Book III
Reading edition
Book III studies the speech itself. Book I having presented rhetoric from the view of the speaker, and Book II from the view of the audience, Book III now applies it directly to the speech.[36]
Since rhetoric must treat systematically three things: (1) what the means of persuasion are to be, (2) the diction,[37] (3) how to arrange[38] the parts of the speech, ... [the first has been discussed]. We have next to speak of the diction. For it is not enough to know what we ought to say; we must also know how we ought to say it, and this contributes much to the effect of the speech. The first subject of our inquiry [(1) above] was naturally that which comes first by nature, the facts themselves—in what aspects they are persuasive. The second is the expression of these in the diction. The third [not (3) above], which is of very great importance, is the delivery.
The threefold division sketched here seems at first sight to coincide, so far as it goes, with the one that afterward became traditional. Classical rhetoric as a whole assumes a fivefold division: (1) εὕρεσις, inventio, the gathering and analysis of the material; (2) τάξις, dispositio, collocatio, the arrangement, sequence, or movement in the large; (3) λέξις, elocutio, the diction, or the choice of words and their combination in phrases, clauses, and sentences, the movement in detail; (4) ὑπόκρισις, pronuntiatio, delivery, or “elocution”; (5) μνήμη, memoria, memory. But Aristotle’s division neither corresponds to this nor is consistent with itself. The first item is the same in both. Aristotle’s second item is clearly the same as the later third, and has the same name (λέξις, diction). The third item of his opening sentence seems equivalent to the traditional second (τάξις), and uses the corresponding verb (τάξαι); but below he makes his third item instead delivery (ὑπόκρισις), which is the fourth item of the traditional division, and then proceeds in the same chapter to include delivery, by implication, under diction.[39] In a word, the opening division of Book III is baffling. But the actual development of the book is quite clear: chapters i-xii on λέξις, diction, or, in the widest sense, style; chapters xiii-xix, on τάξις, or arrangement.
Delivery, after declaring it to have the greatest force (δύναμιν ... μεγίστην), he dismisses in a few sentences. Tantalizing in its brevity, this passage is nevertheless suggestive; for it sketches an analytic division of delivery into voice-placing and volume, pitch, and rhythm; it points to the value for public speaking of the arts of dramatic recital; and, most important of all, it relates delivery to the whole idea of style as concrete presentation versus abstract formulation.
In thus uniting delivery and diction as alike means of effective utterance Aristotle has seemed to some readers to disparage both. He has seemed to express, as in a similar passage of Book I,[40] a philosophic contempt for style. But this impression is not confirmed by scrutiny. Not only can he hardly be thought to despise that to which he devotes himself cordially throughout a large part of his treatise, but his words here hardly yield the inference that has been drawn from prejudicial translation. They may be rendered more precisely as follows:
An art [of delivery] is not yet settled; for even that of diction emerged but late and seems a bore when regarded ideally. But[41] since the whole practise of rhetoric is gauged to actual effect upon hearers (πρὸς δόξαν), we must give delivery our care, not of [abstract] right, but of necessity. [Abstract] justice, indeed, demands of our speech nothing more than that it should neither offend nor propitiate. For [abstractly] just [method] is so to make one’s plea with the facts that everything beyond exposition is superfluous. Nevertheless [delivery] is of great importance, as I have said, because of the human frailty of the hearer. Indeed, the consideration of the hearer is in a degree necessary in all teaching; for even in explanation it makes some difference whether we speak thus or thus—not so much, however [as in active persuasion], all these things [i.e., of diction and delivery] being means of suggesting images[42] and gauged to the hearer. Therefore no one thus teaches geometry.
That art, then [of delivery], when it comes, will produce the same effects as acting; and some authors, as Thrasymachus on the pathetic, have made a slight attempt at it. Acting is both a natural gift and less reducible to art; but diction has its technic. That is why those who have mastered it take prizes regularly, as do the histrionic orators;[43] for written speeches prevail more by diction than by thought.
“No one thus teaches geometry” cannot be taken as a slur on style. It simply reminds us, by applying to diction a distinction made with great fulness in the first two chapters of Book I, that rhetoric is not geometry. Formulation, as in geometry, is colorless because it is abstract; but any actual presentation, even mere information (ἐν πάσῃ διδασκαλίᾳ), demands style, whether concreteness or the arts of delivery, for mere lucidity (πρὸς τὸ δηλῶσαι), much more for any sort of appeal. For any sort of presentation, Aristotle is saying, we must study style; and we must include the study of delivery except in those written addresses which depend on style even more than on thought.
The appeal of style, Aristotle says in this same first chapter,[44] was first discerned by the poets. The method of suggestion, in other words, belongs to poetic. This is more than a critical distinction; it has an application directly pedagogical. The teaching of style, always delicate and difficult, may well begin through poetry, which in descriptive heightening and in harmony of sound with sense, especially of pace with mood, most plainly exhibits style as adaptation. The connection of this with delivery, both with reading aloud and with dramatic recital, though obvious, is often neglected. Elocution in our modern sense may, if rightly related, be one of the gateways to appreciation of style.[45]
Chapter ii,[46] after glancing at the fundamental virtue of lucidity, considers the choice of words for appropriateness and for suggestiveness, i.e., for their connotation, and especially for the descriptive vividness of the concrete, as in metaphors. Chapter iii[47] deals with the inappropriateness arising from bad taste;[48] Chapter iv,[49] with the extension of metaphor into simile. Chapter v[50] passes from single words and phrases to their combination in clauses and sentences. The distinction is important, and is kept throughout the classical rhetoric from Aristotle down. The choice of words is ἐκλογή (electio); the shaping of clauses and sentences is σύνθεσις (compositio). Frequent translation of the latter by our English word composition, which has a meaning so much wider as to be quite different, misses the specific point of technic, and often makes the ancient writers say what they did not mean.[51]
As the primary virtue in the choice of words is precision, so the first consideration in their combination, says Aristotle, is purity, idiom, conformity to usage. This assured, the next considerations of movement in detail are dignity,[52] which he presents as mainly amplification, and appropriateness.[53] Appropriateness here is not merely of the single word, as in Chapters ii and iii, but of the movement, or pace. It is gauged both to the moral habit (ἦθος) of the audience and to the emotion (πάθος) of the occasion.
From this general idea of appropriate movement Aristotle passes[54] to specific consideration of rhythm with his oft-quoted dictum:
The order[55] of the diction must be neither metrical nor unrhythmical. The former, by seeming artificial, is unpersuasive and at the same time distracting. For it makes us think of recurrences and wait for them to come, as children anticipate the answer to the heralds’ “Whom does the freedman choose as his attorney? Cleon.” On the other hand, the unrhythmical is immeasurable, and a measure we must have, though not by metrical recurrence; for the boundless can be grasped neither by the ear nor by the mind. Now measure in the most general sense is number;[56] and number as applied to the order of the diction is rhythm, of which meters are sections. Rhythm, therefore, the speech must have, but not meter, or it will be a poem—rhythm not too nice, that is, not carried too far.
Of the three rhythms the heroic is solemn and lacking in prose harmony.[57] The iambic is the very diction of the crowd; i.e., it is heard oftener than any other measure in speech, and it lacks capacity to lift and startle. The trochaic[58] is too suggestive of comic dancing, as is evident in trochaic tetrameters, which are a skipping rhythm. There remains the pæan.
Having laid down the principle that prose movement should be rhythmical, but not metrical, why does Aristotle proceed immediately to discuss it in terms of meters? Simply, perhaps, because these terms are familiar and definite. How else, indeed, shall we speak of a particular movement specifically? Perhaps also because the consideration of the larger, freer rhythms of prose is best opened through the fixed rhythms of verse, i.e., because meter is the gateway to appreciation of rhythm, as poetry in a wider sense (see page 24) is the gateway to style in a wider sense. For us moderns it is the more significant that the classical doctrine of clauses and sentences deals so largely with rhythm, since our doctrine throws the emphasis on logic. That the σύνθεσις, or compositio, should be idiomatic, dignified, appropriate, Aristotle has urged briefly; that it should be rhythmical he proceeds to set forth in detail, consecutively showing how.[59]
Thus that the diction should be rhythmical, not unrhythmical, what rhythms make it rhythmical, and in what modes, has been set forth.
Now diction[60] [in its sentence-movement] is connected either loosely and only by conjunctions, as the preludes in the dithyrambs, or compactly, as the antistrophes of the old poets. The former movement is the old one, as in Herodotus; for, though once universal, it is now exceptional. By calling it loose I mean that it has no end in itself except as its subject-matter runs out. It is unsatisfying to the ear by its indefiniteness, since we all wish to glimpse the end. That [natural desire] is why [runners] lose wind and heart only at the goal. They do not give out before because they are looking ahead to the finish.
The loose movement, then, of diction is this; the compact, on the other hand, is the one by periods [or definite units]. By period I mean a diction having a beginning and an end in itself and a length to be grasped as a whole. Such sentence-movement is both satisfying to the ear and easily followed by the mind;[61] satisfying as being the opposite of endless and as giving the hearer the sense of always having hold of something, because something has always been ended by itself, whereas the unsatisfying is neither to see ahead nor to get through; easily followed, as being easily held in mind, and that because periodic diction has number, which is the chief aid to memory. That is why verses are more easily remembered than loose prose, because verse has number to measure it. The period should also be completed with the sense, not broken off....
A period[62] is either in members or simple. Composed in members, it is such a diction as makes a rounded whole and yet is distinct in its parts, and such as the breath will carry easily, not by [arbitrary] division, but as a whole. A member is one of its parts; and by simple period I mean a period of one member. Both members and periods should be neither curt[63] nor long. For short [members] often make the hearer stumble, since while he is still surging ahead, if he is pulled up by the stopping of the measure that he carries in his head as a guide, he must stumble as in a collision. Long [members] on the other hand make the hearer feel himself left behind, as by walking companions making the turn beyond the usual stretch.... Over-use of short members, since it precludes the periodic form, drags the hearer headlong.
In other words, the period is a sentence movement forecast and fulfilled by the speaker, divined and held by the hearer, as a definite rhythmical and logical unit. Its characteristic is that conclusiveness which satisfies at once ear and mind. In sound and in syntax it is the opposite of formless aggregation, of the addition of clause to clause as by afterthoughts. Forethought, indeed, is its very note. Thus its typical advantages are rather for oratory than for narrative. Oratory moves by grouping around ideas; narrative, by adding image to image. The style of Herodotus is in this sense aggregative. Its aim being to proceed not from idea to idea in thought, but from fact to fact in time, it is “loosely joined,” “running on,” without other rhythmical value than fluency. That Aristotle means to disparage Herodotus when he calls this movement old and unsatisfying need hardly be inferred. Old it is typically, the movement of all early prose, of Herodotus no more than of Froissart and Villani. Unsatisfying, unpleasing to the ear (ἀηδές) it is not—in its place; but its place is not in oratory, which demands definite measures to mark definite stages of thought. Otherwise the audience is frustrated and loses the way.
What Aristotle means by his comparison of the two movements is that the former is unsatisfying, not absolutely, but for the purposes of the latter, i.e., for oratory. This interpretation is confirmed by what he adds concerning the length of members, or clauses. That staccato habit of short statements which in oratory “drags the hearer headlong,”[64] unsatisfied and uncomprehending, may in narrative be actually superior. To drag the hearer headlong is sometimes precisely what a story-teller desires. Examples abound, for instance, in Victor Hugo, for whom this movement became a mannerism. Neither of the two sentence movements, which from the point of view of Aristotle’s time we may call the historical and the oratorical, has remained through the long development of prose quite the same. Narrative has developed in modern times a movement more and more consciously poetic, while history in our special modern sense has turned more and more to the conscious group-movement which he associated with oratory. To-day we see much the same difference between our prose fiction and our expository history that he saw between Herodotus and Demosthenes. But the change is in application, not in the movements themselves. It remains true, and important, that there is on the one hand a prose movement rhythmically and intellectually loose, indefinite, and current, and on the other hand a prose movement compact, conscious, concluded point by point. The latter, the periodic, remains the typical movement of public address; for the audience, in order to follow, in order “to have hold of something[65] and to get something done,” demands definite measures.[66]
Having laid down as fundamental the distinction between the two typical prose movements, Aristotle proceeds to details: the balance of member against member,[67] and the heightening of the individual member[68] by visualizing metaphor. His recurrence here to metaphor is unexpected, since he has discussed this already in Chapter ii[69] under the choice of words; but here something is added. The connotation of figurative language is explored further as a means to make a whole statement telling. Aristotle is inquiring how such pithy sayings as he has just exemplified in balance and antithesis are made forcible[70] by other means; and he implies that the process is essentially poetic, as being imaginative first in realization and secondly[71] in movement.
Imaginative realization in metaphor and simile is considered here as intellectual suggestion. As enthymemes, so metaphors and similes must steer between the obvious and the subtle. The best images, like the best enthymemes, stimulate the hearer to coöperate, to see the relation for himself.
As to sense such are the popular enthymemes; as to style [they are popular] if the order, or movement[72] is antithetical.... So for the terms; if metaphorical, they must be neither far-fetched, for then they are hard to grasp, nor trite, for then they stir no emotion; and besides [as to movement] they must put [the thing] before our eyes; for we must see it in action rather than in intention (happening rather than about to happen, present rather than future). The essential elements, then, are three: metaphor, antithesis, actuality.[73]
what is meant by “before our eyes” and how this is to be done. I mean that those passages put the thing before our eyes which show it in action.[76] For instance, to say that a good man is “square” is metaphor ... but it does not show him in action, whereas “in flowering vigor” does, and so does “at large.” And in “Straightway the Greeks with bounding feet” the “bounding” is at once actuality and metaphor.... In all these [instances from Homer] by being alive (living, organic) [the subject] seems to be in action.... They all make the subject moving and living; and actuality is movement.[77]
Aristotle’s recurrence, then, to metaphor in the midst of his doctrine of sentence movement is because metaphor has the wider implication. It may be more than a single vivid word; it should extend to a whole habit of realizing a thing in action; and this involves expression in a sentence movement that shall heighten the suggestion by its pace.
In like manner the recurrence to aptness, or adaptation, is not repetition. Chapters ii[78] and iii[79] deal with aptness of single words; chapter xii,[80] with aptness of sentence movement. From the general definition[81] of apt movement as adaptation both to the moral habit of the audience (ἦθος) and to the emotion of the occasion (πάθος) Chapter xii now proceeds to the typical adaptations offered by the several fields of oratory: the deliberative, the forensic, and the occasional.[82]
It must not be forgotten that one style is appropriate to one kind [of oratory], another style to another. Style for writing is not the same as style for debate, nor style for public debate the same as style for legal pleading. Both [style for writing and style for debate] have to be known: the latter, as command of correct (or idiomatic) utterance; the former, as deliverance from the necessity of keeping silence when one wishes to communicate—[an inhibition] which those suffer who do not know how to write. Style for writing is the most precise; style for debate, the most histrionic (the best adapted to delivery). The latter is [adaptation] of two sorts: expression of character, and expression of emotion. This is why actors also seek such plays, and dramatists such personæ [as give expression to character and emotion].
The distinction here between style for writing (to be read aloud) and style for speaking (for immediate utterance) is general, as appears in the following reference to asyndeton and in the comparison of public speaking to the broad brush work of fresco; but it is also particular. It distinguishes occasional, or panegyric oratory as demanding a style more literary. “Style for writing,” above, must from its context refer to panegyric; and below Aristotle adds: “The style of occasional oratory is best suited to writing; for its function is to be read.”[83]
The final section[84] of Book III deals with the larger parts of a speech: exordium, statement of facts, proof, peroration. This discussion of τάξις (dispositio) is both meager and perfunctory, hardly more than a rehearsal of those definitions and counsels which were already familiar in teaching and apparently in manuals,[85] and which were to be handed on by later tradition. Its importance is therefore primarily historical. It has little other significance, little of the Aristotelian discernment and suggestiveness. What the modern teacher of rhetoric misses, both here and throughout the later classical discussions of dispositio, is some definite inculcation of consecutiveness. That consecutiveness was achieved in the best practise there can be no doubt; how it was taught we are left to guess. As to movement in this larger sense, what we commonly mean by composition, Aristotle’s Poetic is more definite and more suggestive than his Rhetoric.
He begins[86] by saying that the only essential parts of a speech are proposition and proof. It is presently apparent that by “parts” here he means components, or elements, of any and every sort of speech. The statement of facts, for instance, is not a part in the sense of a distinct division except in forensic; and Aristotle rightly objects[87] to subdivisions by “parts” which are neither distinct nor applicable generally. Even refutation, as he shows later,[88] is not a distinct part, either in function or in method or in place. The most that can be allowed are four: proposition and proof as essential, exordium and peroration as usual.
With the same common sense he shows that the first function of the exordium[89] is to put the hearers in a position to understand; its second, to win their sympathy.[90] Chapter xvi[91] passes to the recital of facts (διήγησις). The common rendering of this term by narrative has been widely misleading. True, the corresponding Latin term is narratio, and the thing is narrative in the sense of being sometimes, though not always, chronological; but narrative in our modern use, and especially in our modern text-books, is associated with objects and methods which Aristotle is not here considering at all, and which he rightly relegated to poetic. The Greek term διήγησις and the corresponding Latin narratio mean exactly what is called in a modern lawyer’s brief the “statement of facts,” as distinct from the following “argument.” It therefore belongs properly, as a distinct part having a distinct place, to forensic. When used in occasional oratory,[92] it should on the contrary be broken up, not as in forensic continuous. In deliberative oratory[93] it has least scope, i.e., it hardly appears as a separable part.
But the recital of facts, though it corresponds to the “statement” of a brief in substance, need not be so limited in style. A speech is not a brief; and the pleader, ancient or modern, must make his facts live.[94]
Speak also from the emotions, reciting what goes with them (i.e., their physical expression), both what is familiar and what is characteristic of yourself or your opponent: “He left me with a scowl”; or, as Æschines said of Cratylus, “hissing and shaking his fists.” [Such descriptive suggestions are really] elements of persuasion; for the familiar images become tokens of what you are trying to impress. Many such expressions are to be had from Homer: “So she spoke, and the old woman covered her face with her hands,” as we commonly put our hands to our eyes when we begin to weep.
As to persuasion by argument[95] Aristotle begins with a mere hint of that determination of the main issue and character of the case which was afterward elaborated into the classified doctrine of the στάσις (status).
Examples [παραδείγματα, he goes on] are more suited to deliberative oratory; enthymemes, to forensic.[96]... Do not speak in enthymemes seriatim, but mix them in [with other means of persuasion]; or they will impair one another. There is a quantitative limit.... Do not seek an enthymeme for everything; or you will write like some philosophers; ... and do not speak in enthymeme when your immediate aim is emotional ... or ethical.
The peroration[97] consists: (1) of disposing the hearer well toward oneself and ill toward one’s adversary; (2) of enhancing and disparaging; (3) of stirring the hearer to emotion; (4) and of recapitulation.... [For this last function] the primary idea is that what was promised has been given in full, so that we must tell both what [we have said] and why. This is told by comparison [of our own case] with our opponent’s.... Asyndeton[98] befits the final words, that they may be peroration, not oration: “I have spoken; you have heard; you have it; judge.”
With no less abruptness Aristotle’s Rhetoric stops. It can hardly be said to conclude; and certainly it has no peroration.