility that the reader may be put into a suspicious state of mind by finding that a doubtful point is passed over, and so be less easily convinced than he otherwise might be. The writer of an argument must consider these things, and upon the good judgment with which he settles such questions much of his success depends.

In this first sub-proposition there is no need of stating formally the syllogism involved, since, if the first or major premise is successfully defended, the rest follows obviously. As a matter of practical arrangement, then, the sub-proposition and its defense might stand in this order:—

1. The discovery of America by the Norsemen is proved by the allusions in pre-Columbian literature to a land over sea.

a. Citation of passages.

b. Proofs that these do not refer to the Eastern Hemisphere.

c. Proofs that they are not mythical allusions.

d. Proofs that they do refer to America.

e. Proofs that the knowledge shown came from Norsemen.

When these points are established, the first sub-proposition, with its underlying syllogism, may be looked upon as proved, and the next may be taken up. I have not studied the question of the discovery of America by the Norsemen closely enough to know that the line of proof given is the best possible, but it serves well enough to illustrate the general form of the skeleton of an argument. Each of the subordinate propositions must be divided and subdivided if necessary, until the divisions can be handled easily and proved conclusively; and the writer will do well to test the strength of his argument by making a complete chain of the syllogisms involved, seeing which rests upon another, so that the arrangement may be conformed to principles of natural sequence.

One important matter in reasoning is never to claim too much. Care must be taken not to put upon a proof a greater strain than it will bear. It is also an obvious rule that it is wise to insist upon no more than is absolutely needed to establish the proposition in hand. Sometimes it is wise to indicate that more might be proved, but in general the assent of the reader is to be treated as a bank account to be drawn upon as far as it is necessary, but in no case beyond the actual need of the occasion. It is well never to waste strength in proving more than is essential, and always to avoid a side issue as one shuns a road leading to sure destruction. Often it is a wise device in argumentation to establish a point and leave the reader to perceive its import. Here as everywhere a thing which the reader is led to do for himself is a hundred fold more effective than anything which can be done or said for him. The phrase, “Is it not possible that these facts prove this?” has won more converts than the boldest assertion: “These proofs make it impossible to doubt.” Man that is born of woman is born to obstinacy as the sparks fly upward, and if he be assured that he cannot or shall not doubt, he is apt to begin to doubt from simple contrariety. Yet it will not do to run any risk of leaving the reader in doubt as to what has been established by the arguments given. It is often necessary to insist that a proposition is proved. A victory is hardly recognized as a victory until the trumpets are blown, and an argument is scarcely concluded without some sort of a declaration of success.

Where the line is to be drawn between the extreme of leaving to the reader the perception of what is proved and that of insisting that a demonstration has been made must depend upon the audience addressed. The writer of an argument has especial need to be sure to whom he speaks. He must consider the knowledge of his audience, their views, and especially their prejudices. It is in relation to the last that there comes into play what it has been the fashion to call Persuasion. Although Argument is an appeal to the intellect, there are few chains of reasoning which fail to appeal also to the emotions. It is hardly possible to conceive of a discussion which will not to a greater or less degree touch the passions of those addressed. Much is effected by keeping in mind the natural prejudices of the reader, and so framing arguments that they shall appeal directly to the emotions by the personal or selfish nature of their deductions. An illustration is to be found in political harangues, which, while nominally devoted to proving the wisdom or advisability of some party measure, are really only clever attempts to convince that the measure is for the personal advantage of voters. This is, of course, the abuse of this form of argument. The legitimate use of this appeal to the passions is in the putting of reasoning sound and wholesome in itself into a form which shall captivate the hearer or reader. It is the lavishing upon the composition of all the graces of manner, of style, of ornament, which the writer can compass, to the end that the reader shall be attracted and inclined to accept the conclusions set down. Stevenson speaks of a somewhat different matter in words which are strictly applicable here:—

Almost as often as with the establishment of our own reasoning, are we concerned in argument with endeavors to overthrow the logic of an opponent. Frequently it is necessary to refute views opposed to that which is being put forward. It is in general wise not to bring this in too soon. It is well to predispose the reader in favor of the conclusion to be defended, and then to take up contrary opinions. Sometimes a broad statement at the beginning to the effect that objections exist is politic; and in any case it is important that there be no slightest appearance of shirking or evading the issue. When the writer is conscious that the weight of popular sentiment or general opinion is against him, he may sometimes command attention and provoke interest by boldly plunging at once into an attack upon commonly received theories. Audacity always commands attention, and if it be reinforced by ability it is no less sure of admiration. A striking example of this method is to be found in Colonel Ingersoll’s attacks upon the Bible and religion. However one may be shocked by his violations of good taste, and whether one does or does not agree with his methods or his conclusions, it is impossible to deny his success as a speaker. The very boldness with which he has attacked has insured a hearing. This form of discussion calls for dash, courage, and confidence,—and it is sometimes the result of sheer impudence. Only he who has great powers and perfect command of them can reasonably hope to succeed here.

For answering the arguments of others, and indeed for the proper examination of one’s own, it is necessary to give attention to the numerous fallacies which may creep into reasoning, by design or by accident. These are to be completely mastered only by the minute study of logic; but some are so common that they should be considered here.

The first fallacy is that of the confusion of terms, such as that found in the attempt to prove the chef to be an interpreter of the beautiful, or in the question whether Whitman was a poet.

The second fallacy which in practical writing it is well to be on guard against is the non sequitur. There is much advanced as argument—as for instance in political editorials—where consecutiveness is confounded with causality. Post hoc ergo propter hoc is the phrase which sums this up: After this, therefore because of this. “We shall die after eating this meal, therefore this meal is the cause of our death,” is an example of this fallacy. Put in this way the absurdity is evident; but a genuine fallacy, lurking under words as the conventional serpent of school-girl compositions and of temperance orators lurks beneath flowers, is a different thing. Here is part of an editorial from one of the leading New York daily journals:

In the especial case to which this refers there may or there may not have been bribery; but it is well to bear in mind that this editorial proves nothing. It amounts merely to saying that the vote happened after the interview and was therefore the result of it; so that it is in reality one of those fallacies which in a simpler form appear so absurd. Yet readers in abundance accept this sort of thing as proof, especially when political prejudice inclines them to believe it. It would seem that a little common sense and a little care in examination were all that could be needed to dispose of specious errors of this class, yet they every day prevail.

The third fallacy is that of analogy. Analogy proves a probability, but it cannot establish a certainty. If a young woman has refused a dozen suitors, it is manifestly absurd to say that this proves that she will be equally unkind to the thirteenth. Politicians reason by analogy that a State which hitherto has gone Republican or Democratic may be counted upon to give a majority for its old party; when, lo, a change comes suddenly, and the conclusion is found to be false. That we have always liked the novels of a certain author does not insure that we shall be pleased with his next; that the sun has always risen does not prove that it will rise to-morrow morning; that men have from time immemorial been born with one head does not prove that a child may not be born with two,—as testify the freaks of dime museums. It is true that analogy often establishes a probability so strong that it amounts to a moral certainty. We are justified in acting upon the assumption that the sun will rise to-morrow, and in assuming that any given child of whose birth we hear has but one head. It is important in arguing, however, to bear in mind the difference, whether in one’s own reasoning or in that of an opponent, between analogy and absolute proof. Things which are like the same thing are like each other; but things that are like the same thing are not necessarily equal to each other.

The practical rules which may be given for the writing of Argument are chiefly recapitulations of what has been said.

a. Begin with clear understanding and clear statement of proposition and of terms.

b. Plan argument with reference to the especial point to be established and to the audience to be addressed.

c. Proceed generally from the more obvious to the less clear, and from the weaker to the stronger proof.

d. Be acutely alive to fallacies in any reasoning which is to be refuted, but to fallacies in your own work no less.

e. Never force a proposition or a proof beyond its value.

f. Concede all side issues and irrelevant matters if by so doing you do not lessen the chain of reasoning in points really important, and especially if in so doing you can foster a disposition favorable to your position.

g. Always remember that assertion is not argument.

To these rules might not inappropriately be added the saying of Sophocles: “Truth is always the strongest argument.”