Rhetorical Theory: Enargeia and Memory

As has been noted in chapter 1, Latin rhetorical handbooks have a number of things to say on the subject of references to places and monuments that were not part of the visible milieu.^[1] We have seen that these works speak of the importance of considering the subjective and objective qualities of places when crafting a believable narration or a compelling argument. Quintilian's comment that "praise and blame" (5.10.41: commendationem . . . et invidiam ) could be derived from the qualities of a place also suggests that the Roman orator was encouraged to consider various aspects of places in making emotional appeals, although Quintilian goes into little detail in explaining exactly how this technique was to be carried out.^[2]

Of practical use to the orator who intended to speak at length about a place or monument neither seen by nor familiar to his audience was

[1] See above, pp. 19–25.

[2] As mentioned in chapter 1 (pp. 23–24), Quintilian gives two examples of such "praise and blame" connected to place (5.10.41), referring to a passage spoken by Ajax in Ovid's Metamorphoses (13.5–6) and to the charge of Milo's prosecutors in the Pro Milone (17, 18, 91). Also, according to the rhetorical concept of "decorum" or "propriety," the place where a speech was delivered was to be included among the considerations determining the style of a speech. Cf. Quint. 11.1.46–47 and similar remarks in Cic. De or. 3.210–11; Or. 123 (although Cicero does not refer specifically to place). See above, p. 19 n. 5.

the material in the rhetorical handbooks referring to a particular kind of description in which the speaker was advised to use concrete details in order to create a "visual image" in the minds of his listeners. As mentioned in chapter 1, the technique can be found under a variety of names, and it is often identified with ekphrasis, enargeia, hypotyposis, diatyposis, evidentia, repraesentatio, illustratio, demonstratio, descriptio, and sub oculos subiectio . Further, it is subsumed under a number of theoretical categories, including techniques of the narration and of the peroration, aspects of ornate style, and figures of thought.^[3] It seems clear from this proliferation of terms and categories of treatment that Latin rhetoricians and orators well understood the importance of using this kind of description but were unsure how to define it precisely and how to integrate it into the received structure of rhetorical theory. Concerning one point, however, there is no ambiguity. The successful employment of evidentia (to settle on one of the terms) caused the listener to picture what was described with "the eyes of the mind" (Quint. 8.3.62). The subject matter of such descriptions can be found listed in later rhetorical treatises, for the technique became one of the standard exercises (progymnasmata ) of the oratorical schools of the Empire.^[4]

[3] Lausberg, who notes the importance of using enargeia (or evidentia ) in the narration of a speech, categorizes it in the following way (Handbuch, 1:13–16): Evidentia is one of five figures of thought appealing to the emotions. Both figures of thought and figures of speech are subdivisions of ornamentation, which (along with clarity and Latinity) is a stylistic virtue. The stylistic virtues are, in turn, an aspect of style (elocutio ), the third of the five divisions of rhetoric. For a summary of ancient citations, see Lausberg, Handbuch, 1:399–407 (§§810–19). In Latin, see esp. Her. 2.49; 4.68; Cic. Part. or. 20; De or. 3.202; Inv. 1.104, 107; 2.78; Or. 139; Quint. 4.2.63; 6.2.32; 8.3.61–71; 9.2.40–44. Some sources define topographia as the description of a geographical location, and topothesia as the description of an imaginary place (Lausberg, Handbuch, 1:406–7, §819).

[4] For terms used in Quintilian and Ad Herennium, see above, p. 20 n. 6. As mentioned in chapter 1, Zanker ("Enargeia," 298) states that enargeia and ekphrasis are often used almost interchangeably, although he defines enargeia as a heightened form of ekphrasis or description. (See also Krieger, Ekphrasis, 7 n. 8: "In this broad assignment of its function, ekphrasis would seem to overlap, almost totally, the rhetorically encouraged virtue of enargeia, which is also defined as vivid description addressed to the inner eye.") The distinction between a more (enargeia ) or less (ekphrasis ) "ocular" description, barely perceptible in Greek texts, disappears amidst the profusion of terms found in Latin texts. I have, for the most part, avoided using the term ekphrasis as an equivalent for Latin evidentia or illustratio in this chapter since (a ) many scholars understand it to refer only to its later meaning—i.e., the literary description of a work of art (see Krieger, Ekphrasis, 7–8, on the narrowing of the term to include only objects from the plastic arts); and (b ) the term does not seem to have been current in Cicero's own time (see Zanker, "Enargeia," 305). For primary sources on ekphrasis, see Spengel, Rhetores Graeci, 2:16–17 (Hermogenes), 46–49 (Aphthonius), 118–20 (Aelius Theon); 3:251 (Georgios Choeroboscus), 491–93 (Nicolaus the Sophist). See also Downey, "Ekphrasis," esp. cols. 921–30; Norden, Kunstprosa, 1:285–86; Lausberg, Handbuch, 1:399–407 (§§810–19); Friedländer, Johannes von Gaza, 1–103; Zanker, "Enargeia"; Krieger, Ekphrasis . Norden's list of subjects of ekphraseis in Silver Age Latin prose includes "sunrises and sunsets, the calm sea and a pleasant journey,a grove, a charming house, especially a villa (temple, church) or painting, a city, a lovely girl, a splendid animal" or "horrid caves, the ocean and its terrors, . . . storm and shipwreck, torments, murder."

Subjects for vivid description included places (topoi ), times/seasons (chronoi ), persons (prosopa ), occasions/celebrations (panegyreis ), events (pragmata ), and, most familiar from poetry, statues and other artistic representations (agalmata, eikones ). As was the case with earlier treatment of the technique, however, no clear theoretical understanding of vivid description is to be found in the progymnasmata of the later Empire, and only its emphasis on "visualization" separated it from other, closely related, rhetorical techniques, such as narration or characterization.^[5]

Before dealing in more detail with the way evidentia was to be used in oratory, it is revealing first to consider the technique in light of the dominant ideas concerning perception and imagination in Cicero's day. While theories about these aspects of the mind varied among the major

[5] Since vivid description was understood to include the description of actions and events as well as of objects and people, no clear distinction can be drawn between it and narration. See the discussion of description as an aspect of narration in Genette, "Boundaries of Narrative," 5–8. Theoretical discussions of both enargeia and evidentia seem to combine the idea of stimulating an audience to visualize what is described (a potentially static tableau) with the idea of "actualization" (moving tableaux). See the discussion of the related technique of energeia in Arist. Rhet. 3.11.1–4 (1411b22–1412a9), where he speaks of bringing things "before the eyes" by the use of striking, often metaphorical, diction; and in Steiner, Colors of Rhetoric, 10–11. Cf. Krieger's discussion of the implications of ekphrasis, which—according to his definition—ultimately embraces "every attempt, within an art of words, to work toward the illusion that it is performing a task we usually associate with an art of natural signs" (Ekphrasis, 9). For Krieger the most fascinating but impenetrable aspect of ekphrasis consists in its attempt to force language into denying its temporality and creating a kind of stasis.

philosophical schools, a notion shared among them was that the mind was capable of receiving impressions in much the same way that soft wax receives the impression of a sealing stone.^[6] But much controversy revolved around the question of exactly what sort of mental impressions arose from various sources. Cicero addresses the topic in the Academica . In this dialogue, he depicts Lucullus speaking on behalf of the philosophy of Antiochus, who had abandoned the Skepticism of Carneades and claimed to have returned to the position of the so-called Old Academy. Lucullus states that it is possible for an individual to arrive at knowledge of what is true by means of the information received through the senses, and he bases this assertion on the idea that we can distinguish "true presentations" to the senses (visa vera ) from "illusionary" or "false images" (visiones inanes ) by means of the "clarity" (perspicuitas ) of the former. These true presentations, which arise from actually existing objects, can be "grasped" by the mind and form the basis both for memory and for accurate judgment.^[7] Cicero, defending the Skeptical position of the "New Academy" of Carneades, argues that false sensory data are, in fact, often indistinguishable from true. For instance, the dreamer, while dreaming, is aware of no mark that distinguishes the images he or she "sees" from what is real. Cicero goes on

[6] For the metaphor, see Plato Tht. 191c–d; Arist. Mem. 450a31; Theophr. De sens. 52; Sext. Emp. Math. 7.228. Hellenistic theories of sense perception can be divided into three groups. (1) According to the intromission theory of the Epicureans, thin films of atoms (eidola/simulacra ), produced continuously by every object, travel from an object to the eyes of an observer. These films are then capable of entering through the eyes and creating an image (phantasia ) of the object in the mind of the observer. (2) The extramission theory associated with Euclid and the mathematicians assumed that sight depends on visual rays sent out from the eyes and falling on the surfaces of objects. (3) The "mediumistic" theory, associated with the Stoics, stated that sight occurs when a visual pneuma (a mixture of air and fire) originating within the eye causes the air between observer and object to be stretched taut in the shape of a cone. The cone of stretched air, the base of which is constituted by the observed object, then carries back to the eye the aspect of that object.

[7] For Stoic theories of sense perception, see Hahm, "Early Hellenistic Theories of Vision"; von Staden, "The Stoic Theory of Perception"; Long, Hellenistic Philosophy, 123–31, and Problems, 91; Watson, Stoic Theory of Knowledge, 34–37. While judgment of the validity of various sense impressions is seen by the Stoics as volitional, nevertheless the clear and perspicuous presentation "virtually seizes us by the hair, dragging us off into assent" (Sext. Emp. Math. 7.257). For Aristotelian theories of perception, see Modrak, Aristotle: The Power of Perception .

to maintain that although it is impossible to distinguish absolutely whether sensory data are true or false, we may judge—and therefore act—on the basis of probable truth or falsity. He compares this process to a sea voyage. At the beginning of a voyage, the traveler aims at a certain destination and makes informed judgments on the basis of available evidence but cannot know for certain what the course of the voyage will actually be (2.100). Similarly, the careful application of rational judgment to the data provided by the senses can help us to make decisions concerning probable truth and falsity, although absolute certainty remains beyond our powers of discrimination.^[8] Lucullus had not only argued that true and false images could be distinguished; he had also contended that only a true, "graspable" presentation (phantasia kataleptike ) arising from actual objects was available to memory.^[9] Cicero takes issue with this statement as well, asserting that both images that have their origin in real objects and those that are merely illusory could be impressed upon the mind and thus made available to memory.

Although Cicero does not explicitly confront the question of what kind of impression images created through evidentia might make on the mind, it is clear from the Academica that, from the point of view of the audience, such images would be classified as "false" (visiones inanes ) rather than "true" (visa vera ), and therefore comparable to the impressions of dreams, hallucination, or the imagination.^[10] If we follow Ci-

[8] For Skeptical criticism of the Stoic account of perception, see especially von Staden, "The Stoic Theory of Perception," 112–27; Long, Hellenistic Philosophy, 90–97, 128–29.

[9] For the phantasia kataleptike, the "cognitive presentation" of the Stoics, see, for example, Sandbach, "Phantasia Kataleptike"; von Staden, "The Stoic Theory of Perception," 97–99; Long, Hellenistic Philosophy, 126–31; Long and Sedley, Hellenistic Philosophers, 1:241–53.

[10] "False" images include (a ) those that have no counterpart in the real world, such as that of a unicorn; (b ) those that appear to be presentations of objects that are not actually available to the senses at the time perceived, such as the image of an existing object seen in a dream or in a hallucination; and (c ) images of reality that do not accurately represent reality, such as a straight stick that appears bent in water. Cf. definitions of imago in Julius Victor (Halm, Rhetores Latini minores, 339.33–36: Ab imagine; est autem imago, cum ipsi aliquid simile effingimus, quamvis non sit in rerum natura, ut Marcus Tullius in Verrem: nam sicut Scyllam, inquit, nautis infestam accepimus ) and Macrobius (Sat. 4.5.9: et imago, quae est a simili pars tertia, idonea est movendis affectibus. ea fit cum aut forma corporis absentis describitur, aut omnino quae nulla est fingitur ), cited by Schrijvers, "Invention, imagination et théorie des émotions," 401.

cero's (rather than Lucullus's) arguments in the dialogue, we may suppose that verbally induced images, like other illusionary images, could potentially be indistinguishable from the actual in the impressions that they leave on the mind and, therefore, in their availability to processes of memory and feeling. This supposition is reinforced by the similarity of vocabulary used in passages dealing with the two kinds of impressions. When Lucullus argues that the "real" object creates an image that can be distinguished from the illusory, the terms he uses to describe the special, distinguishing characteristic of the real are "clarity" and "perspicuity."^[11] Even a philosophical Skeptic like Cicero, who argues in the dialogue that this apparent clarity could not serve as an unshakable criterion of the real, does not attempt to deny the affective power on the mind of the clear and perspicuous presentation. When we turn from philosophy to rhetorical theory, we find that it is precisely this quality of "perspicuity" that is the chief characteristic of vivid description: evidentia involves a mode of verbal expression that produces a clear, "visual" impression on the listener through "the eyes of the mind." It seems reasonable to infer, then, that ancient orators who, like Cicero, were familiar with both Greek philosophical and rhetorical theory might have seen in the technique of evidentia a verbal counterpart to the sensory reception of clear and striking images. While no one could claim that a listener might mistake the mental image created by verbal description for the visual experience of an existing object or objects, the use of the concept of clarity or enargeia to describe both visiones hints at the belief that these two avenues to the memory and emotions had a similarly potent effect.

[11] Enargeia (cf. Cic. Acad. 2.17: enargeia/perspicuitas/evidentia ) was defined by Epicurus as the quality of clarity in a sense impression that allows us to trust its reliability (Ep. Hdt. 52). The Stoics argued that the "cognitive presentation" (phantasia kataleptike ) could be distinguished from other presentations by the fact that it is "striking" (plektike: Sext. Emp. Math. 7.257, 258, 403) and "evident" (enarges: Sext. Emp. Math. 7.257, 403). Apparent clarity was, for the Skeptics, one—but only one—of the criteria that might lead one to conclude that a sense impression was probably—but not absolutely—true (cf. Long, Hellenistic Philosophy, 96–98). Zanker, "Enargeia," 308–10, traces the use of enargeia by the philosophers, particularly Epicurus, for whom it is the "indispensible condition for those sense impressions (phantasiai ) which alone can be trusted as evidence for the external properties of the senses." He suggests that Hellenistic literary critics borrowed the term from contemporary philosophical writings, especially Epicurean texts, where the connection between enargeia and sight is emphasized.

Of the connection of such verbally produced images with the emotions the ancient rhetoricians had little doubt.^[12] In a passage introduced by the statement that it was his intention to reveal to his readers the hidden secrets of his art (6.2.25), Quintilian asserts that in order to stir the emotions of others an orator must first be moved himself (6.2.27). He then explains how the orator may learn to generate feelings in himself:

There exist those things that the Greeks call phantasiai and that we term visiones, by means of which the images of absent things are represented in our minds in such a way that we seem to see them with our eyes as if they were actually present.^[13] Whoever is extremely responsive to these (6.2.30: has . . . bene conceperit ) will have the greatest power over the emotions. . . . In fact, we may acquire this power easily if we desire it.^[14] When our minds

[12] See, for example, Her. 2.49 (decimus locus ); 4.51, 69; Cic. Inv. 1.104 (decimus locus ); De or. 3.202; Quint. 6.2.32–34; 8.3.61–71; [Longinus] On the Sublime 15.1.

[13] As von Staden ("The Stoic Theory of Perception," 127 n. 11) remarks, "Phantasia . . . is a translator's nightmare." According to Lee, Plato believed phantasia to be "nothing less than the master-concept for our entire perceptual relation with the physical world" ("The Sense of an Object," 46). For Aristotle phantasia refers to a process or "movement" (kinesis ) by which images are presented to the mind. On the one hand, it is related to perception (aisthesis ), since all phantasmata originate in sense perception; on the other, it is related to judgment (hypolepsis ), which relies on it in forming opinions. It is, however, identical with neither perception nor judgment. (For Aristotelian phantasia, see, for instance, Most, "Seming and Being," 27–30, 32 n. 30; Rees, "Aristotle's Treatment of phantasia "; Engmann, "Imagination"; Schofield, "Aristotle on the Imagination"; Watson, "Phantasia in Aristotle," and Phantasia, 14–37; Modrak, Aristotle: The Power of Perception, passim.) Epicurus used the same word to describe an aspect of the mechanics of perception—the creation of an image in the mind by the reception through the eyes of successions of eidola (Epicurus Ep. Hdt. 49.6–50.4). The Stoics usually defined phantasia as the presentation in the psyche of an observed object, frequently (but not always—see Sext. Emp. Math. 8.56–57 [SVF 2:29, no. 88]) reserving emphasis, phantastikon, and phantasma for images of nonexisting objects or objects not present at the time of perception (SVF 2:21–22 [nos. 54, 55], 24 [no. 61]). For the Stoic phantasia kataleptike, the clear and striking presentation virtually demanding assent, see above, p. 93 n. 9. For citations of ancient evidence, translation, and commentary, see Long, Hellenistic Philosophers, 1:72–90; 2:75–93. See Watson, Phantasia, for a clear and useful review of evidence on phantasia from Plato to the Greek Neoplatonists.

[14] Quintilian here makes clear the difference between poetic phantasia, through which the poet is emotionally transported and is thereby able to depict vividly scenes beyond the realm of reality, and rhetorical phantasia, which— although rousing the emotions of the orator—is a rationally controlled exercise. One has the sense that, for Cicero and Quintilian, the successful orator can never be so moved by emotion that he does not constantly observe himself and the effect he is having on his audience. See Russell, Criticism in Antiquity, 108–10, and Schrijvers, "Invention, imagination et théorie des émotions," for differences between poetic and rhetorical phantasia implied by Quintilian and discussed by the author of On the Sublime (15.1–12).

are unoccupied and open to empty wishes and daydreams, the images about which I am speaking appear, so that we seem to be making a journey or sailing or fighting or addressing the people or spending riches we do not actually possess, and it seems to us that we are not thinking these things but acting. Why shall we not make practical use of this defect of our minds? Suppose that I am complaining that a man has been killed; shall I not keep before my eyes everything that well might actually have occurred? . . . Will not the blood, the pallor, the terrible groan, and finally the death rattle of the victim reside in my mind? From this follows enargeia, which Cicero calls illustratio and evidentia, by which things seem not so much to be said as to be shown; and our emotions are aroused no differently than if we were actually present at an event.

(6.2.29–32)

Quintilian begins with the assumption that false images (i.e., visiones inanes, which do not come from an object actually available to the senses at the moment they are perceived) often exist in the mind, although he makes no attempt to explain exactly how these images originate. In the ordinary person these simply occur when the mind is at rest (inter otia ). The orator, however, can train himself to produce particular images in order to bring about the emotional response that accompanies them. When he hopes to arouse sympathy in his audience for a murdered man he will first create that emotion in himself by imagining in vivid detail the scene of a pathetic and brutal attack. Quintilian does not say that he will think of what he knows did happen but, rather, will summon up a "believable" scenario for the murder (6.2.31: quae in re praesenti accidisse credibile est ). So far, Quintilian is speaking only of the orator and the process by which he could summon remembered images to his own mind. Then, however, he makes a crucial intellectual leap by connecting this process, which takes place within the mind of the orator, with his use of verbal descriptions in order to stimulate visualizations in the minds of his audience. The reader of the passage is therefore led to see the entire process as a connected sequence: the

speaker first summons images from his memory, where they are stored; if the orator is skillful and imaginative, these stimulate the particular emotional response that he had hoped to create in himself; the orator then, through vivid description, stimulates corresponding visiones in the minds of his audience; and these, in turn, produce a seemingly inevitable emotional reaction in the listeners. The process by which the mind of the orator is moved and that by which his audience is moved is, in essence, the same. A particular image (visio ), summoned to mind, sets in motion a predictable emotional response (pathos ).^[15] The difference lies chiefly in the control exercised by the orator, who—through conscious exercise of the imagination—creates the vision and, seemingly, is able to transmit it in words to his listeners, who then passively receive it.

Much of this appears fairly peculiar to the modern reader, and yet those aspects of the theoretical discussion of evidentia that appear most strange to us are perhaps most revealing of differences between Cicero's audience and ourselves. Implicit in these discussions of the images derived from experience of real objects is the idea that such images were able to produce a cohering and lasting physical impact on the mind by means of the eyes, as if the persisting image produced from staring several moments at a dark shape on a white ground were not a fleeting illusion but a lasting impression that could be retained by the mind and reproduced from memory. Even those phantasiai that manifestly originate within the mind, as in daydreams or sleep, were usually conceived of as pictures preserved by memory from previous sensory experience.^[16] This made mental images that had no corresponding existence in the real world somewhat difficult to explain, but a number of theories were evolved to deal with the problem; no explicit attempt was made, however, to explain the perceptual mechanics of evidentia, by which the orator, through words, produced images not previously perceived in the

[15] Cf. [Longinus] On the Sublime 15.1–2, where the process is understood in exactly the same way: "For the term Imagination (phantasia ) is applied in general to an idea which enters the mind from any source and engenders speech, but the word has now come to be used predominantly of passages where, inspired by strong emotion (enthousiasmou kai pathous) you seem to see what you describe and bring it vividly before the eyes of the audience " (trans. W H. Fyfe, Loeb Classical Library edition of Longinus [Cambridge, Mass., 1953]; italics added).

[16] See, for example, Arist. De insomn. 460b1–8; 461b22–23 (images in dreams are the residue of actual sense impressions).

minds of those he addressed.^[17] Most striking in the ancients' attempts to wrestle with such problems is their tendency to describe much of what goes on in the human mind as the creation or remembrance of coherent, retrievable pictorial images rather than as a manipulation of abstractions.^[18]

It is also to be observed that the accounts of evidentia or enargeia found in ancient sources assume an extraordinary degree of correspondence between words and images. The effect in such descriptions of

[17] Aristotle (De insomn. 461a15–25) describes how the confusion (through illness, for instance) of movements in the mind can produce monstrous and distorted images. For the Stoics as well, altered states (madness, illness, drunkenness) explained the propagation of certain kinds of false images in the mind. The Epicureans, with their intromissionist theory of perception, explained the ability to conceive of false images (in the sense of those mental images that did not correspond to any previous sensory perception of an actual object) by our reception of flimsy, composite films created by the accidental adhering of eidola from different objects (Lucr. 4.722–55). Quintilian's statement (8.3.64) that those who hear ekphrastic description imagine details beyond those specified by the orator suggests that each listener, stimulated by the words of the orator, creates images out of his or her previously existing store of remembered phantasiai . The emphasis in rhetorical ekphrasis on realism (e.g., Quint. 8.3.70; [Longinusl On the Sublime 15.8) also suggests that the orator's descriptions were meant to summon scenes with which all were familiar, whether from their own experience or from dramatic or artistic representations. The description of a captured city, for instance—a standard ekphrastic topic (Her. 4.51; Quint. 8.3.67–70)—would probably have summoned to the minds of an audience images from well-known paintings. Cf. Jos. Addison, The Spectator, no. 416, 27 June 1712: "When I say the ideas we receive from statues, descriptions or such-like occasions, are the same that were actually once in our view, it must not be understood that we had once seen the very place, action, or person, that are carved or described. It is sufficient that we have seen places, persons, or actions in general which bear a resemblance, or at least some remote analogy, with what we find represented; since it is in the power of the imagination, when it is once stocked with particular ideas, to enlarge, compound, and vary them at her own pleasure."

[18] Cf. Arist. De an. 3.8 (432a7–11), trans. W S. Hett, Loeb Classical Library edition: "No one could ever learn or understand anything without the exercise of perception, so even when we think speculatively we must have some mental picture of which to think; for mental images (phantasmata ) are similar to objects perceived except that they are without matter." Cf. L. Wittgenstein, Philosophical Investigations no. 396, trans. G. E. M. Anscombe (New York, 1953), 120: "It is no more necessary to the understanding of a proposition that one should imagine anything in connexion with it, than that one should make a sketch from it."

words as words, each freighted with emotional resonance for the listener, is scarcely acknowledged by ancient rhetorical theorists. Only in Quintilian (8.3.64) is there any recognition that the words of description, once filtered through the individual consciousness of the listeners, might stimulate different images, thoughts, and emotions in the mind of each individual. The supposition is rather that the words of the description are irrelevant except for their efficacy in creating for the listeners a mental picture of what is described; and, further, that essentially the same mental picture would be present in the mind of each listener. Without embarking on a full-scale review of the implications of evidentia for the ut pictura poesis thesis, we may simply note that these assumptions by ancient rhetoricians about the way vivid description worked in the minds of an audience suggest that ancient, nonliterate society may well have possessed powers of pictorial visualization much greater and more intense than our own.^[19]

Of what use was evidentia to the ancient orator? The material in the rhetorical textbooks emphasizing the effectiveness of the technique in rousing the emotions of an audience is useful, considering the crucial role played by emotional appeals in Ciceronian oratory, and yet unsatisfying, when we compare it with the complexity of Cicero's use in the Catilinarians of images the audience actually had before their eyes.^[20] As has been demonstrated, when Cicero directed his listeners to consider the cult statue of the Temple of Jupiter Stator in the first Catilinarian or the statue of Jupiter Optimus Maximus on the Capitolium in the third speech, in each case he aimed not simply at a straightforward appeal to the emotional impact of what the audience saw but at a complex series of thoughts and feelings dependent both on what he said and on what was before their eyes. This leads us to consider whether rhetorical theory might not suggest a more subtle and far-ranging employment of

[19] Cf. Yates, Art of Memory, 4: "Ancient memories . . . could depend on faculties of intense visual memorization which we have lost." There is an extensive bibliography on the relationship between verbal description and visual representation. See, for instance, Mitchell, Iconology (with full bibliography on the ut pictura poesis question); Brilliant, Visual Narratives, 73–76; Steiner, Colors of Rhetoric, esp. 1–69; Leach, Rhetoric of Space, esp. 3–24.

[20] The idea that such passages were emotionally powerful is extremely important, since it was Cicero's belief that of the three offices of oratory—to teach, to delight, to move—it was the third on which the success of a speech ultimately depended (see Cic. De or. 1.17, 53, 60; 2.115, 178, 215; Or. 128; Brut. 276, 279, 322).

the technique, analogous to strategies of persuasion observed in Cicero's exploitation of the visible milieu.

I believe that ancient rhetorical theory does indeed point the way to a broader use of vivid description through what it has to say about the art of memory. In the passage from the De finibus quoted in chapter 1, the experience of seeing a sequence of memorable places, each with vivid and moving associations for Cicero and his friends, had caused Piso to exclaim at "how great a power of association resides in places" (5.2: tanta vis admonitionis inest in locis ) and to recall that upon this fact depended the system of artificial memory (memoriae . . . disciplina ). According to this system, which was believed to have been created by the poet Simonides (the same poet, it should be remembered, who was also reputed to have called poetry "speaking painting"), an orator was taught to recall the various ideas or even words for his speech by associating them with images of things (imagines ) and to remember the order of these ideas by picturing the images in various loci or backgrounds.^[21] The imagines were compared to letters, the loci to writing tablets, and the latter were meant to be impressed on the mind in an unvarying sequence and reused on different occasions with a variety of images. The Ad Herennium suggests that the orator choose these backgrounds in locations that will not be confusing because of the coming and going of large numbers of people (3.31). Furthermore, they must be distinct, one from the other, and neither too close together nor too far apart, neither too bright nor too dim (3.31–32). The axially and hierarchically ordered rooms of the Roman atrium house, memorized in a fixed order, made an excellent series of mental loci . The Ad Herennium also notes that these backgrounds did not all have to be memorized through actual experience, since "thought" (3.32: cogitatio ) could provide an endless supply of suitable places.

What are the implications of this system of artificial memory for the use of evidentia? First, training in the ars memoriae not only accustomed the orator to create "visual" pictures in his mind but taught him to associate abstract ideas with these pictures. As Frances Yates has pointed out, "The word 'mnemotechnics' hardly conveys what the artificial memory of Cicero may have been like, as it moved among the

[21] For Simonides's statement, see Plut. Mor. 346f (cf. 17f; 58b). On the art of memory, see Quint. 11.2.1–51; Cic. De or. 2.351–60; Her. 3.28–40; Lausburg, Handbuch, 1:525–27 (§§1083–90); Yates, Art of Memory, 1–49; Caplan, "Memoria"; Kennedy, Art of Rhetoric, 123–26.

buildings of ancient Rome, seeing the places, seeing the images stored on the places, with a piercing inner vision which immediately brought to his lips the thoughts and words of his speech."^[22] And, just as Quintilian had connected the orator's practice of visualizing a scene in order to rouse his own emotions with that same orator's effort to induce his audience also to visualize emotionally charged scenes, so it seems likely that, in this instance as well, a speaker would realize that a technique he had employed as part of the private propaedeutic of speech making could also be used to manipulate the thoughts and feelings of his audience. Thus Cicero may well have understood that the mnemonic technique by which he impressed ideas on his own mind could be employed to impress concepts on the minds of his listeners.^[23] The first step in this process would be to create a series of distinct and memorable settings in the minds of his listeners. This he could do in the same way he himself had learned to keep in mind "backgrounds": either by direct visual experience or by the creation in the mind's eye of vivid images of various locations. On the one hand, he could make use of the familiar monuments and buildings visible to the audience in the Forum, leading them to connect with these loci symbolic values supporting his rhetorical goals. We have seen how in the Pro Scauro, for instance, Cicero directed the attention of his audience to the key monuments of the Forum and tied to each a meaning relevant to the status and character of the accused. On the other hand, the orator could induce his audience to imagine a locus through his employment of vivid description. This locus and the images created within it could then be used by the orator not simply to elicit an emotional reaction but at the same time to trigger associations with various larger and more complex ideas. The ars memoriae, therefore, is able to provide us with the theoretical link between the use of evidentia to create emotionally potent images in the minds of listeners and the much more far-reaching use of the same technique to attach symbolic values to those images. At the same time, through its implicit acknowledgment that both imagination and experience are effective ways of impressing loci on the mind, the ars memoriae points to a theoretical connection between the kinds of meanings developed by Cicero

[22] Art of Memory, 4.

[23] Speaking of artistic representation of memory images, Yates writes (Art of Memory, 81): "For when people were being taught to practice the formation of images for remembering, it is difficult to suppose that such inner images might not sometimes have found their way into outer expression."

for elements of the visible milieu and those meanings tied to places and things merely described to an audience.

This hypothesis, of course, leaves many questions unresolved. If the orator were to follow the suggestions in the rhetorical handbooks, we assume that he would (a ) attempt to engage his own emotions in treating the subject matter of his speech by means of imaginative reconstructions of events; (b ) stimulate visualizations in his audience through vivid description; (c ) impress ideas and their sequence on his own mind through memorization of a large number of images set within various loci; and also, perhaps, (d ) attempt to impress image-filled memory loci on the minds of his audience, either by the manipulation of actual visual milieux or by vivid descriptions. But would the images that aroused the orator's own emotions (a ) be identical with those he hoped to produce in the minds of his audience through vivid description (b )?^[24] Would either of these be identical with the loci and images he kept in mind to remember what he wished to say and when (c )? And what was the relationship between these memory loci (c ) and those he created in the minds of his audience (d )?^[25]

[24] Yates, Art of Memory, 86, commenting upon the artificial memory techniques used by Renaissance preachers, notes that "the similitude spoken in the sermon is not strictly speaking the similitude used in artificial memory. For the memory image is invisible, and remains hidden within the memory of its user, where, however, it can become the hidden generator of externalized imagery."

[25] The orator's need to avoid confusion by minimizing the number of images kept in mind suggests that a great deal of overlap would have had to exist in the use to which images were put. To return once more to the Pro Scauro, when Cicero wished to stimulate within himself feelings of pity for the danger in which his client, Aemilius Scaurus, stood, he might have pictured to himself the outstanding deeds and character attributed to various of the Scauri (cf. Antonius's description of exactly this process in De or. 2.194–95). It would, in turn, advance his case to mention some of these same events to the audience as well, since a celebration of Scaurian dignity, generosity, and bravery would serve to attach the audience to the defendant. Furthermore, his mention of these events could be calculated to have a greater impact if it took the form of a vivid description of memorable scenes that stimulated striking images in the minds of the audience. The images that originated in the mind of the orator would thus form the basis for those he hoped to create in the minds of his listeners. Since certain of these images—for instance, that of Lucius Metellus saving the Palladium from the burning shrine of Vesta—could be associated with monuments visible in the Forum, Cicero could memorize these places in a fixed order and connect with them imagines he wished to describe, as well as broader concepts to which the imagines were wedded. These private memory loci and their connection with various associative ideas could then, in turn, form the basis for the external structure of this section of the speech.

If this habit of mnemonic visualization was as important a part of the orator's training as we have assumed, it seems probable that it would have been reflected in other literary works as well as in speeches. Perhaps the "associative" itinerary of Athens in Cicero's De finibus (5.2) and the poetic itinerary of the center of Rome in the Tristia (3.1.27–74) of Ovid, the most rhetorical of Augustan poets, provide evidence of the adaptation of rhetorical training and technique to nonrhetorical uses.

I hope that we may begin to explore some of these problems through an analysis of In Verrem II.4, but, whatever the answers to specific questions may be, a general principle ought here to be affirmed: namely, that this technique, like any technique of Ciceronian oratory, was not an end in itself but was a means to an end. Ignoring this principle has led many to classify vivid description in oratory merely as an ornament of style and to analyze its rhetorical use in the same way one would analyze the use of enargeia or ekphrasis in poetry.^[26] Whether or not the source of the technique lay in poetry, its aims in rhetorical texts were not the same as in poetry. The stylistic beauty or affective power of a descriptive passage in a deliberative or forensic speech and the emotion it must have engendered in those who heard it were never, in themselves, the raison

[26] Most modern discussions of ekphrasis focus on its long poetic history and its frequent use as an epideictic technique in imperial school and public rhetoric, and there is a corresponding tendency to interpret its use in Ciceronian oratory in terms either of poetry or of epideictic. Norden (Kunstprosa, 1:285), for instance, agrees with Rohde's interpretation of ekphrastic passages in rhetoric as motivated by a desire to imitate and "compete with" ekphrastic passages in Hellenistic poetry. A similar conception of the ornamental function of description in classical rhetoric appears in the more recent study of Genette, "Boundaries of Narrative" (6). A useful corrective to this emphasis on techniques of style is found in Brink's discussion of differences between poetry and rhetoric in "Cicero's Orator and Horace's Ars Poetica ." Brink (99–100) points out that the concept of flectere, missing from poetry, is always related in rhetoric to a practical purpose. See also Neumeister, Grundsätze der forensischen Rhetorik, 156–92, for the impossibility of separating "style" from the practical goals of persuasion in forensic and deliberative rhetoric. In "Ekphrasis and the Theme of Artistic Failure," 136 n. 5, Leach warns that "although imitation within the declamatory schools is a token of the admiration with which ekphrasis in poetry was regarded, it should not be taken to indicate that the purposes of the poets and orators were precisely the same." Leach's intent is to rescue poetic ekphrasis from misinterpretation, but the same sentiment should also be applied to rhetorical ekphrasis .

d'être of such a passage. Rather, vivid description in Ciceronian oratory was always part of an overall strategy meant to lead an audience to a predictable emotional and intellectual stance vis-à-vis the particular subject at issue before them, whether their role was as jurors, voters, or simply members of the body politic. Before analyzing the description of places and things in the text of In Verrem II.4, therefore, it will be useful to consider the chief factors that informed Cicero's rhetorical strategy in the work.