likely to accrue, as of being likely to prove really a good; while by one who is indifferent and reluctant, the opposite view is taken.

Now, there are three causes of a speaker's deserving belief; for so many in number are the qualities on account of which we lend our credit, independently of proof adduced; and these are prudence, moral excellence, and the having our interests at heart (for men are fallacious in what they allege or advise by reason, either of all, or some, of these causes; for either, from want of ability, they do not rightly apprehend the question; or, rightly apprehending it, from their depravity, they do not tell you what they think; or, being men both of ability and moral excellence, they have not your interests at heart, on which account it is possible they should not give you the best advice, though fully known what is best); and besides these there is no other: it follows therefore, of course, that the speaker who appears to possess all these qualities is considered by his audience as deserving credit. Now, the means by which men may appear virtuous and prudent are to be derived from what has been laid down on the subject of the virtues; for it is by help of the very same things that an orator may invest himself, and any one else, in a certain character. The subject of feeling an interest, and of friendliness, must be discussed in my treatise of the passions, commencing henceforth. Passions, however, are all emotions whatsoever, on which pain and pleasure are consequent, by whose operation, undergoing a change, men differ in respect to their decisions: for instance, anger, pity, fear, and whatever other emotions are of such a nature, and those opposed to them. But it will be fitting to divide what I have to say respecting each into three considerations: to consider, respecting anger, for example, how those who are susceptible of anger are affected; with whom they usually are angry; and on what occasions. For, granted that we be in possession of one, or even two of these points, and not of them all, it will be impossible for us to kindle anger in the breast; and in the case of the rest of the passions in a similar way. In the same way, then, as on the subjects treated of above, I have separately drawn up the several propositions, so let me do in respect of these also, and make my distinctions according to the manner specified.

IT

THE PROPER HANDLING OF AN ARGUMENT

T FOLLOWS that I speak of the modes of disengaging oneself from arguments. They are either the meeting them with contradictory arguments, or starting an objection. Now as to the meeting them with counter arguments, it is evident that we may do it on the ground of the same topics [as were given for refutation]; for the arguments arise out of probabilities, and many things which appear likely are opposed to each other.

Objections, however (as was stated in the "Topics »), are started in four ways: 1. For either it may be from the same subject; 2. from a similar; 3. or an opposite [to that from which the adversary argues]; 4. or from points already decided. By deriving your objection from the same source, I mean that, supposing the enthymeme were respecting love, "that it was a good feeling»; there would be a twofold objection; for it [might be started] asserting generally that "every want is bad”: or, particularly, that "the proverbial expression 'Caunian loves' would not have arisen had there not been some wicked loves." Again, an objection is alleged on the ground of a contrary fact; as if the enthymeme was this, "The good man benefits all his friends; » [and the objection,] "But the bad man does not hurt all his.» And on the ground of similar cases; if the argument be, "Those who have been treated ill always

hate;» [the objection,] that "those who have been treated well do not always love.» And again, the decisions of men of celebrity: thus, suppose one brought forward the argument, that "we ought to have some feeling for those who are intoxicated, because they err ignorantly"; this objection [may be started], that "Pittacus is not therefore entitled to praise, otherwise he would not in his enactments have imposed higher fines, in case the party committed the error while intoxicated.»

But as all reasonings [of the orator] are derived from four sources, and these four are probability, example, proof positive (тεкμýριov), and signs; and as the reasonings drawn from what is usual, or appears to be so, are drawn from probabilities, while those drawn by inference from similarity of circumstances, whether in one or more instances (when the speaker embracing what is general then infers particulars), exist by virtue of example; while those again which are inferred from what is necessary and fact are founded on proof positive; and lastly, as those drawn from what does or does not hold good, whether universally or particularly, result from signs, [it being remembered] that probability is not what always, but what usually occurs; it is plain that it is, in every case, possible to get rid of reasonings such as these by starting an objection. The solution is, however, [sometimes] apparent, and not always real; for the objector does not do it away on the ground of its not being probable, but on that of its not being necessary. Wherefore it always happens that the defendant has the advantage of the accuser, by means of this piece of sophistry. For as the accuser constructs his proof by means of probabilities (the task of getting rid of the positive certainty being by no means the same [in point of difficulty], as that of getting rid of the probability of the charge); and as that which is merely probable is invariably open to an objection (for it otherwise would not be a probability, but invariable and necessary); and if this method of solution have been adopted, the judge supposes either that it does not amount to a probability, or at least that he ought not to decide, having been imposed upon in the way mentioned above; because [in fact] he is not bound to pronounce on positive proofs alone, but also on probabilities, which is the spirit of the oath, "that he will decide to the best of his judgment »; wherefore that will not be a satisfactory objection which rests merely on the absence of proof positive, but it is further incumbent on the objector to get rid of the probability; this, however, will be the case, if the objection be probable in a higher degree (and it may be so in two ways, either on the score of time, or of the nature of the case; and pre-eminently so, if in both these particulars); for if in the majority of instances it be as you state, then it is a greater probability.

Signs also, and the reasonings deduced from them, may be got rid of, even if they be facts, as was stated in the first book; for it is clear to us from the "Analytics" that every sign is illogical.

The same method of solution applies to reasonings grounded on example as to those on probabilities; for if we have a single instance in contravention, it has been answered [sufficiently to show] that it is not necessary; or that in the majority of instances, and those of more frequent occurrence, the case is otherwise. If, however, it be the case more frequently, and in the majority of instances, we must contend that the present is not the case in point, or that its application is not in point, or that it has some difference at all events.

But proofs positive (TɛKμhpia), and the reasonings grounded on them, we shall not be able to get rid of, at least not on the plea of inconclusiveness; this is clear to us from the "Analytics": it remains for us to show that what is asserted is not the case; if, however, it be clear both that it is true as a matter of fact, and that it is a proof positive of the point, from that moment it becomes irrefragable; for henceforth it is plain from demonstration.

A

OF AMPLIFICATION AND EXTENUATION

MPLIFICATION and extenuation are not elements of enthymemes (by topic and element I mean the same thing), since the element and the topic is that under which many enthymemes fall; whereas amplification and extenuation are [themselves] enthymemes for showing that a thing is great or little, like those for showing that it is good or bad, just or unjust, or falls under either of the other denominations. And these are all the questions about which syllogisms and enthymemes are conversant; so that unless each of these be a topic of an enthymeme, amplification and extenuation are not.

Neither, again, are the topics which are available to solution of enthymemes, at all different in species from those employed in their construction: for it is evident that he effects solution who either proves [something contrary], or states an objection, and they establish a counter proof of the contrary; thus, "If one has argued that a fact has taken place, the other argues that it has not"; or "if one argues that it has not, the other insists that it has." So that this will not amount to a difference; for both employ the same vehicles of proof, inasmuch as they each allege enthymemes to show that it is, or is not the case.

But the objection is not an enthymeme, but is, as was stated in the "Topics," the stating some opinion from which it will appear that no legitimate inference has been arrived at, or that [the opponent] has assumed some false proposition.

Thus much, then, on the subject of examples, and maxims, and enthymemes, and, in a word, all the means of persuasion which address themselves to the understanding, both the sources whence we may furnish ourselves plentifully with them, and the means by which we may effect their solution. It remains for us to go over the subject of style and arrangement.

L

ON EXCELLENCE OF STYLE

ET excellence of style be defined to consist in its being clear (a sign of this is this, that the diction, unless it make the sentiment clear, will not effect its purpose); and neither low, nor above the dignity of the subject, but in good taste; for the style of poetry, indeed, is not low, yet it is not becoming in prose.

Of nouns and verbs those which are in general use produce the effect of clearness: to prevent its being low, and to give it ornament, there are other nouns which have been mentioned in the "Poetics," for a departure [from ordinary acceptations] causes it to appear more dignified; for men are affected in respect of style in the very same way as they are towards foreigners and citizens. On which account you should give your phrase a foreign air; for men are admirers of things out of the way, and what is an object of admiration is pleasant. Now in the case of metrical compositions, there are many things which produce this effect, and there are very becoming, because both the subject and the person stand more apart [from ordinary life]; in prose, however, these helps are much fewer, for the subject is less exalted: since even in that art were a slave, or a mere youth, or [any one, in fact, in speaking] of mere trifles to express himself in terms of studied ornament, it would be rather unbecoming; but here too [as in poetry] the rule of good taste is that your style be lowered or raised according to the subject. On which account we must escape observation in doing this, and not appear to speak in a studied manner, but

naturally, for the one is of a tendency to persuade, the other is the very reverse; because people put themselves on their guard, as though against one who has a design upon them, just as they would against adulterated wine. [Let your style then be such] as was the case with the voice of Theodorus as compared with that of the other actors; for it appeared to be that of the character which was speaking, theirs, however, were foreign from the character. And the deceit is neatly passed off if one frame his nomenclature upon a selection from ordinary conversation; the thing which Euripides does, and first gave the hint of.

As, however, nouns and verbs are [the materials] of which the speech is made up, and as nouns admit so many species as have been examined in the "Poetics," out of the number of these we must employ but sparingly, and in very few places, exotic and compound words, and those newly coined; where they may be employed I will state hereafter: the reason [of the restriction] has been mentioned, viz., because they remove your style [from that of common life] more than is consistent with good taste. Words, however, of ordinary use, and in their original acceptations, and metaphors, are alone available in the style of prose: a proof [that this is the fact is] that these are the only words which all persons employ; for everybody carries on conversation by means of metaphors, and words in their primary sense, and those of ordinary use. Thus it is plain that if one should have constructed his style well, it will be both of a foreign character, and that [the art of the orator] may still elude observation, and [the style itself] will have the advantage of clearness; this, however, was laid down to be the perfection of rhetorical language. But of all nouns, those which are equivocal suit the purposes of the sophist, for by their help he effects his fallacies, while synonyms are of use to the poet; I mean these which are both synonyms and of common usage, as яopevŋobai and ẞadičew, for these two are both of common usage and synonymous to each other.

The nature then of each of these varieties, and how many species of metaphor there are, and also that this ornament is of the greatest effect, as well in poetry as prose, has been explained (as I have observed above), in the "Poetics." In prose, however, we should bestow the greater attention on them, in proportion as an oration has to be made up of fewer adjuments than a metrical composition. Moreover, the metaphor possesses in an especial manner [the beauties of] clearness and sweetness, with an air of being foreign; and it is not possible to derive it from any other person.

You must, however, apply, in the case both of epithets and metaphors, such as are appropriate; and this will depend on their being constructed on principles of analogy, otherwise they will be sure to appear in bad taste; because contraries show themselves to be such, particularly when set by each other. But you must consider, as a purple garment becomes a youth, what is equally so to an old man; since the same garment does not become [both].

And if you wish to embellish your subject, see you deduce your metaphor from such things coming under the same class as are better; and if to cry it down, from such as are worse: I mean, as the cases are opposed and come under the same genus, that the saying, for example, of a beggar, that "he prays," and of one who is praying, that "he begs" (both being species of asking), is to do the thing which has been mentioned; just as Iphicrates called Callias "a mere collector to the goddess, and not a bearer of the torch.» He, however, replied, "that he must needs be uninitiated himself, or he would not call him a collector, but a bearer of the torch." For these are both services connected with the goddess; the one, however, is respectable, while the other is held in no repute. And some one [speaks of the courtiers of Dionysius as] Dionysian parasites; they, however, call themselves artificers. And these expressions are both metaphors; the one of persons who would

depreciate, the other the contrary.

Even robbers, nowadays, call themselves purveyors. On which principle we may say of a man who "has acted unjustly," that he "is in error»; and of one who "is in error," that he "has acted unjustly.» Again, of one who has stolen, both that has taken, [in way of diminution,] and that has ravaged [in exaggeration.] But the saying, as the Telephus of Euripides does, "that he lords it o'er the oars, and landing in Mysia," etc., is out of taste; for the expression, "lording it o'er," is above the dignity of the subject; [the rhetorical artifice] then, is not palmed off. There will also be a fault in the syllables, unless they are significant of a grateful sound; for instance, Dionysius, surnamed Chalcous, in his elegies, calls poetry, "the clangor of Calliope," because both are vocal sounds; the metaphor, however, is a paltry one, and couched in uncouth expressions.

Again, our metaphors should not be farfetched; but we should make the transfer, on the principle of assigning names out of the number of kindred objects, and such as are the same in species, to objects which are unnamed, of which, however, it is clear, simultaneously with their being uttered, that they are akin, as in that approved enigma,—

for the operation is undesignated by any name, and both are species of attaching; wherefore the writer called the application of the cupping instrument, a gluing. And, generally speaking, it is possible out of neatly constructed enigmas to extract excellent metaphors: because it is on the principles of metaphor that men construct enigmas; so that it is evident that [if the enigma be a good one] the metaphor has been properly borrowed.

The transfer also should be made from objects which are beautiful; beauty, how. ever, of words consists, as Licymnius observes, in the sound or in the idea conveyed; as does also their inelegance. And there is, moreover, a third, which does away the sophistical doctrine; since it is not the fact, as Bryso argues, "that no one speaks inelegantly, if, indeed, the using one expression instead of another carries with it the same meaning”: for this is a fallacy; because some words are nearer in their ordinary acceptations, more assimilated, and have more peculiar force of setting the object before the eyes than others. And what is more, one word represents the object under different circumstances from another; so that we may even on this principle lay it down that one word has more or less of beauty and inelegance than another; for although both words, [at the same time,] express [properties which are] beautiful, as well as such as are inelegant; yet they either express them not qua they are beautiful, or not qua they are inelegant; or granting they do, yet they express them, the one in a greater, the other in a less degree. But we are to deduce our metaphors from these sources;—from such as are beautiful either in sound, in meaning, or [in the image they present] to the sight, or any other sense. And there is a difference, in the saying, for instance, "the rosy-fingered Aurora,» rather than "the purple-fingered," or, what is still worse, "the crimson-fingered.»

Also, in the case of epithets, it is very possible to derive one's epithets from a degrading or disgraceful view of the case; for instance, "the murderer of his mother": and we may derive them from a view on the better side; as, "the avenger of his father." And Simonides, when the victor in a race by mules offered him a

* A reference to the operation of "cupping."