Institutes of Oratory



Book I
Book II
Book III
Book IV
Book V
Book VI
Book VII
Book IX
Book X
Book XI
Book XII



1. Of Invention, I think, enough has been said; for I have not only treated of the mode of informing judges, but have touched on the art of exciting their feelings. But as it is not enough for those who are erecting edifices, to collect stones and materials, and other things useful for the architect, unless the hand of the workman be also applied to the disposition and collocation of them, so, in speaking, however abundant be the quantity of matter, it will form but a confused mass and heap, unless similar arrangement bind it together, disposed in regular order, and with its several parts connected one with another.

2. It is therefore not without reason that arrangement is considered the second of the five parts of oratory; for though all the limbs of a statue be cast, it is not a statue until they are united; and if, in our own bodies, or those of any other animals, we were to displace or alter the position of any part, they would, though they had the same number of parts, be but monsters. Even our joints, if but in the least degree dislocated, lose their whole use and power of action; and disorder in an army is an impediment to its efficiency. 3. Nor do those appear to be in the wrong, who think that the system of the world is maintained by order, and that, if its order were broken, it would cease to exist as a whole.

So speech, if deficient in that quality, must necessarily be confused, and float like a ship without a helm; it can have no coherence; it must exhibit many repetitions, and many omissions; and, like a traveller wandering by night in unknown regions, must, as having no stated course or object, be guided by chance rather than design.

4. The whole of this book, therefore, shall be devoted to arrangement; a quality, which, if it could be taught by rules adapted to every kind of subject, would not have fallen to the lot of so small a number of speakers. But as the forms of causes have been, and will ever be, infinite in variety, and as no one cause, during so many ages, has been found in all respects similar to another, the pleader must exercise his sagacity, his discernment, his invention, and his judgment, and must ask counsel from himself. Yet I do not deny that there are some things that may be taught by precept, and of these I shall not fail to treat.


1. Let division, then, as I signified above, be the distribution of a number of things into its component parts; partition, the regular distribution of parts into their members, and a just disposition connecting those that follow with those that precede; and arrangement a due distribution of things and their parts in their proper places. 2. But let us remember that arrangement is often altered to suit the interest of a cause, and that the same question is not always discussed first by both parties; a point of which, to omit other examples, Demosthenes and Aeschines may afford us an instance, who, in the trial concerning Ctesiphon, adopt a very different order, as the accuser commences with the question of law, on which he thought himself the stronger, while the defendant introduces all other particulars, or almost all, before touching on the question of law, in order to prepare the judges for considering the point of legality at the conclusion.

3. For it may be to the interest of one side to state one point first, and of the other to state another; else the pleading would always be conducted at the pleasure of the prosecutor; and, in a case of mutual accusation, when each party defends himself before he accuses his adversary, the order of everything on either side must be different. [In anticategoria, that is, recrimination and mutual accusation, the arrangement on each side is different; for instance, if you accuse me of homicide, and I accuse you of sacrilege, I should speak first of homicide, in order to clear myself, while you would speak first of sacrilege.] I shall therefore set. forth the method which I myself have followed, and which I have adopted partly from the rules of others, and partly from my own reasoning; nor have I ever made any mystery of it.

4. It was my great care, in forensic pleadings, to ascertain, in the first place, all the points that were concerned in any cause; as in the schools there are certain particulars, and but few, that are laid down previous to the declamation; the Greeks call them θεματα, Cicero [Topic. c. 21] proposita. When I had placed these, as it were, full in my view, I contemplated the cause not less with reference to the opposite side than to my own.

5. First, then, (what is not difficult to be ascertained, but is above all to be regarded,) I settled what each party wished to establish, and then by what means, in the following way. I considered what the prosecutor would state first; this would either be an admitted, or a contested point. If it were admitted, the question could not lie in it. 6. I passed therefore to the answer of the defendant, and considered it in the same way. Sometimes, too, what was elicited from thence was admitted.

But as soon as there began to be any disagreement, the question arose. The process was of this nature: You killed a man; I did kill him; the fact is admitted; I pass on. 7. The defendant ought to give a reason why he killed him. It is lawful, he may say, to kill an adulterer with an adulteress. It is admitted that there is such a law. We may then proceed to a third point, about which there may be a dispute. They were not guilty of adultery; they were.

Hence arises the question. It is a controversy about fact, a matter of conjecture. 8. Sometimes, however, a third point is admitted, that they were guilty of adultery. But, the accuser may say, it was not lawful for you to kill them; for you were an exile, or infamous. There is, then, a question about law. But if, when the prosecutor says at first, You have killed, the defendant reply, I have not killed, the dispute commences at once. It is thus that we must ascertain when the controversy begins; and we must consider what forms the first question.

9. The accusation may be simple: Rabirius killed Saturninus; or complex: Lucius Varenus has incurred the penalty of the law respecting assassins; for he is guilty of killing Caius Varenus, of wounding Cneius, and also of killing Salarius [see the fragments of the speech Pro Vareno in Ernesti]; since there will thus be three distinct propositions. The same may be said of civil suits [as distinguished from criminal prosecutions]. But out of a complex accusation may arise several questions and states, if the accused denies one point, justifies another, and endeavors to set aside another by taking exception at the form of process. In this case the accuser must consider carefully what he ought to refute, and in what parts of his speech.

10. As to what concerns the accuser, I do not altogether dissent from Celsus, who, doubtless following Cicero, persists in maintaining somewhat too positively, on this head, that strong arguments should be advanced in the first place, the strongest of all in the last, and the weaker in the middle; because the judge requires to be moved at the beginning, and pressed forcibly at the end. 11. But on the side of the accused, the strongest argument against him must first be attacked, lest the judge, looking to that point, should regard with too little favor our establishment of other points.

Yet this order may occasionally be changed, if the lighter points be evidently false, and the refutation of the heaviest charge extremely difficult; so that, after thus detracting from the credit of the accusers, we may proceed to the last point, when the judge is ready to suppose that all the charges may be false. It will be necessary, however, to make some preliminary remarks, in which a reason may be given for putting off the consideration of the principal charge, and a refutation of it may be promised, in order that we may not seem to fear that which we do not at once overthrow, 12. Attacks on the past life of the accused must generally be refuted first, that the judge may be inclined to hear with favor the question on which he is to give a decision. But Cicero, in his speech for Varenus, has delayed the consideration of such charges to the conclusion, regarding, not what is expedient generally, but what was expedient on that occasion.

13. When the accusation is simple, we must consider whether we will give our answer in one proposition or in several. If in one, whether we build our case on fact, or on written law. [Quintilian here looks to the primary division of general states, according to which some are de re, and are called status rationales; others de scripto.] If on fact, whether what is charged against us is to be denied or justified. If on written law, on what point of law the question stands, and whether it regards the letter or the intention. 14. This we shall discover, if we ascertain what law it is that gives rise to the suit, that is, on what the point for decision rests. In the exercises of the schools, some laws are laid down merely to connect a series of circumstances in a case: thus, Let a father [this law we find in Seneca the Rhetorician, pp. 286, 479; and in the 278th of the Declamations that go under the name of Quintilian], who recognizes a son that he has exposed, take him back on paying for his subsistence; Let it be lawful for a father to disinherit a son who is disobedient to his admonitions. A father who has taken back a son that he had exposed, requires him to marry a rich relation; the son wishes to marry the daughter of the poor person that brought him up.

15. The law regarding children exposed is a subject for moving the feelings; but the decision depends on the law concerning disinheritance. Nor does the question always rest on one law only, but sometimes on more than one, as in a case of αντινομια, or contradictory laws. This matter being considered, it will be seen about what points the question is.

A complex defense is such as that in Cicero's speech for Rabirius; If he had killed Saturninus, he would have acted rightly: but he did not kill him. 16. But when we advance many arguments against one proposition, we must consider, in the first place, all the points that can be advanced, and must then settle in what part of our speech it is expedient that each should be stated. 17. In regard to this matter, I do not hold the same opinion which I expressed a little above concerning propositions, and to which I assented in respect to arguments, (in the place in which I treated of proofs,) that we may sometimes begin with the stronger; for, in refutation, the force of our questions ought always to increase, and to proceed from the weakest to the strongest, whether they be of the same or a different kind. 18. But questions of law may sometimes arise from one ground of dispute after another; those of fact look always to one point; in both, however, the order is the same. But let us speak first of such as are of different kinds, the weakest of which ought to be discussed first.

Hence it is that, after considering some questions, we concede or grant them to the opposite party; for we cannot pass to others unless by dismissing those that come first. 19. This ought to be done in such a manner, that we may not appear to have despaired of them, but to have set them aside, because we can establish our cause without them. An agent demands money from a person for interest on an inheritance: a question may arise whether he who is acting as agent has a right to be an agent. 20. Suppose that we, after we have discussed this question, give it up, or are defeated upon it, the next question may be, whether he in whose name the action is brought, has a right to have an agent. Suppose that we give way on this point also, the cause may admit of the question whether he, in whose name the suit is instituted, is heir to the person to whom the interest is due, and sole heir.

21. If these points also be granted, it may be asked whether the money is really due. On the other hand, nobody would be so foolish as to yield what he considered his strongest point, and pass on to others of minor importance. Similar to the preceding case is one that is given in the schools: You must not disinherit an adopted son; though you may disinherit this adopted son, you must not disinherit one who hat deserved well of his country; though you may disinherit one who has deserved well of his country, you may not disinherit whatever deserving son has not obeyed your will; though he may have been bound to obey your will in all other things, you may not disinherit him for not having obeyed it in regard to an option, or, if you may disinherit him for an option, not for such an option as this. Such is the dissimilarity in questions of law. 22. But in matters of fact there may be several questions all tending to the same object: as if, for instance, a person who is on trial for theft, should say to the accuser, Prove that you had the property; prove that you lost it; prove that you lost it by theft; prove that you lost it by my theft. The first three points may be conceded; the last cannot.

23. I used also very frequently to adopt this method. I went back from the last species (for it is that which commonly contains the point for decision) to the first general question, or descended from the genus to the last species; and that even in deliberative causes. 24. Suppose, for example, that Numa deliberates whether he shall accept kingly power when the Romans offer it. First arises the general question, Whether he ought to reign at all; then follow the particular questions, Whether he ought to reign in a country not his own; whether at Rome; whether the Romans will tolerate such a king as himself.

The case is similar in matters of controversy: Suppose a man who has deserved well of his country makes choice of another man's wife: the last special question is, Whether a man can make choice of another's wife: The general question is, Whether he who has deserved well of his country ought to receive whatever he makes the object of his choice; then follow the inquiries, whether he can choose from the property of a private person; whether he can demand a woman in marriage; whether he can demand one who has a husband. 25. But these questions are not set forth in our speech in the same order in which they occur to us [in our speech questions are not disposed in the same order in which they occur to us in meditation and investigation; for what occurs to us first is generally the hypothesis, or ground of controversy; but it is to be placed last, other questions being put before it]; for that in general occurs first which is to be expressed last, as thus, You ought not to make choice of another man's wife. Hence haste spoils division.

We should not, therefore, content ourselves with what offers; but should inquire something further, as, whether he may not even make choice of a widow; something further still, as, whether he may not choose anything belonging to a private person; or last of all, going back to what is next to the general question, whether he may not make choice of anything unlawful. 26. Examining the proposition of our adversary, therefore, as is very easy, let us decide, if possible, what it is natural should be answered first; and this, if we but contemplate the cause as being actually pleaded, and the necessity laid upon us of replying at once, will readily occur to us.

27. But if it should not occur, let us set aside that which occurs to us first, and reason with ourselves thus: What, if it were otherwise? questioning ourselves a second and a third time, until nothing remain for consideration. Thus we shall examine even the minutest points, which, if well treated, will make the judge more inclined to listen to us on the main point. 28. With this process the rule that "we should descend from what is common to what is particular," is not much at variance; for what is common is mostly general. Some person has killed a tyrant, is a common or general proposition; a certain person has killed a tyrant, a woman has killed him, his wife has killed him [the allusion is to Alexander of Pherae, who was killed by his wife Thebe. See Cicero de Inv. ii. 49], are particular propositions.

29. I used also to select those points in which I agreed with my opponent, provided they were to my purpose, and not only to press such matters as he admitted, but to multiply them by division; as in this case [see Seneca, Controv. xxii.]: "A general, who, in a competition for public honors, had come off superior to his father, was taken prisoner by the enemy; certain deputies, going to ransom him, met the father on the road as he was returning from the enemy's camp, who said to them, 'You are going too late.' 30. The deputies searched the father, and found a sum of money in gold concealed in the breast of his robe; they then proceeded to their place of destination, and found the general fixed to a cross, who uttered the words, 'Beware of the traitor.' The father was accused." What is admitted on both sides? That treason was signified, and signified by the general. We try to find the traitor. You admit that you went to the enemy, and went secretly; that you returned in safety, brought away gold, and had the gold concealed.

31. What the accused has done, is sometimes set forth very forcibly in the statement of the case, and, if it takes possession of the mind of the judge, his ears are almost closed against the defense. In general, it is to the advantage of the accuser to amass facts, and of the defendant to separate them. I used also to do, with regard to the whole subject of a cause, that which I noticed as being done in regard to arguments; that is to say, stating all the particulars that could possibly be urged against me, and overthrowing them one  after another, I left nothing remaining but that which I wished to be believed.

32. Thus, in charges of prevarication [those advocates are said to prevaricate who set aside true charges and proofs, and put forward such as are groundless, acting in collusion with the accused person. We call those prevaricators, says Ulpian, who betray the cause, which they profess to support, to their adversaries], it may be argued, The accused could have been acquitted only by the establishment of his innocence, or by the intervention of some authority, or by force or bribes having been offered to the judges, or through the difficulty of proof, or through prevarication: That he was guilty you admit; no authority interposed; there was no force offered; you do not complain that the judges were bribed; there was no difficulty in the way of proof; and what remains, then, but that there must have been prevarication? 33. If I could not set aside all the points against me, I at least set aside the greater number. For instance, It is acknowledged that a man was killed; not in a solitary place, to lead me to suspect that he was killed by robbers; not for the sake of booty, for he was not rifled; not in the hope of inheriting anything, for he was poor; Malice must then have been the cause: But who was his enemy?

34. This method, of examining everything that can be said, and of rejecting as it were one particular after another, in order to arrive at the strongest point, not only facilitates the art of division, but also that of invention. Thus, Milo is accused of killing Clodius: He either killed him or did not kill him: It would be safest to deny that he killed him, but if that cannot be done, it must be allowed that he killed him either justly or unjustly; and we must doubtless say justly: He killed him then either intentionally or through necessity; for ignorance cannot be pretended: 35. Whether there was intention, is doubtful, but, as people think that there was, we must attempt some defense of it, and say that the intention was to serve his country. Or shall we say that he killed him through necessity? The encounter with him was then accidental, and not premeditated; one of them therefore was lying in wait: Which of the two? Assuredly Clodius. Do you see how the necessary chain of circumstances leads us to the ground of defense?

36. Let us consider further: He certainly either wished to kill the lier-in-wait Clodius, or he did not; it is safer if we can say that he did not: Then the attendants of Milo must have done the deed, without orders from Milo, and without his knowledge. But this timid mode of defense detracts from the credit of our assertion, that Clodius was justly killed. 37. We must therefore add, The attendants acted in such a way as each of us would wish his own attendants to act. This kind of practice is the more useful, as it often happens that nothing that presents itself pleases us, and yet something must be said. We should accordingly contemplate the cause under every aspect; and thus either that which is best will be discovered, or that which is least bad. Occasionally we may turn to advantage the statement of our adversary; for that it is sometimes equally to the purpose of both parties, has been observed in the proper place.

I know that it is discussed in some authors, in many thousands of lines, how we may discover which party ought to speak first; but this is decided in the forum either by the rigor of formula, or by the nature of the process, or, finally, by lot. 38. In the schools such inquiries are of no importance, since it is allowable to make a charge and to refute it, in the same declamation, as well on the side of the prosecutor as on that of the defendant. [Possessor is here used, in a strange and surprising sense, for defensor.] But in most suits it cannot even be determined which party has a right to precedence; as in the case, A father, who had three sons, one an orator, another a philosopher, and a third a physician, divided his property by his will into four parts, and gave one part to each of the three, directing that the fourth part should go to him who should be of most service to his country. 39. They go to law; who ought to speak first, is uncertain; though the statement of the case is clear; for we must begin with him whose part we take. Such are the directions that may be given about division in general.

40. But how shall we find out questions that are more obscure? Just as we discover thoughts, words, figures, style; namely, by the exercise of our ability, and by care and practice. Scarcely anything, however, will escape a speaker, unless he be inattentive, if he will, as I remarked, but take nature for his guide. 41. But many orators, affecting a character for eloquence, are content with arguments that are merely showy, or that contribute nothing to the establishment of their case.

Others think that they let nothing escape them, while they merely contemplate what presents itself to their own eyes. That what I say may be the better understood, I will give a case from the schools, one not very difficult or new, as an example: 42. Let the son who neglects to plead for his father on a trial for treason, be disinherited. Let the man who is found guilty of treason, be banished with the advocate who pleads for him. A father was accused of treason; one of his sons, who was a man of eloquence, appeared as advocate for his father; the other, an illiterate man, did not appear at all; the father was found guilty, and went with the son who pleaded for him into exile. The illiterate son, after distinguishing himself by his bravery, obtained of his country, as a reward, the recall of his father and his brother. The father returned and died intestate; the illiterate son sues for a portion of his property; the eloquent son claims the whole of it.

43. In this case those men of eloquence, to whom we appear ridiculous, as being anxious about causes that rarely occur [such as the imaginary cases of the schools], will seize upon the favorable characters. Their pleading will be for the illiterate against the eloquent son; for the brave against the unwarlike; for the benefactor against the ungrateful; for him who desires only a part of his father's property, against him who would allow no portion of it to his brother. 44. All these are points in the cause, and a great support to it, but they do not secure victory. In such a cause, the thoughts sought by such orators will be, if possible, daring or obscure, (for obscurity is now a virtue,) and they will think that they come off well in the matter if they distinguish themselves with sufficient clamor and noise.

Those, again, whose object is better, but whose regard is confined to that which readily presents itself, will see the following points as it were swimming on the surface: 45. That the illiterate son was excusable for not appearing at the trial, as he could have been of no assistance to his father; that the eloquent son has little ground for blaming the other for his absence, as the father was found guilty; that he who procured his father's recall deserves to inherit his father's property; and that the other son is of a covetous, unnatural, and ungrateful disposition, as he refuses to share the inheritance with a brother to whom he owes so much; they will see also that a question may be raised as to the letter and intent of the law, and that, unless this question be settled, there can be no room for anything else to follow.

46. But he who shall follow nature, will doubtless reflect thus: that the illiterate son will say, in the first place, My father, dying intestate, left two sons, my brother and myself; and I claim part of his property by the common law of nations. Who indeed is so thoroughly foolish and ignorant, that he would not commence thus, even though he knows not what a proposition is? 47. This common law of nations the pleader will moderately commend, as being extremely just. It then follows, that we consider what can be replied to so equitable a claim. A reply presents itself at once: The law directs that a son who does not defend his father when accused of treason is to be disinherited; and you did not defend your father. On this proposition will naturally follow some praise of the law, and some censure of the son for not defending his parent.

48. Hitherto we have had to do only with what is admitted. Let us again turn our attention to the claimant: will he not, unless he be utterly senseless, plead thus? If the law stands in the way, there is no ground for an action; the trial is a mere form. But that there is a law in the way, and that it punishes that of which the illiterate son was guilty, is undoubted. What then shall we say on his behalf? I was illiterate. 49. But the law was in force; it comprehends all men; it will be of no use to allege want of education. Let us inquire, then, whether the law can be invalidated in any point. What does nature suggest, (for to nature I must frequently appeal,) but that when the letter of a law is against us, we must look to the intention of it?

The general question then arises, Whether we ought to rest on the letter, or on the intention, of any law? But concerning law in general we may dispute for ever; nor has this point ever been fully decided. We must inquire, therefore, whether in this particular law, about which we are concerned, anything can be found that is at variance with the letter of it. 50. The law says, then, Whatever son has not defended his father, shall be disinherited. What? Whatever son, without exception? Considerations such as these will then present themselves of their own accord: Suppose that a son who was but an infant, or one who was sick, or one who was out of the country, or in the army, or on an embassy, did not defend his father, would he be disinherited? Something considerable has now been gained; a son may not have defended his father, and yet not be disinherited.

51. Let him, however, who has so far meditated on the case, "pass over," as Cicero says, "after the manner of a Latin flute-player," [Cic. Pro Muraen, c. 12. The flute-player on the stage turned from one actor to another, as each proceeded to speak. That the flute-players at Rome were Latins appears from Livy, ix. 30] to the side of the eloquent son. He will say, Though I allow the reasonableness of such exceptions, you were not an infant, or out of the country, or serving in the army. Will anything else occur to the other son, but to say, I am illiterate. 52. But the eloquent son will make the obvious reply, Though you could not plead for your father, you might have appeared at his side; and the remark is just.

The illiterate son must consequently recur to the intention of the lawgiver: He intended, he will say, to punish unnatural conduct, but I have not behaved unnaturally. 53. In reply to which the eloquent son will say, You did act unnaturally, as you incurred the penalty of being disinherited, though penitence or desire of distinction has since gained you the privilege of this kind of option. [So that, when you might have asked a reward for yourself, you asked the recall of your father and brother.] Besides, it was through you that your father was found guilty; for you seemed to have already decided on his case. To this the illiterate son will reply, you rather were the cause that he was found guilty; for you had offended many people, and excited enmity against our family; conjectural allegations, as will also be that which the illiterate son may say in the way of excuse for his absence, that it was the object of his father not to expose his whole family to danger. Such are the considerations that come under the first question as to the letter and intent of the law.

54. Let us direct our attention further, and examine whether anything more may be found out, and, if so, how it may be discovered. I purposely imitate the manner of one inquiring, that I may teach the student how to inquire, and laying aside all regard to ornaments of style, lower myself to promote the advantage of my pupils. Hitherto we have drawn all our questions from the person of the claimant; why should we not ask some questions regarding the father? The words of the law are, Whatever son has not defended his father, shall be disinherited. 55. Why may we not ask this question, Whether such is the case, whatever be the character of the father whom he has not defended? We ask such a question frequently in those cases in which sons are prosecuted, as liable to the penalty of imprisonment, who do not support their parents; as the son that did not support his mother who had given evidence against him when he was accused of not being a Roman citizen; and he that did not maintain his father who had sold him to a dealer in slaves. But with regard to the father of whom we are speaking, of what argument can we lay hold? He was found guilty.

56. Does the law then relate only to fathers who are acquitted? A hard question at first sight. But let us not despair. It is probable that the intention of the legislator was, that the aid of children should not be wanting to innocent fathers. But the illiterate son would be ashamed to allege this intention, because he acknowledges that his father was innocent. [Were he to make such allegation, he would represent himself as culpable in not appearing in support of his father.] 57. The law, however, Let him who is found guilty go into exile with his advocate, furnishes another argument in the cause. It seems scarcely possible that a penalty should have been directed against a son, in reference to the same father, whether the son appeared in his defense, or did not appear. Besides, no law has any relation to exiles. [It cannot therefore be supposed that the law (about disinheriting) was intended to affect the advocate of the condemned person, who was necessarily an exile.] It is not, therefore, probable that this law was intended to refer to the advocate of the person condemned; for can any property be possessed by an exile?

The illiterate son, whether he looks to the letter or to the intention of the law, makes it doubtful whether he was called upon to defend his father. 58. The eloquent son will both cling to the words of the law, in which no exception is expressed, and will say that it was from this very consideration that the penalty of being disinherited was denounced against sons who should not defend their fathers, lest they should be deterred from defending them by fear of banishment, and that the illiterate son did not appear on behalf of an innocent father. It is well deserving of remark that from one state may spring two general questions: Whether every son is obliged to defend his father? and Whether every father has a right to expect defense from his son?

59. All our questions hitherto have arisen from two persons [the father and the illiterate son]; as for the third, which is that of the adversary, no question can arise about him, because there is no controversy about his share of the property. Let our investigations, however, be still pursued; for all that has been said might have been said even though the father had not been recalled from exile. Nor let us fix immediately on the reflection which readily presents itself, that his recall was procured by the illiterate son. He that shall sagaciously consider that point, will find his view directed to something further; for as species follow genus, so genus precedes species. [For though a man is an animal, it does not follow that an animal is a man.]

60. Let us suppose, therefore, that his recall had been procured by another; a question of ratiocination and syllogism will arise, Whether the recall is equivalent to a repeal of the sentence, and places the father in the same position as if judgment had not been pronounced against him? Here the illiterate son will proceed to say that he could not have obtained the restitution of his property, being entitled to one reward only [by setting aside the sentence, not by obtaining pardon; for pardon, if granted, must have been extended to the father and his advocate, and would thus have been accounted as two rewards], by any other means than by procuring the recall of his father on the same understanding as if he had never been accused; an understanding which also annulled the penalty of the advocate, as completely as if he had not appeared on behalf of his father.

61. We then come to that which presented itself to us at first, that the father's recall was procured by the illiterate son. Here we again proceed to reason whether he who procured the recall ought not to be regarded in the light of an advocate, as he effected that which the advocate sought to effect; and that it is not unfair that that should be received as equivalent which is more than equivalent. 62. What remains is a question of equity: which of the two makes the more rightful claim. This question, too, admits of division: Even if each claimed the whole property,— and, surely, when the one claims but half, and the other the whole, to the exclusion of his brother. [Even if each claimed for himself the whole inheritance, the claim of the illiterate eon would seem the more just; and certainly, when the illiterate son claims but half, and the eloquent son the whole, the claim of the illiterate son must appear as just again as that of the other.] But, even when these points are discussed, the memory of the father will have great influence with the judges, especially when the question is about the disposal of his property.

It will, therefore, be a subject for conjecture, what intention the father had in leaving no will at his death. But this relates to quality, which is matter for another state [that of quality]. 63. It is, however, at the conclusion of causes that questions of equity are generally considered, because there is nothing to which judges listen with greater [for then they seem to be more at liberty in forming their decision, being less restricted by the rigor of the law] readiness. Yet expediency will occasionally cause a change in the order; for instance, if we have but little confidence that the law will be in our favor, we may work on the minds of the judges at the commencement by considerations of equity.

On this head I have no further directions to give in general. 64. But let us now proceed to consider the several parts of judicial causes; and though I cannot pursue them to the last species, that is, to every particular form of question and process, I may yet treat of them in a general way, so as to show under which state each kind of cause commonly falls. And as it is naturally the first question in a case whether what is alleged occurred, it is with this that I shall begin.


1. All conjecture has reference either to fact or intent. To each belong three parts of time, the past, the present, and the future. Concerning fact there are both general and particular questions; that is, such as are not limited to the consideration of certain circumstances, and such as are so limited. 2. About intent there can be no question, unless where there is a person concerned, and a fact is admitted. When the question, then, is about a fact, it is to be considered either what has been done, or what is being done, or what is going to be done. Thus, in general questions, we inquire whether the world was formed by a fortuitous concourse of atoms; whether it is ruled by a providence; whether it will one day fall to pieces; in particular questions, whether Roscius has committed parricide [Cicero pro Rosc. Amerino]; whether Manlius is aspiring to sovereignty; whether Caecilius will justly prosecute Verres [Cicero Div. in Q. Caecilium].

3. In judicial pleadings past time is most concerned; for no man accuses another but for something that has been done; while what is actually taking place, or is likely to take place, is inferred from the past. It is a subject for inquiry, also, whence a thing has proceeded, as, concerning a pestilence, whether it arose from the anger of the gods, from the bad state of the atmosphere, or from the corruption of the waters, or from noxious exhalations from the ground. Concerning a fact, too, what was the cause of it; as, why did fifty princes sail to Troy, whether from being bound by an oath, or from being led by example, or from a desire to oblige the sons of Atreus? These two kinds of questions [whence a thing arose; and what was the cause of an act] are not very different.

4. As to matters that concern the present time, if they are not to be discovered by proofs, from circumstances which must have preceded, but by the senses, they have nothing to do with conjecture; for example, if it should be asked at Lacedaemon whether walls are in the course of erection at Athens. But the state of conjecture, which may seem foreign to this head, has also a place under it [though some might be disposed to refer such questions rather to the state of quality], as when it is inquired respecting any individual, who he is; as it was a question, in the action against the heirs of Urbinia, whether he who laid claim to the property as a son was Figulus or Sosipater: 5. The person of the man was under the eye of the court, so that it could not be inquired whether he was, (as we inquire whether anything is beyond the ocean,) nor what he was, nor of what nature, but who he was.

This kind of question, however, depends for decision on the past, as whether this Clusinius Figulus was born of Urbinia. Such causes have been tried in my time, and some of them have come under my advocacy. 6. Conjecture with regard to intent has reference doubtless to all the parts of time, as with what intent was Ligarius in Africa? With what intent does Pyrrhus solicit peace? How will Caesar feel, if Ptolemy kills Pompey?

Questions of conjecture and quality are made with regard to magnitude, species, and number, as, whether the sun is greater than the earth; whether the moon is spherical, plane, or conical; whether there is only one world or several. 7. Nor are such questions confined to physical subjects; for we inquire, whether the Trojan or Peloponnesian war was the greater; what sort of shield was that of Achilles; whether there was but one Hercules.

But in judicial causes, which consist of accusation and defense, there is one kind of question, that of conjecture, in which the inquiry is about an act, and the author of it. This sometimes embraces the two questions in one, and both are alike denied; sometimes considers them separately, as when it is first inquired, whether the act was committed or not, and, if it was, by whom it was committed. 8. The consideration of the act itself, also, sometimes embraces a single question, as whether a man died, sometimes two, as whether he died of poison or disease of the stomach.

There is another kind of conjectural question which regards the act only, when, if the act be admitted, there can be no doubt as to the author of it; and a third, which has reference only to the author when the act is acknowledged, but it is disputed by whom it was committed. 9. But that which I have specified in the third place, is not always confined to one question, for the accused person may either simply deny that he himself committed the act, or may assert that another committed, it; nor is there one mode only of throwing the charge upon another person, for sometimes there arises mutual accusation, which the Greeks call αντικατηγορια, and some of our writers accusatio concertativa; sometimes the guilt is thrown upon some person not implicated in the cause, which person is sometimes known and sometimes unknown; and when it is thrown upon one that is known, it may be imputed to one out of the question, or to the deceased, as having killed himself intentionally.

10. In these cases there is a comparison of persons, motives, and other things, similar to that which there is in αντικατηγορια: as Cicero, for example, in pleading for Varenus, throws a suspicion of guilt on the slaves of Ancharius; and, in speaking for Scaurus, with reference to the death of Bostar, turns the imputation of it on Bostar's mother. 11. There is also a kind of comparison contrary to this, in which each party claims the credit of some act, and another in which persons are not opposed, but only facts, that is, when it is inquired, not which of two persons did a thing, but which of two things was done. When the question is settled about the act and the agent, we may then inquire about the intention.

I now proceed to speak of particulars. When a charge is denied, both as to the act and the agent, it is denied in this way: I have not committed adultery; I have not aspired to regal power. On trials for murder and poisoning, such a distinction as the following is very common: The deed has not been committed, or, if it has been, l am not guilty of it. 12. But when the accused says, Prove that the man was murdered, the weight of the argument falls wholly on the accuser, for nothing else will be said against him on the part of the accused, except perhaps some suspicions, which he ought to throw out as vaguely as possible; because, if we fairly assert a point, we must make it good, or be in danger of being found guilty; for as, while the question lies between what is advanced by our opponent and what is advanced by ourselves, the statement of either party may be presumed to be true, so, when the point on which we take our stand is overthrown, we may be hard pressed on all the remaining points.

13. But when a cause turns on the ambiguous symptoms of disease of the stomach or poisoning, there is no third point, and, therefore, each side must hold to what it has alleged. Sometimes the question is about the nature of the thing itself, whether it was poisoning or disease of the stomach, when arguments are drawn from circumstances, independently of any consideration of the person. 14. For it is of importance to inquire, whether a banquet preceded the death, or any serious transaction; whether toil or ease, wakefulness or sleep.

The age of the deceased, too, may have some influence on the decision; and it is of consequence to know whether he died suddenly, or was wasted with long illness. If it be sudden death only that calls for consideration, there will be a still wider field for discussion for both parties. 15. Sometimes proof respecting an act is sought from the character of the accused party; as, it is credible that poisoning was the cause of death, because it is credible that poisoning was committed by the accused; or, it is incredible that the accused was guilty of poisoning, therefore it is incredible that poisoning was the cause of death.

But when there is a question at the same time regarding the person accused, and the deed of which he is accused, the natural order of things is for the accuser to prove first of all that the deed was committed, and then that it was committed by the accused. If, however, he find more proofs bearing on the person, he may change that order. 16. The accused, on the other hand, will make it his first object to deny that the deed was committed; because, if he succeeds in establishing that point, he has no need to say anything further; while, if he is defeated on it, there may remain some other means for him to establish his innocence.

In cases, also, where there is a dispute about fact only, and where, if the fact is proved, there can be no doubt as to the agent, arguments are in like manner drawn from persons and from circumstances, though with regard to the question of fact simply. 17. This is the case (for I must adduce such examples as are most familiar to learners) in the following subject of controversy: A son, who had been disinherited by his father, devoted himself to the study of medicine. His father falling sick, and every other physician despairing of saving his life, the son, being consulted, said that he would cure him, if he would take a draught which he would give him. The father, after drinking part of the draught that he had received, said that poison had been given him; the son drank what was left; the father died; the son was accused of parricide. 18. Here it is known who gave the draught; and, if it was poison, there can be no doubt as to the author of the poisoning; but whether it was poison must be decided by arguments arising from the character of the accused.

There remains a third kind of conjectural causes, in which it is admitted that a deed has been done, but there is a question about the author of it. Of such cases it is superfluous to give an example, since abundance of trials on such points occur; as when it is acknowledged that a man has been killed, or that sacrilege has been committed, but the person who is accused of the deed denies that he is guilty of it.

Hence arises αντικατηγορια, or recrimination; it being admitted that a deed has been done, while each party charges the other with the commission of it. 19. As to this kind of cause, Celsus tells us that it cannot occur in the forum; a fact of which I suppose that nobody is ignorant. The judges are assembled to decide the case of one accused person; and if the accused and the accuser bring charges against each other, the judges must choose which of the two cases they will try. [The judges must give the preference to one of those, who come forward with reciprocal accusations, over the other, and direct him alone to assume the character of accuser; and thus all anticategoria is excluded from the forum.]

20. Apollodorus also says that αντικατηγορια includes two causes; and doubtless, according to the practice of the forum, there are two distinct cases. Yet this kind of conjectural cause may come under the cognizance of the senate or the emperor. But even on ordinary trials it requires no difference in the pleadings; for the decision that is given affects both parties, though sentence is pronounced only on one. [When one gains his cause, the other loses; when one is condemned, the other is acquitted.] 21. In this kind of cause defense must always have the precedence; first, because to protect ourselves is of more importance to us than to injure our adversary; secondly, because we shall have greater weight in accusing, if our own innocence be first established; and, lastly, because it is only by this order of things that the cause can become double; for he who says, I did not kill, leaves it free for himself to add you killed, but he who first says, you killed, renders it superfluous to say afterwards, I did not kill.

22. Such causes, moreover, depend on comparison, which is managed in more than one way; for we either set the whole of our cause against the whole cause of our adversary, or particular arguments on our side against particular arguments on his. Which of these two modes ought to be adopted in any case, can only be decided by considering which is the more likely to be of service to it. Thus Cicero, in pleading for Varenus, compares, in regard to the first head of accusation, argument with argument; for he has the advantage, as the person of a stranger is but rashly compared with that of a mother. It is best, therefore, that particular arguments should, if possible, be overthrown by particular arguments; but if we find a difficulty as to certain parts, we must fight with the whole force of our cause in a body.

23. But whether the parties accuse one another; whether the accused turns the guilt on the accuser without any formal accusation; (as Roscius throws it on his accusers, though he does not bring them before the judges;) or whether a deed be attributed to persons whom we assert to have perished by their own hand, the arguments of the two parties are matched in the same way as in causes which involve recrimination. 24. That species, however, of which I spoke last, is often handled not only in the schools, but also in the forum; for, in the case of Naevius of Arpinum [an occurrence of a similar nature is mentioned by Tacitus, Ann. iv. 21], the question was merely whether his wife had been thrown down by him, or had thrown herself down of her own accord. My pleading in that cause is the only one that I have hitherto published; and I acknowledge that I was induced to publish it by a youthful desire for fame. As for the other pleadings, which are circulated under my name, they are so corrupted by the carelessness of the short-hand writers who took them down to make profit of them, that they contain very little genuine matter of mine.

25. There is also another kind of conjectural cause, involving two questions, differing from αντικατηγορια, and relating to rewards; as in the following case: A tyrant, suspecting that poison had been given him by his physician, put him to the torture. As he persisted in denying that he had given poison, the tyrant sent for another physician, who said that poison had been given him, but that he would administer an antidote; he then gave the tyrant a draught, and the tyrant, immediately after drinking it, died. The two physicians dispute about the reward for tyrannicide; and as, in a case of αντικατηγορια, where each party endeavors to throw the blame on the opposite, so in this case, where each party makes a claim, persons, motives, means, opportunities, instruments, and evidence are brought into comparison.

26. Another kind of case also, though there is no recrimination in it, is treated in the same manner as one of recrimination; I mean that in which it is inquired, without accusing any one, which of two things has taken place; for each side makes its own statement, and supports it; as, in the suit concerning the property of Urbinia, the claimant says that Clusinius Figulus, the son of Urbinia, finding the army, in which he was serving, defeated, fled, and after being thrown into various adventures, and even kept prisoner by a king, made his way at length into Italy, and arrived at his native place Margini, where he was recognized: Pollio, on the other hand, asserts that he was a slave to two masters at Pisaurum; that he practiced medicine; and that, being set free, he joined himself to another person's company of slaves, and requesting permission to serve with them, was purchased. 27. Does not the whole action consist of a comparison of two allegations, and two distinct questions for conjecture? But the mode of proceeding for those who claim property or resist claims to it, is the same as that for persons prosecuting and defending in civil suits.

Grounds for conjecture are drawn in the first place from the past, in which are comprehended persons, motives, intentions. For the order in which we have to consider evidence as to any act, is, whether the party charged with the commission of it had the will to do it, had the power to do it, and whether he actually did it.

28. Hence we must consider, first of all, what sort of character he is against whom a charge is brought; and it is the business of the accuser to make whatever he imputes to the accused appear not only disgraceful, but as consistent as possible with the crime for which he is brought to trial. For instance, if he reproaches a man accused of murder with being incontinent, or adulterous, such dishonorable imputations will indeed hurt him, but will be of less avail to support the charge than if he prove him to be daring, headstrong, cruel, or rash.

29. The advocate of the accused, on the other hand, must make it his object, if possible, to refute, justify, or extenuate such allegations; or, if he find it impracticable to do so, the next thing is to separate them from the question before the court; and many imputations of that nature are not only irreconcilable with the charge, but tend to overthrow it; for example, if a man who is accused of theft be represented as prodigal or careless of his property; for disregard of money, and covetousness, do not seem likely to meet in the same character. 30. If such means of defense fail, he must have recourse to the remark, that the question has no reference to the imputation; that he who has committed one offence has surely not been guilty of all kinds of offences; that the accusers had the audacity to make such false charges only because they hoped that the accused, being injured and wounded by them, would be overwhelmed by a weight of slander.

31. Other allegations may be made by the accusers, against which common-place arguments rise in opposition. In such a case, the advocate of the accused may commence with arguments drawn from his character; and this sometimes generally, as, It is incredible that a father should have been killed by his son; or that a general should have betrayed his country to the enemy. To such arguments it is easily answered, either, that every sort of crime may be committed by the bad, and is, indeed, daily detected among them, or, that it is monstrous that charges should be denied on the ground of their atrocity. 32. Sometimes particularly; a mode which may have various results; as dignity, for example, sometimes supports an accused person, and at other times is turned into a proof of his guilt, on the representation that the hope of impunity was conceived from it; and in like manner poverty, humility, wealth, are set in different lights according to the ability of each party.

33. Good morals, however, and integrity in the past time of life, must always be of great influence in favor of an accused party. If no attack is made on his character, his advocate will dwell strongly on that circumstance; while the accuser will try to confine the attention of the court to the question before it, on which alone judgment is to be pronounced, and will observe that every offender must have committed a first offense, and that the commencement of guilt is not to be celebrated by a feast of glorification. [Encoenia was a feast at the dedication of a temple, or at the opening of any new building.] 34. Such will be the observations which the accuser will make in reply; but in the early part of his pleading he will impress the mind of the judge in such a way as to be thought rather to have been unwilling to throw out imputations than to have been unable.

Hence it is better for the accuser to abstain from casting any reflection on the past life of the accused, than to attack him with light or frivolous charges, or such as are manifestly false, because the credit of his other statements would thus be diminished; and he who throws out no imputations may be thought to have abstained from them as being superfluous, while he who throws out groundless imputations shows that his only chance of success lay in attacking the past life of the accused, a point on which he chose rather to be defeated than to be silent. 35. Other considerations, derived from the character of individuals, I have fully noticed where I have treated of the sources of arguments.

The next sort of proof is derived from motives,'^ in which ' are chiefly to be regarded anger, hatred, fear, avarice, hope; for all others fall under some variety of these. If any of them be attributable to the accused, it is the part of the accuser to make it appear that motives may stimulate a person to any act whatever, and to exaggerate the force of those motives on which he lays hold for the support of his arguments. 36. If none of them are attributable to him, he may shape his speech in such a way as to insinuate that there may have been latent motives, or may observe that it is to no purpose to consider from what motive the accused committed the crime, if it is apparent that he did commit it; or he may say that the crime is the more detestable from there having been no motive for it. The advocate of the accused, on the other hand, will insist, as often as possible, on this point, that it is incredible that any / crime can have been committed without a motive. On this consideration Cicero dwells with great force in many of his speeches, and especially in that for Varenus, who had everything else against him, and was in consequence condemned.

37. But if a motive is alleged by the accuser for the crime, the advocate of the accused may say that the motive is false, or frivolous, or was unknown to the accused. Motives may sometimes be imputed to the accused to which he must be a stranger; for instance, it could not be known, he may say, whether the deceased intended to make him his heir by whom he is said to have been killed, or designed to prosecute him. [The Romans often made several wills, and we may on that ground say that it was unknown to the accused whether he was heir to the deceased; for though he might have been aware that he was named as his heir in one will, he could not have been certain that he was named as heir in his last will.] If other grounds of defense fail, we may say that motives are not necessarily to be regarded, for what person can be found that does not fear, hate, and hope, but that most entertain those feelings without violating the moral duties? 38. Nor must the advocate omit to observe that all kinds of motives do not prevail with all kinds of persons; for though poverty may have incited some persons to steal, it could have had no influence with a Curius or a Fabricius.

39. Whether we should speak of the motive or of the person first, is a question; and different courses have been adopted by different orators; with Cicero motives generally take the precedence. But to me, unless the nature of a cause gives a preponderance to either, it seems more natural to commence with the person; since for the accuser, for instance, to say either the charge is credible of no one, or it is credible of the accused, is a more general proposition, and a more just division. 40. Yet regard to convenience may change that order, as it changes many other things.

Nor are motives for the willful commission of an act only to be sought, but motives that may have misled to the commission of it, as drunkenness, or ignorance; for as these lessen the culpability when the quality of an act is considered, so they tend greatly to establish a question regarding fact. [In speaking of the quality of an act, we may often make a concession, and say that it was done through ignorance, imprudence, intoxication; and such considerations may tend to make it seem pardonable; but in a question of fact considerations of that nature tend rather to strengthen the evidence of it.] 41. However, I know not whether a person can ever be the subject of a charge, (I mean in a real cause,) without one or other party speaking of him; but concerning motives it is often superfluous to inquire, as in cases of adultery and theft, because the crimes themselves carry their motives on the face of them.

42. In the next place, it seems necessary to look to views, which open a wide field for consideration: as, whether it be probable that the accused hoped that such a crime could be executed by him; that when he had committed it, it would not be known; or that, if it were known, it would be forgiven, or visited with a light or tardy punishment, or one from which he would feel a less portion of inconvenience than he would experience of gratification from the commission of the deed; or whether he thought it worth so much to undergo the penalty.

43. Afterwards it may be considered whether he might have done the deed at another time, or in another way, or with greater facility or security; a method adopted by Cicero in defense of Milo, when he specifies the number of occasion on which Clodius might have been killed by Milo with impunity. Besides, we may ask why the accused should have preferred to make an attack in that place, or at that time, or in that manner, (arguments which are also most ably enforced in the same pleading,) 44. or whether, if he was led by no design, he was hurried away by impulse, and without reason, (for it is a common saying, that crimes have no reason, [scelera non habere consilium]) or whether he was led away by a habit of vice.

The first point, whether he had the will, being discussed, the next consideration is, whether he had the power. Under this head are contemplated place and time; as, with respect to a theft, whether it was committed in a solitary or frequented place; in the daytime, when there might have been many witnesses, or in the night, when the difficulty of proof is greater. 45. All obstacles and opportunities, indeed, will be taken into consideration; they are numerous and well known, and require no examples. This second head is of such a nature, that, if the crime could not have been committed, the trial comes to nothing; if it could, the question follows, Did the accused commit it? But these considerations respect also conjecture as to intention, for it is inferred from these whether he hoped to effect his purpose. In consequence means ought also to be regarded, as the suites of Clodius and Milo.

46. The question, whether the accused committed the crime, commences with the second division of time, that is, the present, and that which is closely connected with it, to which belong noise, cries, groans, or anything similar; to subsequent time belong concealment, terror, and such circumstances. To these are to be added all kinds of signs or indications, of which I have already treated; as well as words and acts, both such as preceded and such as followed. 47. These words and acts are either our own or those of others. But some words hurt us less than others; our own words hurt us more and profit us less than those of others; those of others profit us more and hurt us less than our own. As for acts, sometimes our own profit us more, and sometimes those of others, as when our adversary has done anything that appears in our favor; but our own always hurt us more than those of others.

48. There is also this difference to be observed in words, that they are either plain or equivocal; but whether they are our own or those of others, those which are equivocal must necessarily be less effective either to benefit or to injure. Our own, however, are often injurious to us, as in the well-known case, A son being asked where his father was, replied, wherever he is, he is alive [ubicunque est, vivit. What is equivocal here? The word vivit, which might be taken for bibit. That there was a continual confusion of the letters b and v is known to every one who is conversant with the winters of antiquity]; but he was found dead in a well.

49. The words of another which are equivocal, can never hurt us, unless when the author of them is uncertain or dead; as in the cases, A voice was heard in the night, Beware of tyrannical power: and, A dying man being asked from whom he received the poison of which he was dying, replied, It is not expedient for you to know; for if there be any one that can be questioned as to the meaning, he will put an end to the ambiguity. 50. But while our own words and acts can be justified only by reference to the intention, those of others may be refuted in various ways.

In what I have said, I have spoken, I think, chiefly with reference to one kind of conjectural causes; but something of these is applicable to all kinds of causes; for in questions respecting theft, deposits, and loans of money, arguments are derived both from possibilities, as whether there was any money that could have been deposited, and from persons, as whether it was credible that such a person deposited money with such another person, or whether it was credible that he lent money to such a person; whether it is probable that the prosecutor is a slanderer, or that the defendant is an impostor or a thief.

51. But even in the case of a person accused of theft, as in cases of murder, there is an inquiry about the deed and the author of it. In regard to cases of loan and deposit there are two questions, but always separate [in regard to a deposit it may sometimes be inquired whether the accuser really deposited money, and whether it was ever returned; but these questions will hardly be asked at the same time], whether the money was given, and, whether it was returned. Cases of adultery have this peculiarity, that two parties are generally imperilled in them, and that something must be said of the past life of both; a question, however, may arise, in some cases, whether both ought not to be defended together; but the decision of this point must depend on the nature of the case; for if the defense of one party will support the other, I should take them together, if it is likely to be injurious to it, I would separate them.

52. But lest any one may think me inconsiderate in saying that adultery is generally a charge against two persons, but not always, I would add that a woman alone may be accused of adultery with an unknown person: Presents, it may be said, have been found in her house, and money, of which the giver has not been discovered; and love letters, of which it is doubtful to whom they were written. 53. In regard to forged writing the case is similar; for either several persons may be charged with the crime, or one only. The writer of an instrument however will always find it necessary to guarantee the signature of the person who has signed it; but the person who has signed it cannot always guarantee the handwriting of him who is said to have written it; for he may be deceived. But he who is said to have engaged their services, and for whom the instrument is alleged to have been written, will have to support both the writer and all who signed the writing. The sources of proof are similar in cases of treason and of aspiring to sovereignty.

54. But the custom in the schools, of considering everything in our favor that is not in the argument laid down for us, may be prejudicial to young men proceeding to the forum. You accuse me of adultery: who is witness? who testifies to the fact? of treason: what reward have I received? who was privy to the transaction? of administering poison: where did I buy it? from whom? when? for how much? through whose hands did I convey it? Or we plead in defense of one accused of aspiring to tyranny, where were his arms? what guards had he assembled? 55. I do not deny that such questions may be asked, or that we may urge them on behalf of the party whom we defend; for I myself would call for such proofs in the forum, if I should find my adversary not in a condition to give them.

But in the forum we miss the facility for asking such questions that there is in the schools, where scarcely a single cause is pleaded in which some argument of this kind, or perhaps several, are not advanced. 56. Similar is the ease with which some declaimers, in their perorations, assign parents, children, or nurses, to whomsoever they please. Yet we may more reasonably allow a speaker to call for proofs that are not offered than to discuss them as if they were offered.

How we must examine as to intention, was sufficiently signified when we distinguished the three points of inquiry, whether a person had the will, whether he had the power, and whether he did the deed; for in the same manner as it is inquired whether a person had the will, so it is inquired with what intent he acted, that is to say whether he intended to do an evil act. 57. The order in which circumstances are stated, also, either adds to the credit of the statement, or detracts from it; and so much the more as the circumstances are more or less consistent or inconsistent with each other. But these qualities are not discovered but by reference to the connection of a cause throughout. Yet we must always observe what particular agrees or suits with any other particular.


1. Next to conjecture respecting a fact comes definition of it, for he who is unable to prove that he has done nothing, will try, in the next place, to make it appear that he has not done that which is laid to his charge. Definition is accordingly managed, for the most part, by the same methods as conjecture, the kind of defense only being changed [for in conjectural cases the fact is simply denied; in cases of definition an objection is made to the term applied to the fact], as we may see in cases of theft, deposits, or adultery; for as we say, I have not been guilty of theft, I did not receive a deposit, I have not committed adultery, so we say, what I did is not theft, what I received was not a deposit, what I committed is not adultery. 3. Sometimes we proceed from quality to definition, as in actions regarding madness, bad treatment of a wife, and offenses against the state, in which, if it cannot be said that what is laid to the charge of the accused was rightly done, it remains to say, that to act thus is not to be mad, to treat a wife ill, to injure the state.

Definition, then, is an explication of something in question, proper, clear, and concisely expressed. 3. It consists chiefly, as has been said, in the notification of genus, species, differences, and peculiarities; as, to define a horse (for I shall adopt a well known example), the genus is animal, the species mortal, the difference irrational (for man is also a mortal animal,) and the peculiarity neighing. 4. Definition is frequently used in pleading causes, for many reasons; for sometimes parties are agreed upon the term, but differ as to what is to be included under it; and sometimes the thing is clear, but there is a doubt as to the term to be applied to it. When there is an agreement about the name, and a doubt about the thing, the decision sometimes depends upon conjecture; as when it is asked, what is God?

5. For he [the Stoic] who denies that God is a spirit, diffused through every part of the universe, does not say that the term divine is improperly applied to his nature, like Epicurus, who has given him a human form, and a place in the spaces between the worlds. Both [the Stoic and the Epicurean] use one term, but are in doubt which of the two natures [a diffused spirit or a human form] is consistent with the reality. 6. Sometimes it is quality that is to be considered, as, What is oratory? is it the power of persuading, or the art of speaking well? This kind of question is very common in civil causes; thus it is inquired, whether a man found with another man's wife in a brothel is an adulterer? because the question is not about the name, but the quality of the act, and whether the man has been guilty of any offence at all; for if he has committed any offence, he can be nothing else but an adulterer.

7. It is a definition of a quite different kind when the question is wholly about a term, the application of which depends on the letter of the law, and which would not be discussed in a court of justice, but for the words which give rise to the dispute. Thus it is inquired, whether he who kills himself is a murderer; whether he who forced a tyrant to kill himself is a tyrannicide; and whether the incantations of magicians are poisons; for about the thing itself there is no controversy, as it is known to all men that it is not the same to kill one's self as to kill another, to kill a tyrant as to drive him to suicide, to recite incantations as to administer a draught of poison, but it is a question whether they do not, respectively, come under the same denomination.

8. Though I hardly dare to dissent from Cicero, who, following many authorities, says that definition is always concerned about a thing itself and something else, (as he that denies that a certain term is applicable to a certain thing, is obliged to show what term would be more applicable,) yet I consider that there are, as it were, three species of it. 9. For sometimes it is convenient to put a question thus: Is that adultery which is committed in a brothel? When we deny that it is adultery, it is not necessary to show by what term it ought to be called, for we deny the whole charge. Sometimes the inquiry is made thus: Is this act theft or sacrilege? Not but that it is sufficient for the defense that it is not sacrilege; still it is necessary to show what else it is: and consequently both theft and sacrilege must be defined.

10. Sometimes, again, it is a question with regard to things of different species, whether one can come under the same denomination as the other, when each has its own proper appellation, as a philtre, and a dose of poison. But in all disputes of this kind the question is whether this also comes under the same term, because the term, about which the dispute is, is acknowledged to be applicable to something else. It is sacrilege to steal what is sacred from a temple; is it also sacrilege to steal from it private property? It is adultery to lie with another man's wife in her own house? is it also adultery to lie with her in a brothel? It is tyrannicide to kill a tyrant; is it also tyrannicide to drive a tyrant to kill himself?

11. Accordingly syllogism, of which I shall speak hereafter, is, as it were, a weaker kind of definition; because in the one, it is inquired whether the same term is applicable to the thing in question as is applicable to something else; and, in the other, whether one thing is not to be reasoned about in the same manner as another. 12. There is also so much difference in definitions, that it is doubtful, as some think, whether the same thing can be defined in more than one form of words: as, whether rhetoric can be defined not only as the art of speaking well, but also as that of conceiving and expressing thoughts well, and of speaking with the full force of language, and of saying what is to the purpose. Yet we must take care that different definitions, though not at variance in sense, be expressed in a different form of words. But this is a subject for discussion among philosophers, not in courts of justice. 13. Sometimes words that are obscure, and but little known, require definition, as clarigatio [properly " a demand of satisfaction from an enemy under penalty of a declaration of war." Plin. H. N. xxii. 2. See Livy, i. 32], proletarius [from proles, "offspring;" one of the lowest class of the people, who were not usually called to serve in war, and were regarded as able to serve the state only by producing children. See Festus, and Aul. Gell. xvi. 10]. Sometimes also words that are well known in common speech, as what is the exact meaning of penus, "provisions, [as being laid up penitus, in the inner part of the house]" and litus, "a shore."

This variety is the reason that some authors include definition in the state of conjecture, others in that of quality; others even rank it among legal questions. 14. Some have not been at all pleased with that subtilty of definition which is adapted to the manner of logicians, thinking it rather fitted for cavilling about the niceties of words in the discussions of philosophers than likely to be of any service in the pleadings of orators. For though, they say, definition is of avail, in discussion, to hold in its fetters him who has to reply, and to force him to be silent, or to admit, even against his will, that which is to his prejudice, yet it is not of the same use in legal arguments; for in them we have to persuade the judge, who, though he may be fettered by our words, will yet, unless he is satisfied with our matter, mentally dissent from us altogether.

15. What great necessity, indeed, has a pleader of such preciseness of definition? If I do not say Man is an animal mortal and rational, can I not, by setting forth his numerous qualities of body and mind, in words of a wider scope, distinguish him from the gods or from brutes? 16. Is it not generally allowed, too, that one thing may be defined in more ways than one, (as Cicero [Pro Muraen., c. 35] shows, in saying, quid enim vulgo? universos, "for what are we to understand by publicly? All men,") and with a freedom and variety of manner, such as all orators have ordinarily adopted? Since, assuredly, the slavery of binding ourselves to certain words, (for slavery it indisputably is,) which has its origin in the practice of the philosophers, is very seldom to be seen in them; and Marcus Antonius, in the books of Cicero de Oratore [II. 25], expressly cautions us against attempting such exactness; 17. for it is even dangerous, since, if we err but in one word, we are likely to lose our cause entirely; and the best course is that middle one which Cicero adopts in his oration for Caecina [C. 15], and in which things are set forth, but exactness of terms is not hazarded.

For, says he, judges [recuperatores, judges appointed by the praetor in private or civil causes, and named from recupero, "to recover," because people might recover property by their means], that is not the only kind of violence which is offered to our persons and our lives, but there is a far more atrocious kind of violence, which, threatening us with the peril of death, often unsettles the mind, alarmed with terror, from its proper state and condition. 18. Or we may be secure, I may add, by letting proof precede definition; as when Cicero, in his Philippics [IX. 3], first establishes that Servius Sulpicius was killed by Antony, and then, in conclusion, defines thus: for he certainly killed who was the cause of death. I would not deny, however, that such rules are to be observed in pleading only as far as they are serviceable for our cause; and that if a definition can be made, at once strong, and expressed in a concise form of words, it is not only an ornament to our speech, but has very great effect, provided that it be impregnable.

19. The invariable order in definition is what a thing is, and whether it is this; and in general there is more difficulty in establishing the definition than in applying it to the matter in hand.

As to the first point, what a thing is, there are two objects to be regarded; for our own definition is to be established, and that of the opposite party is to be overthrown. 20. Hence in the schools, where we imagine contradiction offered to us, we have to lay down two definitions as exact as is possible for each party. But what we have to observe in the forum is, that our definition be not, possibly, superabundant, or nothing to the purpose, or ambiguous, or inconsistent, or equally favorable to both sides; faults that cannot happen except through the unskillfulness of the pleader. 21 . But, if we would define accurately, we shall be likely best to effect our object, if we first settle in our mind what we wish to establish; for our words will thus be exactly suited to our purpose.

That this point may be the clearer, let us still adhere to our familiar example: A man who has stolen private property from a temple, is accused of sacrilege. 22. The fact is not disputed; the question is, whether the term sacrilege, which is in the law, is applicable to the offence. It is accordingly disputed whether the act is sacrilege. The prosecutor adopts the term, because the money was stolen from a temple; the defendant denies that it is sacrilege, because he stole private property, but admits that it was theft. The advocate of the prosecutor, therefore, will define thus, It is sacrilege to steal anything from a sacred place; while the advocate of the defendant will define in this way, It is sacrilege to steal anything sacred.

23. Each, too, will try to overthrow the definition of the other; and a definition is overthrown in two ways, by being proved to be false or incomplete. It may indeed have a third fault, that of having no relation to the matter under consideration, but it will hardly be made faulty in this respect, except by fools. 24. We make a false definition, if we say, A horse is a rational animal; for a horse is indeed an animal, but irrational. That, again, which is common to anything else, will not be peculiar to the thing defined. Thus, then, the accused will say that the definition of the accuser is false; while the accuser cannot say that that of the accused is false; for it is sacrilege to steal anything sacred; but he will say that it is incomplete, since he ought to have added, or from a sacred place.

25. But for establishing and overthrowing definitions, one of the most effective modes is to have recourse to the consideration of peculiarities and differences, and sometimes also to etymology. All these particulars equity, as in other matters, will assist to support, and sometimes, also, conjecture. Etymology is but rarely introduced. We have one example of it in Cicero: For what is a tumult [Cic. Philipp. viii. 1. Tumultus is generally supposed to be from tumeo, "to swell;" Cicero seems to derive it from timeo, "to fear"], but such a perturbation that greater fear (timor) arises? whence also the term tumult is derived. 26. But about peculiarities and differences great subtility is displayed; as when it is inquired whether an addictus, whom the law condemns to serve until he pays his debts, is a slave. The one party will define thus: He is a slave who is legally in slavery; the other: He is a slave who is in slavery under the same legal conditions as a slave; or, as the ancients said, qui servitutem servit, "who serves as a slave."

Yet this last definition, though it differs somewhat from the other, is feeble, unless it be supported by the aid of peculiarities and differences; for the opponent will say that the addictus does serve as a slave, or under the same legal conditions as a slave. 27. Let us look, then, to the peculiarities and differences on which I touched lightly, in passing, in the fifth book: A slave, when he is set free, becomes a freedman; an addictus, when he recovers his liberty, is ingenuus [a free-born citizen. From ingeno, or ingigno, ingenui; born In the country, or born among other citizens, and in the same condition with them]; a slave cannot obtain his liberty without the consent of his master; a slave has no benefit of law [as they are not cives, the jus civile does not extend to them]; an addictus has. What is peculiar to a freeman, is that which no one has who is not free, as a pronomen, nomen, cognomen, and tribe; and these an addictus has.

28. When it is decided what a thing is, the question, whether it is this, is almost settled. However, we have to take care that our definition be favorable to our own cause. But what is most influential in a definition is the question of quality, as whether love be madness? To this question belong such proofs as Cicero says are proper to definition; proofs from antecedents, consequents, adjuncts, contraries, causes, effects, similitudes; of the nature of which arguments I have already spoken. 29. Cicero, in his speech for Caecina [C. 15], gives a concise example of arguments from beginnings, causes, effects, antecedents, consequents: Why then did they flee? For fear. What did they fear? Violence, doubtless. Can you then deny the beginning, when you have admitted the end? He has also recourse to similitude; Shall not that which is called violence in war, be called by the same name in peace? 30. But arguments are also drawn from contraries; for instance, if it be inquired whether a philtre be poison or not, because poison is not a philtre.

That the other kind of definition may be better known to my young men, (for I shall always think them my young men,) [his old pupils] I shall here give an example of a fictitious case. 31. Some youths, who were in the habit of associating together, agreed to dine on the sea-shore. One of them being absent from the dinner, the others erected a sort of tomb to him, and inscribed his name upon it. His father returning from a voyage across the sea, landed at that part of the coast, and, on reading his son's name, hanged himself. 32. These youths are said to have been the cause of the father's death. The definition of the accuser will be, He that commits any act that leads to the death of another, is the cause of the other's death. That of the accused will be, He who knowingly commits any act by which the death of another must necessarily be caused, etc. But setting aside definition, it is enough for the accuser to say, You were the cause of the man's death; for it was through your act that he died, since, if you had not acted as you did, he would now be alive.

33. To this the advocate of the accused will reply, He by whose act the death of a person has been caused, is not necessarily to be condemned for it; else what would become of accusers, witnesses, and judges, in cases of life and death? Nor is there always guilt in the person from whom the cause proceeded; for instance, if a person recommends a voyage to another, or invites a friend from over the sea, and he perishes by shipwreck, or if he invites a person to supper, and he dies of a surfeit committed at it, would he be guilty of the death of any of those persons? 34. Nor was the act of the young men the sole cause of death, but also the credulity of the old man, and his weakness in enduring affliction; for if he had had more fortitude or wisdom, he would be still living. Nor did the young men act with any bad intention; and he might have judged, either from the place of the supposed tomb, or from the marks of haste in its construction, that it was no real sepulchre. How then ought they to be punished, who, though they may seem to be homicides in every other respect, are evidently not so in intention?

35. Sometimes there is a settled definition, in which both parties agree. Thus Cicero says, Majesty resides in the government and in the whole dignity of the Roman people. But it may sometimes be a question whether this majesty has been injured, as was the case in the cause of Cornelius. [There are many observations on the majesty of the people, and the means by which it may be violated, in Cicero de Invent. ii. 17, seqq., and Partit. Orat. c. 30.] 36. But even if such a cause be thought similar to one dependent on definition, yet, as there is no dispute in it about definition, the point for decision must be one of quality, and must be included in that state which we happen to have had occasion to mention. It was however the subject next in order.


1. As to quality, it is sometimes considered in the most comprehensive sense, and in reference to more points than one; for it is sometimes a question what is the nature of a thing, and what is its form, as whether the soul is immortal, and whether God is of human shape; sometimes the inquiry is about magnitude and number, as what is the size of the sun? Are there more worlds than one? All such questions are indeed to be solved by conjecture, but they involve the question of quality.

2. They are also often treated in deliberative questions, as, if Caesar should deliberate whether he should make war on Britain, he would have to inquire what is the nature of the ocean there; whether Britain is an island (for the point was then unknown); what extent of land there is in it; and with what number of forces it would be proper to attack it. Under the head of quality, too, falls the consideration of what ought to be done or not done; to be sought or to be avoided; matters which enter most into deliberative questions, but also present themselves frequently in judicial pleadings, the only difference being that in one case the question regards the future, in the other the past.

3. All that relates, too, to the demonstrative kind of oratory falls under the consideration of quality; it is admitted that something has been done; it is to be shown what sort of a thing has been done. All judicial causes relate either to reward, or to punishment, or to the measure of one or the other. The first kind of cause is accordingly either simple or comparative; in the former we inquire what is just; in the latter, what is more just; or what is most just.

When the point for decision has respect to punishment, there is offered on the part of him who is accused, either justification of the charge, or extenuation of it, or excuse, or, as some think, deprecation.

4. Of these the most efficient is justification, by which we make it appear that the act, which is laid to the charge of the accused, was unobjectionable. A son is disinherited because he has served in the army, or been a candidate for office, or taken a wife, without the consent of his father; the father justifies what he has done. The followers of Hermagoras call this kind of defense κατ αντιληψιν, "contrasumption," using that term with reference to the mind of the pleader. [The term αντιληψις, which
signifies contrasumptio, the followers of Hermagoras understand, not of any corporeal act or sumptio, but of the action of the mind of the advocate, who, in contemplating the matter in question, thinks that it is to be regarded in a different light from that in which his adversary views it.]
I find no literal translation of the word in Latin; but it is called defensio absoluta, "absolute defense."

5. The sole question, however, is respecting the act, whether it is just or not. Whatever is just, is founded either on nature or on human institutions. 6. On nature is based whatever is consonant to dignity of anything, in which designation are comprehended the virtues piety, integrity, continence, etc. Some also add to render like for like; but that is not to be lightly admitted; for though violence opposed to violence, or retaliation, may offer no injustice towards him who has been the aggressor [Cicero de Inv. ii. 22], yet it does not follow that, because the act of each party is similar, the first act was just. Where there is strict justice on both sides, there is the same law, and the same condition; and perhaps no acts can be regarded as equal that are in any respect dissimilar. Human institutions consist of laws, customs, decisions, arguments.

7. There is another mode of defense, in which we justify an act in itself indefensible by aids drawn from without; this the Greeks call the defense κατ αντιθεσιν, "by opposition." The Latins also do not render this literally, for among them it is called causa assumptiva, "defense by assumption." 8. In this kind of defense the strongest plea is when we justify the act by the motive of it; such is the plea of Orestes, Horatius, and Milo. It is also called αντεγκλημα, "recrimination," because all our defense depends on accusing the person who is indicated by the other party: He was killed, but he was a robber; he was emasculated, but he was a ravisher.

9. There is also another kind of defense derived from the motives of an act, which differs from that just mentioned, and in which a deed is neither justified on its own ground, as in the absolute defense, nor by opposing another deed to it, but on the representation of its having been of some service to our country, or to some body of men, or even to our adversary, or sometimes to ourselves, provided it be such a deed as we might lawfully do for our own benefit; an argument which can be of no profit with regard to an accuser who is a stranger to us, and who prosecutes us according to the letter of the law, but which may be of use in reference to family disputes. 10. For a father may without presumption say to his children, on a trial for renouncing them, or a husband to his wife, if he is accused of treating her ill, or a son to his father, whom he seeks to prove insane, that what he has done was for his own interest; but, in such cases, the plea of escaping loss is much better than that of aiming at advantage.

11. Cases similar to those of the schools have to be conducted in trials about real occurrences; for the case of the renounced children in the schools is in the forum a case of children actually disinherited by their parents, and seeking to recover their property before the centumviri; the case of ill-treatment in the schools is in the forum a case of restoring a wife's dowry, in which the question is, through whose fault the divorce was caused; and that which in the schools is a case of insanity is in the forum a suit for appointing a guardian. 12. Under the head of advantage comes also the plea that something worse would have happened if the defendant had not acted as he did; for in a comparison of evils the less is to be regarded as a good; for example, if Mancinus should justify the treaty with Numantia on the ground that, if it had not been made, the whole Roman army would have been destroyed. This species of defense is called in Greek αντιστασις, "balancing." Some rhetoricians call it comparison.

13. Such are the modes of proceeding in defense of an act; but if a defense can neither be sustained on the motive of the act itself, nor by extrinsic aid, our next course is to transfer the charge, if we find it possible, on another party. Hence translation, or "exception," has been regarded as forming one of those states which have been previously mentioned. Sometimes, then, the blame is thrown on a person, as if Gracchus, being accused of concluding the Numantine treaty, (through fear of which accusation he seems to have passed his popular laws in his tribuneship,) should say that he was sent to conclude it by his general. 14. Sometimes it is cast on some circumstance, as if a person who had been directed to do something in the will of another, and had not done it, should say that it was rendered impossible by the laws. This the Greeks call μεταστασις, "transference."

Should these modes of defense fail us, there remains excuse, founded either on ignorance or on necessity. On ignorance: as if a person should brand another as a fugitive slave, and, after he is decided by law to be a freeman, should excuse himself by saying that he did not know that he was free. On necessity: as when a soldier does not present himself at the end of his furlough, and alleges that he was hindered by floods or by sickness. 15. Chance, also, is sometimes represented as the cause of a fault. Sometimes, too, we state that we have certainly erred, but that we acted with a good intention. Of both these kinds of excuses examples are so numerous and obvious that to offer any here is unnecessary.

If, again, none of those means which have been mentioned can avail us, we must see whether the charge can be extenuated. This is what is by some said to be the state of quantity [status quantitatis]. 16. But quantity, when it relates either to penalty or reward, is decided by the quality of the deed, and accordingly appears to me to fall under the state of quality, as well as quantity used with reference to number. The Greeks have the terms πηλικοτης, "magnitude," and ποσοτης, "multitude;" we include both under the same term.

17. The last method of all is deprecation; a mode of address which most rhetoricians do not allow to be admissible into judicial pleadings; and Cicero himself, in his speech for Quintus Ligarius [C. 10], seems to declare himself of the same opinion, when he says, I have pleaded many causes, Caesar, and even in conjunction with yourself, while regard to your public duties retained you in the forum, but I certainly never stooped so far as to say, "Forgive him, judges, he has erred, he has offended, he did not think of what he was doing; if he ever do so again," etc. 18. But in the senate, before the people or the emperor, and wherever there is power to relax the law, deprecation finds its place. It sometimes derives its greatest efficacy from the character of the accused himself, if it appear, from his previous life, that there are these three kinds of merit in him: that he has lived innocently, that he has been serviceable to others, and that he affords expectation that he will conduct himself blamelessly for the future, and make himself of some use to the world; and if, in addition, he seems to have already suffered sufficient, either from other inconveniences, from his present imminent peril, or from penitential feelings.

Sometimes, too, external circumstances give weight to deprecation, as nobility, dignity and the support of relatives and friends. 19. Most dependence however is to be placed on the judge, if we can make it appear that, should he spare the accused, commendation for clemency, rather than blame for weakness, will attend him. But even in common causes [as distinguished from pleadings before the senate or the emperor], deprecation is often introduced, though not through the whole course of a cause, yet in a great portion of it; for there is frequently such a distinction as this made: Even if he had been guilty of the charge, he ought to be pardoned; a method which has often had great effect in doubtful questions; and all perorations contain some portion of entreaty.

20. Sometimes, too, the accused rests the whole of his cause on this ground; for instance, if a father has disinherited his son [parents might disinherit their children if they led an immoral life; but children who had been disinherited, as well as those not named in the will, might go to law after the father's death to set aside the will], and testified, by an express declaration [elogium was a testimony concerning any person, whether for the purpose of blame or praise; of blame, as in this passage; of praise, as in Cic. Tusc. Quaest. i. 14], that he did so because he had formed a connection with a courtezan; for the whole question, in this case, is whether the father ought not to have pardoned such a fault, and whether the centumviri ought not to be indulgent to it. But even under strict forms of law, and in penal prosecutions, we make the distinction in pleading whether the penalty has been incurred, and whether it ought to be inflicted. However, what the rhetoricians thought, is true, that a criminal cannot be rescued from the law solely by this mode of defense.

21. With respect to rewards, two questions are to be considered; whether a party claiming a reward is deserving of any; and whether he is deserving of so great a reward as he claims. If there be two claimants, which of them is the more worthy; if several, which is the most worthy. 22. The decisions of such questions depend on the species of merit in the claimants. We have accordingly to regard, not only the act of any claimant, whether as represented to the judges, or as compared with the acts of others, but also his character; for it makes a great difference whether the person who has killed a tyrant is a young or an old man; whether a man or a woman; whether a stranger or a relative; and we must consider the place, too, on several accounts; whether it were in a state accustomed to tyranny, or one that had been always free; in the citadel, or at his own house; as also the manner, whether by sword or by poison; and at what time, whether during war or in peace; when he was about to resign his power, or when he was going to attempt some new wickedness. Among the recommendations of such an act, also, are to be reckoned the loss of popularity, the risk incurred, and the difficulty surmounted.

23. In regard to liberality, likewise, it is important to consider from what sort of character it has proceeded; for it is more pleasing in a poor than a rich man; in one who confers, than in one who requites, an obligation; in a person who has children than in one who is childless. We ought to inquire, too, what degree of benefaction he has bestowed, at what time, and with what object, that is, whether with any expectations of advantage to himself. Similar points are to be considered in a similar manner. The question of quality, accordingly, calls for the greatest resources of the orator; for there is a vast field for ability, whichever side the speaker takes, and the feelings have nowhere greater influence. 24. Conjecture also frequently admits proofs adduced from extrinsic circumstances, and employs arguments derived from the nature of the subject; but to show the quality of an act is the business of eloquence; and it is here that she reigns, predominates, and triumphs.

Under this head Virginius [IV. 1, 23] puts cases of disinheritance, insanity, ill-treatment of a wife, and those of female orphans suing for marriage with relatives [a theme for declamation in the schools, taken from the laws of the Athenians, among whom female orphans might sue for a marriage with a relation, as appears from the Phormio of Terence]. For the most part, indeed, such cases actually come under the consideration of qualities, and some writers have been found to call them questions of moral obligation. 25. But the laws respecting these matters sometimes admit also other states; for conjecture enters occasionally into many such questions, as when the accused parties, for instance, maintain that they have not done what is laid to their charge, or that they did it with a good intention.

Examples of such cases are abundant; and those of insanity and ill-treatment depend on definition. For laws often give rise to considerations of equity, when it has to be shown for what reasons equity would not be observed by a strict adherence to the law. 26. What is not justifiable as a legal act, may be defended on the ground of equity. We have to consider, too, in how many and what cases it is unlawful for a father to disinherit his son; under what charges a suit for ill-treatment is inadmissible; and in what circumstances a son is not allowed to accuse his father of insanity.

27. Of disinheriting there are two forms; the one for a crime completed, as when a son is disinherited for having committed rape or adultery; the other for a crime as it were incomplete, and still dependent on a condition, as when a son is disinherited because he continues disobedient to his father. The one is attended with rigorous proceedings on the part of the father; (for what is done is irrevocable;) the other is in some degree mild, and of an admonitory nature; for the father shows that he is more inclined to correct his son than to renounce him; but in either case the pleading on the part of the son ought to be in a submissive tone, and adapted to make due satisfaction to the father.

28. I know that those pleaders, who are ready to make attacks on fathers under cover of a figure of speech [it often happens that orators are ready to attack a father under cover of a figure, that is, to throw out something against him which, at first sight, does not appear calumnious, but in which there is some secret assault on his character], will not allow the justice of this remark; attacks which I would not say should never be made, (for cases may occur that demand them,) but they should certainly be avoided when it is possible to proceed in any other manner. But of figures I shall treat in another book.

29. The suits of wives on account of ill-treatment are similar to those of sons in regard to disinheritance; for they require the same moderation in stating charges.

As to actions on account of insanity, they are brought either on the ground of something that has taken place, or something that may or may not hereafter take place. 30. In regard to what has taken place, the pleader for the son has an open field for attack, but he should make his attack in such a way as exposes only the conduct of the father, while he should manifest pity for the father himself, as being disordered in mind from weakness of body. But in regard to that which has not taken place, and which admits of a change of purpose, he should use much solicitation and persuasion, and at last express his regret that infirmity, not immorality, obscures his reason; and the more he praises his previous goodness of character, the more easily will he prove that it has been changed by disease.

31. The accused party himself, as often as the case allows, should observe calmness in making his defense; for anger and excitement are indications of insanity. What is common to all such causes, is, that the accused parties do not always attempt a justification of their conduct, but frequently have recourse to apology and entreaties for pardon. For in family disputes it is often sufficient to secure acquittal, if it can be shown that a person has offended but once, or through mistake, or less gravely than is laid to his charge.

32. But many other kinds of suits come under the consideration of quality; as those for assault; for though the accused sometimes denies that he committed any assault, yet the decision generally depends on the nature of the act and the apparent intent. 33. Another kind of questions are those about appointing an accuser, which are called divinations; as to which Cicero, who accused Verres at the solicitation of the allies, adopts the following division: that we must consider by whom those, for whom redress is sought, would most desire the cause to be conducted, and by whom the party, who is accused, would least wish it to be conducted.

34. Such questions as the following, however, are most frequent: which of two pleaders has the stronger reasons for desiring to be the accuser; which of the two will bring the greater energy or ability to support the impeachment; which will carry it forward with greater integrity. 35. To these are to be added also questions respecting guardianship, in which it is usual to inquire whether regard ought to be had to anything else besides accounts; whether honesty only is required to be observed, and not also care as to speculations and consequences. Similar to these are cases of mismanagement of agency, or, in the forum, cases of misconduct of business; for an action may be brought for the mismanagement of anything entrusted to another.

36. Besides these, there are imagined in the schools cases of crimes not mentioned in the laws [see the Declamations attributed to Quintilian, 252, 344, 370, 371; and Seneca the Rhetorician, p. 428]; cases in which it is either inquired whether the act in question is really not mentioned in the laws, or whether it be really a crime. Both these inquiries rarely occur in the same case. Among the Greeks there were often prosecutions, and not in imaginary cases, for misconduct on embassies; where it was a common question, on the ground of equity, whether it is at all allowable for an ambassador to act otherwise than he has been instructed; and for how long a period the accused was an ambassador; since some ambassadors terminate their office with the delivery of their message; as in the case of Heius [with regard to Heius, who was at the head of the deputation sent by the Sicilians to Rome, (Cic. in Verr. Iv. 8) it was made a question whether he should not have returned to Sicily, and reported the result of his embassy, before he proceeded to give testimony against Verres], who, after his message was delivered, gave his testimony against Verres. But much depends on the quality of the act with which the ambassador is charged.

37. Another sort of accusation is that of having acted contrary to the interests of the state. [Examples of this sort of cause may be seen in Seneca the Rhetorician, ii. p. 21; also 344, 355, 492, 495; and in Fortunatianus, Pithoean. p. 40.] From such accusations arise a thousand legal cavillings: as, what it is to act contrary to the interests of the state; whether the accused has injured the state; or merely neglected to serve it; and whether it was injured by him or only on his account. But in these cases, again, much depends on the nature of the supposed act. Another charge is that of ingratitude; and in cases of that kind it is inquired whether the party against whom the charge is brought really received any kindness; an inquiry which is rarely to be answered in the negative; for he who denies the receipt of a kindness which he has received, fixes the charge of ingratitude on himself. 38. Additional inquiries are, what was the extent of the kindness that he received; whether he made any return at all; whether he who has made no return ought necessarily to be convicted of ingratitude; whether he could have made any return; whether he ought to have made that return which was demanded of him; and what is his general disposition.

Such as follow are of a more simple kind, as that of unjust divorce [see the Declamations attributed to Quintilian, 251, 262]; cases of which, as regards the law, have this peculiarity [the woman, proceeding against her husband, defends her own character; the husband, justifying himself for having divorced her, accuses the wife], that the defense is on the side of the accuser, and the accusation on that of the defendant. 39. That, too, in which a person makes a statement to the senate of the reasons that prompt him to kill himself; where the only point of law is, whether he who desires to die, that he may withdraw himself from legal proceedings against him, ought not to be prevented from killing himself [a subject for declamation in the schools, taken from a law at Marseilles, where poison was publicly kept for those who wished to drink it, provided that they stated to the senate their reasons for determining to die. See Val. Max. ii. 6]; all other considerations depend on quality.

Cases are also imagined regarding wills, in which the question has reference to quality alone, as in the case which I have detailed above, where a philosopher, a physician, and an orator, contend for the fourth part of their father's property, which he had bequeathed to the most worthy of his children. It is a similar case when suitors, equal in rank, claim marriage with a female orphan relative, and when the only question is about the most deserving among the competitors. 40. But it is not my intention to go through all such sorts of cases; (for more might still be imagined;) nor are the questions that arise from them common to all alike, but are varied by circumstances. I only wonder that Flavus [the same whom in sect. 24 he calls by his other name, Virginius], whose authority is deservedly great with me, restricted the subject of quality, when he was composing a, work merely for schools, within such narrow limits.

41. Quantity also, whether with respect to measure or number, falls generally, as I said, if not always, under the head of quality; but measure is sometimes determined by the equitable estimation of an action, as when it is inquired, how great an offence has been committed, or how great an obligation received, and sometimes by strict legality, as when it is disputed, under what law a person is to be punished or rewarded. 42. Thus, If he who has insulted a youth ought to pay ten thousand sesterces (which is the penalty appointed for such an offence,) ought he, if the youth whom he has dishonored hangs himself, to be punished capitally, as being the cause of his death? In such a case, those are deceived who plead as if there were a question between two laws; for, in regard to the ten thousand sesterces there is no controversy, since they are not claimed. 43. The point to be decided is, whether the accused was the cause of the young man's death.

The same sort of question, regarding measure also, resolves itself, at times, into a question of fact, as when it is disputed, whether a person, who has killed another, should be condemned to perpetual banishment, or to banishment for five years; the point for decision is, whether he committed the murder intentionally or not. 44. Such a question as the following, too, which arises from number, depends for decision on law: whether thirty rewards be due to Thrasybulus for removing thirty tyrants: and when two thieves have carried off a sum of money, whether each of them ought to restore fourfold or only double. But in such cases the act [that is qualitas facti] is taken into estimation, and yet the question of law [that is juris quaestio, or legalis quaestio] depends on quality.


1. An accused person who can neither deny that he has committed an act, nor prove that the act which he has committed is of a nature different from that which is attributed to it, nor justify the act, must necessarily take his stand on some point of law that is in his favor; whence generally arises a question about the legality of the process against him [actionis est quaestio], a question which does not, as some have thought, always present itself in the same manner. 2. For it sometimes precedes the trial, as in the case of the nice examinations of the praetor, when there is a doubt about the right of a person to be an accuser, and sometimes it occurs in the progress of the trial itself. The nature of such a question is twofold, as it is either intention or prescription that gives rise to it.

There were some who made a state of prescription, as if prescription were not concerned in all questions in which other laws are concerned. 3. When a cause depends on prescription, it is not necessary that there should be any inquiry about the fact itself. A son, who has been disinherited by his father, raises the question of prescription against him, as being infamous; and the dispute is then merely on the point whether the father has the right to disinherit. As often as we can, however, we must take care that the judge may conceive a favorable opinion of the main question, for he will thus be more inclined to listen to our arguments on the point of law; as in cases respecting sponsions, which arise from interdicts of the praetor, though the question may not be about right to possession, but merely about possession itself, yet it will be proper to show not only that we were in possession, but that that of which we were in possession was our own.

4. But the question occurs still more frequently with regard to intention. Let him who has saved his country by his valor choose whatever he pleases as a recompence. I deny that whatever he chooses ought to be given to him; I have no formal prescription; but I try to set the intention of the lawgiver, in the manner of prescription, against the written letter. In either case the state [that of exception] is the same.

5. Moreover every law either gives, or takes away, or punishes, or commands, or forbids, or permits. It gives rise to dispute either on its own account, or on account of another law; and to inquiries either with regard to its wording, or to its intention. As to its wording, it is either clear, obscure, or equivocal. 6. What I say of laws, I wish to be understood of wills, agreements, contracts, and every sort of written instruments; and even of verbal bargains. And as I have made four states or questions on this head, I will touch upon each of them.


1. The question of most frequent occurrence among lawyers is concerning the written letter of a law, and the intention of it; and it is about such questions that a great part of legal discussion is employed. It is, therefore, not at all wonderful that they prevail in the schools, where cases involving them are purposely invented. One species of this kind of question, is that in which there is a dispute about the letter of a law as well as the intention of it. 2. This occurs where there is any obscurity in a law, of which each party supports his own interpretation, and tries to overthrow that of his adversary; as in this case: Let a thief pay fourfold what he has stolen: Two thieves stole in company ten thousand sesterces; forty thousand are demanded from each; they represent that they ought to pay only twenty thousand each: here the prosecutor will say that what he demands is fourfold; and the thieves will say that what they offer is fourfold; and the intention of the law is alleged by each side in its own favor.

3. Or a dispute of this kind may occur when the wording of the law is clear in one sense, and doubtful in another; as, Let not the son of a harlot be allowed to make a speech to the people [this law is noticed by Hermogenes, p. 16]; A woman who had had a son by her husband, began to play the harlot: Her son is prohibited from addressing the people. Here the letter of the law evidently refers to the son of a woman who was a harlot before he was born, and it is doubtful whether the case of the son in question does not come under the law, because he is the son of the woman named, and she is a harlot. 4. It is a common question, too, how the following law, Let there be no second pleading about the same matter [see Declam. Quint. 266], is to be understood; that is, whether the term second pleading refers to the pleader, or to the suit. Such are the questions that arise from the obscurity of laws.

But there are others that arise, and this is the second class, where the words of the law are clear; and those who have particularly attended to this point, have called it, the state of what is expressed and what is intended. In this case, the one party makes a stand on the letter, and the other on the meaning. 5. But the literal interpretation may be combatted in three ways. One is, when it is shown from the law itself that it cannot be observed invariably, as is the case with regard to the law, Let children maintain their parents, or be put in prison; for an infant will surely not be put in prison. From this exception there will be a possibility of proceeding to others, and of making a distinct inquiry whether every one who does not maintain his parents is to be put in prison, and whether the particular person in question.

6. For this reason some masters in the schools propose a sort of cases in which no argument can be drawn from the law itself, and in which the only question is concerning the matter that is the subject of controversy. For example, Let a foreigner, if he mounts the wall, he punished with death: The enemy having scaled the walls, a foreigner repulsed them: It is demanded that he be put to death. 7. Here there will not be distinct questions whether every stranger, or whether this stranger, should be put to death, because no stronger objection can be brought against the literal interpretation of the law than the act which is the subject of dispute. It is sufficient merely to ask whether a foreigner may not mount the walls even for the purpose of saving the city. The defense of the foreigner, therefore, must rest on equity and the intention of the law.

It may happen, however, that we may be able to adduce examples from other laws, by which it may be shown that we cannot always adhere to the letter; a method which Cicero has adopted in his pleading for Caecina [C. 14, and especially 18, 19]. 8. There is a third mode, when we find something in the very words of a law to prove that the legislator intended something different from what is expressed, as in this case: Let him who is caught with steel in his hand at night, be sent to prison: A magistrate sent to prison a man who was found with a steel ring. Here, as the word in the law is "caught," it appears sufficiently clear that nothing is meant in it but steel intended for mischief.

9. But though he who rests on the meaning of the law, will endeavor, as often as he can, to explain away the letter of it; yet he, who adheres to the letter, will try at the same time to gain support from the intention of it. In wills it sometimes happens that the intention of the testator on a point is manifest, even though there be nothing written upon it, as in the case of Curius, when the well-known contest between Crassus and Scaevola occurred [Cic. De Orat. i. 39; ii. 32]. 10. A second heir was appointed, if a posthumous son should die before he was past the years of tutelage: No posthumous son was born. The near relatives laid claim to the property. Who could doubt but that it was the will of the testator that the same person should be heir if a son was not born who was to be heir if a son died? But he had written nothing on the point.

11 . A case exactly the reverse of this lately occurred, when something was written in a will which it was evident that the testator had not intended. A person who had bequeathed five thousand sesterces, having, in making a correction, erased the word "sesterces," inserted "pounds weight of silver," leaving the words "five thousand" standing. Yet it was apparent that he meant to leave five pounds weight of silver, for such a sum of silver as five thousand pounds weight for a legacy was unheard of and incredible. 12. Under this head fall the general questions, whether we ought to adhere to letter or intent; and what was the intention of the writer under consideration. The methods of treating them are to be sought from quality or conjecture [the only foundations for arguments in such cases, are, either to show that what you advance is probable, which belongs to conjecture, or just, which belongs to quality], of which I think that enough has been said.


1. The next head to be considered is that of contradictory laws, because it is agreed among all writers on rhetoric that in antinomia, there are two [one state regarding the letter, and one regarding the intent, in reference to each law] states regarding letter and intent; and not without reason; because, when one law is opposed to another, there arise, on both sides, objections against the letter, and questions regarding the intention; and it becomes a matter of dispute, with respect to each law, whether we ought to be guided by that law. 2. But it is obvious to everybody that one law is not opposed to another in strict equity; for, if there were two kinds of equity, the one must be abrogated by the other; but that the laws clash with each other only casually and accidentally.

The laws that interfere with one another may be of a like nature, as, if the option of a tyrannicide, and that of a man who has saved his country, occur at the same time, liberty being granted to each of choosing what he pleases, there would hence arise a comparison of their respective services, of the conjunctures in which they acted, and of the nature of the recompences on which they fix their thoughts. 3. Or the same law may be opposed to itself; as in the case of two deliverers of their country, two tyrannicides, two women who have been violated [a common case in the schools, where it was an imaginary law that a woman who had been forcibly violated might demand that the ravisher should be put to death, or that he should marry her without receiving any dowry. In the case to which allusion is made in the text, one man is represented as having violated two women in one night, one of
whom demands his death, and the other his hand in marriage]
; in regard to whom there can be no other question but that of time, whose claim had the priority, or of quality, whose claim is the more just.

4. Dissimilar or similar laws, also, are sometimes in conflict. Dissimilar laws are such as may be attacked by arguments of a different kind even though no law be opposed to them; as in this case [it will be an example of antinomia, if the commandant of a citadel, who has saved his country, demand permission to quit his post], Let not the commandant quit the citadel; Let the man who has saved his country choose what he pleases. [Suppose that the commandant and the deliverer of his country are the same person; then, with respect to him in his character of deliverer, (the words in brackets are supplied as necessary to the sense. The text seems defective)] though no law stand in the way, it may be inquired whether he ought to receive whatever he chooses; and, in regard to him in his character of commandant, many arguments may be urged by which the letter of the law is overthrown; for instance, if there be a fire in the citadel, or if it be necessary to sally forth against the enemy.

5. Similar laws are those to which no opposition can be made but that of another similar law. Suppose that one law says, Let the statue of a tyrannicide be placed in the gymnasium; that another law says, Let not the statue of a woman be placed in the gymnasium; and that a woman kills a tyrant; it is plain that neither under any other circumstances can the statue of a woman be placed there, nor the statue of a tyrannicide be prevented from being placed there. 6. Two laws are of unlike nature, when many arguments may be used against the one, and nothing can be said against the other but what is the subject of the controversy; as in the case where the deliverer of his country demands impunity for a deserter; for against the law regarding the deliverer of his country many arguments may be brought, as I have just remarked, but the law concerning deserters can be set aside only by the option allowed to the deliverer of his country.

7. In addition, the point of right involved in conflicting laws, is either admitted by both parties, or doubtful. If it is admitted, there commonly follow such questions as these: Which of the two laws is the more binding; whether it relates to gods or men; to the state, or to private individuals; to reward or to punishment; to great or small matters; whether it permits, forbids, or commands. 8. It is a common subject of inquiry, too, which of the two laws is the more ancient; but the most important consideration is, which of the two laws will suffer less; as in the case of the deserter and the deliverer of his country; for if the deserter be not put to death, the whole law regarding deserters is set aside; but if he is put to death, another choice may be allowed to the deliverer of his country.

It is, however, of great consequence which course is the better and more equitable; a point on which no direction can be given but when the case is proposed for consideration. 9. If the point of right be doubtful, then arises a question on one side, or on both sides reciprocally, respecting it; as in such a case as this: Let a father have the power of seizing the body of his son, and a patron that of seizing his freedman: Let freedmen belong to the heir: A certain person made the son of his freedman his heir; after his death the right of seizure is claimed both by the son of the freedman and the freedman himself, each over the other; and the son, now become patron, denies that his father was possessed of the rights of a father, because he was subject to a patron.

10. Two provisions in a law may interfere with one another, as well as two laws. Thus, Let an illegitimate son, born before a legitimate one, be to his father as legitimate; if born after an illegitimate, only as a citizen.

What I say of laws, is also to be said of decrees of the senate. If they contradict each other, or are at variance with the laws, there is still no other name [it will always be called antinomia] for the state of the question.


1. The state called syllogism has some resemblance to that of letter and intention, inasmuch as one party, under it, always takes its stand on the letter; but there is this difference, that in the state of letter and intention, arguments are brought against the letter, in that of syllogism the meaning is carried beyond the letter; in the former, he who adheres to the letter, makes it his object that at least what is written may be carried into effect; in the latter, that nothing may be done besides what is written. Syllogism has also some affinity to definition [there is a very great affinity between syllogism and definition; for definition decides that the name of two things is the same; syllogism proves that two things are to be regarded in the same light]; for if syllogism be weak, it often has recourse to definition.

2. For suppose that there be this law: Let a woman who administers poison be put to death. And this case: A woman several times gave a philtre to her husband who had neglected her; afterwards she procured a divorce from him; being solicited by her relatives to return to him, she did not return; the husband hanged himself; the woman is accused of poisoning. The strongest argument of the accuser will be to say that a philtre is poison; this will be a definition; but if it fail to produce sufficient effect, the syllogism will be attempted, (to which he may proceed, giving up, as it were, the definition,) to decide whether she does not deserve to be punished as much as if she had actually poisoned her husband.

3. The state of syllogism, therefore, deduces from that which is written that which is uncertain; and, as this is collected by reasoning, it is called the ratiocinatory state. The following are the points which it mostly embraces: Whether what is lawful to be done once, is lawful to be done more than once: A woman found guilty of incest, and precipitated from the Tarpeian rock, is found alive; she is required to undergo the punishment a second time. [See this case in Seneca Rhet. p. 92.] Whether what the law grants with regard to one person or thing, it grants with regard to several: A man who has killed two tyrants at once, claims two rewards. 4. Whether what was lawful before a certain time, was also lawful after it: A woman is forcibly violated; the ravisher flees; the woman is married, and, on his return, demands her option. Whether what is forbidden with regard to the whole, is forbidden with regard to part: It is not lawful to receive a plow in pledge; a man received a plowshare. Whether what is forbidden with regard to part, is forbidden with regard to the whole: It is not lawful to export wool from Tarentum; a person exported sheep.

5. In these cases of syllogism the one party rests on the letter; the other alleges that no provision is made in the law against the act in question. "I demand," says the accuser, "that the woman guilty of incest be thrown headlong from the rock; for such is the law." On the same ground the woman who has been forcibly violated claims her option; and "in exporting sheep," it is said, "wool is exported;" and it is the same with other cases. 6. But it may be replied, "It is not written in the law that a woman condemned should be twice thrown headlong; that a woman forcibly violated should have her option whenever she pleases; that a tyannicide should receive two rewards; that there is no provision in the law about a plowshare, or about sheep;" and what is doubtful is then to be collected from what is certain.

To deduce from what is written that which is not written, is a matter of greater difficulty: Let him who has killed his father be sewn up in a sack; A man kills his mother. Let it be unlawful to drag a man from his house to the judgment-seat; A man drags another from his tent. 7. In such cases, the questions are, whether, when there is not a particular law for a case, we must have recourse to a similar law; and whether the matter in question is similar to that to which the letter of the law refers.

But what is similar may be either greater, or equal, or less. [Similitude is three-fold; a thing may be like another, but less; like it, and equal; like it, but less. Hence three species of arguments are derived from comparison; from equality, from the greater to the less, and from the less to the greater.] In the first case, we inquire whether sufficient provision has been made with regard to the matter in question in the law to which we refer it, and whether, if sufficient provision has not been made, we ought to apply that law to it. In the two other cases, we inquire concerning the intention of the legislator. But arguments founded on equity are the strongest.


1. Of ambiguity the species are innumerable; insomuch that some philosophers think there is not a single word that has not more significations than one. [See Aul. Gell. xi. 12.] But the genera of it are very few; for it arises either from words taken singly or in connection.

2. A single word gives rise to ambiguity, when it is a denomination for more things or persons than one, (the Greeks apply to such ambiguity the term homonymy,) as Gallus; for as to this word, taken by itself, it is uncertain whether it means a bird, a native of a certain country, a proper name, or a person in a certain condition of body [cock; Gaul; and the Galli were emasculated priests of Cybele]; and it is uncertain whether Ajax means the son of Oileus. Some verbs also have different meanings, as cerno. 3. Such ambiguity presents itself in many ways; whence often arise disputes, especially with regard to wills, when persons who have the same name contend about their liberty [when slaves are set free by a will], or succession to an inheritance; or when, from ambiguity in the expression, it becomes a matter of doubt what is bequeathed to any person.

4. Another source of ambiguity is, when a word has one signification if taken entire, and another if divided, as ingenua, armamentum, Corvinum. Such words can only give rise to silly cavillings, but the Greeks make them the origin of controversies in the schools: hence comes the well-known dispute about the αυλητρις, whether a hall that had fallen three times, or a female flute-player, if she fell, was to be sold. [See Diog. Laert. vii. 62; Theon. Progymn. p. 35. Αυλη τρις is "a hall three times;" αυλητρις, "a female flute-player."]

5. A third kind of ambiguity arises from compound words; for example, if a person should direct by his will, that his body should be buried in occulta loco, "in a sequestered spot," and should bequeath a portion of land round his tomb, to be taken from his heirs, as is usual, for the protection of his ashes, the expression in occulto, if taken as a compound word, inocculto, "unsequestered," might be the origin of a law-suit. 6. So, among the Greek rhetoricians, Λεων and Πανταλεων have a contention, as it is doubtful whether the letter of a will signifies that all the possessions are left to Leon, παντα Λεοντι, or that the possessions are left Πανταλεοντι, to Pantaleon.

7. But ambiguity is more frequent in words put together; it sometimes arises from uncertainty with respect to cases, as in the verse,

Aio te, Aeacida, Romanos vincere posse,
[A verse from Enuius; see Cicero de Divin. ii. 56.]
I say that you, offspring of Aeacus,
The Romans can defeat.

Sometimes from collocation, when it is doubtful to what a word or words ought to be referred; and this very frequently happens when that which is in the middle may be connected either with what precedes or with what follows, as in the words of Virgil [Aeneid. i. 477] respecting Troilus,

Lora tenens tamen,

Holding still the reins,

Where it may be asked, whether Troilus is dragged because he still holds the reins, or whether, though he still holds the reins, he is nevertheless dragged. 8. Hence is that case in the schools, that a man in his will ordered to be erected statuam auream hastam tenentem, where it is a question, whether it was to be a golden statue holding a spear, or a golden spear, with a statue of some other material. Ambiguity is caused still more frequently by an improper inflexion [an improper mode of delivering a phrase or sentence, so as to connect such parts of it as ought to be separated, and to separate such as ought to be connected, would misrepresent the sense] of the voice, as in the verse,

Quinquaginta ubi erant centum inde occidit Achilles.
[A verse translated, with a slight change, from the Greek. See Aristot. Sophist. i. 4.]

9. Sometimes it is doubtful to which of two antecedents a word is to be referred; hence the scholastic case, Let my heir be bound to give my wife a hundred pounds of the plate, "quod elegerit," where it is doubtful to which of the two elegerit should be referred. But of the three last examples of ambiguity, the first may be corrected by a change of cases, the second by a separation or transposition of the words, and the third by some addition. 10. Ambiguity, caused by the doubling of an accusative, may be removed by the introduction of an ablative, as in the words

Lachetem audivi percussisse Demeam,
[These words may be translated either, "I heard that Laches had struck Demea," or, "I heard that Demea had struck Laches."]

may be altered to a Lachete percussum Demeam. There is, however, in the ablative, as I remarked in the first book, a natural ambiguity, as in Coelo decurrit aperto, it is doubtful whether per apertum Coelum, "through the open heaven," is meant, or quum Coelum apertum esset, "when the heaven was opened." 11. We may divide words from one another in pronunciation by taking breath or pausing; thus we may pause after statuam, and then say auream hastam, or we may pause after statuam auream, and then add hastam. An addition, in the third example, may be made by inserting ipse after elegerit, " quod elegerit ipse," that the heir may be understood, or ipsa, that the wife may be understood. An ambiguity caused by the insertion of a superfluous word, may be removed by withdrawing it, as in the phrase nos flentes illos deprehendimus. [It is the word illos that may be withdrawn.]

12. Where it is doubtful to what a word or phrase should be referred, and where, perhaps, the word or phrase itself is ambiguous, we may have to alter several words to make a correction; as in Haeres meus dare illi damnas esto omnia sua. [It is doubtful whether the word sua is to be referred to haeres or to illi.] Cicero runs into this kind of fault, in speaking of Caius Fannius [Cicero Brut. c. 26]: He, by the direction of his father-in-law, of whom, as he had not been elected into the college of augurs, he was not extremely fond, especially as he had preferred Quintus Scaevola, his younger son-in-law, sibi, to him, &c.; for this sibi may be referred, either to the father-in-law or to Fannius.

13. The lengthening or shortening of a syllable, too, left dubious in any writing, may be a cause of ambiguity, as in the word Cato, for it means one thing in the nominative, when its second syllable is short [Quintilian speaks as if the o in the nominative case of Cato were uniformly shortened. Compare Varro vocat, viii. 6, 73], and another in the dative or ablative, when that syllable is made long. There are many other species of ambiguity besides, which it is not necessary to specify.

14. Nor is it of importance whence ambiguity arises, or how it is removed; it is sufficient that it presents two senses to the mind; and the mode of understanding the writing or the pronunciation, is a matter of equal consideration for both parties in a suit. It is a useless precept, therefore, that we should endeavor, in this state, to turn the word or phrase in our own favor, for, if that be possible, there is no ambiguity.

15. Every question of ambiguity, however, has respect to the following points; sometimes, which of two interpretations is the more natural; and always, which of the two is the more equitable; and which was the meaning attached to the words by him who wrote or spoke them. But the manner of treating these questions, for or against, has been sufficiently shown in what I have said on conjecture and quality.


1. Between the states [not the two, of which he has just been speaking, but states in general, and especially the four legal states] there is a certain affinity, for in definition the question is, what is the meaning of a term; in the syllogism, which is the most nearly related to definition, the object is to ascertain what the writer meant; and from antinomia, or the contradiction of laws, it appears that there are two [one in regard to each law] states of writing and intention of the writer. Definition, again, is itself a kind of ambiguity, as the meaning of a word may be regarded in two lights. 2. The state of what is written and the intention of the writer has regard also to the signification of terms, and the same object is kept in view in antinomia. Hence some rhetoricians have said, that all these states merely constitute that of letter and intention; others think that in that of letter and intention lies the ambiguity which gives rise to dispute [Cicero de Orat. i. 26].

But all these states are in reality distinct; for an obscure law is one thing, and an ambiguous law another. 3. Definition is concerned with a general question regarding the nature of a term; which question may be unconnected with the scope of a cause; the state of letter and intention discusses the meaning of the very word which is in the law; syllogism tries to settle what is not in the law; ambiguity considers a word under two senses; antinomia is a comparison between two contradictory laws. 4. This distinction, accordingly, has been justly admitted by the most learned rhetoricians, and continues to be observed among the generality of the wisest.

As to discussions of this kind, though directions on all points could not be given, yet it has been practicable to give some. 5. There are other particulars which allow facility for instruction concerning them only when the subject, on which we have to speak, is propounded; for not only must a whole cause be divided into its general questions and heads, but these divisions themselves must also have their own distribution and arrangement of matter. In the exordium there is something first, something second, and so on; and every question and head must have its own disposition of particulars, as single theses [general or indefinite questions].

6. For is it possible that he can be thought sufficiently skilled in arrangement, who, after dividing his case into these points, whether every kind of reward ought to be granted to the deliverer of his country; whether he should be permitted to take private property; whether a marriage with whom he pleases should be allowed him; whether a married lady should be given him; and whether the lady whose case is before the court, should then, when he has to speak on the first point, mix up his observations indiscriminately, just as each happens to come into his head, not knowing that he should consider first whether we should hold to the letter of a law, or to the intent of it, that he should make a commencement on this question, and then, connecting with it what follows, should arrange the whole of his speech with the same regularity as the parts of the human body, of which, for example, the hand is a portion, the fingers a portion of the hand, and the joints portions of the finger?

7. It is this nicety of arrangement that a writer on rhetoric cannot teach, unless when he has a certain and definite subject before him. 8. But what would one or two examples avail, or even a hundred or a thousand, in a field that is boundless? It is the part of a teacher to demonstrate day after day, sometimes in one kind of causes and sometimes in another, what is the proper order and connection of particulars, that skill may gradually be acquired by his pupils, and the power of application to similar cases. 9. All cannot be taught that art is able to accomplish [Cicero de Orat. ii. 16]. What painter has learned to copy every object on the face of the earth? But when he has once acquired skill in copying, he will produce a representation of whatever he takes in hand. What artist in fashioning vases has not produced one such as he had never seen?

Some things, however, depend not on the teachers, but on the learners. 10. A physician will teach his pupil what is to be done in every sort of disease, and what is to be conjectured from certain symptoms; but it is the pupil's own genius that must acquire for him the nice faculty of feeling the pulse, of observing the different degrees of heat, and the alterations in respiration and complexion, and of noting what tokens are significant of any particular malady. In like manner, let us seek most aid from ourselves, and meditate our own causes, reflecting that men discovered the art of oratory before they taught it.

11. For that is the most effective arrangement of a pleading, an arrangement justly called economic, which cannot be made but when the whole cause is spread as it were before us, and which tells us when we ought to adopt an exordium, and when to omit it; when we should make a continuous statement of a case, and when a statement subdivided into heads; when we should begin at the beginning, and when, after the manner of Homer, in the middle or towards the end; 12. when we should make no statement at all; when we should commence with our own allegations, and when with those of our adversary; when with the strongest proofs, when with the weaker; in what sort of cause questions should he propounded in the introduction; in what causes the way should be prepared for them by prefatory hints; what the mind of the judge will be likely to admit if expressed at once, and to what he must be conducted gradually; whether our refutation should oppose the arguments of the adversary one by one, or in a body; whether our appeals to the feelings should be reserved for the peroration, or diffused through our whole speech; whether we should speak first of law, or of equity; whether we should first charge our opponent with past offences, or repel them if advanced against ourselves, or confine our remarks to the points for decision; 13. and, if a cause be complex, what order should be observed in our conduct of it, and what oral or written evidence, of any kind, should be set forth in our regular pleading, or reserved.

This is the virtue as it were of a general, dividing his forces to meet the various events of war, retaining part to garrison fortresses or defend cities, and distributing other parts to collect provisions, to secure passes, and to act by land or by sea as occasion may require. 14. But such merits in oratory he only will display, to whom all the resources of nature, learning, and industry, shall be at hand. Let no man expect, therefore, to be eloquent only by the labor of others. Let him who would be an orator be assured that he must study early and late; that he must reiterate his efforts; that he must grow pale with toil; he must exert his own powers, and acquire his own method; he must not merely look to principles, but must have them in readiness to act upon them; not as if they had been taught him, but as if they had been born in him. 15. For art can easily show a way, if there be one; but art has done its duty when it sets the resources of eloquence before us; it is for us to know how to use them.

16. There remains then only the arrangement of parts; and in the parts themselves there must be some one thought first, another second, another third, and so on; and we must take care that these thoughts be not merely placed in a certain order, but that they be also connected one with another, cohering so closely that no joining may appear between them; so that they may form a body, and not a mere collection of members. 17. This object will be attained, if we take care to observe what is suitable for each place, and study to bring together words that will not combat but embrace each other. Thus different things will not seem hurried together from distant parts, all strangers one to another, but will unite themselves, in a sure bond and alliance, with those that precede and those that follow; and our speech will appear not merely a combination of phrases, but all of a piece. But I am perhaps proceeding too far, as the transition from one part to another beguiles me; and I am gliding imperceptibly from the rules for arrangement into those for elocution, on which the next book shall formally enter.



1. In the observations which are thrown together in the last five books, is comprehended the method of inventing, and of arranging what we invent; and though to understand this method thoroughly and in all its parts is necessary to the attainment of the height of oratorical skill, yet to beginners it is fit that it be communicated rather in a shorter and simpler way. 2. For otherwise learners are apt to be deterred by the difficulties of a study so various and complicated; or their faculties, at an age when they require to be strengthened, and to be fostered with some degree of indulgence, are debilitated by being devoted to a task too burdensome for them; or they think that, if they acquire skill in these matters only, they are sufficiently qualified to become truly eloquent; or, again, as if they were bound to certain fixed laws of speaking, they shrink from every attempt to do anything for themselves.

3. Hence it has happened, as some think, that those who have been the most diligent writers of rules on the art, have been farthest from attaining true eloquence. Yet it is necessary to point out a way to those who are entering on the study; but that way should be plain to be pursued, and easy to be shown. Let the able teacher, therefore, such as I conceive in my mind, choose the best precepts out of all that have been given, and communicate at first only such as he approves, without occupying his time in refuting those of an opposite kind. Pupils will follow where the master leads, and, as their minds are strengthened by learning, their judgment will also increase.

4. Let them suppose at first that there is no other road than that by which they are conducted, and discover afterwards that it is the best. The principles, however, which writers, by a pertinacious adherence to their respective opinions, have rendered embarrassing, are in themselves by no means obscure or hard to be understood. 5. In the whole treatment of this art, accordingly, it is more difficult to decide what to teach, than to teach it when a decision is made upon it; and in these two departments, especially, invention and arrangement of matter, there are but very few general rules, and if he, who is under instruction, shows no repugnance or inability to attend to them, he will find the way open to the acquirement of everything else.

6. I have already spent much labor on this work, with a view to show that oratory is skill in speaking well; that it is useful; that it is an art, and a virtue; that its subjects are everything on which an orator may be required to speak; that those subjects lie mostly in three species of oratory, the demonstrative, the deliberative, and the judicial; that all speech consists of matter and words; that, as to matter, we must study invention, as to words, elocution, and as to both arrangement; all which particulars memory must guard and delivery recommend. 7. I showed that the duty of an orator was comprised in the three arts of persuading, exciting, and pleasing; that, for persuading, statement and argument are most efficient, and for exciting, appeals to the feelings, which may be dispersed through the whole of a speech, but should be used chiefly at the beginning and the end; while to please, though it depends on both matter and words, belongs chiefly to elocution. 8. I observed that some questions are indefinite, others definite, or limited to the consideration of persons, places, and occasions; that in regard to every thing there are three points to be considered, whether it is, what it is, and of what nature it is.

To these remarks I added that demonstrative oratory consists in praising and blaming; that, in speaking of a person's character, we must notice what was done by the person himself of whom we speak, and what took place after his death, and that this kind of oratory was employed about the honorable and the useful. 9. To deliberative oratory I observed that a third part [deliberative oratory has two parts common to it with the demonstrative or epideictic kind, namely, concerning what is honorable and what is useful] is added, dependent on conjecture, as when we inquire whether that, which is the subject of our deliberation, is possible, and whether it is likely to happen. In this department of oratory, too, I said that it ought above all to be considered what is the character of the speaker, before whom, and on what subject, he speaks. As to judicial causes, I remarked that some depend on one point, some on several; that in some a mere attack or defense is sufficient; and that all defense consists either in denial, (which is of two kinds, as we may dispute whether the fact in question really happened or whether that which happened was of the nature attributed to it,) or in justification, or in exception.

10. I added that questions in a cause relate either to something done, or something written; that in regard to anything done, we consider its probability, its nature, and its quality, and in respect to anything written, the meaning or intention of the words; in contemplating which, the nature of whole causes, criminal and civil [causae are properly subjects of oratory of any kind, demonstrative, deliberative, or judicial; actiones are properly causae forenses, or subjects for oratory of the judicial kind, whether of a criminal or civil nature], has to be regarded; all of which are included under the heads of letter and intention, syllogism, ambiguity, or contradictory laws.

11. I stated, moreover, that in every judicial cause there are five parts; that the judge is to be conciliated in the exordium; that the cause is set forth in the statement of facts, supported by evidence, and overthrown by refutation; and that the memory is to be refreshed, or the feelings excited, in the peroration. 12. To this I added the topics for argument and addresses to the passions, and showed the means by which judges must be roused, appeased, or amused. Last of all was subjoined the method of division. But let him who shall read this work for improvement feel assured that the course of proceeding laid down in it is one in which nature ought to do much of herself even without learning; so that the various heads of which I have spoken should seem not so much to have been invented by teachers, as to have been noticed by them according as they presented themselves.

13. What is to follow requires more labor and care, since I have now to treat of the art of elocution, which is, as all orators are agreed, the most difficult part of my work; for Marcus Antonius, of whom I have spoken above [this well-known saying of his is to be found in Cic. de Orat. i. 21; Orat. c. 5; Plin. Ep. v. 22], when he said that he had seen many good speakers, but none of them truly eloquent, understood that it is sufficient for a good speaker to say just what is proper, but to speak in an ornate style belongs only to the most eloquent. 14. If such excellence, accordingly, was found in no speaker down to his time, and not even in himself or in Lucius Crassus, it is certain that it was wanting in them and in preceding speakers, only because it was extremely difficult of acquirement. Cicero himself, indeed, is of opinion, that invention and arrangement are in the power of any sensible man, but eloquence only in that of the complete orator; and it was on this account that he gave his chief attention to the rules for that accomplishment.

15. That he acted rightly in doing so, is shown by the very name of the art of which we are speaking; for eloqui, "to speak forth," is to express whatever has been conceived in the mind, and to communicate it fully to the hearers; an art, without which all preceding attainments are useless, like a sword sheathed and clinging to its scabbard. 16. Eloquence, therefore, requires the utmost teaching; no man can attain it without the aid of art; study must be applied to the acquirement of it; exercise and imitation must make it their object; our whole life must be spent in the pursuit of it; it is in this that one orator chiefly excels another; it is from this that some styles of speaking are so much better than others.

17. For we are not to suppose that the Asiatics [the distinction of orators into Asiatic, Attic, and Rhodian, is frequently noticed in Cicero. See also Dionysius Halicarnassensis, vol. v. p. 447, ed. Reisk], or other speakers in any way faulty, were unable to invent matter or to arrange it; or that those whom we call dry [he seems to mean some who wished to be called Attics, and who, seeing that Lysias the Attic orator had adopted a close and dry style, studied to resemble him. Cicero passes a censure on them in his Orator] were void of understanding or perspicacity in their pleadings; but the truth is that the first wanted judgment and moderation in expressing themselves, and the second energy; and hence it is evident, that in expression lie the faults and excellences of oratory.

18. Yet it is not to be understood that regard is to be paid, only to words, for I must meet and stop those in the very vestibule as it were, who would take advantage of what I have just admitted, and who, neglecting to attend to the study of things, which are the nerves of all causes, consume their lives in an empty application to words, making it their object to attain elegance, which is, indeed, in my judgment, an excellent quality in speaking, but only when it comes naturally, not when it is affected. 19. Bodies that are in health, with the blood in a sound state, and strengthened by exercise, have their beauty from the same causes from which they have their vigor [see Cic. de Orat. iii. 45], for they are well-complexioned, of a proper tension, and with muscles fully developed; but if a person should render them artificially smooth, and paint and deck them in an effeminate fashion, they would be made eminently repulsive by the very labor bestowed in beautifying them. 20. A becoming and magnificent dress, as it is expressed in the Greek verse, adds dignity to men; but effeminate and luxurious apparel, while it fails to adorn the person, discovers the depravity of the mind. In like manner the transparent and variegated style of some speakers deprives their matter, when clothed in such a garb of words, of all force and spirit.

I would, therefore, recommend care about words, and the utmost care about matter. 21. The best words generally attach themselves to our subject, and show themselves by their own light; but we set ourselves to seek for words, as if they were always hidden, and trying to keep themselves from being discovered. We never consider that they are to be found close to the subject on which we have to speak, but look for them, in strange places, and do violence to them when we have found them. 22. It is with a more manly spirit that Eloquence is to be pursued, who, if she is in vigor throughout her frame, will think it no part of her study to polish her nails and smooth her hair.

23. It generally happens that the more attention is paid to such niceties, the more oratory is deteriorated; for the best expressions are such as are least far-fetched, and have an air of simplicity, appearing to spring from truth itself. Those which betray care, refuse to appear otherwise than artificial and studied; they fail to exhibit grace, and do not produce conviction; besides that they obscure the sense, and choke the crop as it were, with a superabundance of herbage. 24. What may be said simply, we express paraphrastically, from fondness for words; what has been told sufficiently, we repeat; what may be clearly signified in one word, we envelope in a multitude; and we often prefer to intimate our thoughts rather than express them.

Indeed no natural expression now satisfies us, since none appears elegant that another speaker has used. 25. We borrow tropes or metaphors from the poets most corrupt in taste, and think that we are witty only when there is need of wit to understand us. Yet Cicero [De Orat. i. 3] had plainly enough told us, that to depart from the ordinary style of language, and from the practice sanctioned by universal reason, is, in speaking, even the greatest of faults. 26. But Cicero, forsooth, was a harsh and unpolished orator; and we, to whom all that nature dictates is contemptible, and who seek, not ornament, but meretricious finery, know how to speak better than he; as if there were any excellence in words except as far as they agree with things; and if we are to make it the object of our whole life, that our words may be nice, and splendid, and ornate, and properly arranged, the whole fruit of our studies comes to nothing.

27. Yet we see most of our speakers hesitating about every word, seeking for expressions, and weighing and measuring them when they are found. Even if the sole object of their solicitude were that they might always use the best words, yet their unhappy care would deserve to be execrated, as it retards the course of their speech, and, from hesitation and diffidence, extinguishes the ardor of imagination. 28. He is but a wretched, and, I may say, a poor orator, who cannot endure to lose a single word. Yet not a single word, assuredly, will he lose, who shall first of all have learned the true principles of eloquence, and shall, by a long and judicious course of reading, have acquired a copious supply of words, and attained the art of arranging them; and who, besides, shall have made himself master of his stores by constant exercise, so that they may always be at hand and before his eyes.

29. To him who shall have done this, things and their names will present themselves at once; but for such excellence there is need of previous study, and of ability acquired, and, as it were laid up; for anxiety in seeking, judging, and comparing words, should be used while we are learning, not after we have become speakers. Otherwise, as men who have not secured a fortune have recourse to occasional expedients, so such speakers, from not having previously studied sufficiently, will be at a loss for expressions. 30. But if resources for speaking have been acquired beforehand, they will be ready for our use, not seeming merely to answer exigencies, but to attend on our thoughts, and to follow them as a shadow follows the substance.

31. Yet in this kind of care we should set bounds to ourselves; for when our words are good Latin, significant, elegant, and properly arranged, why should we labor for anything more? But some speakers make no end of dissatisfaction with themselves, and of hesitating at almost every syllable; speakers who, when they have found the best terms, are anxious for something still more antique, far-fetched, and surprising; and who do not understand that in a speech of which the language is much extolled, the sense is too little regarded.

32. Let the greatest possible care, then, be bestowed on expression, provided we bear in mind that nothing is to be done for the sake of words, as words themselves were invented for the sake of things, and as those words are the most to be commended which express our thoughts best, and produce the impression which we desire on the minds of the judges. 33. Such words undoubtedly must make a speech both worthy of admiration and productive of pleasure; but not of that kind of admiration with which we wonder at monsters; or of that kind of pleasure which is attended with unnatural gratification, but such as is compatible with true merit and worth.


1. What the Greeks, then, call φρασις, we call in Latin elocutio, "elocution." We judge of it in regard to words taken either singly or in conjunction. In reference to words considered singly, we must take care that they be Latin, intelligible, elegant, and appropriate to that which we wish to express; in regard to words in conjunction, we must see that they be correct, well arranged, and diversified occasionally with figures. 2. What was necessary to be said, however, on the subject of speaking in pure Latin and with correctness, I stated in the first book, when I was treating on grammar.

But there I only observed that words should not be impure; here it will not be improper to intimate that they should have nothing provincial or foreign about them; for we may find many authors not deficient in the arts of style, who, we should say, express themselves rather affectedly than in pure Latin; as the Athenian old woman called Theophrastus, a man otherwise of great eloquence, a stranger, from observing his affected use of a single word, and being questioned on the subject, replied that she had discovered him to be a foreigner only from his speaking in a manner too Attic. 4. In Livy, again, a writer of extraordinary elegance, Asinius Pollio thought that a certain Patavinity was discoverable [I. 5, 56]. Let all our words, therefore, and even our tone of voice, if possible, declare us to be natives of this city, that our speech may appear truly Roman, and not merely to have been admitted into citizenship.


1. Perspicuity in words arises from a certain propriety; but the word propriety itself is taken in more than one sense; for its first acceptation signifies the exact term for a thing, which term we shall not always use; for we shall avoid such as are obscene, or offensive, or mean. 2. Mean terms are such as are beneath the dignity of a subject, or of the persons to whom we address ourselves. But in avoiding meanness some speakers are in the habit of running into a very great error, as they shrink from all terms that are in common use, even though the necessity of their subject calls for them; as he, for example, who, in pleading a cause, spoke of an Iberian shrub, of which he himself would alone have known the meaning, had not Cassius Severus, in derision of his folly, observed, that he meant to say Spanish broom. 3. Nor do I see why an eminent orator should have thought that duratos muria pisces, "fishes preserved in pickle," was more elegant than the very word which he avoided.

But in that sort of propriety, which uses the exact word for everything, there is no merit; though that which is contrary to it is a fault, and is called with us improprium, and in Greek ακυρον, "impropriety," as in Virgil [Aen. iv. 419], tantum sperare dolorem, "to hope so great pain." 4. Or the expression in the speech of Dolabella [Cicero was his instructor in the art of speaking, Ep. ad Div. ix. 16], which I have found corrected by Cicero, mortem ferre; or such as are now extolled by some people, decernere, verba ceciderunt. Yet a word which is not proper will not necessarily be chargeable with the fault of impropriety; because, above all, there are many things, both in Greek and Latin, that have no proper term. 5. He who hurls jaculum, "a javelin," is said jaculari, but he who hurls pilum, "a lance," or sudes, "a stake," finds no word peculiarly assigned to the act; and though it is manifest that lapidare means "to throw stones," the throwing of clods or tiles has no peculiar term. Hence what is called catachresis, the abuse of words, becomes necessary.

6. Metaphor, too, in which much of the ornament of speech consists, applies words to things to which they do not properly belong. Hence the propriety of which we are speaking, relates, not to a word absolutely, but to the sense in which it is used, and is to be estimated, not by the ear, but by the mind.

7. In the second place, when several things come under the same term, that is called the proper sense of the term from which all the other senses are derived; as the word vertex signifies water whirling round, or whatever is whirled round in a similar manner; hence, from the twisting round of the hair, it means the top of the head; and, from its application to the head, it came to signify the highest peak of a hill. We very rightly call all these things vertices, but properly that to which it was first applied. So it is with soleae and turdi, names of fishes. [The solea, "sole," is a flat fish, named from its similarity to the sole of the foot; the turdus was a fish found about rocks, and was so named from some resemblance to the thrush.]

8. There is also a third sort of propriety, the reverse of the second, when a thing, common to many purposes, has a peculiar sense as applied to one of them; as a funeral song is called noenia [noenia was properly any rude and trifling song, but was at length peculiarly restricted to the sense of funeral dirge] and a general's tent augurale. [Various things were done in the general's tent, but it had its name from augurium, "augury," because the auspices were taken there.] Also, a term which is common to many things, may be applied in a pre-eminent sense to some one of them; as we say "the city," for Rome, "venales" for newly-purchased slaves, and "Corinthian" for Corinthian brass; though there are many other cities, many other things to which venales may be applied, and there are Corinthian gold and silver as well as brass. But in such a use of terms there is no peculiar exhibition of the ability of the orator. 9. There is a kind of propriety, however, which is greatly to be admired, and for which anything is extolled that is said with peculiar effect, that is, with the utmost possible significancy; as Cato said that Julius Caesar applied himself soberly to overthrow the republic [Suetonius, in his Life of Julius Caesar, c. 53, says that he used to drink but little wine, and that Cato said he was the only man who applied himself soberly to overthrow the government]; or as Virgil [Ecl. vi. 5. Deductum means below the dignity of epic poetry] says deductum carmen, "a humble strain," and Horace acrem tibiam, "a shrill pipe," and Hannibalem dirum, "dire Hannibal." [Hor. Od. i. 12, 1; ii. 12, 2; iii. 6, 36.]

10. Under this kind of propriety is mentioned by some the appositeness of characteristic words, which are called epithets, as dulce mustum, "sweet new wine," and Cum dentibus albis, "with white teeth." [Virgil, Georg. i. 295; Aen. xi. 681.] Of this species of propriety I shall speak in another place. Terms that are happily applied in metaphor are also frequently called proper. 11. Sometimes, too, a term that is eminently characteristic of a person is called proper to him; thus Fabius, among his many military virtues, was called Cunctator, "the Delayer."

Words that signify more than they actually express, might seem to be fitly mentioned under the head of perspicuity, as they assist the understanding; but I would rather place emphasis among the ornaments of speech, because it does not merely tend to make what is said understood, but causes more to be understood than what is said.

12. On the other hand, obscurity arises from the adoption of words remote from common use; as, for example, if a person should search into the commentaries of the pontiffs, the most ancient treaties, and the writings of obsolete authors [comp. i. 6, 40; Cic. de Orat. ii. 12; Hor. Epist. ii. 1, 23], and make it his object that what he extracted from thence should not he understood. By such means some affect a character for erudition, endeavoring to prove themselves the only persons who comprehend certain subjects. 13. Words, too, that are more familiar to certain districts than to others, or peculiar to certain arts, produce obscurity, as, the wind Atabulus [In Apulia, Hor. Sat. i. 5, 77. Plin. H. N. xvii. 24; Senec. Nat. Quaest. iv.; Aul. Gell. xi. 22], the ship Saccaria [properly a vessel for conveying sacci, bags, filled perhaps, in general, with corn]; and In malaco sanum; such expressions must either be avoided before a judge who is ignorant of their meaning, or must be explained, as is the case with terms that are called homonyma; as with regard to the word Taurus, for example, it cannot be understood, unless it be specified whether it signifies a mountain, a constellation in the heavens, the name of a man, or the root of a tree.

14. Yet still greater obscurity arises in the construction and concatenation of words, and there are still more sources of it. Let our periods, therefore, never be so long that attention cannot sustain itself throughout them; nor so clogged by transpositions of phrases, that the end of the sense is not to be discovered till we reach the end of a hyperbaton. A still worse fault than these is a confused mixture of words, as in the verse,

Saxa vocant Itali mediis quae in fluctibus aras, [Aen. i. 109.]

Rocks which th' Italians altars call, amid
The waves.

15. By parenthesis, also, (which both orators and historians frequently use, to interpose some remark in the middle of a period, [Quintilian at once describes and exemplifies the parenthesis]) the sense is generally embarrassed, unless what is inserted be very brief. Thus Virgil, in the passage where he describes a young horse, after having said,

Nec vanos horret strepitus,

Nor dreads he empty noises,

and after having interposed some remarks in another form [Georg. iii. 79], returns, at the fifth verse following, to his first thought,

Tum si qua sonum procul arma dedere,
Stare loco nescit

Then, if but distant arms give forth a clang,
How to stand still he knows not.

16. But above all we must avoid ambiguity, not only that species of it of which I have spoken above, and which renders the meaning doubtful, as Chremetem audivi percussisse Demeam, but also that sort which, though it cannot perplex the sense, yet, as far as words are concerned, runs into the same fault with the other; for instance, if a person should say, visum a se hominem librum scribentem; for though it is certain that the book was being written by the man, yet the speaker would have put his words badly together, and rendered them ambiguous as far as was in his power.

17. In some writers, also, there are clouds of empty words; for while they shrink from common forms of expression, and are attracted by a fancied appearance of beauty, they involve all their thoughts, which they are unwilling to express straight-forwardly, in verbose circumlocutions, and joining one of these tissues of words to another of a similar character, and mixing up others with them, they extend their periods to a length to which no breath can hold out. 18. Some labor even to attain this fault, a fault by no means of recent date; as I find in Livy that there was a teacher in his day who exhorted his scholars to obscure what they said, using the Greek word σκοτισον: and from whom, I should suppose, proceeded that extraordinary eulogium, So much the better; even I myself cannot understand it.

19. Some, again, too studious of brevity, exclude from their periods words necessary, even to the sense; and, as if it were enough for themselves to know what they wish to say, are regardless how far it concerns others to understand them; for my own part, I call that composition abortive, which the reader has to understand by the exertion of his own ability [not through the perspicuity of the writer or speaker]; others, interchanging words perversely, secure the same fault through the aid of figures. 20. But the worst kind of obscurity is that which the Greeks call αδιανοητον, that is, when words that are plain in one sense, have another sense concealed in them; as Conductus est caecus secus viam stare ["A blind man was led (or hired) to stand by the way"]; and as he who tore his body with his teeth is represented in the schools, supra se cubasse, as having lain upon himself.

21. Such ingenious and daring phraseology is thought eloquent because of its ambiguity; and there is an opinion now prevalent with many, that they ought to think that only elegantly and exquisitely expressed which requires to be interpreted. But it is pleasing also to certain hearers, who, when they find out the meaning of it, are delighted with their own penetration, and applaud themselves as if they had not heard but invented it.

22. With me, however, let the first virtue of composition be perspicuity; let there be proper words, and a clear order; let not the conclusion of the sense be too long protracted; and let there be nothing either deficient or superfluous. Thus will our language both deserve the commendation of the learned, and be intelligible to the unlearned.

These observations refer to perspicuity in our words; for how perspicuity in our matter is to be secured, I have shown in my rules concerning the statement of cases. 23. But the case is similar with regard to both; for if we say neither less nor more than we ought, nor anything ill-arranged or indistinct, what we state will be clear, and intelligible even to the moderately attentive hearer. We must bear in mind, indeed, that the attention of the judge is not always so much on the alert as to dispel of itself the obscurity of our language, and to throw the light of his intellect on our darkness, but that he is often distracted by a multiplicity of other thoughts, which will prevent him from understanding us, unless what we say be so clear that its sense will strike his mind as the rays of the sun strike the eyes, even though his attention be not immediately fixed upon it.

24. We must, therefore, take care, not merely that he may understand us, but that he may not be able not to understand us. It is for this reason that we often repeat what we fancy that those who are trying the cause may not have sufficiently comprehended; using such phrases as, That part of our cause, which, through my fault, has been stated but obscurely, etc., on which account I shall have recourse to plainer and more common language; since, when we pretend, occasionally, that we have not fully succeeded, the admission is sure to be well received from us.


1. I come now to the subject of embellishment, in which doubtless, more than in any other department of oratory, the speaker is apt to give play to his fancy. For the praise of such as speak merely with correctness and perspicuity is but small; since they are thought rather to have avoided faults than to have attained any great excellence. 2. Invention of matter is often common to the orator and to the illiterate alike; arrangement may be considered to require but moderate learning; and whatever higher arts are used, are generally concealed, or they would cease to deserve the name of art; and all these qualities are directed to the support of causes alone.

But by polish and embellishment of style the orator recommends himself to his auditors in his proper character; in his other efforts he courts the approbation of the learned, in this, the applause of the multitude. 3. Cicero, in pleading the cause of Cornelius, fought with arms that were not only stout, but dazzling; nor would he, merely by instructing the judge, or by speaking to the purpose and in pure Latin and with perspicuity, have caused the Roman people to testify their admiration of him not only by acclamations, but even tumults of applause. It was the sublimity, magnificence, splendor, and dignity of his eloquence, that drew forth that thunder of approbation.

4. No such extraordinary commendation would have attended on the speaker, if his speech had been of an every-day character, and similar to ordinary speeches. I even believe that his audience were insensible of what they were doing, and that they gave their applause neither voluntarily nor with any exercise of judgment, but that, being carried away by enthusiasm, and unconscious of the place in which they stood, they burst forth instinctively into such transports of delight.

5. But this grace of style may contribute in no small degree to the success of a cause; for those who listen with pleasure are both more attentive and more ready to believe; they are very frequently captivated with pleasure, and sometimes hurried away in admiration. Thus the glitter of a sword strikes something of terror into the eyes, and thunderstorms themselves would not alarm us so much as they do if it were their force only, and not also their flame, that was dreaded. 6. Cicero, accordingly, in one of his letters to Brutus [now lost], makes with good reason the following remark: That eloquence which excites no admiration, I account as nothing. Aristotle [Rhet. iii. 2, 5; see also Longinus, sect. 1],  also, thinks that to excite admiration should be one of our greatest objects.

But let the embellishment of our style (for I will repeat what I said) be manly, noble, and chaste; let it not affect effeminate delicacy, or a complexion counterfeited by paint, but let it glow with genuine health and vigor. 7. Such is the justice of this rule, that though, in ornament, vices closely border on virtues, yet those who adopt what is vicious, disguise it with the name of some virtue. Let no one of those, therefore, who indulge in a vicious style, say that I am an enemy to those who speak with good taste. I do not deny that judicious embellishment is an excellence, but I do not allow that excellence to them. 8. Should I think a piece of land better cultivated, in which the owner should show me lilies, and violets, and anemones, and fountains playing, than one in which there is a plentiful harvest, or vines laden with grapes? Should I prefer barren pine-trees, or clipped myrtles, to elms embraced with vines, and fruitful olive-trees? The rich may have such unproductive gratifications; but what would they be, if they had nothing else?

9. Shall not beauty, then, it may be asked, be regarded in the planting of fruit-trees? Undoubtedly; I would arrange my trees in a certain order, and observe regular intervals between them. What is more beautiful than the well-known quincunx [see Virgil, Georg. ii. 277, 283; Cicero de Senect. c. 17; Xen.
Oecon. c. 4, sect. 20]
, which, in whatever direction you view it, presents straight lines? But a regular arrangement of trees is of advantage to their growth, as each of them then attracts an equal portion of the juices of the soil. 10. The tops of my olive, that rise too high, I shall lop off with my knife; it will spread itself more gracefully in a round form, and will at the same time produce fruit from more branches. The horse that has thin flanks is thought handsomer than one of a different shape, and is also more swift.

The athlete, whose muscles have been developed by exercise, is pleasing to the sight, and is so much the better prepared for the combat. 11. True beauty is never separate from utility. [See Xen. Mem. Soc. B. iii. c. 8.] But to perceive this requires but a moderate portion of sagacity.

What is of more importance to be observed, is, that the graceful dress of our thoughts is still more becoming when varied with the nature of the subject. Recurring to our first division, we may remark that the same kinds of embellishment will not be alike suitable for demonstrative, deliberative, and judicial topics. The first of these three kinds, adapted only for display, has no object but the pleasure of the audience; and it accordingly discloses all the resources of art, and all the pomp of language; it is not intended to steal into the mind, or to secure a victory, but strives only to gain applause, and honor.

12. Whatever, therefore, may be attractive in conception, elegant in expression, pleasing in figures, rich in metaphor, or polished in composition, the orator, like a dealer, as it were, in eloquence, will lay before his audience for them to inspect, and almost to handle; for his success entirely concerns his reputation, and not his cause. 13. But when a serious affair is in question, and there is a contest in real earnest, anxiety for mere applause should be an orator's last concern. Indeed no speaker, where important interests are involved, should be very solicitous about his words.

I do not mean to say that no ornaments of dress should be bestowed on such subjects, but that they should be as it were more close-fitting and severe, and thus display themselves less; and they should be, above all, well adapted to the subject. 14. In deliberations the senate expects something more elevated, the people something more spirited; and, in judicial pleadings, public and capital causes require a more exact style than ordinary; but as for private causes, and disputes about small sums, which are of frequent occurrence, simple language, the very reverse of that which is studied, will be far more suitable for them. Would not a speaker be ashamed to seek the recovery of a petty loan in elaborate periods? Or to display violent feeling in speaking of a gutter? [See Cicero Orat. c. 21. From the ancient writers it would appear that there were constant lawsuits about stillicidia. See Martial vi. 19; Lactant. de Fals. Relig. c. 1.] Or to perspire over a suit about taking back a slave? [Redhibitio took place in consequence of an aedile's edict, when a person had sold another an unhealthy or ill-conditioned slave, and the buyer required that the money should be returned to him, and the slave taken back.]

15. But let us pursue our subject; and, as the embellishment, as well as the perspicuity of language, depends either on the choice of single words, or on the combination of several together, let us consider what care they require separately, and what in conjunction. Though it has been justly said that perspicuity is better promoted by proper words, and embellishment by such as are metaphorical, we should feel certain, at the same time, that whatever is improper cannot embellish. 16. But as several words often signify the same thing, (and are called synonymous,) some of those words will be more becoming, or sublime, or elegant, or pleasing, or of better sound, than others; for as syllables formed of the better sounding letters are clearer, so words formed of such syllables are more melodious; and the fuller the sound of a word, the more agreeable it is to the ear; and what the junction of syllables effects, the junction of words effects also, proving that some words sound better in combination than others.

17. But words are to be variously used. To subjects of a repulsive character words that are harsh in sound are the more suitable. In general, however, the best words, considered singly, are such as have the fullest or most agreeable sound. Elegant, too, are always to be preferred to coarse words; and for mean ones there is no place in polished style. 18. Such as are of a striking or elevated character are to be estimated according to their suitableness to our subject. That which appears sublime on one occasion, may seem tumid on another; and what appears mean when applied to a lofty subject, may adapt itself excellently to one of an inferior nature. In an elevated style a low word is remarkable, and, as it were, a blemish; and in like manner a grand or splendid word is unsuited to a plain style, and is in bad taste, as being like a tumor on a smooth surface.

19. Some words are to be estimated, not so much by reason, as by taste; as in the phrase,

Caesa jungebant foedera porca.
[Aen. viii. 641. But the word porca occurs in Horace, Od. iii. 23, 4; also in Cato R. R. 134; and Cicero de Leg. ii. 22.]

which the invention of a word has rendered elegant; for if porco had been used, it would have been mean. In others the reason for their use is plain. We lately laughed, and with justice, at a poet who said

Praetextam in cista mures rosere camilli,

The dwarfish mice the gown within the chest-
Had gnawed;

20. but we admire the expression of Virgil,

Saepe exiguus mus,
[Georg. i. 181.]

Oft has the tiny mouse, &c.;

for the epithet exiguus, happily applied, causes us not to expect too much; the singular number, also, is preferable to the plural; and the monosyllabic termination, which is uncommon, gives additional beauty. Horace has accordingly imitated Virgil in both these points:

Nascetur ridiculus mus, [A. P. 139.]

A paltry mouse will be produc'd.

21. Our language Indeed is not always to be elevated, but sometimes to be depressed. Humility in our words sometimes gives of itself greater force to what we say. When Cicero, in speaking against Piso, exclaims, "When your whole family is drawn in a tumbril," does he not seem to have purposely adopted a mean word, and to have thrown, by the use of it, increased contempt on the man whom he wished to humble? And in another place he says, "You present your head to your adversary, butting with him." [A word properly used of rams. See Lucret. ii. 320. Compare also Virgil, Georg. ii. 526.] 22. From such sources come jokes that please the illiterate; as these in Cicero, "The little boy, that slept with his elder sister [Cicero pro Caelio, c. 15];" "Cneius Flavius, who put out the eyes of the crows [a proverbial expression, as appears from Macrobius, Sat. vii. 5];" and, in the speech for Milo, "Ho! you, Ruscio," [Pro Milone, c. 22. Ruscio is the name of a slave whom Cicero supposes to be under examination as to whether Clodius lay in wait for Milo] and in that for Varenus, Erutius Antoniaster, "Erutius a puny Antony."

Such humiliation of style is however still more remarkable in our school declamations, and, when I was a boy, such expressions as "Give your father bread," and, in reference to the same person, "You keep even a dog," used to be extolled [putting him to shame that kept a dog, but would not keep his father]. 23. But the practice, though frequently the cause of laughter, is dangerous, especially in the schools, and more than ever at the present time, when the exercise of declamation, being greatly at variance with reality, suffers from a ridiculous fastidiousness about words, and has excluded from its language a great portion of the Latin tongue.

24. Words are proper [that is, such as were in common use among all classes of people], newly coined, or metaphorical. To proper words antiquity adds dignity; for old words, such as every writer would not think of using, render language more majestic and venerable; and of this kind of ornament Virgil, an author of extremely fine taste, has pre-eminently availed himself. 25. The words olli, and quianam, and mis [for mei, genitive of ego, as appears from Eunius apud Priscian. lib. xiii], and pone, strike the reader of his poetry, and throw over it that authority of antiquity, which is so highly pleasing in pictures, and is unattainable by art.

But we must use such words with moderation, and not extract them from the remotest darkness of past ages. Satis is old enough; what necessity is there, I would ask, for substituting oppido, of which preceding writers, even in our own day, made use occasionally? I suspect that nobody would now allow us to use it; antigerio assuredly, of which the signification is the same, no writer would use, unless he wished to make himself remarkable. 26. What need is there for aerumnae, as if to say labor was not sufficient? Reor is repulsive; autumo just endurable; prolem ducendam fit only for tragedy; universam ejus prosapiam tasteless.

In short, almost all our language has undergone change. 27. Some old words, however, still appear more pleasing from their antiquity; some are at times adopted from necessity, as nuncupare and fari; and many others may be introduced with a little venturesomeness, provided that no affectation be apparent in the use of them; a fault which Virgil ridicules with wonderful effect in the following epigram:

28. Corinthiorum amator iste verborum,
Thucydides Britannus, Atticae febres,
Tau Gallicum, min, al, spinae male illisit.
Ita omnia ista verba miscuit fratri

[. .."So he mixed all those words for his brother."]

29. The person on whom it was made was Cimber, by whom it was signified in the words of Cicero, Germanum Cimber occidit, that his brother was killed. [A play on the word Germanus. The phrase will signify either "Cimber killed a German," or "Cimber killed his brother." Compare Vell. Pat. ii. 67. This Cimber was Titus Annius Cimber, not the assassin of Caesar. See Cicero Philipp. xi. 6.] Sallust is also attacked in an epigram equally well known:

Et verba antiqui multum furate Catonis,
Crispe, Jugurthinae condifor historiae

" And thou, O Crispus, the author of the history of Jugurtha; who hast plentifully stolen words from old Cato." 30. It is an offensive kind of affectation; it is easy to any one; and it is so much the worse, as he who indulges in it will not suit his words to his matter, but will seek extraneous matter to which his obsolete words may be applied.

To invent words, as I observed in the first book, has been more largely allowed to the Greeks, who have not hesitated to form words significant of certain sounds and impressions on the senses, using a liberty like that with which the earliest inhabitants of the earth gave appellations to things. 31. But our countrymen, though they have made some few attempts in composition and derivation, have scarcely attained full success in their efforts. I remember that when I was very young, it was a subject of discussion between Pomponius [this is doubtless tho Lucius Pomponius Secundus to whom Pliny
alludes, H. N. vii. 19, as a tragic poet, who had been consul]
and Seneca even in their prefaces, whether gradus eliminat, "advances his steps over the threshold," [Nonius Marcellus, i. 179] was a proper expression. But our forefathers did not hesitate to say expectorat [Cicero De Orat. iii. 38]; and exanimat is certainly of the same stamp.

32. Of derivation and formation there are such examples as beatitas and beatitudo in Cicero [De Nat. Deor. i. 34], which he himself indeed considers harsh, but thinks that they may grow less repulsive by use. Certain derivatives have been formed, too, not only from common words, but from proper names, as Sullaturit by Cicero [Ad Att. ix. 10], and Fimbriatum and Figulatum by Asinius Pollio. 33. Many new words have also been formed from the Greek, and a great portion of them by Sergius Flavius, of which some seem rather harsh, as ens and essentia; yet why we should so much dislike them, I do not see, except that we are disposed to be unjust judges in our own case, and suffer in consequence from poverty of language. 34. Some words of this kind, however, keep their ground; for those which are now old were once new; and some have but recently come into use, since Messala first used reatus, "the condition of a person under accusation," and Augustus Caesar munerarius, "belonging to presents or shows of gladiators."

My teachers were still in doubt whether piratica, "piracy," musica, "music," and fabrica, "the art of construction," could properly be used. 35. Favor, "favor," and urbanus, in the sense of "witty," Cicero thought new; for, in a letter to Brutus [now lost], he says, Eum amorem, et eum (ut hoc verba utar) favorem; in consilium advocabo, and in a letter to Appius Pulcher [Epist. ad Div. iii. 8], Te hominem, non solum sapientem, verum etiam (ut nunc loquimur) urbanum. Cicero also thinks that obsequium, "compliance," was first used by Terence [See Cicero de Amicit. c. 24]; and Caecilius says that the expression albenti coelo, "the heaven growing clear," originated with Sisenna. [It was doubtless Caecilius the Rhetorician that made this observation with regard to the historian Sisenna.] Hortensius appeals to have first used cervix in the singular [see Varro L. L. viii. 5; x. 4]; for the ancients have it in the plural.

We may, therefore, make attempts; for I do not agree with Celsus, who does not admit that words may be invented by the orator. 36. Since,— when some words, as Cicero says [Orat. Partit. c. 5], are primitive, that is, used in their original sense, and others derived, or formed from the primitive,— though it may not be allowable for us to coin new words, different from those which the first men, however barbarous, invented, yet at what time did it cease to be allowable to derive, vary, and compound words, a privilege which was surely granted to the immediate successors of the first men?

37. If we ever think, moreover, that we are coining a word too venturously, we may defend it with some apologetical phrase, as, that I may so express myself; if I may be allowed so to speak; in some way; permit me to use the word; a mode of excuse that may be serviceable when we use expressions which are too daringly metaphorical [Cicero de Orat. iii. 41. Longinus, sect. 32], and which can hardly be hazarded with safety; for it will thus be evident, from our very caution, that our judgment is not at fault. In regard to this point there is a very elegant Greek saying, in which we are directed προεπιπλησσειν τη υπερβολη, "to be the first to blame our own hyperbole." [Aristot. Rhet. iii, 7, 9.]

38. A metaphorical use of words cannot be commended except in the contexture of discourse.

Enough, then, has been said of words considered singly, which, as I showed in another place, have no beauty in themselves; yet they are not inelegant unless when they are below the dignity of the subject on which we have to speak; always excepting the expression of obscenities by their exact terms. 39. Let those attend to this remark, who think that obscene expressions need not be avoided, because there is no word indecent in itself [see Aristot. Rhet. iii. 2, 13], and because, as they say, whatever indecency there is in any act, the idea of it is still conveyed to the intellect under whatever other phraseology it may be veiled. For my own part, satisfied with the observance of Roman modesty, I shall, as I have already replied to such reasoners, vindicate decorum by silence.

40. Let us then proceed to consider the nature of connected discourse, the embellishment of which requires, above all, attention to two points; what language we conceive in our minds, and how we express it. In the first place, we must settle what we would wish to amplify or extenuate; what we would express vehemently or calmly, floridly or austerely, verbosely or concisely, roughly or mildly, grandly or simply, impressively or attractively. 41. We must also consider with what kind of metaphors or other figures, with what thoughts, in what style, and with what arrangement of matter, we may be likely to effect the object which we wish to accomplish.

But in attempting to show by what means a style may be rendered elegant, I shall first touch on the faults which are opposed to elegance; for the beginning of excellence is to be free from error. 42. We must first of all, then, not expect that a style will be elegant which is not appropriate. What Cicero [Orat. Partit. c. 6] calls appropriate is that kind of style which is neither more nor less in any respect than is becoming; not that it should not be neat and polished, (for that is a part of elegance,) but because wherever there is excess there is faultiness. 43. He would accordingly have authority in the words, and thoughts that are either impressive in themselves, or suited to the opinions and manners of the audience. For if these particulars be observed, we may adopt those forms of expression by which he considers that style is rendered ornate, select terms, metaphorical and hyperbolical phrases, epithets, repetitions, synonymes, and all such phraseology as is not unsuitable to the subject of our speech, or to the representation of things.

44. But since I have undertaken first to point out faults, let me observe that one sort of fault is that which is called κακεμφατον: whether the words which we use have by bad custom been distorted to an obscene meaning, as ductare exercitus [Plaut. Asin. i. 3, 12; v. 11, 13; Terent. Phorm. iii. 2, 15] and patrare bellum have been, by those who laugh, please the gods, at phrases which Sallust used in their pure and antique sense; (and I consider that the blame lies, therefore, not with writers but with readers; 45. yet such expressions are to be avoided, inasmuch as we have perverted pure words through corruption of morals, and we must yield even to prevailing vices;) or whether the junction of two words suggests by its sound something obscene, as, for instance, if we say cum hominibus notis loqui, unless the word hominibus be placed between cum and notis, we appear to fall into that which requires some prefatory excuse; for the last letter of the preceding syllable cum, which cannot be pronounced without the lips meeting together, either obliges us to pause most unbecomingly, or, if it be united with the following letter n, partakes of the objectionable sound of it. [See the same objection to saying cum nobis mentioned by Cicero Orat. c. 45.]

46. There are other junctions of words that produce a similar effect, but it would be tedious to specify them, and, in doing so, I should dwell upon the fault which I say should be avoided. Let me observe, however, that the division of a word sometimes gives the same offence to modesty; as in the use of the nominative case of intercapedinis. 47. Nor is such misrepresentation made of what is written only; for many readers will try, unless we are very cautious, to intimate that something of an obscene nature is suggested, (like him in Ovid,

Quoeque latent, meliora putat,
"Whate'er is hid, he more attractive deems,)

and to extract from words which are as free from indecency as possible, some reason for a charge of indecency. Thus Celsus finds the κακεμφατον in the words of Virgil,

Incipiunt agitata tumescere; [Georg. i. 357.]

but if we allow this to be the case, it is not safe to say anything.

48. The next fault to unseemliness of expression is that of meanness, which the Greeks call ταπεινωσις, and by which the greatness or dignity of a thing is depreciated, as Saxea est verruca in summo mantis vertice, "It is a stone wart on the top of the mountain's head." To this fault the opposite in nature, but equal in departure from judgment, is to apply to little things terms of extravagant meaning, unless to excite laughter be our object in doing so.

We should not, therefore, call a parricide a vicious man, nor a man attached to a harlot, a villain; for the former appellation expresses too little, and the latter too much. 49. From such errors in judgment composition is rendered dull or mean, or dry, or flat, or disagreeable, or slovenly; faults which are easily understood by reflecting on the opposite excellences; for the first is opposed to the spirited, the second to the elegant, the third to the rich, and the others to the cheerful, attractive, and correct.

50. We must also avoid the fault called μειωσις, "diminution," when something is wanting to an expression, so that it is not sufficiently full; though this indeed is rather a fault of obscurity than of neglect of ornament in style. But when diminution is adopted by writers designedly, it is called a figure, as is the case with ταυτολογια, "tautology," that is, the repetition of the same word or phrase. 51. The latter, though not wholly avoided even by the best authors, may yet be considered a fault; but it is one into which Cicero [Pro Cluent. c. 35] himself often falls, through inattention to such petty carefulness; as in the words, Non solum illud judicium judicii simile, judices, non fuit, "Not only that judgment, judges, was not like a judgment." Sometimes it is called by another name, επαναληψις, and is also numbered among the figures, of which I shall give examples in that part of my work where the beauties of style are to be noticed.

53. A worse fault than this is 'ομοιολογια, "sameness of style," which relieves the weariness of the reader with no gratification from variety, but is all of one complexion, by which it is fully proved to be deficient in oratorical art; and, from the tameness of its thoughts and figures of speech, as well as from the monotony of its phraseology, it is most disagreeable not only to the mind but also to the ear. 53. We must beware too of μακρολογια, that is, the use of more words than is necessary, as in Livy, Legati, non impetrata pace, retro domum, unde venerant, abierunt, "The ambassadors, not having obtained peace, returned back home, from whence they had come." But periphrasis, which is akin to macrology, is thought a beauty.

Another fault is πλεονασμος, "pleonasm," when a sentence is burdened with superfluous words, as I saw with my eyes; for I saw is sufficient. 54. Cicero humorously corrected a fault of this kind in Hirtius, who having said, in a declamation against Pansa [see Cicero, Epist. ad Div. ix. 18], that a son had been borne ten months by his mother in her womb, What, exclaimed Cicero, do other women bear their children in their cloaks? Sometimes, however, that kind of pleonasm, of which I gave an example just before, is used for the purpose of affirming more strongly; as,

Vocemque his auribus hausi, [Virg. Aen. iv. 359.]

And with these very ears his voice I heard.

55. But such addition will be a fault whenever it is useless and redundant, not when it is intended. There is also a fault called περιεργια, superfluous operoseness, if I may so express myself, differing from judicious care, just as a fidgety man differs from an industrious one, or as superstition from religion; and, to make an end of my remarks on this point, every word that contributes neither to the sense nor to the embellishment of what we write, may be called vicious.

56. Κακοζηλον, injudicious affectation, is a fault in every kind of style; for whatever is tumid, or jejune, or luscious, or redundant, or far-fetched, or unequal, may come under this term; all, indeed, that goes beyond excellence, all that is produced when imagination is not guided by judgment, and is misled by the appearance of some fancied beauty, may be characterized as affected; a fault which is the worst of all faults in oratory; for other faults are merely not avoided, but this is pursued. But it lies wholly in language.

57. Faults in matter are, that it is void of sense, or common, or contradictory, or redundant; corruption of style arises chiefly from the use or words that are improper, superfluous, or obscure in meaning, or from feebleness in composition, or puerile seeking for similar or equivocal expressions. 58. But all affectation is something false, though everything false is not affectation. To be affected in style is to speak otherwise than nature directs, or than is proper, or to use more words than are sufficient.

Language is corrupted in as many ways as it is improved. But on this head I have spoken more fully in another work [On the Causes of the Corruption of Eloquence]; it is noticed also frequently in this, and will be noticed occasionally hereafter; for, in speaking of ornament, I shall speak from time to time of such faults as border on excellences, and are to be avoided.

59. The following blemishes also spoil the beauty of composition: Want of proper arrangement, which the Greeks call ανοικονομητον: unskillful use of figures, which they call ασχηματον: inelegant junction of words or phrases, which they term κακοσυνθετον. Of arrangement, however, I have already treated; of figures and composition I shall treat hereafter. Another kind of fault which the Greeks notice is κοινισμος, the compounding of a style from different dialects; as, for example, if a writer should mix Doric, Ionic, and Aeolic words with Attic.

60. A fault similar to this in our writers, is to mix grand words with mean, old with new, such as are poetical with such as are common. This produces such a monstrosity as Horace imagines at the commencement of his book on the Art of Poetry,

Humano capiti cervicem pictor equinam
Jungere si velit

If to a human head a horse's neck
A painter chose to join,

and to add other parts from different animals.

61. Ornament is something superadded to perspicuity and propriety. The first two steps towards it consist in a vigorous conception and expression of what we wish to say; the third requisite is, to render what we have conceived and expressed more attractive, and this is what we properly call embellishment. Let us, therefore, number εναργεια, which I noticed in my directions respecting narration, among the ornaments of style, because distinctness, or, as some call it, representation, is something more than mere perspicuity; for while perspicuity merely lets itself be seen, εναργεια forces itself on the reader's notice.

62. It is a great merit to set forth the objects of which we speak in lively colors, and so that they may as it were be seen; for our language is not sufficiently effective, and has not that absolute power which it ought to have, if it impresses only the ears, and if the judge feels that the particulars, on which he has to give a decision, are merely stated to him, and not described graphically, or displayed to the eyes of his mind. 63. But as this art of depiction is contemplated by writers under several heads, I shall divide it, not indeed into all the parts which they specify, and of which the number is ambitiously augmented by some of them, but into the principal, on each of which I shall say something.

There is, then, one kind, by which the whole figure of an object is painted as it were in words:

Constitit in digitos extemplo arrectus uterque, [Aen. v. 426.]

Forthwith erect upon their toes both stood,

with the other particulars described, which set before us the appearance of the contending champions with such exactness, that it could not have been plainer even to the spectators themselves. 64. In this quality of style, as in all others, Cicero displays the highest excellence. Is any one so incapable of conceiving images of objects, that, when he reads the description in the oration against Verres [V. 33], The praetor of the Roman people, with sandals, with a purple cloak after the Greek fashion, and a tunic reaching to his feet, stood upon the shore leaning on a courtezan, he does not seem to behold the very aspect and dress of the man, and even to imagine for himself many particulars that are not expressed? 65. I, for my part, seem to myself to see his countenance, the look of his eyes, the repulsive dalliance of him and his mistress, and the tacit disgust, and shrinking modesty, of those who witnessed the scene.

66. Sometimes the picture, which we endeavor to exhibit, is made to consist of several particulars, as is seen in the same orator (for he alone affords examples of every excellence in embellishment) in the description of a luxurious banquet: I seemed to myself to see some entering, others going out, some tottering from the effects of wine, some yawning from yesterday's carousal. The ground was polluted, muddy with spilt wine, and covered with faded garlands and fish-bones. [From a speech of Cicero in defense of Gallius, when he was accused of bribery, as appears from Aquila Romanus. The speech is lost.] What more could a person who had entered the place have seen? 67. It is thus that commiseration for captured cities is excited; for though he who says that a city is captured, doubtless comprehends under that expression all the circumstances with which such a calamity is attended, yet this short kind of announcement makes no impression on the feelings.

68. If you expand, however, what was intimated in the single word, there will be seen flames spreading over houses and temples; there will be heard the crash of falling edifices, and a confused noise of various outcries; there will be seen some fleeing, and others clinging in the last embrace of their relatives; there will be the lamentations of women and children, and old men preserved by an unhappy fate to see that day; 69. there will be the pillaging of profane and sacred treasures; the hurrying of soldiers carrying off their booty and seeking for more; prisoners driven in chains before their captors; mothers struggling to retain their infants; and battles among the conquerors wherever the plunder is most inviting. For though, as I said, the idea of the city being taken includes all these circumstances, yet it is less impressive to tell the whole at once than to specify the different particulars; and the particulars we shall succeed in making vivid if we but give them a resemblance to truth. 70. We may also invent some circumstances, such as are likely to occur on such occasions.

A similar vividness will be given to description by the mention of adjuncts or consequences; as,

      Mihi frigidus horror
Membra quatit, gelidusque coit formidine sanguis,

      A shiv'ring horror stakes
My limbs, and my cold blood congeals with fear; [Aen. iii. 29]

Trepidae matres pressere ad pectora natos,[Aen. vii. 518]

Trembling mothers clasp'd
Their infants to their breasts.

71. To the attainment of this excellence, (an excellence, in my opinion, of the highest order,) the way is very easy. We must look to nature, and follow her. All eloquence relates to the transactions of human life; every man refers what he hears to himself; and the mind easily admits what it recognizes as true to nature.

72. To throw light upon descriptions similes have been very happily invented; some of which, as they strengthen proof, are numbered among arguments; others are adapted to give a lively representation of things; and it is this sort that is applicable to the present head of our subject:

      Inde lupi ceu
Raptores atra in nebula,
[Aen. ii. 355]

      Thence like rav'ning wolves,
In a dark mist,


Avi similis, quae circum litora, circum
Piscosos scopulos, humilis volat aequora juxta
, [Aen. iv. 254]

—Like the bird that flies
Around the shores, around the fishy rocks,
Low, near the sea.

73. In the use of this kind of illustration we must take the greatest care that what we introduce by way of similitude may not be obscure or unknown; for that which is offered as an illustration of something else, ought to be plainer than that which it is meant to illustrate. Similes of the following kind we may accordingly leave to the poets:

Qualis ubi hybernam Lyciam, Xanthique fluenta
Descrit, aut Delon maternam invisit Apollo
; [Aen. iv. 143]

As when Apollo wintry Lycia quits
And Xanthus' stream, or visits Delos' isle,
His birth-place.

74. It would not become an orator to demonstrate something plain by a reference to something obscure. But that kind of simile also, of which I spoke in treating of arguments, contributes to the ornament of style, and helps to render it sublime, or florid, or attractive, or striking. The more distant, indeed, is the subject from which any illustration is drawn, the more novelty it has, and the more surprise it causes.

75. Such as the following may seem common, adapted only to aid in enforcing conviction: As ground is made better and more fertile by culture, so is the mind by learning: and, As surgeons amputate limbs rendered useless by disease, so base and mischievous persons, though intimately allied to us by blood, must be cut off from our society. This from the speech for Archias [C. 8] is more sublime; Stones and deserts reply to the voice; fierce wild beasts are often moved, and stand still, at the song of the poet, etc.

76. This kind of similes is often greatly abused by the licentiousness of declaimers; for they adopt such as are false; and they do not apply them fairly to the things to which they wish them to seem applicable. An example of both faults is afforded in some that were everywhere repeated when I was a young man: Of great rivers the sources are navigable; and, The generous tree bears fruit as soon as it springs up.

77. In every comparison, either the simile precedes and the subject of it follows, or the subject precedes and the simile follows. But sometimes the simile stands by itself and is unconnected; sometimes, as is preferable, it is joined with the object of which it is the representation, resemblances in the one answering to resemblances in the other; an effect which what we call redditio contraria, and the Greeks ανταποδοσις, produces. 78. The simile which I mentioned just above, precedes the subject:

      Inde lupi ceu
Raptores atra in nebula;

that in the first book of the Georgics [Ver. 512], after the long complaint concerning civil and foreign wars, follows its subject:

Ut quum carceribus sese effudere quadrigae,
Addunt in spatia; et frustra retinacula tendens
Fertur equis auriga, nec audit currus habenas.

As when the chariots from the barriers start,
And speed athwart the plain; the charioteer,
Tight'ning in vain the curb, is borne away
By his own steeds, nor heeds the car the rein.

But this simile is without any ανταποδοσις. 79. Such mutual correspondence, however, brings under the eye as it were both objects of comparison, and displays them together. I find many noble examples of it in Virgil; but I must take them from the orators in preference. Cicero, in his speech for Muraena [C. 13], says, As they say that those, among the Greek musicians, who cannot become players on the lyre, may become players on the flute, so we see that those who cannot become orators betake themselves to the study of the law.

80. There is is another example in the same speech [C. 17], animated with a spirit almost of poetry, and with an antapodosis, which renders it more effective as an embellishment: For as tempests are oftentimes excited by the influence of some particular sign in the heavens, and oftentimes arise suddenly, without any assignable cause, and from some undiscoverable origin, so in regard to such a tempest of the people at the comitia, though we may often understand by what influence it has been raised, yet its origin is often so obscure, that it seems to have arisen without any cause at all. 81. There are also short similes of this kind; as, Wandering through the woods after the manner of wild beasts; and that of Cicero in reference to Clodius, From which trial he escaped naked, as from a house on fire. But similes like these will occur to the recollection of every one, from every-day conversation.

With this kind of simile is connected the power of setting a thing before the eye, not only with plainness, but concisely and quickly. 82. Brevity, indeed, to which nothing is wanting, is justly extolled, but that kind of brevity which says nothing more than is necessary, (the Greeks call it βραχυλογια, and it shall be noticed among the figures of speech,) is less deserving of commendation. Yet it is very happy when it comprises much in few words, as in the phrase of Sallust, Mithridates corpore ingenti, perinde armatus, "Mithridates, of vast stature, and suitably armed." But obscurity attends on those who attempt such conciseness injudiciously.

83. A beauty akin to the preceding, but of higher merit, is emphasis, which intimates a deeper meaning than the words used actually express. There are, however, two kinds of it; one which signifies more than is said; the other which signifies something that is not said. 84. Of the former kind there is a specimen in Homer [Odyss. xi. 522], where Menelaus says, that the Greeks descended into the horse; for by that one word he shows the vastness of the horse; and there is a similar specimen in Virgil,

Demissum lapsi per funem, [Aen. ii. 262]

Descending by a rope let down,

for thus also the height of the horse is signified. Virgil, too, when he says that the Cyclops lay stretched through the cave [Aen. iii. 631], measures the prodigious hulk of his body by the space of ground that it occupied.

85. The latter kind consists either in the entire suppression of a word in what we say, or in the omission of it at the close. As to the suppression of a word or thought, Cicero has given an instance of it in his speech for Ligarius [C. 5]. But if, Caesar, in your present height of power you had not so much clemency in your own disposition as you have; in your own disposition, I say; I know how I am expressing myself; for he suppresses that which we nevertheless understand, that there were not wanting men to incite him to cruelty. An omission at the close is by αποσιωπησις, which, as it is a figure, will be noticed in its proper place. 86. There is emphasis also in many common expressions; as, You must be a man; and, He is but a mortal [an excuse for faults]; and, We must live [we must make the
most of life]
. So like is nature in general to art.

It is not enough, however, for eloquence to set forth the subjects of discussion clearly and vividly; but there are many and various modes of embellishing language. 87. The αφελεια of the Greeks, "simplicity" pure and unaffected, carries with it a certain chaste ornament, such as is so much liked in women; and there is a certain pleasing delicacy of style that arises from a nicety of care about the propriety and significancy of words. Of copiousness there is one kind that is rich in thought, and another that abounds with flowers.

88. Of force there is more than one species; for whatever is complete in its kind, has its proper force. Its chief manifestation, however, is δεινωσις, "vehemence" in exaggerating an indignity; in regard to other subjects a certain depth; in conceiving images of things, φαντασια; in fulfilling as it were a proposed work, εξεργασια; to which is added επεξεργασια, a repetition of the same proof, or superabundant accumulation of argument.

89. Allied to these qualities is ενεργεια, which has its name from action, and of which the chief virtue is to prevent what is said from being ineffective. There is also a kind of bitter force, which is commonly employed in invective, as in the question of Cassius [Cassius Severus], What will you say when I shall invade your domain, that is, when I shall teach you that you do not know how to revile? A sort of sharp force, also, as in the saying of Crassus [Lucius Licinius Crassus, the celebrated orator, uttered these words in the senate, when he repulsed the lictor that Philippus sent to him. See Val. Max. vi. 2], Should I consider you a consul, when you do not consider me a senator?

But the chief power of an orator lies in exaggeration and extenuation. Each has the same number of expedients, on a few of which I shall touch; those which I omit will be of a similar character. 90. But they all have their sources in matter or in words. Of the invention and arrangement of matter, however, I have already treated; my present business is to show how expression may contribute to elevate or depress a subject.


1. The first mode of amplifying or extenuating, then, lies in the nature of the term which we apply to anything, as when we say, that a man who was beaten, was murdered; that one who is disingenuous, is a thief; or, on the other hand, that one who beat another, touched him, or that one who wounded another, hurt him. Of both there is an example in one passage of the speech for Caelius [C. 16]; If a woman, being a widow, lives freely; being bold, lives without restraint; being rich, lives luxuriously; being wanton, lives like a courtezan; should I, if a man salutes her somewhat familiarly, consider him as an adulterer? 2. For he calls a woman who was rather immodest, a courtezan; and says, that he who had been long connected with her, saluted her somewhat familiarly.

This sort of amplification becomes stronger and more remarkable, when the terms of larger meaning are compared with those for which we substitute them; thus Cicero says in his speech against Verres [I. 3], We have brought before your tribunal, not a thief, but an open robber; not a simple fornicator, but a violator of all chastity; not a person guilty only of sacrilege, but an open enemy to everything sacred and religious; not a mere assassin, but a most cruel executioner of our countrymen and allies. 3. By the first term much is signified; by the second still more.

I see that amplification, however, is effected chiefly in four ways; by augmentation, by comparison, by reasoning, and by accumulation.

Augmentation is most effective, when even things of which we speak as inferior to others, are made to seem of importance. This may be done either by one step or by several. By augmentation we reach, not only the highest point, but sometimes, as it were, beyond that point. 4. To exemplify all these remarks a single instance from Cicero [In Verr. v. 56] will suffice: It is an offence to bind a Roman citizen, a crime to scourge him, almost treason to put him to death; and what shall I say that it is to crucify him?

5. For had the Roman citizen only been scourged, Cicero would have exaggerated the guilt of Verres one degree, by saying, that even a less kind of punishment than scourging was an offence; and had he only been put to death, the guilt would have been aggravated by another degree; but after having said, that to put him to death was almost treason, a crime than which there is no greater; Cicero adds, what shall I say that it is to crucify him? When he had come to that crime, which is the greatest of all, words were necessarily wanting to express anything beyond it.

6. An advance, beyond what seems highest, may also be made in another way; as in what Virgil says concerning Lausus:

      Quo pulchrior alter
Non fuit, excepto Laurentis corpore Turni,
[Aen. vii. 649.]

      Than whom
Was none more beautiful, except the form
Of the Laurentian Turnus.

To say, than whom was none more beautiful, was to go apparently as high as possible, but something was afterwards added. 7. There is also a third way, in which we do not advance by steps, there being no more and most, but proceed at once to something than which nothing greater can be named: You killed your mother; what shall I say more; you killed your mother. For this is a kind of augmentation, to represent anything as so great that it cannot be augmented. 8. Language is amplified less evidently, but perhaps for that very reason more effectively, when, without any breaks, but in one continuous series and course, something always follows greater than what goes before; thus Cicero [Philipp. ii. 25] reproaches Antony with his vomiting, In an assembly of the people of Rome, when holding a public office, when master of the horse.

Every particular is an advance on that which precedes: To vomit from excessive drinking would have been of itself disgusting, even if not before a public assembly; it would have been disgusting before a public assembly, even if not of a whole people; before a whole people, even if not the people of Rome; even if he had held no office, or not a public office, or not that of master of the horse. 9. Another speaker might have distinguished these steps, and dwelt upon each of them; Cicero hastens to the summit at once, and gains it, not by climbing, but at the utmost speed.

But as this kind of amplification looks always to something higher, so that which is made by comparison seeks to raise itself on something lower. For by elevating that which is beneath, it must of necessity exalt that which is placed above. 10. Thus Cicero, in the passage just quoted, says, If this had happened to you at a banquet, and over those immense cups of yours, who would not have thought it disgraceful? But when it occurred before an assembly of the Roman people, etc. And in one of his speeches against Catiline [In Cat. i. 7], If, assuredly, my slaves feared me, as all your fellow-citizens fear you, I should think that I must quit my house.

11. Sometimes, by mentioning an instance of something similar, we may make that which we wish to exaggerate appear greater; thus Cicero, in his speech for Cluentius [C. 11], having related that a woman of Miletus had received a bribe from the heirs in reversion to cause abortion in her own person, exclaims, Of how much greater punishment is Oppianicus deserving for a crime of a similar nature? The woman of Miletus, in doing violence to her own body, tortured only herself; Oppianicus effected a like object by violence and torture to the body of another. 12. Nor let any one think that this sort of amplification, though of a like character, is the same with the mode of proceeding in regard to arguments, where the greater is inferred from the less; for in the one case to prove is the object, in the other to magnify; as, in regard to Oppianicus, the purpose of the comparison is not to show that he committed a crime, but that he committed a greater crime than another person.

13. In the two cases, however, though different, there is a certain affinity; and I shall therefore have recourse to the same example of which I made use in the other place, though not for the same purpose; for what I have here to show is, that, for the sake of amplification, not only a whole is compared with a whole, but parts with parts; as in this passage [Cicero in Cat. i. 1]: Did that illustrious man, and chief pontiff, Publius Scipio, kill, in his private character, Gracchus, when he was making only moderate changes in the commonwealth, and shall we consuls bear with Catiline, who is seeking to devastate the whole earth with fire and sword? 14. Here Catiline is compared to Gracchus; the commonwealth to the whole world; the moderate change to slaughter, fire, and devastation; a man in his private character with the consuls; and if a speaker should wish to dilate on these points severally, each would furnish ample matter for the purpose.

15. As to the amplifications which I said were made by reasoning, let us consider whether I designated them by a sufficiently appropriate term; though I am not indeed very anxious as to that point, provided that the thing itself be clear to those who wish to understand it. I have, however, adopted that term, because this sort of amplification is introduced in one place and produces its effect in another; so that one thing is magnified in order that another may be corroborated; and thence we arrive by reasoning at that which is the object of our amplification.

16. For instance, Cicero [Cicero Philipp. ii. 25], designing to reproach Antony with his wine-bibbing and vomiting, says. You, with such a throat, with such sides, with such strength in your whole body, fit for a gladiator, etc. What has the mention of the throat and sides to do with the intoxication? It is by no mean's without effect; for, looking to their capacity, we may estimate how much wine he swallowed at the marriage of Hippia, which he could not bear and carry off even with that strength of body fit for a gladiator. If, therefore, one thing is concluded from another, the term reasoning is neither improper nor extraordinary; and it is a term which I have introduced for the same reason among the states.

17. So likewise amplification arises from ensuing circumstances, as, in the case of Antony, such was the force of the wine bursting from him, that it produced no trifling effect, or inclination to vomit, but an absolute necessity of doing so, where it least of all became him; and the food which he cast up was not fresh, as sometimes happens, but such as remained in his stomach from the feast of the preceding day. 18. Circumstances that have preceded an act, too, lead to a similar conclusion; for when Aeolus, at the request of Juno,

   —cavum conversa cuspide montem
Impulit in latus; ac venti, velut agmine facto
Qua data porta, ruunt
, [Aen. i. 8.]

   —turn'd his spear, and struck
The hollow mountain's side, and forth the winds
Rush, as in banded throng, where'er a way
Was giv'n,

it is signified how great a tempest would follow. 19. Is it not amplification by reasoning, also, when we purposely extenuate the most atrocious crimes, (which we ourselves have previously represented as meriting the utmost detestation,) in order that the charges which are to follow may appear more enormous?

This is done by Cicero [In Verr. v. 44], when he said, These are but trifling charges against such a criminal. The captain of a vessel, from a most honorable city, purchased exemption from the terror of scourging with a sum of money; to allow him to do so was humanity in Verres. Another, that he might not be beheaded, sacrificed also a sum of money; this was but an ordinary occurrence. 20. Has not Cicero used amplification from reasoning, in order that the audience might estimate how enormous what was to be inferred must be, when such transactions, compared with it, were humane and ordinary? In this manner one thing is frequently enhanced by a reference to another; as when the merit of Scipio is magnified by dwelling on the military excellences of Hannibal; and when we extol the bravery of the Gauls and Germans, in order that the glory of Julius Caesar may appear the greater.

21. It is also a kind of amplification, when something is said of one thing with reference to another, with a view to which, however, it does not appear to be said. The chiefs of Troy thought it nothing discreditable that the Greeks and Trojans should endure so many calamities for so long a period for the sake of the beauty of Helen [Hom. Il. iii. 156]; how great, then, must that beauty be supposed to have been! It is not Paris, who carried her off, that says this; nor any young man; nor one of the multitude; but old men, the wisest of the people, the counsellors of Priam. 22. And even the king himself, exhausted by a ten years' siege, deprived of so many children, with utter destruction hanging over him, he, to whom it might have been thought that that face, which had been the cause of so many tears, would have been odious and detestable, not only listens patiently to this remark, but calling her "daughter," places her at his side, and even exculpates her, and denies that she is the cause of his misfortunes.

23. Nor does Plato, in his Symposium [Sect. 41], when he represents Alcibiades as confessing, on his part, how he wished to have been treated by Socrates, appear to have given this account in order to blame Alcibiades, but in order to show the incorruptible morality of Socrates, which could not be shaken even by the obvious advances of the most attractive of mankind. 24. It is thus, too, that the extraordinary stature of the ancient heroes is left to be inferred by us from the weapons which they used; as instances, may be mentioned the shield of Ajax [Il. vil 219]; and the spear of Achilles [Il. xvi. 140].

Of this kind of artifice Virgil has admirably availed himself in his description of the Cyclops [Aen. iii. 659]; for how huge must we conceive the body to be, the hand of which trunca pinus regit, "a pine-tree lopped of its branches supports?" How great also must have been the size of Demoleos, when two men, with their united efforts, could scarcely support his coat of mail on their shoulders, and yet he, clad in it,

cursu palantes Troas agebat, [Aen. v. 260]

The scatter'd Trojans at full speed pursued!

25. Cicero himself, again, could hardly have imagined anything so descriptive of the luxury of Mark Antony as he intimates when he says [Philipp. ii. 27], You might have seen the couches of slaves in their bed-rooms, decked with Pompey's purple quilts. Slaves in their bed-rooms use purple quilts, and those the quilts of Pompey; nothing stronger can be said; and yet we must consider that there was infinitely greater extravagance in the master than in his slaves. 26. This species of amplification is of a similar nature with what is called εμφασις: but the one suggests an inference from a word, the other from a thing; and the latter is as much more effective than the former as things are more impressive than words.

There remains to be noticed under amplification the accumulation of a number of words or thoughts having the same signification; for though they do not ascend by steps, yet they are heaped up, as it were, by coacervation. 27 What did your sword do, Tubero, that was drawn in the field of Pharsalia? At whose body was the point of it aimed? What was the object of your appearance in arms? To what were your thoughts, your eyes, your hands, directed? What ardor inspired your breast? What did you wish or desire? [Cicero pro Ligario, c. 3.] This is similar to what the Greeks call συναθροισμος: but in the Greek there is an amassing of many things; in the other figure there is an aggregation of particulars relating to one.

This kind of amplification is often produced by a series of words rising higher and higher in meaning; as [Cicero in Verr. v. 51], There was present the door-keeper of the prison, the praetor's executioner, the death and terror of the allies and citizens of Rome, the lictor Sextius.

28. The art of extenuation is nearly similar; for there are as many steps when we go up as when we go down. I shall content myself, therefore, with one example of it, taken from that passage where Cicero speaks thus of a speech of Rullus [Cicero de Lege Agr. ii. 5]: Some few, however, who stood nearest to him, suspected that he wished to say something about the Agrarian law. If this is considered to signify that the speech was not understood, it is extenuation; if that it was obscure, it is exaggeration.

I know that hyperbole may also be thought by some a species of amplification; for it either magnifies or diminishes; but as the meaning of hyperbole is larger than that of amplification, it must be reserved for consideration under the head of tropes. Of these I should proceed to treat at once, if they were not a form of speech distinct from other forms, consisting in words used, not in their proper, but in a metaphorical sense. Let me grant a little indulgence, therefore, to a desire which is almost universal, and not omit to speak of that ornament of style which most regard as the principal and almost only one.


1. The ancient Latins called whatever they conceived in the mind, sententia, "a thought." This acceptation of the word is not only very common among orators, but retains some hold of a place in the intercourse of ordinary life. For when we are going to take an oath, we speak ex animi nostri sententia, "from the thought of our mind," and when we congratulate our friends, we express ourselves ex sententia, "from our thought." [See Cicero de Orat. ii. 64.]

Not unfrequently, however, they spoke of uttering their sensa; as to the word sensus, it seems to have been applied by them only to the bodily senses. 2. But a custom has now become prevalent of calling the conceptions of the mind sensus, and those striking thoughts, which are introduced chiefly at the close of periods, sententiae. Such thoughts were far from being common among the ancients, but in our day are used to excess. I think it necessary for me, therefore, to say a few words concerning the different kinds of them, and the methods in which they may be used.

3. Though they all come under the same appellation, those that are properly called sententiae are the most ancient of their kind; the Greeks call them γνωμαι, and they received their name, both in Greek and Latin, from their similarity to counsels or decrees. The word is one of general meaning and reference; and a sententia may be deserving of praise in itself, without being applied to any particular subject. Sometimes it relates merely to a thing; as, nothing contributes so much to popularity as goodness; sometimes to a person, as that saying of Domitius Afer, A prince who would look into all things, must of necessity overlook many things.

4. Some have called it a part of an enthymeme [Aristot. Rhet. ii. 21, 2], and some the beginning or end of an epicheireme; and it sometimes is so, but not always. It is remarked with more truth, that it is sometimes simple, as in the two examples which I have just given; sometimes accompanied with a reason; as, For in all disputes, he that is the stronger, even though he receive the injury, appears, because his power is greater, to have inflicted it [Sallust, Jug. c. 10]; and sometimes double, as

Obsequium amicos, veritas odium parit,[Terence, Andr. i. 1, 41]

Obsequiousness makes friends, plain truth breeds hate.

5. Some have even made ten kinds, but in a way in which many more might be made, distinguishing them into sententiae of interrogation, of comparison, of negation, of similitude, of admiration, etc.; for a thought might thus have a name from every form of language. One of the most remarkable kinds is that which consists in an opposition of particulars:

Mors misera non est; aditus ad mortem est miser, [cited by Lactantius, iii. 17]

Death is not grievous, but th' approach to death.

6. Sometimes thoughts are enunciated in a direct manner:

Tam deest avaro quod habet, quam quod non habet, [from Publius Syrus]

The miser wants as much that which he has,
As that which he has not;

but they receive additional force from a change in the form of expression; as,

Usque adeone mori miserum est? [Virg. Aen. xii. 646]

Is it then such a grievous thing to die?

For this is more spirited than the direct expression, Death is not grievous. The same effect may be produced by the adaptation of a general sentiment to a particular case; thus Medea in Ovid, instead of saying in a direct manner, Nocere facile est, prodesse difficile, "It is easy to do harm, difficult to do good," expresses herself with more animation thus:

Servare potui: perdere ac possim rogas? [From Ovid's lost play of Medea.]

I have had power to save, and do you ask
Whether I can destroy?

7. Thus Cicero [Pro Lig. c. 12] makes a personal application of a common thought: Your height of fortune, Caesar, carries with it nothing greater than the power, and nothing better than the will, to save as many persons as possible; attributing that to Caesar which belonged properly to the circumstances in which Caesar was placed. But in the use of such sentiments, we must take care, as we must indeed with regard to all thoughts, that they be not too frequently introduced, or be evidently inapplicable, (as is the case with many that are used by some speakers, who call them καθολικα, and utter all that make for their cause as incontrovertible,) and that they be not employed everywhere, or put into the mouth of all characters indiscriminately. 8. For they are more suitable to persons of authority, whose character may give weight to what they say. Who indeed would listen patiently to a boy, or a young man, or a person of no estimation, if he spoke decisively, or uttered precepts with the air of a master?

9. Whatever we conceive in the mind, also, is an enthymeme, but that which is properly called so, consists of two thoughts in opposition, because it seems to be as pre-eminent among other thoughts as Homer among poets and Rome among cities. Of this enough has been said in the part where I spoke of arguments. [See Cicero, Topic, c. 13.] 10. It is not, however, always used for the purpose of argument, but sometimes merely for embellishment; as, Shall the language of those, Caesar, whose impunity is an honor to your clemency, incite you to cruelty? [Cicero Pro Lig. c. 4.] Cicero adds this question, not because it contains a new reason, but because it had been already shown, by other arguments, how unjust such conduct would be; 11. and it is subjoined at the close of the period by way of epiphonema, not as a proof, but as a triumphant blow to the adverse party; an epiphonema being, as it were, the concluding attestation to something already related or proved; as in Virgil,

Tantae motis erat Romanam condere gentem, [Aen. i. 33.]

Such was the task to found the Roman state!

Or as in the words of Cicero [Pro Mil. c. 4], The young man, being of honorable disposition, chose rather to incur danger, than to endure what was disgraceful.

12. There is also to be noticed what is called by the moderns νοημα: a term which may be taken as implying any thought whatever; but our rhetoricians have distinguished by it that which they do not express, but wish to be understood; as in what was said to the young man, whom his sister had several times redeemed when he had enlisted among the gladiators, and who brought an action against her under the lex talionis, because she had cut off his thumb while he was asleep, You deserved, exclaimed she, to have your hand whole; intimating, that he deserved to be all his life a gladiator.

13. What is called a clausula, too, requires to be mentioned; and if it be in the sense of what we term a conclusion, it is proper, and, in some places, necessary; as, You must, therefore, make confession concerning your own conduct, before you blame anything in that of Ligarius. [Pro Lig. c. 1.] But our modern speakers use it in another sense, and intimate that every thought at the conclusion of a period should fall pointedly on the ear. 14. They think it unbecoming, and almost a crime, to take breath at any passage which is not intended to call forth acclamations. Hence those small witticisms, uttered in bad taste, and forced into the service of the subject; for there cannot be as many happy thoughts as there must be conclusions to periods.

15. The following kinds of sententiae among the moderns may also be noticed. That which consists in something unexpected; as the retort of Vibius Crispus [V. 13, 48] to a man who, when he was walking about the forum in a coat of mail, pretended that he did so from fear: Who, exclaimed Crispus, has given you permission to be so much in fear? Or as the remarkable address of Africanus [Julius Africanus the orator is meant; see Tacitus Dial. de Orat. c. 15] to Nero on the death of his mother, Your Gallic provinces, Caesar, entreat you to bear your good fortune with firmness.

16. Others consist in some indirect allusion; as, when Domitius Afer was pleading for Cloantilla, whom Claudius had pardoned when she was accused of having buried her husband, who had been killed among the rebels, he remarked in his peroration, addressing himself to her sons, Nevertheless, young men, do not fail to bury your mother. [The sons themselves, therefore, appear to have accused their mother, and Domitius' admonition about filial piety appears to have been to this effect: Do not fear the displeasure of Caesar for burying your mother, for he loves affection, and has pardoned your mother for burying her husband contrary to the law; filial piety will please him more than your unnatural accusation.] 17. Some are aliunde petita, that is, transferred from one thing to another; as Crispus, in pleading for Spatale, whose lover, after making her his heiress, had died at the age of twenty-two, exclaimed, O youth of extraordinary forethought, who thus gratified himself!

18. A mere repetition makes some of these sententiae; as that of Seneca in the letter which Nero sent to the senate after killing his mother, wishing to make it appear that he had been in great danger, That I am safe, neither, as yet, do I believe, nor do I rejoice. But this is better when it is strengthened by an opposition of clauses; as, I know where to flee, but whom to follow I do not know. [Cicero ad Att. viii. 7.] Why need I add that the miserable man, though he could not speak, could not hold his peace? [Speech of Cicero against Piso. See letter of St. Jerome to Oceanus.] 19. It is most striking, however, when it is vivified by some comparison; as in the remark of Trachalus against Spatale, Do you desire, therefore, O ye laws, most faithful guardians of chastity, that tenth parts of estates should be awarded to wives, and fourth parts to mistresses? [By the lex Julia et Papia Poppaea, it was appointed that to a wife, who had no children, no more than a tenth part of her husband's estate should be bequeathed, and for every child that she might have surviving from a former marriage another tenth. See Jurisprud. Ante-Just. Schult. p. 609; Gothofr. ad Cod. Theod. tom. ii. p. 648.]

Of such kinds of sententiae, however, some may deserve to be called good, and some bad. 20. Those are always bad that are mere plays on words, as, Conscript Fathers, for I must commence thus, to remind you of what is due to fathers,—. A still worse kind, as it is more false and far-fetched, is such as that which was attributed to the gladiator, (whom I mentioned just above,) as a retort to his sister, I have fought to my finger. [See an epigram attributed to Martial, Spectac. xxix. 5.] 21. But perhaps the most execrable of all is when ambiguity in the words is joined with something that conveys a false notion as to the matter. I remember, when I was a young man, hearing a famous pleader, who had given a mother some splinters of bone, picked out of a wound in her son's head, merely for the sake of a sententia, exclaim, Unhappy woman, you have not yet conveyed your son to his funeral pile, and yet you have collected his bones.

22. Thus many delight even in the pettiest attempts at wit, which, if examined, are merely ridiculous, but which, when first produced, please the hearer with a show of ingenuity. For example, there is an imaginary case in the schools of a man who, having been shipwrecked, after being previously ruined by the barrenness of his grounds, hanged himself; and it is said of him, Let him whom neither earth nor sea sustains, hang in the air. 23. A similar witticism was made on the young man that I mentioned above, to whom, when he was tearing his flesh, his father gave poison: He who eats that, ought to drink this. To a luxurious man, also, who is said to have pretended a resolution to die by hunger, the admonition was offered, Make a rope for yourself; you have reason to be angry with your throat; or take poison; a toper ought to die drinking.

24. Some are mere inanity, as that of the declaimer exhorting the generals of Alexander to bury him in the ashes of Babylon, and exclaiming, I celebrate the obsequies of Alexander, and will any one behold them from the window of his house? as if the absence of spectators from windows were more to be deplored than anything else relating to the ceremony. Some are extravagant; as what I heard a speaker say of the Germans, I know not where their head is placed [this is generally supposed to refer
to the tallness of the Germans]
; and of a brave man, He repelled wars with his shield. 25. But there would be no end, if I were to attempt to enumerate all the species of tasteless witticisms.

Let us rather attend to a point which is of more importance. There are two different opinions respecting sententiae; that of those who set the highest value on them; and that of those who entirely reject them. With neither of these opinions do I exactly concur. 26. If brilliant thoughts are too crowded, they interfere one with another; as in crops of corn, and fruits on trees, nothing can grow to its just size that wants space in which to expand itself. Nor does a figure in a picture, which has no shade surrounding it, stand out in relief; and accordingly painters, when they combine several objects in the same piece, keep them distinct by intervening spaces, that shadows may not fall on the objects.

27. This pursuit of fine thoughts, also, makes style too curt; for every thought makes as it were a stand, as being complete in itself; and after it there must necessarily he the commencement of another sentence. Hence language is rendered too unconnected, and being composed, not of members, but of bits, has no proper construction; for these round and polished portions refuse to unite with each other. 28. The complexion, too, of the style, is variegated with spots, which, however brilliant, are of many and diverse hues; and, although a band and decorations of purple, put on a dress in their proper place, give a radiance to it, yet certainly a garment bedecked with various patches would be becoming to nobody.

29. However, therefore, such ornaments may seem to glitter and stand out, as it were, in composition, yet we may well compare them, not to the light of flame, but to sparks appearing among smoke; for they would not be noticed if the whole composition were luminous, any more than the stars are seen in the light of the sun; and the eloquence that tries to raise itself, as it were, with frequent little bounds, presents an unequal and broken surface to the view, neither gaining the admiration paid to lofty objects, nor exhibiting the attractions of level ground.

30. To this is to be added another evil, that the speaker who is always hunting for striking thoughts, must necessarily produce many that are trifling, vapid, and impertinent; for he can make no proper distinction where he is overwhelmed with numbers. Hence you may witness, among such orators, even the division of their subject set off with the air of a fine thought, as well as their arguments, if they be delivered at the close and fall of a period. 31. You, yourself an adulterer, have killed your wife; I could not have tolerated your conduct, even if you had but divorced her, is with them a mode of division; and, Would you be convinced that the philtre was poison? The man would be now alive, if he had not drunk it, is a form of argument. Most of them, indeed, may be said not to utter fine thoughts, but to utter everything as if it were a fine thought.

32. Some, again, make the contrary practice their study, shunning and shrinking from all such charms of composition, and approving nothing but what is plain, and humble, and without effort. Thus, while they are afraid that they may sometimes fall, they are always creeping on the ground. But what crime do they suppose that there is in producing a fine thought? Does it not strike the judge? Does it not recommend the speaker? 33. It is a fashion of speaking, they reply, which the orators of antiquity did not follow. How far back in antiquity, let me ask them, do they refer us? If to a remote period, Demosthenes produced many fine thoughts, such as no one had produced before him. If to a more recent period, how, let me ask, can they approve even Cicero, when they think that there ought to be no deviation from the manner of Cato and the Gracchi? Before the Cato and the Gracchi, too, there was a still plainer way of speaking.

34. For my own part, I consider such ornaments of style to be the very eyes, as it were, of eloquence; but I should not wish eyes to be spread over the whole body, lest other members should be obstructed in their functions; and, if I were compelled to make a choice, I should prefer the rudeness of the ancients to the affectation of the moderns. But a middle course is open between them; as, in our mode of living and dress, a certain elegance may be observed which is free from blame. Let us add, therefore, as far as we can, to the merits of our style; but let it be our first care to avoid faults, lest, while we wish to be better than the ancients, we make ourselves merely unlike them.

35. I shall now proceed to the consideration of tropes, which I mentioned as the next head of my subject. The illustrious orators of our times call them motus, "movements" [because they move a word, as it were, from its proper signification to another] or "changes." Rules concerning them the grammarians generally deliver; but when I was speaking of their duties, I delayed entering on this head, because, as it refers to the embellishment of style, it seemed to me that it would demand more attention, and that it should be reserved for a more important place in my work.


1. A trope is the conversion of a word or phrase, from its proper signification to another, in order to increase its force. Concerning tropes grammarians have carried on interminable disputes among themselves and with the philosophers; disputes as to what genera there are of them, what species, what number, and which are subordinate to others. 2. For myself omitting all such subtilties as useless to form an orator, I shall speak only of those tropes which are most important and most in use; and in regard to these, too, I shall content myself with observing, that some are adopted for the purpose of adding to significance, others for the sake of ornament; that some take place in words used properly, and others in words used metaphorically; and that tropes occur, not only in single words, but also in thoughts, and in the structure of composition. [Some tropes take place in single words, as metaphor, synecdoche, metonymy, &c.; others in thoughts or sentences, as allegory, periphrasis, &c.]

3. Those, therefore, appear to me to have been in error, who thought that there were no tropes but when one word is put for another; nor am I insensible, that in the tropes which are used with a view to significance, there is also embellishment; but the reverse is not the case, as there are some that are intended for embellishment only.

4. Let us commence, however, with that species of trope, which is both the most common and by far the most beautiful, I mean that which consists in what we call translatio, and the Greeks μεταφορα, "metaphor."

Metaphor is not only so natural to us, that the illiterate and others often use it unconsciously, but is so pleasing and ornamental, that, in any composition, however brilliant, it will always make itself apparent by its own lustre. 5. If it be but rightly managed, it can never be either vulgar, or mean, or disagreeable. It increases the copiousness of a language, by allowing it to borrow what it does not naturally possess; and, what is its greatest achievement, it prevents an appellation from being wanting for anything whatever. A noun or a verb is accordingly transferred, as it were, from that place in the language to which it properly belongs, to one in which there is either no proper word, or in which the metaphorical word is preferable to the proper. 6. This change we make, either because it is necessary, or because it adds to significance, or, as I said, because it is more ornamental. Where the transference produces no one of these effects, it will be vicious.

From necessity the rustics speak of the gemma [see Cicero de Orat. iii. 38; Orat. c. 24], "bud," of the vines, (for how else could they express themselves?) and say that the corn thirsts, and that the crops suffer. From necessity we say, that a man is hard, or rough, because there is no proper term for us to give to these dispositions of the mind. 7. But we say that a man is inflamed with anger, burning with desire, and has fallen into error, with a view to significance or force of expression, for none of these phrases would be more significant in its own words than in those adopted metaphorically. The expressions, luminousness of language, illustrious birth, storms of public assemblies, thunderbolts of eloquence, are used merely for ornament; and it is thus that Cicero [Pro Mil. c. 3] calls Clodius in one place a source, and in another a harvest and foundation, of glory to Milo. 8. Some things also, which are unfit to be expressed plainly, are intimated metaphorically, as,

Hoc faciunt, nimio ne luxu obtusior usus
Sit genitali arvo, et sulcos oblimel inertes
; [Virg. Georg. iii. 135]

"This they do, lest by too much indulgence the action of the genital field should grow too unenergetic, and obstruct the inert furrows." On the whole, the metaphor is a short comparison; differing from the comparison in this respect, that, in the one, an object is compared with the thing which we wish to illustrate; in the other, the object is put instead of the thing itself. 9. It is a comparison, when I say that a man has done something like a lion; it is a metaphor, when I say of a man that he is a lion.

Of metaphors in general there seem to be four kinds; the first, when one sort of living thing is put for another; as, in speaking of a driver of horses,

gubernator magna contorsit equum vi,

The steersman turn'd his horse with mighty force;

or as Livy [XXXVIII. 54] says that Scipio used to be barked at by Cato. 10. The second, when one inanimate thing is put for another; as,

Classique immittit habenas, [Aen. vi. 1]

He gives his fleet the reins.

The third, when inanimate things are put for things having life, as,

Ferro, non fato, maerus Argivum occidit,

By steel, not fate, the wall of Greece fell down;

and the fourth, when things having life are put for things inanimate,

   —sedet inscius alto
Accipiens sonitum saxi de
vertice pastor, [Aen. ii. 307.]

   The shepherd sits amaz'd,
Listening the sound from the high mountain's head.

11. From the last kind of metaphor, when inanimate things are exalted by a bold and daring figure, and when we give energy and feeling as it were to objects that are without them, extraordinary sublimity is produced; as in Virgil,

Pontem indignatus Araxes, [Aen. viii. 728]

Araxes that disdain'd a bridge;

12. in Cicero, What was your drawn sword, Tubero, doing in the field of Pharsalia? At whose body did its point direct itself? What was the meaning of your arms? Sometimes this beauty is doubled, as in Virgil,

Ferrumque armare veneno, [Aen. ix. 773.]

To arm the steel with poison,

for to arm with poison, and to arm steel, are both metaphors.

13. Those four might be distinguished into more species; as a word may be taken from one sort of rational animal and applied metaphorically to another, and the same may be done with regard to irrational animals; and, in like manner, we may apply a metaphor from the rational to the irrational, or from the irrational to the rational; and from the whole of a thing to a part, or from the part to the whole. But I am not now giving directions to boys, or supposing that my readers, when they understand the genus, cannot master the species.

14. But as a moderate and judicious use of metaphors adorns language, so a too frequent introduction of them obscures it, and renders the perusal of it fatiguing; while a continuous series of them runs into allegory and enigma. Some metaphors, too, are mean, as that which I recently mentioned, There is a wart of stone, etc. 15. Some are repulsive; for though Cicero uses the expression sentina reipublicae, "sink of the commonwealth," with great happiness, to signify a herd of bad characters, yet I cannot for that reason approve of the saying of an old orator, Persecuisti reipublicae vomicas, "You have lanced the ulcers of the commonwealth."

Cicero [De Orat. iii. 41] himself excellently shows that we must take care that a metaphor be not offensive; such as (for I will use his own examples) that the republic was castrated by the death of Africanus, or that Glaucia was the excrement of the senate; 16. that it be not too great, or, as more frequently happens, too little for the subject; and that it be not inapplicable; faults of which he who knows that they are faults will find numerous examples. But an excess, even of good metaphors, is vicious, especially if they be of the same kind. 17. Some are harsh, that is, based on a resemblance not sufficiently close, as "The snows of the head [Hor. Od. iv. 13, 12];" and,

Jupiter hibernas cana nive conspuit Alpes, [Hor. Sat. ii. 5, 41]

Jove o'er the Alps spits forth the wintry snows.

But the greatest source of error in regard to this subject is, that some speakers think whatever is allowed to poets, (who make it their sole object to please, and are obliged by the necessity of the metre to adopt many metaphorical expressions,) is permissible also to those who express their thoughts in prose. 18. But I, in pleading, would never say the shepherd of the people on the authority of Homer, nor speak of birds rowing with their wings, though Virgil [Georg. iv. 59; Aen. vi. 19], in writing of bees and of Daedalus, has used that phrase with great happiness; for a metaphor ought either to occupy a place that is vacant, or, if it takes possession of the place of something else, to appear to more advantage in it than that which it excludes.

19. What I say of metaphor may be applied, perhaps with more force, to synecdoche; for metaphor has been invented for the purpose of exciting the mind, giving a character to things, and setting them before the eye; synecdoche is adapted to give variety to language, by letting us understand the plural from the singular, the whole from a part, a genus from the species, something following from something preceding; and vice versa; but it is more freely allowed to poets than to orators. 20. For prose, though it may admit mucro, "a point," for a sword, and lectum, "a roof," for a house, yet it will not let us say puppis, "a stern," for a ship, or quadrupes, "a quadruped," for a horse. But it is liberty with regard to number that is most admissible in prose; thus Livy often says, Romanus praelio victor, "The Roman was victorious in the battle," when he means the Romans; and Cicero, on the other hand, writes to Brutus [in a letter now lost], Populo imposuimus et oratores visi sumus, "We have imposed on the people, and made ourselves be thought orators," when he speaks only of himself.

21. This mode of expression not only adorns oratorical speeches, but finds its place even in common conversation. Some say that synecdoche is also used, when we understand something that is not actually expressed in the words employed, as one word is then discovered from another; but this is sometimes numbered among defects in style under the name of ellipsis; as,

Arcades ad portas ruare; [Aen. xi. 142]

Th' Arcadians to the gates began to rush;

22. I consider it rather a figure; and among figures it shall be noticed. But from a thing actually expressed another may be understood; as,

Aspice aratra jugo referunt suspensa juvenci, [Virg. Ecl. ii. 66]

Behold the oxen homeward bring their plows
Suspended from the yoke,

whence it appears that night is approaching. I know not whether this mode of expression be allowable to an orator, unless in argumentation, when one thing is shown to indicate another. But this has nothing to do with elocution.

23. From synecdoche metonymy is not very different. It is the substitution of one word for another; and the Greek rhetoricians, as Cicero [Orat. c. 27] observes, call it υπαλλαγη. It indicates an invention, by the inventor, or a thing possessed, by the possessor. Thus Virgil says,

Cererem corruptam undis, [Aen. i. 117]

Ceres by water damag'd,

and Horace,

Terra Neptunus classes Aquilonibus arcet, [Hor. A. P. 63]

      Neptune, receiv'd
Within the land, from north winds shields the fleets.

The reverse [that is, if we were to say the sea for Neptune] would be offensive.

24. It is of great importance, however, to consider how far the use of the trope is permitted to the orator; for though we daily hear Vulcan used for fire, though it is elegant to say vario Marte pugnatum for the fortune of the battle was various, and though it is more becoming to say Venus than coitus, yet to use Bacchus and Ceres for wine and bread would be more venturesome than the severity of the forum would allow. Thus, too, custom permits us to signify that which is contained from that which contains it; as well-mannered cities, a cup was drunk, a happy age; but the opposite mode of expression scarcely any one would use but a poet; as,

   Proximus ardet
, [Aen. ii. 312]

Ucalegon bums next.

25. It may perhaps be more allowable, however, to signify from the possessor that which is possessed; as, a man is eaten up, when his estate is squandered. But of metonymy of this sort there are numberless forms. 26. We adopt it when we say that sixty thousand were killed by Hannibal at Cannae; when we say Virgil for Virgil's poetry; when we say that provisions, which have been brought, have come; that a sacrilege has been found out, instead of the person who committed it; and that a soldier has a knowledge of arms instead of a knowledge of the military art.

27. That kind of metonymy, too, by which we signify the cause from the effect, is very common both among poets and orators. Thus the poets have.

Pallida mors aequo pulsat pede pauperum tabernas, [Hor. Od. il 4, 13]

Pale death, with equal foot, knocks at the gate
Of poor man's cottage, &c.


Pallentesque habitant morbi, tristisque senectus, [Aen. vi. 275]

And pale diseases dwell, and sad old age;

and an orator will speak of rash anger, cheerful youth, and slothful inactivity.

28. The following kind of trope has also some affinity with the synecdoche. When I say vultus homiinis, "the looks of a man," I express in the plural that which is singular; yet I do not make it my object that one may be understood out of many, (for my meaning is evident,) but make an alteration only in the term. When I call, also, gilded ceilings golden ceilings, I deviate a little, as the gilding is but a part, from the truth. To notice all such expressions, however, would be too trifling an employment, even for those who are not forming an orator.

29. Antonomasia, which for a proper name substitutes something equivalent, is very common among the poets, and is sometimes effected by an epithet, which, when the name to which it is applied is set aside, is a sufficient substitute for it, as Tydides, Pelides, for Diomede and Achilles; sometimes by specifying some remarkable characteristic of a person; as,

Divum pater atque hominum rex,

The father of the gods and king of men;

sometimes by mentioning some act by which a person is distinguished; as,

   —Thalamo quae fixa reliquit
, [Aen. iv. 495]
   The arms which in the chamber fix'd
He, impious, left.

30. Among prose writers, though there is not much use of this phraseology, yet there is some; for, though they would not say Tydides and Pelides, yet they would say impius, by itself, for an impious man; and they do not hesitate to say the destroyer of Carthage and Numantia for Scipio, and the prince of Roman eloquence for Cicero. He himself has certainly taken such liberty: You do not commit many faults, said the old master to the hero; [Cicero pro Muraen. c. 29. Perhaps Chiron and Achilles are meant] where the name of neither is expressed, but both are understood.

31. Onomatopoeia, that is, the making of words, which was counted by the Greeks among the greatest merits, is scarcely permitted to us. Many words, indeed, were thus made by those who formed the language at first, with a view to adapt the sound to the impressions produced by the things signified; hence mugitits, "lowing," sibilus, "hissing," murmur, "murmur," had their origin. 32. But now, as if everything that was possible in that way had been accomplished, we do not dare to produce a new word, though many that were formed by the ancients are daily falling out of use.

We scarcely allow ourselves to venture on what are called παραγομενα, words that are derived, in whatever way, from others in common use; such as Sullaturio, "to desire to act like Sulla," proscripturio, "to desire to proscribe," and laureati postes, for "posts decked with laurel," are regarded as of the same nature. [Cicero ad Attic. ix. 10.] 33. The word evaluit was successfully introduced; but vio for eo "to go," was an unfortunate experiment. In regard to the Greek words obelisco coludumo, and others, we are forbidden to make harsh junctions, but we appear to look with satisfaction on septentriones. [Compounded of septem, "seven," and triones, "plow-oxen," according to Varro, Aulus Gellius, and others.]

34. The more necessary, therefore, is καταχρησις, which we properly call abusio, and which adapts, to whatever has no proper term, the term which is nearest; as,

   Equum divina Palladis arte
, [Aen. ii. 16. Aedifico properly means to build a house.]

A horse they build by Pallas' art divine;

and, among the tragic poets, Now a lion will bring forth; but a lion will be a father. 35. There are a thousand examples of the kind. Cruses are called acetabula [acetabulum was properly a cruse for vinegar], whatever they contain; boxes, pyxides [pyxis was properly a box made of boxwood; from πυξος], of whatever material they are made; and he who kills his mother or brother is called parricida. All these catachreses are to be considered distinct from the metaphor; for catachresis is used where a term is wanting; metaphor, where another term is in use. The poets are accustomed, even in speaking of things that have their own proper names, to use, catachrestically, proximate terms in preference; a practice which is rarely adopted in prose.

36. Some also will say that there is a catachresis when we use virtus for rash valor, or liberalitas for luxury; but such misapplications are distinct from the catachresis, for in them it is not one word, but one thing that is put for another; since no one thinks that luxury and liberality mean the same thing; but one calls the thing, whatever it is, luxury, and another liberality, though neither has any doubt about the distinctness of their signification.

37. Of tropes which modify signification, there remains to be noticed the μεταληψις, or transsumptio, which makes a way, as it were, for passing from one thing to another; a trope which is very rarely used, and is extremely liable to objection, but which is not uncommon among the Greeks, who call Chiron the Centaur, and term νησοι οξειαι, "sharp-pointed islands," θοαι, "swift." [Odyss. xv. 298.The islands are the Echinades. Both the Scholiast and Strabo observe that θοαι, "quick," is here used for οξειαιi, "sharp," in reference to the sharp-pointed promontories or rocks. Apollonius Rhodius uses θοος several times in the sense of sharp.]

Who would bear with us, if we should call Verres Sus, "Hog," or Laelius Doctus, "Learned?" 38. For metalepsis is of such a nature, that it is an intermediate step, as it were, to that which is metaphorically expressed, signifying nothing in itself, but affording a passage to something. It is a trope that we affect rather that we may seem to be acquainted with it, than one of which we ever stand in need. The most common example of it is, cano, canto, dico, canto being intermediate between cano and dico. 39. I shall dwell no longer upon it, for I see but little use in it, except, as I said, where one thing is to lead to another.

40. Other tropes are used, not for the sake of adding to significance, but for ornament, such as the επιθετον [this word is not to be taken in the sense of an adjective merely, but as signifying anything attached], which we rightly call appositum; by some it is termed sequens. The poets use it with more frequency and freedom than writers of prose; for it is sufficient for them that it suits the word to which it is applied; and we accordingly do not find fault with their albi dentes, "white teeth," and humida vina, "liquid wine."

But an epithet in a writer of prose, if nothing is added to the meaning by it, is a redundancy. Something is added to the meaning, if that which said is less without it, as, O abominable wickedness! O disgraceful licentiousness! 41. But ornamental epithets are most effective when they are metaphorical; as unbridled desire [Cicero in Catil. i. 10], mad piles of building [Cicero pro Mil. 20, 31]. The επιθετον [epitheton] is usually made a trope by the addition of something else to it, as, in Virgil, Turpis egestas, "base poverty," and Trlstis senectus, "sad old age."

But such is the nature of this ornament, that style, without epithets, appears bare and as it were graceless, yet is overburdened with them if they be too numerous. 42. For thus it becomes heavy and embarrassed, so that you would pronounce it, if used in pleadings, like an army with as many sutlers as soldiers, the number of which is doubled but not the strength. However not merely single epithets, but several together, are often used; as,

Conjugio Anchisa Veneris dignate superbo,
Cura deum, bis Pergameis erepte ruinis
, [Aen. iii. 475]

Anchises, with the stately honor grac'd
Of Venus' nuptial couch, of gods the care,
Twice from Troy's ruins rescued!

43. But, in this way, two words applied to one would not have much grace even in verse. There are some, however, who think that the epithet is not a trope, because it produces no change. [In allusion to the derivation of tropus from τρεπειν, " to turn."] Their reason is that an epithet, if it be separated from the word to which it belongs, must (if it be a trope) have some signification by itself, and form an antonomasia. Thus if we say, by itself, He who overthrew Numantia and Carthage, it is an autonomasia; if we add Scipio, an epithet; and an epithet, consequently, must always stand in conjunction with something else.

44. Αλληγορια, "allegory," a word which our writers interpret by inversio, presents one thing in words, and another in sense; or sometimes a sense quite contrary to the words. Of the first sort the following is an example,

O navis, referent in mare te novi
Fluctus? O quid agis? Fortiter occupa
, [Hor. Od. i. 14, 1]

"O ship, shall new waves bear thee back into the sea? O what art thou doing? Make resolutely for the harbor," and all that ode of Horace, in which he puts the ship for the commonwealth, the tempests of the waves for civil wars, and the harbor for peace and concord. 45. Similar is the exclamation of Lucretius,

Avia Pieridum peragro loca, [Lucret. iv. 1]

    I wander o'er
Th' untrodden regions of the Muses;

and the lines of Virgil,

Sed nos immensum spatio confecimus aequor,
Et jam tempus equum spumantia solvere colla
, [Virg. Georg. ii. 541]

"But we have gone over a plain vast in extent, and it is now time to unyoke the reeking necks of the horses." 46. But in the Bucolics he says without any metaphor,

Certe equidem audieram, qua se subducere colles
Incipiunt, mollique jugum demittere clivo,
Usque ad aquam, et veteris jam fracta cacumina fagi,
Omnia carminibus vestrum servasse Menalcam
; [Virg. Ecl. ix. 7]

"I had indeed heard that your Menalcas had preserved by his verses all those parts where the hills begin to recede, and to bend down their summit with a gentle slope, as far as the water, and the top of the old beech, now broken." 47. For in these verses all is expressed in unallegorical words, except the name, by which it is not the shepherd Menalcas, but Virgil, that is to be understood. Prose frequently admits the use of such allegory, but rarely pure; it is generally mixed with plain phraseology. It is pure in the following passage of Cicero: For I wonder, and am concerned, that any man should be so eager to destroy another by his words, as even to make a leak in the ship in which he himself is sailing.

48. Of the mixed, which is most frequent, this is an example [Cicero pro Milone, c. 21]: I indeed always thought that other tempests and storms were to be borne by Milo only amid the waves of popular assemblies; if he had not added only amid the waves of popular assemblies, it would have been pure allegory, but he has thus rendered it mixed. In this sort of language the beauty proceeds from the metaphorical words, and the intimation of the sense from the natural ones.

49. But by far the most ornamental kind of language is that in which the graces of the three figures comparison, allegory, and metaphor, are united. [Cicero pro Muraen. c. 17.] What sea, what Euripus, do you suppose to be affected with so many motions, such great and such various agitations, changes, fluctuations, as the disturbances and tumults which the proceedings of the comitia present? The intermission of one day, or the interval of one night, often throws everything into confusion, and sometimes the lightest breath of rumor changes the opinion of the whole assembly. 50. Care, as in this passage, is, above all things, to be taken, that, with whatever kind of metaphor we begin, we conclude with the same; but many speakers, after commencing with a tempest, end with a fire or the fall of a building; an incongruity which is most offensive.

51. Allegory is frequently used by the commonest minds, and in daily conversation. Those expressions in pleading causes, to set foot to foot, to aim at the throat, and to draw blood, are allegorical, and, though now so trite, are not displeasing. Novelty and variety in style are indeed pleasing; and what is surprising is, on that account, the more agreeable. But, in our pursuit of novelty, we have lost all sight of moderation, and have disfigured the beauties of style by excessive affectation.

53. There is allegory in examples, if they are not given with an explanation accompanying them; as, Dionysius is at Corinth, is a saying which all the Greeks use, and to which many similar might be mentioned. An allegory that is very obscure is called an enigma, which is, in my opinion, a fault in style, if to speak with perspicuity is a virtue. The poets however use it:

Dic quibus in terris, et eris mihi magnus Apollo,
Tres pateat coeli spatium non amplius ulnas
. [Virgil. Ecl. iii. 104]

"Say in what lands, and thou shalt to me be a great Apollo, the breadth of the sky extends not more than three ells." 53. Sometimes also orators; as Caelius says, Quadrantariam Clytaemnestram, et in triclinio Coam, et in cubiculo Nolam, "A farthing Clytaemnestra, a Coan in the dining-room, and a Nolan in the chamber." [These words were directed against Clodia the wife of Metellus, as appears from the speech of Cicero in defense of Caelius. She was called Clytaemnestra, as being supposed to have killed her husband.] Some such enigmas are now solved, and were, when they were uttered, easier to be interpreted; but they are enigmas nevertheless, and cannot be understood unless they are interpreted.

54. In the other kind of allegory, in which what is expressed is quite contrary to what is meant, there is irony, which our rhetoricians call illusio, and which is understood, either from the mode of delivery, the character of the speaker, or the nature of the subject; for if any of these be at variance with the words, it is apparent that the intention is different from the expression. 55. It is the case, indeed, with regard to most tropes, that it is requisite to consider what is said, and of whom, because it is doubtless allowable, as is observed elsewhere, to censure with pretended praise, and to praise under the appearance of censure. An example of the first is, Caius Verres, the city praetor, that upright and careful man, had no entry in his register of this second choosing of judges. [Cicero pro Cluent. c. 33] Of the second, We pretended to be orators, and imposed upon the people. 56. Sometimes it is with derision that the contrary to what we wish to be understood is uttered; as Cicero, in speaking against Clodius [a passage from a lost speech of Cicero], says, your integrity, believe me, has cleared you; your modesty has rescued you; your past life has saved you.

57. There is, besides, another use of allegory, in enabling us to speak of melancholy things in words of a more cheering nature, or to signify our meaning, for some good purpose, in language at variance with it; . . . . . . . . [hiatus]; these we have already specified. If any one does not know by what names the Greeks call them, let him be informed that they are termed σαρκασμος, αστεισμος, αντιφρασις, παροιμια. 58. There are, however, some rhetoricians who say that these are not species of allegory, but tropes; and they support their opinion by a very forcible reason; namely, that allegory is obscure, but that in all these modes of speaking what we mean is clearly apparent. To this is added the consideration that a genus, when distinguished into species, has nothing peculiar to itself, as tree is distinguished into pine, olive, cypress, etc., without retaining any peculiarity to itself; but allegory has something peculiar; and how could this be the case, if it were not itself a species? But whether it be a species or a genus is of no moment in respect to the use of it.

59. To the four forms just enumerated is to be added μυκτηρισμος, a kind of derision which is dissembled, but not altogether concealed.

When that is said in many words which might be said in one, or certainly in fewer, the Greeks call the figure περιφρασις, "a circuitous mode of speaking," which is sometimes necessary, especially when it veils what cannot be plainly expressed without offence to decency; as in the phrase of Sallust ad requisita naturae, "for the necessities of nature." 60. Sometimes its object is merely ornament, as is very common among the poets:

Tempus erat, quo prima quies mortalibus agris
Incipit, et dono divum gratissima serpit
, [Aen. ii. 268]

"It was the time at which the first sleep commences to weary mortals, and by the kindness of the gods spreads itself most gratefully." 61. It is also not uncommon among orators, but always of a more restricted nature; for whatever might be stated more briefly, but is for the sake of ornament expressed more fully, is περιφρασις, to which the Latin name circumlocutio has been given; a term not indeed very proper for designating a beauty of style. But as this figure, when it gives embellishment to language, is called periphrasis, so, when it has a contrary effect, it is termed περισσολογια, "redundancy of words," for whatever is not of service, is hurtful.

62. Hyperbaton, also, that is, verbi transgressio, "transposition of words," as the harmony and beauty of composition often require it, we rank, not improperly, among the excellences of language. For speech would often become rough and harsh, lax and nerveless, if words should be ranged exactly in their original order, and if, as each presents itself, it should be placed side by side of the preceding, though it cannot be fairly attached to it. 63. Some words and phrases must, therefore, be kept back, others brought forward, and, as in structures of unhewn stones, each must be put in the place which it will fit; for we cannot hew or polish them, in order that they may close and unite better, but we must use them as they are, and find suitable places for them.

64. Nor can anything render style harmonious, but judicious changes in the order of words. It was for no other reason [The Republic] that those four words in which Plato states, in the most noble of his works, that he had gone down to the Piraeus [see Diog. Laert. iii. 37; Dionys. Halicarn. vol. v. p. 209 ed. Reisk], were found written several ways on his tablets, than because he was trying to make order contribute as much as possible to harmony. 65. When hyperbaton takes place only in two words, it is called αναστροφη, or reversio, as mecum, secum, or as, among orators and historians, Quibus de rebus. But what properly takes the name of hyperbaton, is the removal of a word to a distance from its natural place with a view to elegance; as, Animadverti, judices, omnem accusatoris orationem in duas divisam esse partes; for in duas partes divisam esse was the natural order, but would be harsh and inelegant. 66. The poets, indeed, besides transposing words, also divide them; as,

Hyperboreo septem subjecta trioni; [Virg. Georg. iii. 381]

a liberty which prose does not tolerate. But the reason for which such a division of a word is called a trope, is, that the sense cannot be ascertained but by uniting the two separate parts; 67, otherwise, when no alteration is made in the sense, and the structure only is varied, it may rather be called a verbal figure; and many writers diversify their language by long hyperbata of this kind. What inconveniences arise from confusion of figures, I have noticed in the proper place.

To hyperbole, as being a bolder sort of ornament, I have assigned the last place. It is an elegant surpassing of the truth; and is used equally for exaggerating and extenuating. 68. It may be employed in various ways; for we may either say what is move than the truth, as, Vomiting, he filled his lap and the whole tribunal with fragments of undigested food [Cicero Philipp. ii. 25]; and,

      Geminique minantur
In coelum scopuli
, [Aen. i. 162]

Two rocks rise threateningly towards the sky;

or we exaggerate one thing by reference to another; as,

      Credas innare revulsas
, [Aen. viii. 691]

You would have thought the Cyclades uptorn
Were floating on the deep;

69. or by comparison; as,

Fulminis ocior alis, [Aen. v. 319]

Swifter than the wings
Of lightning;

or by something of a characteristic nature:

Illa vel intactae segetis per summa volaret
Gramina, nec cursu teneras Ioesisset aristas
, [Aen. vii. 808]

She o'er the rising tops of untouch'd corn
Would fly, nor in her course the tender ears
Would hurt;

or by a metaphor, as in the word volaret, "would fly.'' 70. Sometimes, too, one hyperbole is increased by the addition of another, as Cicero, in speaking against Antony [Philipp. ii. 27], says, What Charybdis was ever so voracious? what Charybdis, do I say? If such a monster ever existed, it was but one animal, but the whole ocean, by Hercules, would scarcely have been able, as it seems to me, to have swallowed up so many things, so widely dispersed, and lying in places so distant, in so short a space of time! 71 But I have noticed, as I think, an exquisite figure of this kind in Pindar, the prince of lyric poets, in the book [lost work] which he has called 'Υμνοι: for he says, that the impetuosity of Hercules in attacking the Meropes, who are said to have dwelt in the island of Cos, was comparable neither to fire, nor wind, nor the sea, but to lightning, as if other objects were insufficient, and lightning only suitable, to give a notion of his rapidity.

73 This Cicero [In Verr. v. 56] may be thought to have imitated, when he said of Verres, There arose in Sicily, after a long interval of time, not a Dionysius, nor a Phalaris, (for that island, in days of old, produced many cruel tyrants,) but a monster of a new kind, though endued with that ferocity which is said to have prevailed in those parts; since I believe that no Charybdis or Scylla was ever so destructive to ships in those seas as he was. 73. There are also as many modes of extenuating as of magnifying: Virgil says of a flock of lean sheep,

   —Vix ossibus hoerent, [Virg. Ecl. iii. 103]

They scarcely hang together by their bones.

Or, as Cicero says, in a book of jests,

Fundum Varro' vocat, quem possim mittere funda,
Ni tamen exciderit qua cava funda patet

But even in the use of the hyperbole some moderation must be observed; for though every hyperbole is beyond belief, it ought not to be extravagant; since, in no other way do writers more readily fall into κακοζηλια, "exorbitant affectation." 74. I should be sorry to produce the vast number of absurdities that have sprung from this source, especially as they are by no means unknown or concealed. It is sufficient to remark, that the hyperbole lies, but not so as to intend to deceive by lying; and we ought, therefore, the more carefully to consider, how far it becomes us to exaggerate that in which we shall not be believed.

It very often raises a laugh; and if the laugh be on the side of the speaker, the hyperbole gains the praise of wit, but, if otherwise, the stigma of folly. 75. It is in common use, as much among the unlearned as among the learned; because there is in all men a natural propensity to magnify or extenuate what comes before them, and no one is contented with the exact truth. But such departure from the truth is pardoned, because we do not affirm what is false. 76. In a word, the hyperbole is a beauty, when the thing itself, of which we have to speak, is in its nature extraordinary; for we are then allowed to say a little more than the truth, because the exact truth cannot be said; and language is more efficient when it goes beyond reality than when it stops short of it. But on this head I have here said enough, because I have spoken on it more fully in the book in which I have set forth the Causes of the Corruption of Eloquence.

Holy, Holy, HolyThe Philo LibraryHypatia's Bookshelf