In things equal to one another, in this way:—"For as a place without a harbour cannot be safe for ships, so a mind without integrity cannot be trustworthy for a man's friends." In those things which fall under the same principle a probable argument is considered in this way:—"For if it be not discreditable to the Rhodians to let out their port dues, then it is not discreditable even to Hermacreon to rent them." Then these arguments are true, in this manner:—"Since there is a scar, there has been a wound." Then they are probable, in in this way:—"If there was a great deal of dust on his shoes, he must have come off a journey." But (in order that we may arrange this matter in certain definite divisions) every probable argument which is assumed for the purpose of discussion, is either a proof, or something credible, or something already determined; or something which may be compared with something else.

That is a proof which falls under some particular sense, and which indicates something which appears to have proceeded from it, which either existed previously, or was in the thing itself, or has ensued since, and, nevertheless, requires the evidence of testimony, and a more authoritative confirmation,—as blood, flight, dust, paleness, and other tokens like these. That is a credible statement which, without any witness being heard, is confirmed in the opinion of the hearer; in this way:—There is no one who does not wish his children to be free from injury, and happy. A case decided beforehand, is a matter approved of by the assent, or authority, or judgment of some person or persons. It is seen in three kinds of decision;—the religious one, the common one, the one depending on sanction. That is a religious one, which men on their oaths have decided in accordance with the laws. That is a common one, which all men have almost in a body approved of and adopted; in this manner:—"That all men should rise up on the appearance of their elders; That all men should pity suppliants." That depends on sanction, which, as it was a doubtful point what ought to be considered its character, men have established of their own authority; as, for instance, the conduct of the father of Gracchus, whom the Roman people made consul after his censorship, because he had done nothing in his censorship without the knowledge of his colleague.

But that is a decision admitting of comparisons, which in a multitude of different circumstances contains some principle which is alike in all. Its parts are three,—representation, collation, example. A Representation is a statement demonstrating some resemblance of bodies or natures; Collation is a statement comparing one thing with another, because of their likeness to one another; Example is that which confirms or invalidates a case by some authority, or by what has happened to some man, or under some especial circumstances. Instances of these things, and descriptions of them, will be given amid the precepts for oratory. And the source of all confirmations has been already explained as occasion offered, and has been demonstrated no less clearly than the nature of the case required. But how each separate statement, and each part of a statement, and every dispute ought to be handled,—whether we refer to verbal discussion or to writings,—and what arguments are suitable for each kind of discussion, we will mention, speaking separately of each kind, in the second book. At present we have only dropped hints about the numbers, and moods, and parts of arguing in an irregular and promiscuous manner; hereafter we will digest (making careful distinctions between and selections from each kind of cause) what is suitable for each kind of discussion, culling it out of this abundance which we have already displayed.

And indeed every sort of argument can be discovered from among these topics; and that, when discovered, it should be embellished, and separated in certain divisions, is very agreeable, and highly necessary, and is also a thing which has been greatly neglected by writers on this art. Wherefore at this present time it is desirable for us to speak of that sort of instruction, in order that perfection of arguing may be added to the discovery of proper arguments. And all this topic requires to be considered with great care and diligence, because there is not only great usefulness in this matter, but there is also extreme difficulty in giving precepts.

XXXI. All argumentation, therefore, is to be carried on either by induction, or by ratiocination. Induction is a manner of speaking which, by means of facts which are not doubtful, forces the assent of the person to whom it is addressed. By which assent it causes him even to approve of some points which are doubtful, on account of their resemblance to those things to which he has assented; as in the Aeschines of Socrates, Socrates shows that Aspasia used to argue with Xenophon's wife, and with Xenophon himself. "Tell me, I beg of you, O you wife of Xenophon, if your neighbour has better gold than you have, whether you prefer her gold or your own?" "Hers," says she. "Suppose she has dresses and other ornaments suited to women, of more value than those which you have, should you prefer your own or hers?" "Hers, to be sure," answered she. "Come, then," says Aspasia, "suppose she has a better husband than you have, should you then prefer your own husband or hers?" On this the woman blushed.

But Aspasia began a discourse with Xenophon himself. "I ask you, O Xenophon," says she, "if your neighbour has a better horse than yours is, whether you would prefer your own horse or his?" "His," says he. "Suppose he has a better farm than you have, which farm, I should like to know, would you prefer to possess?" "Beyond all doubt," says he, "that which is the best." "Suppose he has a better wife than you have, would you prefer his wife?" And on this Xenophon himself was silent. Then spake Aspasia,—"Since each of you avoids answering me that question alone which was the only one which I wished to have answered, I will tell you what each of you are thinking of; for both you, O woman, wish to have the best husband, and you, O Xenophon, most exceedingly desire to have the most excellent wife. Wherefore, unless you both so contrive matters that there shall not be on the whole earth a more excellent man or a more admirable woman, then in truth you will at all times desire above all things that which you think to be the best thing in the world, namely, that you, O Xenophon, may be the husband of the best possible wife; and you, O woman, that you may be married to the most excellent husband possible." After they had declared their assent to these far from doubtful propositions, it followed, on account of the resemblance of the cases, that if any one had separately asked them about some doubtful point, that also would have been admitted as certain, on account of the method employed in putting the question.

This was a method of instruction which Socrates used to a great extent, because he himself preferred bringing forward no arguments for the purpose of persuasion, but wished rather that the person with whom he was disputing should form his own conclusions from arguments with which he had furnished himself, and which he was unavoidably compelled to approve of from the grounds which he had already assented to.

XXXII. And with reference to this kind of persuasion, it appears to me desirable to lay down a rule, in the first place, that the argument which we bring forward by way of simile, should be such that it is impossible to avoid admitting it. For the premiss on account of which we intend to demand that that point which is doubtful shall be conceded to us, ought not to be doubtful itself. In the next place, we must take care that that point, for the sake of establishing which the induction is made, shall be really like those things which we have adduced before as matters admitting of no question. For it will be of no service to us that something has been already admitted, if that for the sake of which we were desirous to get that statement admitted be unlike it; so that the hearer may not understand what is the use of those original inductions, or to what result they tend.

For the man who sees that, if he is correct in giving his assent to the thing about which he is first asked, that thing also to which he does not agree must unavoidably be admitted by him, very often will not allow the examination to proceed any further, either by not answering at all, or by answering wrongly. Wherefore it is necessary that he should, by the method in which the inquiry is conducted, be led on without perceiving it, from the admissions which he has already made, to admit that which he is not inclined to admit, and at last he must either decline to give an answer, or he must admit what is wanted, or he must deny it. If the proposition be denied, then we must either show its resemblance to those things which have been already admitted or we must employ some other induction. If it be granted, then the argumentation may be brought to a close. If he keeps silence, then an answer must be extracted, or, since silence is very like a confession, it may be as well to bring the discussion to a close, taking the silence to be equivalent to an admission.

And so this kind of argumentation is threefold. The first part consists of one simile, or of several, the second, of that which we desire to have admitted, for the sake of which the similes have been employed, the third proceeds from the conclusion which either establishes the admissions which have been made or points out what may be established from it.