1 Traditio (1971) 1-78, hereafter cited as AHF. I shall repeat primary materials only when it seems necessary for the argument and as little secondary bibliography as possible in hopes that the reader will have the patience to consult the first essay. A great portion of the materials collected by such scholars as Norden, E., Curtius, E., von Arnim, H., Marrou, H., Michel, A., Solmsen, F., de Bruyne, E., Weinberg, B., etc., is too continuously relevant to be cited, except for special topics, without swelling the notes out of all proportion. Theirs and other general works cited in AHF I hope the reader will regard with me as part of a mutually shared inheritance of humanistic studies.Google Scholar

2 Aristotle's Poetics: The Argument, Else, G. F. (Cambridge, Mass. 1963) 238. Subsequent references to the Poetics will be to this edition unless otherwise assigned.Google Scholar

3 Institutio oratoria , trans. Butler, H. E., LCL, 4 vols. (London 1953). Subsequent references will be to this edition. The ‘material’ contribution (ὔλη) of the rhetorical hypothesis to literature resembles the contribution which M. Cohen has suggested the study of jurisprudence might be able to make to a contemporary philosophy reduced to ‘a purely formal discipline,’ that is ‘to such formal problems as the relation of mind or thought to reality’ (‘Jurisprudence as a Philosophical Discipline,’ Jour. Phil. Psych. Sci. Meth. 10 [1913] 226). The ‘philosophy of law,’ a phrase combining in itself philosophical and rhetorical discourse, might elude both the empirical bondage to individually discrete cases and the conceptualist bondage to a deductive application of statutes and precedents to circumstances for which they were not designed (228). Such a concern with equity encourages the least specialized intentions of both law and philosophy to liberalize the more conservative forms of each.Google Scholar

4 For the continuing interrelationship between these disciplines, particularly with respect to faculty psychology, in the Thomist commentaries on Aristotle, see De Bruyne, E., Études d'esthétique médiévale, 3 vols. (Brugge 1946) 3. 324–9. This work will be cited hereafter as De Bruyne.Google Scholar

5 The implications for literary theory discoverable in disciplines which originally contributed to its terminology would remain significant even if such treatises had been written and lost. The situation is analogous to the transmission of aesthetic theory. There, according to Bundy, M. W., the concepts making up the history of the imagination have been ‘molded by philosophy, and, specifically in the order of their importance, by metaphysics, psychology, and ethics. Definitions in these realms largely determined the nature of the terms which were to become so important in the vocabulary of the aesthetician and the critic’ (The Theory of Imagination in Classical and Mediaeval Thought, Univ. Ill. Stud. Lang. Lit. 12.2–3 [1927] 270). Though relatable to many topics of this essay, the materials of this study are more pertinent to the essay immediately to follow on the philosophical transmission of literary theory.Google Scholar

6 The terms ‘practical’ and ‘theoretical’ thesis and their accompanying functions, actio and cognitio, gradually replaced the earlier rhetorical terms of hypothesis and thesis. Quintilian: hi thesin a causa sic distingaunt, ut illa sit spectativae partis, haec activae (3.5.11). See AHF 65-71. Cicero describes for Atticus (9.4) his practicing Θέσεισ πολιτικάισ, in utramque partem tum Graece tum Latine, which perhaps dealt with matters close to his own life. Theon also divides theses into theoretical and practical, and what he calls πρακτικόσ, Hermogenes calls πολιτικάσ (Reichel, G., Quaestiones progynasmaticae [Lipsiae 1909] 106).Google Scholar

7 Two of the questions are identical: ‘Quae sit mundi forma? Quae sit solis magnitudo?’ (De inventione 1.8, trans. Hubbell, H. M., LCL [London 1960]); Quanta sit solis magnitudo, quae forma terrae (De oratore 2.66, trans. Sutton, E. W. and Rackham, H., LCL, 2. vols. [London 1959]). Future citations of these works will be to these editions.Google Scholar

8 The consistency of Cicero's earlier and later views with regard to theses — as indeed to the priority of the prudential over the speculative faculty generally (De off. 1.153-7) — as expressed in this section, is a slight modification of the more conventional interpretation in AHF 25. As von Arnim suggests in his introduction, so important for the history of the thesis and its place in rhetoric, Hermagoras, though welcoming general philosophical questions, may not himself have admitted general questions of the more specialized sciences (Leben und Werke des Dio von Prusa [Berlin 1898] 95). This interpretation conforms to the use of καΘόλον in the Poetics to express, as Else insists (273-4), the ‘universal’ of ‘practical questions of human living,’ not of metaphysical generalities on such subjects as the motions of the stars. C. O. Brink likewise observes that Horace's Socraticae … chartae (A.P. 310) refer to ‘a philosophy which is a guide to life rather than an intellectual discipline’ (Horace on Poetry [Cambridge 1963] 131). St. Thomas restates the distinction, so familiar in Renaissance humanists like Sidney, for the Middle Ages: Prudentia magis convenit cum arte quam habitus speculativi, quantum ad subjectum et materiam. Utrumque enim est in opinativa parte animae et circa contingens aliter se habere (S.T., 1.2. qu.57 art.4 r.2). For Roger Bacon's emphasis upon praxis and the moving of the will as the basis of the superiority of ancient literature, see Massa, E., Ruggero Bacone: etica e poetica nella storia dell' ‘Opus Maius’ (Roma 1955) 144-55. So Dante writes to Can Grande on the Commedia: Genus vero phylosophie sub quo hic in toto et parte proceditur, est morale negotium, sive ethica; quia non ad speculandum, sed ad opus inventum est totum et pars (16); see as well De Monarchia 1.3 and Ficino's translation (Dante Alighieri: tutte le opere , ed. Chiappelli, F. [Milano 1965] 864, 731-2, 791). So Boccaccio, Comento Lez. 1. DeBruyne comments on the gradual extension of civiles quaestiones to include almost all moral concerns in the early Middle Ages (1.44-6) and, in discussing St. Thomas' theory of art (3.316-46), gives a good account of the psychological function of the prudential faculty in relation to that of the speculative and productive faculties.Google Scholar

9 Brutus [and] Orator , trans. Hendrickson, G. L. and Hubbell, H. M. [respectively], LCL (London 1952), to which edition future references will be made. On the breadth of application of theses, see De orat. 1.138-41, 2.104 ff; Orat. 125-7. The ‘qualitative’ question, the status or constitutio generalis, occurs when, the act and its definition now established, ‘there is a question nevertheless about how important it is or of what kind, or in general about its quality, e.g. was it just or unjust, profitable or unprofitable (constat et tamen quantum et cuiusmodi et omnino quale sit quaeritur, hoc modo: Iustum an iniustum, utile an inutile),’ De inv. 1.10-12; cf. De part. orat. 102. Cicero elsewhere illustrates the importance of qualitative considerations with this question: ‘“Then does it make no difference … whether a man murders his father or a slave?” If you posit those cases without qualification, their real nature cannot easily be judged (nuda ista si ponas, iudicari qualia sint non facile possunt).’ It is finally the motive (causa), not the nature (natura), of the crime which counts (Paradoxa Stoicorum 24, trans. Rackham, H. [London 1960]). Quintilian accepts Cicero's view of the three types of inquiry; they have even been prescribed to us by nature (‘ipsa nobis etiam naturam praescribit’) and apply to both indefinite and definite issues (3.6.80-1).Google Scholar

10 Various famous homicides are used to exemplify the question of quality. See Cicero, , De inv. 1.18, De part. orat. 104; Cappella, M., De arte rhet. 7 (Rhetores Latini Minores , ed. Halm, C. [Lipsiae 1863] 455 — hereafter cited as Halm).Google Scholar

11 Iulius Victor suggests this compounding of qualities (id est iniectione qualitatum, cum quasi inicitur et supervenit qualitas qualitati) in illustrating the ‘pragmatic’ (negotialis) division of the qualitative issue — which Quintilian says deals with things in general without reference to specific persons (3.6.57 ff.). The question at issue in his example, Iulius says, could not even have arisen ‘nisi adumbrata esset influentibus in se quodammodo prioribus qualitatibus’ (Halm 379-80). See n. 40. In the plastic arts, adumbrata connotes the preliminary roughing out of a statue (Tusc. 3.3).Google Scholar

12 The extensive use of the Orestes story in illustrating legal questions is mentioned by Quintilian who analyzes its detail (3.11.4-12), as does Cicero (De inv. 1.18 f). The interpretation of the ‘bloody sword’ in the story of Ajax and Ulysses is given by Cicero (De inv. 1.10) and by the author of the Rhet. ad Her. (1.18) as an example of the conjectural issue: Quintilian claims it came from the tragic stage (4.2.13). Interpretations of circumstantial evidence occur in Seneca's controversiae (7.3,5) but predominate in the first nineteen declamations attributed to Quintilian. For an English version of these, see The Declamations of Quintilian, trans. Warr, J. (London 1686). Conjectural issues involving circumstantial evidence are illustrated by situations (De inv. 2.14), which are analyzed by a series of questions not unlike those that a novelist might ask in working out his plot (2.45-6).Google Scholar

13 Iulius Victor says one locates the point of issue by ‘division.’ One begins with the general issues common to many cases, and ‘descends’ to special issues proper to the case at hand (a generalibus gradatim descendet ad speciales, id est ad causae proprias): Ita sensim defluit generalitas ad speciem, id est ad τὸ κρινόμενον (Halm 385-6). The anonymous excerpta rhetorica add that the hypothesis or quaestio specialis will often supply a defense where the thesis or genus causae will not. One cannot defend matricide in general, but if a specific person, Orestes, comes forward ex hypothesi, his defense can be that he was justified in avenging his father (Halm 585 f). See n. 63.Google Scholar

14 For further passages on fictional exemplification, see n. 72.Google Scholar

15 The Categories , trans. Cooke, H. P., LCL (London 1938).Google Scholar

16 The Metaphysics , trans. Tredennick, H., LCL, 2 vols. (London 1933). Future references will be to this edition.Google Scholar

17 The Nicomachean Ethics , trans. Rackham, H., LCL (London 1956). Future references will be to this edition.Google Scholar

18 Matthes, D., ‘Hermagoras von Temnos 1904-55,’ Lustrum 3 (1958) 147–8. This study is the most detailed treatment of the origins and subsequent influence of the Hermagorean stasis system. For a briefer résumé, see Nadeau, R., ‘Classical Systems of Stases in Greek: Hermagoras to Hermogenes,’ Greek, Roman, and Byzantine Studies, 2, 1 (1959) 51-71. In ethics an involuntary action is said to be κατὰ σνμβεβηκόσ (N.E. 5.8) and hence morally neutral in that it lies outside the ‘borders’ of the will. In the Poetics the same phrase is used to describe the error in fact or belief which is ‘accidental’ to mimetic representation (1460b16). By analogy, in law whatever does not lead to the establishment of the central question of the case — prior to a judgment of the moral issues then to be considered — lies outside the case proper. Hence a legal narration of the facts, in its narrowest historical sense as a kind of definition of the given circumstances, may exclude explanatory detail and even motives (De inv. 1.28; Alcuin, , Rhet. 22). It is a kind of outline waiting to be ‘filled in,’ in the confirmatio (De inv. 1.34-43), by qualifying colores. In this it resembles, not only the bare thema of a declamation, but also Aristotle's conception of the dramatic hypothesis (Poet. 17), which excludes from its primary ekthesis accidents of motive, names, and circumstances (cf. AHF 43-9). The accidental qualities will then be brought in to confirm the speaker's or dramatist's interpretation of events on the basis of which he is asking the audience to respond in a certain way.Google Scholar

19 Greek Philosophy: Thales to Plato (London 1924) 245. Diogenes Laertius claims that Plato was the first to employ the term ποιότητα in philosophical discussion (3.24).Google Scholar

20 Plato , trans. Bury, R. G., LCL (London 1952). Later references to Epistles will be to this edition. Compare Aristotle's discussion of accident in Met. 1025a14-35, 1026a33-27a28: ‘“Accident” is only, as it were, a sort of name. Hence in a way Plato [Soph. 254A] was not wrong in making sophistry deal with what is non-existent; because the sophists discuss the accident (σνμβεβηκόσ) more, perhaps, than any other people’ (1026b14-17).Google Scholar

21 I am much indebted in this paragraph to the analysis of 0. Dieter, A. L., ‘Stasis,’ Speech Monographs 17, 4 (1950) 345–69. See n. 28.Google Scholar

22 Epicurus, trans. Bailey, C. (Oxford 1926) 31.Google Scholar

23 Both terms are relative to each other: a ‘property’ of one thing may be an ‘accident’ of another, as slavery is a property of ‘slave,’ but an accident of ‘man.’ The principal distinction seems to be that properties ‘are directly perceived in acts of apprehension, but the σνμπτῶματα only in relation to such acts: e.g. we see a man in a certain attitude, etc., and thus know that he is writing,’ Epicurus, 235, 240. Lucretius translates the terms as coniuncta and eventa (1.449-50). See Bailey, , op. cit., 235-44, and his The Greek Atomists and Epicurus (New York 1964) 300–309, for commentary and general discussion. Diogenes Laertius summarizes Epicurus' position (10.51-73). See n. 49.Google Scholar

24 Ross, David Sir, Aristotle (London 1966) 165. Also Bailey, , The Greek Atomists, 301.Google Scholar

25 Cicero's description of a corpus as a ‘qualified thing’ bears a certain resemblance to the Epicurean predication of ‘an aggregate of qualities’ as a body, which ‘in its totality owes its own permanent existence’ to them (1.69). The idea, however, of a body as a combination of Limit and the Unlimited, in the form of ‘qualified extension or space,’ goes back to the Timaeus. If one might speak of a ‘negotiation’ between the monad and the dyad, which results in the apprehension of the physical world, the metaphor may be applied to the relation between the limit of an abstract principle of law and the unlimited randomness of human actions — a relation which can only be apprehended by a kind of ‘equity.’ As limits imposed upon the continuum can only manifest themselves as images with such and such qualities (Tim. 49-50), so the imposition of legal statutes upon the continuum of human actions can only manifest its most nearly perfect adjustment in the image of fiction. See n. 31.Google Scholar

26 Academica 1.24-33, trans. Rackham, H., LCL (London 1961). The Stoic association of the formal, active cause with the qualifying ‘force’ of the world soul, which seems to be reflected in this passage, holds also for the faculties of the human soul. M. van Straaten comments that ‘les anciens Stoïciens considéraient ces δννάμεισ comme des ποιότητεσ, qui reposent dans l'âme comme dans un substrat, de façon que l'âme serait composée de ce substrat et d'un nombre de qualités,’ Panétius (Amsterdam 1946) 122. Later, St. Augustine will associate the activity of the Holy Spirit with the status qualitatis in its power to instill in men a delight in the preservation of that knowledge of the Father which is conveyed by the Son (dulcedo in ista cognitione permanendi, Ep. 11). For the activity of the Holy Spirit with respect to the medieval conception of the poet as theologian, see below 108-109.Google Scholar

27 For example, included among the rhetorical ‘Schemata Dianoeas’ edited by Halm is: Ἰδέα est, cum speciem rei futurae velut oculis offerentes moto animo concitamus. Cicero: Videor mihi videre hanc urbem, lucem orbis terrarum atque arcem omnium gentium, subito uno incendio concidentem' (73). Cf. St. Augustine De trin. 8.4-6. R. Harriott suggests that when Aristophanes (Clouds 547-8) refers to ‘dramatic ideas,’ he perhaps means vividly constructed scenes, or stage pictures, such as that of Socrates in a basket (Poetry and Criticism Before Plato [London 1969] 136 n. 3). The function of optics, the imagination, and the memory in the formation of vivid ideas as prudential intentiones in literary or pictorial exemplification will be treated in the psychological materials of the essay to follow. See nn. 115 and 126; AHF nn. 36, 42-3; and my paper ‘The meaning of Horace's Ut Pictura Poesis,’ JWCI 36 (1973) 31.Google Scholar

28 The epistemological implications of the status system are strikingly suggested by Plato's observations on names, which ought to indicate the essence (οὐσίαν) of things which are in flux: ‘Let us first take up again the word ἐπιστήμη (knowledge) and see how ambiguous it is, seeming to indicate that it makes our soul stand still (Ἱστησιν) at things, rather than that it is carried round with them, … And ἱστορία (inquiry) means much the same, that it stops (Ἱστησι) the flow. And πιστόν (faithful) most certainly means that which stops (ἱστάν) motion. Then again, anyone can see that μνήμη (memory) expresses rest μονή in the soul, not motion’ (Crat. 436E-37B, trans. Fowler, H. N., LCL [London 1926]).Google Scholar

29 Plato's Statesman, trans. Skemp, J. B. (London 1952); the Greek is cited from the Loeb edition (London 1952).Google Scholar

30 Laws , trans. Taylor, A. E., The Collected Dialogues of Plato, ed. Hamilton, E. and Cairns, H. (New York 1963). Future references will be to Laws , trans. Bury, R. G., LCL, 2 vols. (London 1926). For opposing views on the nature of ‘analogy’ (ratiocinatio) arising from these general considerations, see Cicero, , De inv. 2.148-53.Google Scholar

31 It is precisely in terms of the relation between the determinate and the indeterminate that St. Thomas comments upon Aristotle's definition of equity discussed below. Law as universal deficit in particularibus, since non omnia possunt determinari secundum legem; it must be through a sententia that the universale dictum legis applicatur ad particulare negotium. Because the materia humanorum operabilium est indeterminata, the regula, quae est lex, must be indeterminata and not fixed in itself (In decern libros Ethicorum Aristotelis ad Nicomachum expositio, ed. Spiazzi, F. M. [Torino 1964] 298. See following note.Google Scholar

32 Stroux, J., ‘Summum ius summa iniuria,’ Römische Rechtswissenschaft und Rhetorik (Potsdam 1949) 9–66, emphasizes the role of equity as supplement or completion: ‘Aus den “Lücken im Gesetz” resultiert ihm das Wesen der Billigkeit als Besserung des Gesetzes durch Ergänzung: ἔστιν αἔτη ή φύσισ ἡ τοῦ ἐπιεικοῦς • ἐπανόρΘωμα νόμον ῄ ἐλλείπει διὰ τὸ καΘόλον (Eth. Nic. 1137b25). … Auch für Aristoteles ist Billigkeit Notbehelf zur Verwirklichung richtigen Rechts gegenüber der Unvollkommenheit, die dem Gesetz als Menschenwerk anhaftet’ (19). For St. Thomas' analogy of the function of equity with the completion of a sketched outline, see n. 42. His comment on Lesbos is detailed: In Lesbia enim insula sunt lapides duri qui non possunt de facili ferro praescindi ut dirigantur ad omnimodam rectitudinem; et ideo aedificatores utuntur ibi regula plumbea. Et sicut illa regula complicata adaptatur ad figuras lapidis et non manet in eadem dispositione, ita oportet quod sententia iudicis adaptetur ad res secundum earum convenientiam (loc. cit). For the place of equity in Aristotle's ethical thought and its subsequent influence, see Hamburger, M., Morals and Law: The Growth of Aristotle's Legal Theory (New York 1965) esp. 89-110; Barker, E., The Politics of Aristotle (Oxford 1969) lxiii-lxxvi, 146, 362-72; and Kisch, G., Erasmus und die Jurisprudenz seiner Zeit (Basel 1960) and Melanchthons Rechtsund Soziallehre (Berlin 1967) esp. 168-84.Google Scholar

33 The ‘Art’ of Rhetoric , trans. Freese, J. H., LCL (London 1926). See nn. 41, 43, and 117. In a later passage in the Rhetoric (3.16; 1417a15-22) it is clearly the qualitative issues of moral purpose which distinguish the narrative of oratory from ‘narratives’ of mathematics (i.e. specialized treatises): to give the narrative a moral character we should ‘make clear our moral purpose, for as is the moral purpose, so is the character, and as is the end, so is the moral purpose (ποιὸν δὲ τὸ ἦΘοσ τῷ ποιὰν ταύτην ἡ δὲ προαίρεσισ ποιὰ τῷ τέλει). For this reason mathematical treatises (μαΘηματικοὶ λόγοι) have no moral character (ἤΘη), because neither have they moral purpose; for they have no moral end.’ As in the Socratic dialogues, ‘accompanying peculiarities of each individual character’ become ‘ethical indications.’ Hence, characterization will particularize the narrative in quality. In contrast, the process of ‘demonstration involves neither moral character nor moral purpose’ (3.17; 1418a15-6). As Richard of Bury is to repeat in the Middle Ages, Cicero does not go to Euclid for his demonstrations nor Euclid to Cicero (Philobiblon 2). It is precisely their inflexibility in seeing things morally, legally, and epistemologically as either true or false that leads Cicero to criticize the older Stoics, while reasserting his allegiance to Plato and Aristotle, in his Pro Murena (61-5). They have no means by which to ascertain the balance of equity between the ‘quantities’ of fact, on the one hand, or of statute on the other (see below 37-40, 43-51). His criticism of their stringency is in accord with that of jurisconsults who lose sight of equity, the spirit of the law, in grasping only onto the verba (Pro Mur. 27-9). See n. 123.Google Scholar

34 Else's note on the opposition between ἐξ ὑποΘέσεωσ and ἁπλῶσ is significant (308 n. 25). He cites Met. 1082b32: ‘in relation to their hypothesis (ὑπόΘεσιν) they are right, but absolutely (ὄλωσ [= ἁπλῶσ) they are wrong.’ The connotations of singleness or purity (as distinct from a composite), exclusiveness, and objective authority associate ἁπλῶσ more with historical (or preceptive) verification than with the probable universal of poetry.Google Scholar

35 For the substitution of ‘paradeigma’ for ‘hypothesis’ as a term for fiction, see AHF 65-71. The illustrative and heuristic functions of ‘paradeigma’ are clearly related by Plato to each other and to the use of μύΘοσ to fill in the details of an outlined sketch περιγραφήν of an ideal king (Statesman, 277–78).Google Scholar

36 For this interpretation of καιρόσ, I am indebted to the commentary of G. Lanata in her edition of Poetica Pre-Platonica, testimonianze e frammenti (Firenze 1963) 93–7, and to E. L. Bundy who first called my attention to its significance in Pindar.Google Scholar

37 There will be a smaller place for equity in a purely conjectural issue, since, as Plato observes (875E), the occurrence or non-occurrence of an alleged act must always be decided by the individual jury whatever the penalty fixed by legislation might be. Cicero says the same thing about the thesis (Orat. 127). Compare the observations of M. Cohen on the possible contribution of jurisprudence to epistemology: ‘Consider how much would our controversy over the nature of truth have been enriched if, instead of our easy dichotomous division of propositions into the true and false, we had taken notice of what lawyers call legal fictions. … These propositions like the statement of the actor, ‘I am thy father's spirit,’ are not adequately characterized when we say merely that they are true or that they are false’ (op. cit. 227).Google Scholar

38 The ideal of consistency is similar here to the coherent probability sought in the dramatic plot. That such consistency can only be realized in wax or dreams is suggestive for the later development of the literary genre of the dream-vision. See also Plato's analogy of the malleability of gold with regard to the depiction of qualities (Tim. 49D-50B). It is interesting that Dio Chrysostom, in enumerating the various materials within which artists may depict characteristics of the gods (12.44), says that moulding in wax ‘most readily answers the artist's touch and affords the greatest opportunity for change of intention (μετανοίασ),’ that is, for the correction of error in the original design (Dio Chrysostom , trans. Cohoon, J. W., LCL, 5 vols. [London 1939]).Google Scholar

39 Cicero intends to draw his conception of the ideal orator as a likeness taken from a face which is perceived only in the mind (ut ex ore aliquo quasi imago exprimatur). Though the ideal itself, like Plato's Form, does not ‘become’ but remains forever, the conception men have of it will fade (see n. 46). Nevertheless, that conception remains more stable and capable of being transmitted than the particulars which, while partially embodying it, continually cease to be (Orat. 7-10, 18-9; cf. Acad. 1.30-2). The law as an ‘image’ of natural justice (Top. 141a21) resembles the image (εἰκὼν) that Isocrates wished to give of his thought and life in the interests of equity (cf. AHF 58-60). Very similar motivations lie behind the Lives of Plutarch. D. A Russell comments that ‘Thus to describe the bios of a great man was to say “what sort of man he was” (poios tis ēn) and to regard him, in a sense, as one of ourselves’ (Plutarch [London 1973] 102). Other characteristics of the portraits conform to the objectives of equity (ibid. 101-3). With respect to the Lesbian ruler, an amusing example might be Xenophon's account of an armourer who claimed the proportions of his suits were better than those of other artisans (Mem. 3.10.9-12). Socrates asks him how he can make a well-proportioned breastplate for an ill-proportioned man. By making it fit! is the reply. Well proportioned, Socrates responds, does not mean in relation to absolute proportions, then, but to the wearer (πρὸσ τὸν χρώμενον).Google Scholar

40 De partitione oratoria, trans. Rackham, H., LCL (London 1960). The three methods of inquiry (coniectura, definitio, consecutio) associated with the three types of status pertain to either the acquisition of knowledge or to the performance of action. With respect to knowledge, the qualitative inquiry (consecutio) is deductive, ‘investigating a particular thing's consequence* such as ‘is it occasionally the duty of a good man to tell a lie?’ Consecutio has two forms: 1) simple questions concern what things are to be desired or avoided (de expetendis fugiendisve rebus), what things are right and what wrong (de aequo aut iniquo), and what is honourable and what dishonourable (de honesto aut turpi); 2) comparative questions concern whether two things are the same or different and which of two things is preferable. With respect to action, the inquiry either involves duty (officium), which asks what conduct is right and proper and includes the whole range of virtues and vices, or it is involved in ‘producing or in allaying or removing some emotion (aliqua permotione aut gignenda aut sedanda tollendave).’ This summary by Cicero (De orat. 3.111-18) of the topics of qualitative inquiry embraces the main concerns of literature of preceding and subsequent centuries, and suggests, in associating quality with the deductive process, the synthesis of fictional representation — see below 52-58, and De part. orat. 66, 101-4; Top. 89 f; De inv. 1.12. In order to adapt these commonplaces to law, Cicero divides the qualitative issue into legal (negotialem) and equitable (iuridicialem) questions. The legal question involves only a dispute over a point of law. The equitable question deals with the nature (natura) of justice and the principle (ratio) of reward and penalty. It is divided in turn into absolute (absoluta) questions which concern rightness or wrongness when clearly apparent in the act itself, and assumptive (assumptiva) questions, which involve an act which in itself cannot be approved but might be defended on the grounds of extraneous circumstances (De inv. 1.14; 2.62 ff.). When Quintilian redefines negotialis and iuridicialis (3.6.57 ff.), the distinction between them comes to resemble that between indefinite and definite questions, the former considering questions without particular persons, the latter considering the same questions with respect to them. Iulius Victor treats the distinction in detail (Halm 378-82). See nn. 11 and 108.Google Scholar

41 The ‘relativity’ of affectio is brought out by contexts in which we would translate the word as relation (cf. Lewis and Short, I). Considerations of equity arise particularly when ‘quality’ is involved (‘quale sit id, de quo consideretur’): ‘For it often happens, owing to exceptional circumstances (tempore), that what is accustomed under ordinary circumstances to be considered morally wrong is found not to be morally wrong.’ The example Cicero gives is the traditional exoneration of murder if the victim happens to be a tyrant (De officiis 3.19, trans. Miller, W., LCL [Londen 1961]). Elsewhere (De part. orat. 42-3) Cicero says that actions done owing to emotional and mental disturbance (motu animi et perturbatione) cannot be defended in court but in open debate they can on the basis of quantitative issues (in quo quale sit quaeritur). The author of the Rhet. ad Her. comments that ‘according to circumstances (tempore) and a person's status (dignitate) virtually a new kind of Law (novum ius) may well be established’ (2.20) — trans. Caplan, H., LCL (London 1964). This relation to individual circumstances is not only fundamental to Aristotle's definition of the ‘mean’ but also to his judgment of the morality expressed in plays: ‘As to the question whether anything that has been said or done is morally good or bad, this must be answered not merely by seeing whether what has actually been done or said is noble or base, but by taking into consideration also the man who did or said it, and seeing to whom he did or said it, and when and for whom and for what reason; for example, to secure a greater good or to avoid a greater evil’ (1461a6-11 — trans. Fyfe, W. H., LCL [London 1953]). His terms are almost identical to Cicero's definition of affectio .Google Scholar

42 Cf. Prot. 326CD. Aristotle says (N.E. 1.7; 1098a21-6) that the Good itself can only be described ‘in outline — for no doubt the proper procedure is to begin by making a rough sketch, and to fill it in afterwards. If a work has been well laid down in outline (περιγραφῆ), to carry it on and complete it in detail may be supposed to be within the capacity of anybody; and in this working out of details Time seems to be a good inventor or at all events coadjutor. This indeed is how advances in the arts have actually come about, since anyone can fill in the gaps.’ He continues by repeating his earlier warning that various arts are capable of varying degrees of certainty (N.E. 1.3). As earlier the mathematician and the orator achieved different degrees of exactitude in demonstration, so here the carpenter and the geometer seek the right angle by different methods and with different intentions (cf. Plato. Rep. 527-30, Phil. 56-7). What is important in both passages is that the process of moving from sketch to completed portrait is particularly apt for the ‘inexact’ ethical sciences. (It is the ethical content of its narrative, among other things, that distinguishes oratory from mathematical treatises [Rhet. 3.16]). In his commentary on Ethics 1.7 St. Thomas refers to the sketch as a circumscriptio which ‘figuratively’ (figuraliter) offers an external outline of bonum finale hominis, quod est felicitas. This sketch presents a notification of a thing by means of characteristics it shares with other things rather than by means of its own special attributes. For this reason, one speaks ‘figuratively’ first, that is secundum quandam similitudinarium et extrinsecam quodammodo descriptionem, and then fills in later what fuit prius figuraliter determinatum. In commenting on the earlier passage (N.E. 1.3), he distinguishes the subject matter of ethics (from that of the speculative sciences) by saying that its truth must be shown figuratively, id est verisimiliter; et hoc procedere ex propriis principiis huius scientiae. Nam scientia moralis est de artibus voluntariis: voluntatis autem motivum est, non solum bonum, sed apparens bonum. St. Thomas describes the appearances of good as deceptive because the materia moralis est varia et difformis. What appears in certain circumstances to be right appears in others to be wrong, and in hoc multiplex error contingit (op. cit. 9-10, 35-6). For the ability of a fictional paradeigma (μύΘοσ) to ‘fill in’ the outline sketch περιγραφὴν of an ideal in order to achieve a knowledge of the most serious matters, see AHF 65-71. Such adumbrations as that of Cicero's ideal orator, which cannot be delineated exactly, aim particularly at shadowing forth qualities which the ideal ‘ought’ to possess: at qualis esse debeat poterimus fortasse dicere (Orat. 19). Qualitas, then, becomes a ‘ground’ of negotiation between an ideal and its reflection in disparate phenomena. See n. 33. The connection between the ‘prefiguring* of Biblical exegesis and the presentation figuraliter in an outline to be later completed or ‘brought to perfection’ will be treated briefly in the ‘Conclusion’ and in more detail in the succeeding essay. For beauty, according to Suger of St. Denis, as a plaga between earth and heaven, see De Bruyne, 2.142-3.Google Scholar

43 A visual metaphor is also suggested in leges nobis caras esse non propter litteras, quae tenues et obscurae notae sint voluntatis (De Inv. 2.141). What Cicero says of oaths — semper autem in fide senseris, non quid dixeris, cogitandum (De Off. 1.40) — might apply to any oral or written agreement. The interpretation of wills might fall under sanctitas, i.e. the relation of men with departed spirits (manes) — as distinct from that with gods (pietas) or with men still living (aequitas) — as Cicero divides the topics of equity (lato sensu) in his Topica (89 f.). For further treatment of scriptum vs sententia in the interpretation of wills or laws, see Aristotle, Rhet. 1.15, 1375a25-75b25; Rhet. ad Alex. 21b36-22a4; Rhet. ad Her. 1.19, 2.13-4; De inv. 1.55 f., 1.68 ff., 2.138-41; Topica 96; De part Oral 136-7; De orat. 2.110; Inst. orat. 3.6.43,5.10.95-9, 7.6.1 ff. Sententia was interchangeable with voluntas for the original intention of the testator (De part. orat. 136; Brut. 198). In his essay, ‘Summum ius summa iniuria,’ J. Stroux maintained the thesis that Greek rhetorical theory influenced the formation of Roman jurisprudence particularly through questions of equity such as scriptum vs sententia (see esp. 33-8). His thesis has been widely debated. Schulz, F., History of Roman Legal Science (Oxford 1953) stressed the remoteness of such Greek rhetorical topoi (and of the legal philosophy which Cicero derived from them) from the actual practice of jurisprudence (71-5). See also Jolowicz, H. F., Historical Introduction to the Study of Roman Law (Cambridge 1939) 423 f. and Caplan's, H. notes to his edition of the Rhetorica ad Herennium (LCL) 90 f., 94. For support of the rhetorical influence, see Hamburger, M., op. cit, 89-110, esp. 108 ff. with bibliography on the controversy. Bonner, S. F. reviews the problem in connection with the common division of cases by declaimers into ius and aequitas. Implying his support of Stroux, he gives Quintilian the last word (7.6.1): scripti et voluntatis frequentissima inter consultos quaestio est, et pars magna controversi iuris hinc pendet; quo minus id accidere in scholis mirum est, ubi etiam ex industria fingitur (Roman Declamation [Berkeley 1949] 46-8). See, as well, 7.4.10-2. That treatments of equity in later legal codes may only be fictional interpolations has, in itself, interesting implications for literary theory (see Hamburger, 108 n. 3). The problem, finally, of interpreting legal documents is not markedly different from that of interpreting literary texts. We must understand what the author originally intended to say, his voluntas, before considering what the text can mean for us today if its integrity is to be preserved, and, by means of its integrity, our own. But the only access we have to the writer's voluntas is through his scripta, which must be interpreted, like the written law, by an act of the historical imagination and the disciplines at its command. That act starts with the text itself as all investigation starts with the present reality to be undersood. We ‘analyze’ it backward in time until the writer's meaning becomes apparent as a beginning, and then, following the laws of the historical elenchus, we come forward again in synthesis to the present. Only if no contradictions emerge to refute our assumption about its original meaning can we understand it in relation to ourselves — that is, understand how the author might explain his words to us were he living today. (See my ‘The Practice of Historical Interpretation and Nashe's “Brightnesse Falls from the Ayre”,’ JEGP 66, 501-18 and ‘The Definition and Practice of Literary Studies,’ NLH 2, 187-92.) So too with the visual arts. Given the metaphor imago voluntatis for the intention of the testator, it is amusing how closely Quintilian's questions concerning the voluntas of artists, with respect to the interpretation of their works, resemble those of modern iconographers: cur armata apud Lacedaemios Venus and quid ita crederetur Cupido puer atque volucer et sagittis ac face armatus (2.4.26). See nn. 123 and 128.Google Scholar

44 Among Cicero's contemporaries there was nemo qui dilatare posset atque a propria ac definita disputatione hominis ac temporis ad communem quaestionem universi generis orationem traducere' (Brut. 322). Quintilian stresses the importance of equity, quality, and general questions which underlie all particular cases just as strongly (1.pr.16-8, 12.2.15-28). Michel, A. observes that arguments for voluntas tend toward generic questions, for the scriptum toward particular causae: ‘Les premiers sont philosophiques, les seconds dépendent de la philosophie comme toutes les “hypothèses” qui introduisent la “thèse” dans la réalité, le droit dans le fait’ (Rhétorique et philosophie chez Cicéron: essai sur les fondaments philosophiques de l'art de persuader [Paris 1960] 463).Google Scholar

45 Pro Archia 2: Etenim omnes artes, quae ad humanitatem pertinent, habent quoddam commune vinculum et quasi cognatione quadam inter se continentur. The power of literature (Latinae litterae) to preserve the past is symbolized in Cicero's Laws by the Marian Oak, which lives forever because it is planted by the poet's imagination (sata est enim ingenio). The olive tree of the Acropolis and Ulysses's palm of Delos, also, symbolize men's recognition that things may last beyond nature through recollection (ἀνάμνησισ). When the individual tree is destroyed, still there will be another to take its place, and it will be both a different tree and yet the same tree. The ‘historian’ will inquire about the individual tree as he will about whether these or those things in a poem fictane an vera sint. He misses the point, however, if he expects the veritas to be given in the manner of a witness in court (a teste) rather than of a poet (a poeta). For there are indeed alias in historia leges observandas than in poëmata. The standard of one is factual truth, of the other, delight — albeit that Herodotus and Theopompus mixed in innumerabiles fabulae (De leg. 1-5). This passage is one of Cicero's most nostalgic appeals for the cultural continuity he defines in his Pro Archia. His comments on the different kinds of witnesses are repeated by Petronius (Satyr. 118). Both writers are foreshadowed by Aristotle when he contrasts recent with ancient witnesses such as the poets (Rhet. 1.15). Those recent witnesses who themselves risk something in the trial will be reliable in conjectural testimony about whether an act was or was not committed, but ‘if it is a question of the quality (ποῖον) of the act, for instance, whether it is just or unjust, expedient or inexpedient, they are not competent witnesses.’ The passage reveals an important connection between history as empirical knowledge of recent facts and the ‘novelistic’ literary genres, which goes back to the Odyssey. Both are distinguished from poetry, not as prose from verse, but as recent from ancient testimony: history and the novel involve the status conjecturalis, poetry the status qualitatis. The former fall under the strictures of ‘exact’ verification; they can be proven to be true or false. The latter claim only the approximate truth of men's continuing ethical experience. The equitable testimony of the poets, their true voluntas expressed in the written ‘testaments’ of their poems may account for much of the feeling and some of the formal structure of the late medieval genre of the Testament. In it the poet expresses his sententia in the scriptum as if he were there, alive, interpreting it for the reader. It is interesting that the distinction between a novel and a poem may ultimately be a legal one.Google Scholar

46 The desire to find exemplary models in literature and in nature is characteristic also of Cicero's treatment of the state. While Plato described a republic which could not exist and his successors discussed types of constitutions in the abstract (sine ullo certo exemplari formaque rei publicae), Scipio seeks to combine both methods in an account of the founding and development of the Roman state (De republica 2.21-22, trans. Keyes, C. W., LCL [London 1948]). Though embodying the same principles as Plato, he will avoid the shadowy image of a commonwealth (non in umbra et imagine civitatis, 2.52) and use the actual history of the greatest republic in order to show what kind of a state it was that the discussion is trying to make clear (quale esset id, quod ratio oratioque describeret, 2.66). Cicero's exemplary practice is very much like that of Isocrates' use of historical ‘paradeigmata’ (Dem. 34, Nic. 35, Antid. 277) and is central to Renaissance literary methods: he takes as his models (exempla) distinguished people and ages (inlustribus in personis temporibusque) and to these his discussion will conform (ad quae reliqua oratio dirigatur mea, 2.55). A similar concern with historical exemplification is apparent in his essays: De sen. 3, De amic. 18-21, Parad. Stoic. 10. Cicero wishes, furthermore, to produce in the mind of his readers a mental image of such men as Crassus, as Plato did of Socrates, which is on an even grander scale than the portrait he actually presents (De orat. 3.15). It is interesting that Africanus appears in the younger Scipio's dream ‘taking that shape which was familiar to me from his bust rather than from his person (ea forma, quae mihi ex imagine eius quam ex ipso erat notior [De rep. 6.10]).’ Owing to the equity of ancient customs and to the eminent men who protected their institution, Cicero claims the republic once ‘was like a beautiful painting (picturam egregiam).’ Its ‘colors (coloribus), however, were already fading with age,’ and, as Plato warned must be done (769BC), the present time ‘not only has neglected to freshen it by renewing the original colours, but has not even taken the trouble to preserve its configuration and, so to speak, its general outlines (formam saltern eius et extrema tamquam liniamenta [5.2]).’ Such pictorial exemplification is equally responsible for the affective power of history and fiction (De fin. 5.61-4). Though questioning the value of Academic probability (Contra acad. 3.18), St. Augustine also stresses the hortatory and emotive efficacy of literary imagery (De doct. Christ. 2.6; Ep. 55.11.21; De trin. 8.4-6). The capacity of the image, while deriving its existence from sensation, to survive any particular period of the temporal flux perhaps derives from, among other sources, Plato's own late formulation of time itself as a moving image of permanence (Tim. 37D-38B). When we speak of motion, as J. Burnet comments, our language is ‘unscientific and pictorial. It can only convey an “image” of the truth,’ but, nevertheless, Plato ‘had shown already in the Sophist that to be an image was not to be nothing’ (op. cit. 342, 349). Cf. St. Augustine, Solil. 2.10 and Retract. 1.1.11. On the capacity of the various arts (and faculties) to express imagines of that which cannot be imitated by the senses, see Xenophon Mem. 3.10, Cicero Orat. 7-19, Dio Chrysostomus Disc. 12, and Philostratus Life of Apollonius 6.19. I have pointed out relations of these passages to Plato, Horace, and Longinus in JWCI 36 (1973) 1-34.Google Scholar

47 Humanas actiones, atque appetitus considerabo perinde ac si quaestio de lineis, planis, aut de corporibus esset (Ethices, III, Opera Posthuma [1667] 94, quoted from Arber, A., The Mind and the Eye [Cambridge 1964] 86). Miss Arber, as a biologist, is observing that a ‘principle of uniformity’ in nature can never be proved scientifically by induction or deduction. She suggests that such a uniformity ‘should be treated neither as a datum, nor as a conclusion from an argument, but as an hypothesis,’ for hypotheses ‘advance merely to higher and higher degrees of probability as they are found to “work”.’ Originating in intuition, they ‘take their rise through processes including but transcending discursive thought.’ This is one of many passages on the use of hypotheses in the biological sciences in her book which are suggestive for fictional structures. With regard to a uniformity of nature, Aristotle remarked (Met. 1090vb19-20) that ‘observed facts (ἐκ τῶν φαινομένων) show that nature (φύσισ) is not a series of episodes (ἐπεισοδιῶδησ) like a bad tragedy’ (trans. Webster, T.B. L., Art and Literature in Fourth Century Athens [London 1956] 54). Compare the Epicurean criticism of a Stoic providential Intelligence who must be brought in to ‘save’ nature like a deus ex machina in a tragedy to save the plot (Cicero, , De nat. deor. 1.53).Google Scholar

48 Ibid. 47.Google Scholar

49 ‘On the Fame of the Athenians,’ Plutarch's Moralia, trans. Babbitt, F. C., LCL (London 1936). So Acro on Horace A.P. 311: Menander, cum iam fabulam disposuisset, etiam si nondum uersibus adornasset, dicebat, se iamen iam conplesse (Acronis et Porphyrionis Commentarii in Q. Horatium Flaccum , ed. Hauthal, F. [Berlin 1866] 2.628). For the analogy of poetry with mathematics and its contribution to literary formalism, see AHF 34-5, 71-8. Among other passages in Buchanan's, S. Poetry and Mathematics (New York 1962), esp. 56 and 108, the following is particularly relevant: ‘Mathematics deals with relations, and poetry deals with qualities. A sphere results when we can see the relations holding between qualities. Then the two series can be correlated. Mathematical functions find elementary values in qualities. Qualities find their relations in the functions of mathematics. Whenever this happens, a system is recognized, and it takes on a quasi-independence and reality. Often the effect in the thinker is a conviction. Belief attaches itself only to such systems. The further expansions and the wider assumptions are ignored and there is a resting point for thought in a mathematico-poetic allegory’ (146-7). Within the dramatic situation the hero develops, in terms of mathematics and poetry, ‘a system of relations, his idea; and the events have supplied a corresponding set of qualities’ (148). In place of ‘relation’ I should put ‘quantitative’ measurement, whose relative representation Plato says differs so markedly from the qualitative presentation of linguistic images (Crat. 432A-D). See n. 115. The interrelationship between quantitative and qualitative measurement bears an interesting analogy to that between the dual efforts of the mind to interpret the external world objectively in itself, on the one hand, and to understand the private, subjective processes of perception on the other. In his Quantum Physics and Ordinary Language (London 1972), T. Bergstein has pointed out that the concept of complementarity in quantum physics might be profitably applied to this duality: that is, that the greater the preoccupation with measuring the object as it exists independently of the mind that perceives it, the less the possibility for understanding the interaction between the mind and the object perceived, and vice versa. A concern with the interaction, that is, will always tend to dissolve the object into the conditions under which perception takes place; the viewer will remake it in his own image. Objective (quantitative) measurement will tend to be conceptual and mathematical; psychological (qualitative) measurement will tend to be experiential and intuitive. In this form the distinction has relevance for the rhetorical relation between thesis (or statute) and hypothesis (or the particular circumstances demanding judgment). The relation is complementary: the greater the attention given to one (the strict interpretation of the statute), the less is given to the other (the equitable adjustment of the statute to qualitative considerations). This relation might be considered a modern instance of that between a point of observation and a continuum observed, which O. A. L. Dieter points out must be established by both ancient physics and the system of legal status (see above 19). The idea of complementarity might have an equally interesting bearing upon the phenomenon of ‘projection’ in the visual arts as described by E. H. Gombrich in Art and Illusion (New York 1960).Google Scholar

50 For the relation of such a model to imagines see above 31-41. Unlike the more exact specialized disciplines, such as geometry, Cicero says, philosophy demands only a mind keen in pursuing the probable answer to every problem (in quoque verisimile est eliciendum acutis). If one could speak, like Aristotle, in utramque sententiam, or, like Arcesilas and Carneades, argue against any thesis (contra omne quod propositum sit disserat), ‘he would be the one and only true and perfect orator.’ Those who teach by manuals simply sketch (depingere) their juvenile trifles, whereas Cicero intends to unfold (explicemus) the total office of an orator (De Orat. 3.79-81). Of the many other passages treating the close relationship of the Academic and Peripatetic methods of argumentation to oratory, see Acad. 1.15-7, 1.46; De Off. 2.7-8; De fin. 2.1-2, 4.1-7; Tusc. 1.17, 5.10-11; De orat. 3.105-10, 3.122-5, 3.145; Orat. 46-9; De fato 3; De div. 2.150; Inst. orat. 10.1.35. In Tacitus' Dialogus, Maternus' preference for poetry over oratory is really based upon his agreement with Messalla that the past ‘general culture,’ which rhetoric shared with philosophy, is no longer possible. The more contemplative discipline is chosen when the active is no longer feasible in order to preserve the best of the earlier tradition. The importance of the Dialogus in the rhetorical transmission of literary theory, particularly in the Renaissance, has not been sufficiently appreciated. See Michel, A., Le ‘Dialogue des Orateurs’ de Tacite et la philosophic de Cicéron (Paris 1962).Google Scholar

51 Though Quintilian feels the orator need not swear allegiance to any one philosophical code, his description of the Epicurean, Sceptic, Stoic, Academic and Peripatetic attitudes toward rhetorical training make it clear that the last two would be most useful to the most important topics of rhetoric: de virtute, de re publica, de providentia, de origine animomm, de amicitia. Such themes, which elevate both the mind and the language, deal with quae vere bona, quid mitiget metus, coerceat cupiditates, eximat nos opinionibus vulgi animumque caelestem erigat (12.2.23-9).Google Scholar

52 Such a lack of probability could only exist in a fable such as Aesop's, which the progymnasmata describe as naturally impossible (κατὰ φύσιν). Isidore, too, comments that fabulous things are those which did not happen and could not have happened because they are contra naturam (Etym. 1.44.5).Google Scholar

53 Cf. AHF 51-55 and Plato, Rep. 523B ff. A representative of the Middle Stoa, Panaetius greatly modified the older Stoic views on psychology in the direction of the Academy. He realized one might have a distinct ‘impression’ of a ghost and yet at the same time, be prevented from ‘believing’ in it by the knowledge that the dead do not return. The impression must be ‘clarified,’ as in Cicero's account above, by the assurance that no obstacle has presented itself to call the image's truth into question (μηδὲν ἔχονσα ἔνστημα). This assurance of the probability of the image can only be given by the reason through its ‘elenchus’ of impressions. This experimental rationalism in finding criteria for objective truth, which replaced the more intuitive criteria of the older Stoics, closely resembled the system of Carneades. As M. van Straaten concludes, ‘ce que Carnéade trouvait nécessaire pour la réalisation de la plus grande probabilité qui soit possible, Panétius estimait indispensable pour une conviction de la vérité’ (Panétius 133-5). Such modifications by the Middle Stoa are most important for Cicero's views of the literary imago and its power to represent such virtues as he takes up in the De Officiis. The image could utilize the psychological power of clarity (ἐνάργεια, σαφήνεια) of the Stoic φαντασία καταληπτική within the academic structure of argumentation in utramque partem so necessary for fiction. A selection of the clearest ‘inductive’ materials from sensory impressions could be accepted as ‘given’ and cast immediately in a ‘deductive’ demonstration of persuasive argument. Carneades’ view of ‘probable presentation’ or ‘probable and unhampered presentation (φαντασία πιΘανή καὶ ἀπερίσπαστοσ)’ is discussed in Acad. 2.33-4. The formal beauties of literary construction are placed in the service of the Stoic representation of virtue, whose value A. Michel points out: ‘La vertu, dans un monde sans certitude, ne se suffit plus à elle-même. Elle a besoin de l'art pour paraître belle’ (op. cit. 115). Among the tendencies of the Middle Stoa which might be studied in relation to the qualitative concerns with equity are the preoccupation with motive in the analysis of states of mind, as distinct from the act itself (cf. Panétius 147 f.); the increasing social responsibility (176); the emphasis upon natural law as a bond between all men as distinct from laws of national or religious communities (203-11); and the central position of the concept of humanitas in its ethical system (220-22). These attitudes are fundamental to such Renaissance writers as Montaigne and have in part been obscured by an overemphasis upon their more flamboyant skeptical statements.Google Scholar

54 Cicero comments on how moved we are when hearing or reading about some magnanimous deed (cum magno animo aliquid factum) like that of Orestes' offer to sacrifice himself. Such noble examples fill both history and legend (talibus exemplis non fictae solum fabulae verum etiam historiae refertae sunt) and reveal actions done for the sake of an illustrious virtue (dignitatis splendore ductos), not for self-interest (utilitatum) but out of honestate (De fin. 5.63-4). The arrangement of images in ‘places’ in an inner perspective of memory (De orat. 2.357-9) and their power to move the emotions will be taken up with the psychological materials of the essay to follow.Google Scholar

55 Tota rerum natura umbra est aut inanis aut fallax: so Seneca concludes his tragic epistle to an intellectually exhausted world (Ep. 88.46). In words similar to Macbeth's disillusioned comparison of life to a shadow and an idiotic tale, he describes the failure of the liberal arts to achieve ethical significance. (Cf. similar Stoic attitudes in Philo Judaeus' De congressu, esp. 61-2 for the possible source, via a Latin translation, of Macbeth's lines [5.5.24-8]). Even in the philosophical schools, disputes in utramque partem (43) have led only to a Platonic world of appearances behind which no Forms can be perceived. So A. Michel: ‘Cicéron choisit une attitude platonicienne dans un monde où l'on doute des Idées. … L'éloquence est la langue des sages dans un monde qui doute de la sagesse’ (Rhétorique et philosophic chez Ciceron 99). So St. Augustine will comment, with respect to his Contra academicos, how by his time the methods of Academic discussion had led men to a skeptical indifference toward the liberal arts and to a moral ‘lethargy so profound, that not even by the heavenly trumpet can they be aroused’ (Ep. 1, trans. Cunningham, J. G., A Select Library of the Nicene and Post-Nicene Fathers, ed. Schaff, P. [New York 1892] 1.219). In sharp contrast, see Cicero's depiction of Ulysses as the philosophical explorer in quest of a knowledge of the arts (De Fin. 5.48-64).Google Scholar

56 The Geography of Strabo 1.2.17, trans. Jones, H. L., LCL (London 1917). De Lacy, P. reaches the same conclusion on diathesis in observing the Stoic need to mix truth with falsehood in the poetic elaboration of doctrine (‘Stoic Views of Poetry,’ AJP 69 [1948] 267-9). For Strabo's adherence to the Stoic sect (1.2.34, 2.3.8), see Jones’ introduction, xvi.Google Scholar

57 Plutarch's Lives, trans. Perrin, B., LCL (London 1914). The connection here between ἐκκαΘαιρόμενον λόγφ and catharsis, as discussed by Else (224-32, 423-50), may appear whimsical. In relation, however, to Plutarch's comparison of fiction to ‘color’ which gives a desirably illusionistic coherence to fabulous events in poetry as well as biography, it is quite suggestive with regard to the psychological response of the audience (see below 67-71). The act of the hero, in Else's words, is plausibly presented to the spectator ‘and his conscience accepts and certifies it to his emotions, issues a license, so to speak, which says: “You may pity this man, for he is like us, a good man rather than a bad, and he is καΘαρόσ, free of pollution” (438).’ The hero's (given) actions in the story must be made credible for the audience of tragedy to believe them ‘purified’ as the given events of fable must be made probable for the reader of history to believe them reliable. The rationalistic separation of fact from fable is given an interesting further dimension in Cicero's De diuinatione where the miraculous coincidence is clearly distinguished from ‘scientific’ cause and effect (see n. 115).Google Scholar

58 The moral superiority of actual deeds over historical and artistic representations of them — and hence of the prudential over the productive faculty — is argued by Plutarch (345C-51B). See n. 83. The reaction against ‘rhetorical history,’ as well, tended to sharpen the cleavage between truth and falsehood, utility and pleasure. In rejecting panegyric exaggeration, Lucian asserts repeatedly that history aims only at truth (ἀλήΘεια) and can admit no lie (ψεῦδόσ), whereas in poetry the only law is the will (δόξαν) of the poet. History aims solely at being useful (τὸ χρήσιμον); pleasure is purely incidental (Lucian, ‘How to Write History’ 7-9, 38-44, 50-1, trans. Kilburn, K., LCL [London 1959]).Google Scholar

59 The justification of literature — indeed of the liberal arts — as a preparation for later, more mature studies is perhaps the fundamental bequest of the Hellenistic period to literary history. Be they rhetorical, philosophical, or theological, or preparatory for law or for medicine, such studies tended to be specialized, practical, and intolerant. From the beginning, writers such as Strabo (1.2.3) and Plutarch (14-37) tried to defend literature pedagogically on the specialists' terms rather than to analyze the operations of the mind which literary discourse uniquely combined. Their efforts conform to the Stoic desire, in Michel's words, for a ‘science certaine du langage correct (ἐπίστημη τοῦ εὖ λέγειν)’ so that each art may be ‘un système de certitudes pratiques’ (op. cit. 114-6).Google Scholar

60 In his ‘An Essay of Dramatic Poesy’ Dry den has Lisideius quote these lines and comment that the French poet ‘so interweaves truth with probable fiction that he puts a pleasing fallacy upon us; mends the intrigues of fate, and dispenses with the severity of history, to reward that virtue which has been rendered to us there unfortunate’ (Criticism: The Major Texts, ed. Bate, W. J. [New York 1952] 141). That imitation avoids the severity of what it imitates, as the Poetics states (1448b4-19), is a claim also attributed to Aristippus by Diogenes Laertius (2.90). Though Horace's phrase veris falsa remiscet resembles Cicero's omnibus veris falsa quaedam adiuncta (see above 45), its meaning differs. In the Odyssey, to which Horace refers here, ‘true’ manners combined with false incidents (speciosa miracula) do not yield the ‘probable.’ The narrative remains a mixture of two ‘terms,’ fact and fable. Cicero, on the other hand, is saying that true and false impressions can resemble one another so closely (tanta similitudine) that our judgment of any one of them can only claim probability and must await later verification. The mixing of fiction with fact has a structural motivation in Horace: that the middle follow from the beginning and the end from the middle. As in a syllogism, one must ‘invent’ the middle terms, for the beginning and the conclusion are ‘given’ (cf. AHF 40-2). Logically speaking, facta or vera will simply represent the ‘given’ in any form; the falsa or ficta are what must be supplied by the mind. The distinction corresponds to that between nature and art, and more specifically in rhetoric, between inartificial and artificial proofs. F. Solmsen has shown the relation of the development of the second kind of proof from the first in rhetoric to the increasing concern with causation and its analysis in dialogue in Greek drama (Antiphonstudien: Untersuchungen zur Entstehung der Attischen Gerichtsrede [Berlin 1931] 53-9]. Horace's emphasis upon a mixture of falsa with vera also reflects the rhetorical device which Quintilian says can make a fictional narratio plausible: etiam verae alicui rei cohaereat aut argumento, quod sit in causa, confirmetur (4.2.89). Horace's clear parallel between miscuit utile dulci and veris falsa remiscet, in which the true is useful and the fictional attractive, is expressed later as ioca seriis miscere (traced by Curtius, E., European Literature and the Latin Middle Ages , trans. Trask, W. R. [New York 1963] 417-35). A variety of passages defending ioca in medieval literature are cited by Olson, G., ‘The Medieval Theory of Literature for Refreshment and its Use in the Fabliau Tradition,’ to appear in SP (1974). In the Carolingian attitudes toward the visual arts the utile lay in the commemorative, historical function of the work while the ornamental function provided the dulci (see De Bruyne 1.275-305).Google Scholar

61 For Epicureanism, see De Lacy, P., ‘The Epicurean Analysis of Language,’ AJP 60 (1939) 85–92; The Rhetoric of Philodemus , trans. Hubbell, H. M., Trans. of the Conn. Acad. of Arts and Sciences, 23 (1920) 243-382; and the material on the terms poesis, poema, and poeta cited in AHF 68. With regard to the probable, Philodemus' is similar to the older Stoic view: ‘The relation between truth and its opposite is not the same as between two probabilities, one more probable than the other. We must have either truth or falsehood. Would one accept probability in place of truth except in cases where truth is impossible of attainment ? A man should examine carefully and search for truth, and not use vain enthymemes’ (321). Despite their conviction that no truth can be found, the Skeptics placed no more faith in the probable than the Epicureans with respect to the liberal arts. Compare Defin. 1.71-2 with the following passages from Sextus Empiricus: ‘Against the Geometers’ (1-17), ‘Against the Grammarians’ (265), ‘Against the Rhetoricians’ (60-71), ‘Against the Logicians’ (1.166-89), and 'Outlines of Pyrrhonism” (1.226-31). For similar views of the Epicurean Colotes, who says fictions are inept for instructing us in the habitum animarum and rerum caelestium notionem, see Macrobius (Comm. 1.2.3-4). However different his solutions, St. Augustine's Contra Academicos uses similar arguments against the verisimilar.Google Scholar

62 Quintilian adds ‘that we should descend from the common to the particular (a communibus ad propria) is much the same, since what is common is usually general (communia generalia sunt). For example, “He killed a tyrant” is common, while “A tyrant was killed by his son, by a woman or by his wife” are all particular’ (7.1.28). The procedure resembles that of reaching a definition of a particular object experienced: a genere perveniendum ad ultimam speciem (5.10.56). To the same theme of a hero's choosing another's wife as a reward, Fortunatianus adds additional details to illustrate how qualities can be compounded: cum qualitas qualitati supernascitur (Halm 95; see also 379-80).Google Scholar

63 Solebam et hoc facere, ut vel ab ultima specie (nam ea fere est, quae continet causam) retrorsum quaererem usque ad primam generalem quaestionem, vel a genere ad extremam speciem descenderem, etiam in suasoriis. The vel … vel would seem to mean here little more than both … and (= et … et: Lewis and Short, vel I B 2 a). However the division of relevant questions is grasped, their order in oral presentation must conclude with the specific question being examined. Compare Iulius Victor, De divisione, partially cited in n. 13: in omni statu divisio rectius ordietur a generalibus quaestionibus, quae in unaquaque causa generales esse potuerint, et a generalibus gradatim descendet ad speciales, id est ad causae proprias, … Ita sensim defluit generalitas ad speciem, id est ad τὸ κρινόμενον. Eadem ratio est divisionis et in ceteris omnibus statibus. Intuendum tamen est, ne nimis immoremur generalibus quaestionibus, sed praecipuam congressionem ad causae proprias conferamus species. Disputatio enim generalium apud animos iudicum praeparabit tantummodo aditum, ut magis persuadere possis, cum de specialibus disputabis. Ceterum omnis causae firmitas in τῷ κρινομένφ est, id est specie iudicii propria (Halm 385 f.). Among certain excerpta rhetorica included by Halm (585-9) is a passage stressing the usefulness of special questions for the defense, thus exemplifying the association of hypothesis with the iudicatio of a particular case (cf. AHF 65-9): Thesis est quaestio generalis, hypothesis specialis. In thesi, id est in generali quaestione, tantummodo de rebus ipsis sine persona quaeritur: adhibitis personis res quoque communiter; nam quos genus causae premit, species saepe defendit. Si enim secundum thesin, id est generalem quaestionem, quaeratur an animadvertendum sit in matricidam, dubium non erit quin debeat animadverti; sed si ex hypothesi persona accesserit, invenit defensionem: ut si quaeratur, animadvertendum sit in Oresten matricidam, dubium non erit, quam debet animadverti; hac enim se Orestes ratione defendit, quod patrem vindicaverit .Google Scholar

64 St. Thomas comments that prudential judgment concerns the particular case, which he calls an extremum: quod dicitur extremum, quia et ab eo incipit nostra cognitio ad universalia procedens, et ad ipsum terminatur in via descensus. To reach the universalia, one takes a via ascensus (De Bruyne, 3.323). De Bruyne insists that ‘l'allégorisme peut s'élever du sensible au spirituel, il peut aussi descendre du spirituel au sensible.’ Allegory cannot exist ‘sans l'union du spirituel et du sensible dans l'involucrum symbolique’ (2.336 ff.). In a slightly similar manner, the portrait involves for St. Bonaventure, the ‘montée du modèle à l'idée que s'en fait l'artiste’ and the ‘descente de l'idée à l'œuvre qu'il exécute’ (ibid. 3.210 f.). However different the arts involved, these examples reflect, in a medieval context, the procedures of analysis and synthesis. The relation of these procedures to the dialectical use of hypothesis in the Socratic dialogues, treated in AHF, is exemplified by Xenophon (Mem. 4.6.13-5). In search of grounds for a definition Socrates would lead the discussion back to a suitable premise (ἐπὶ τὴν ὑπόΘεσιν), clearly underlying other instances easily admitted, and then, bringing out their similarity to the less clear instance in question, would reveal the general issues latent within it. See the following note.Google Scholar

65 With respect to logic, the inductive process of establishing middle terms, or rather the relation of a first term to a middle term, occurs within the writer's mind as the analysis of the possible ways to construct a story. Then since, as Aristotle says, 'the last step in the analysis seems to be the first step in the execution of the design' (N.E. 3.3.12), the writer, assuming the middle terms to have been established already, presents the conceived construction in the form of the work which the audience actually experiences. The ‘third term,’ the catastrophe, appears to be deductively concluded by syllogism. These distinctions are set forth in the Prior Analytics (2.23-4). They depend, as does much of my discussion of fiction, on the fact that the human mind seems to proceed from particular to universal whereas its logical processes descend from universal to particular (Met. 7.4; 1029b3-12). This accounts for the paradox of fiction which I described in AHF 51-5: the generic issue is implicit in and prior to the specific issue. That is, in practical terms of composition, the more detailed the analysis of particular events, the more precisely articulated the significance of those events can be. The process by which we understand their significance is by going ‘backward’ in analysis from (given) individual particulars to the genera implicit in them by virtue of which, after we have come ‘forward’ again in synthesis, these same particulars may be considered specific. In rhetorical terminology, which refers to subject matter, the genera are described as theses, the particular situations as hypotheses. In dialectical terminology, which refers to the formal procedure by which we discover and present that subject matter, the genera are described as hypotheses (literally that which underlies the theses), the particular ‘situations’ (i.e., the initial propositions to be tested) as theses. The rhetorical thesis corresponds to the dialectical hypothesis, and they combine with one another in the paradoxical term thema, a ‘proposed hypothesis,’ of declamation. See below 61-62. With respect to Socrates' inductive method of inferring a conclusion on the basis of its similarity to questions to which his opponent has already responded affirmatively, Quintilian says that in an oration, where such questions, ordinarily expressed in a dialogue, cannot be asked, it is usual to assume them and their rejoinders and to proceed to a conclusion (5.11.3-5).Google Scholar

66 Georgio Valla glosses quasi retro as an argument quae non directe proponit, assumit, & concludit: sed quae vult concludere, sibi primo assumit, deinde confirmat (De partitione oratoria M. T. Ciceronis Dialogus [Lugduni 1545] 223). Earlier in the De part. orat. Cicero points out that in indefinite questions one aims at conviction but that in definite cases one seeks not only to convince but to excite emotion: Nam est in proposito finis fides, in causa et fides et motus. Quare cum de causa dixero, in qua est propositum, de utroque dixero (9). This reassertion that the general question is latent in the specific case associates the power to move with the particular circumstances. Compare Bartholomaeus Latomus‘ commentary on this passage in the edition cited above (69): Natura [definitae quaestionis] enim talis est, quia omnis motus ex opinione boni vel mali nascitur. Nihil autem bonum vel malum nisi cum persona coniunctum. Personam autem sola causa continet, cui deinceps caeterae circumstantiae, res, causa, locus, tempus, modus adiunguntur. Itaque fit, ut sola etiam causa motus sit capax.’ This associates the concept of qualitas again with increasing specification. A. Michel comments that ‘la passion, dans une cause, se rattache à l'“hypothèse.” Elle concerne donc res et personas. …. La “thèse” des discours veut rester rationelle. La passion subsiste: mais elle est au niveau de l'“hypothèse.” Elle conduit les esprits vers l'idée général, elle les pousse à l'admettre dans les circonstances données’ (ibid. 244-56).Google Scholar

67 The withholding of one's own opinion (i.e. the propositio) is of course, necessary for any argumentum in utramque partem. Cicero attributes this method to Socrates and claims to imitate him 'in trying to conceal my own private opinion, to relieve others from deception and in every discussion to look for the most probable solution (ut nostram ipsi sententiam tegeremus, errore alios levaremus et in omni disputatione quid esset simillimum veri quaereremusy — Tusculan Disputations 5.11, trans. King, J. E., LCL (London 1960). That one may ‘relieve’ others of self-deception is an echo of Plato's description of the ‘catharsis’ of error achieved through a process of elenchus in dialectic (Soph. 230B-E). The cross-questioning (ἔλεγχον) is the greatest of purifications (καΘαρσεών), since by being forced to recognize his self-contradictions, a person gains intellectual humility. I suggested the relation of this form of catharsis to that of the Poetics in AHF 60. The passages suggest the close connections between a dramatic argument, the method in utramque partem, and the emotional effect, brought about by an elenchus in which the propositio is withheld, as it is in Cicero's argument ad motum. The representation of Socrates' inductive, conversational analogies reveal a similar effect by withholding the central question until the person he questions admits the analogies, which that person might not do could he foresee the contradictions into which they were to lead him (De inv. 1.51-4). The purification of this spontaneous elenchus may be partially intended in such dramatically presented dialogues as Cicero's Tusculan Disputations, which he offers, not as a narrative, but eisdem fere verbis, ut actum disputatumque est (2.9).Google Scholar

68 In the Renaissance Iacobus Strebaeus, commenting on Cicero's distinction, says the argument ad fidem looks to the removing of doubts (about gaining a certain verdict) by proof, while the argument ad motum looks to moving the listeners to accept the desired verdict. The structure of the first is syllogistic, while that of the second avoids syllogism because si initio rem proponeret, eamque simpliciter confirmaret, non tam vehementer arderet affectus: ut in argumentatione Dialectica nullae graves excitantur tragoediae, quando rectissimum argumentandi ductum prosequitur (De partitione oratoria M. T. Ciceronis Dialogus [Lugduni 1545] 220). The relation between ethos, as proper to comedy, and pathos, to tragedy, as Quintilian discusses them (6.2.8-36), bears upon the ways of handling the proofs. That appealing ad motum would utilize that force of eloquence which ‘not merely compels the judge to the conclusion toward which the nature of the facts leads him, but awakens emotions which either do not naturally arise from the case or are stronger than the case would suggest’ (24). Bonner, S. F., on the other hand, comments on the similarity of materials in comedy and the controversiae (Roman Declamation [Berkeley 1949] 37 ff).Google Scholar

69 Similar is the setting forth (ἔκΘεσισ) of the materials to be treated in the construction of a geometrical diagram and of a dramatic plot (AHF 43-55). The Rhet. ad Her. says the propositio est per quam ostendimus summatim quid sit quod probari volumus (2.28), and the editor glosses propositio and expositio as πρότασισ or λῆμμα — a word corresponding, in its rhetorical meaning, to hypothesis and later to thema (AHF n. 58).Google Scholar

70 Following the analysis of Ulysses' motivation into the five parts of a speech, comes a description of how the argument in each part may be defective (31 ff.). The defects of the ratio lie primarily in insufficient causality (35 ff.), while those of the rationis confirmatio lie chiefly in arguing from assumptions which are themselves open to question (38 ff). In the confirmatio lies the strongest support of the total argument, and most of its illustrations come from the drama. The names of the five parts of a speech in the Rhet. ad Her. might be compared to Cicero's terms for the five steps in deductive proof (De inv. 1.67).Google Scholar

71 Compare De inv. 1.19 and 1.33. The reverse is also true since in the narratio we should ‘present the outcome in such a way that the facts that have preceded can also be known, although we have not spoken of them’ (Rhet. ad Her. 1.14). The narration, furthermore, must be handled with equal care when it is true as when it is fictitious (1.16). For Cicero, as well as Quintilian, narratio est rerum explicatio et quaedam quasi sedes ac fundamentum constituendae fidei (De part. orat. 31). He makes interesting distinctions between geometrical and philosophical demonstration in accordance with which the rhetorical and poetic argument might more closely resemble the former (Tusc. 5.18; cf. De off. 3.33).Google Scholar

72 In addition to passages cited in this essay and the bibliography given in AHF 69-71 on the use of the historical and fictional example, see Aristotle's Rhet. 1.2, 2.20; Prob. 18.3; Rhet. ad Alex. 28a25-6, 29a25-7, 38b1-5, 39a1-12; Rhet. ad Her. 4.2-6, 62 ff. and Caplan's notes; Cicero, Top. 44 f; De inv. 1.49, 2.118; De part. orat. 40; Quintilian 5.11.1-35, 12.4.1-2. For the rhetorical distortions of history for emotional effect, see Brut. 42-3.Google Scholar

73 Of the extensive literature on the schools of declamation, the most useful studies for our purposes are Bornecque, H., Les déclamations et les déclamateurs d'après Sénèque le Père (Lille 1902); Bardon, H., Le vocabulaire de la critique littéraire chez Sénèque le Rhéteur (Paris 1940); and Bonner, S. F., Roman Declamation in the Late Republic and Early Empire (Berkeley 1949) — with their bibliographies. References to Bonner and Bardon will be given in the text. The edition of Seneca used is that of H. Bornecque, Sénèque le Rhéteur: controverses et suasoires, 2 vols. (Paris 1932). My paraphrases from the Latin are indebted to his translation. In his concise and detailed account of the origins and development of declamation, Bonner indicates its extensive debt to both philosophy and rhetoric by two of his main headings: 1. ‘Philosophical theses and their relation to controversiae and suasoriae’ (2-11); 2. “Greek rhetorical studies as a background to suasoriae and controversiae”' (11-16). He describes, as well, the relation of declamatory procedures to the status rationales, to the status legates, and to the illustration of status by fictional examples in rhetorical treatises (12-16). His first four chapters are an excellent general background. In addition, Parks, E. P., The Roman Rhetorical Schools as a Preparation for the Courts under the Early Empire (Johns Hopkins University Studies in Historical and Political Science 63, 2 [Baltimore 1945] 9-122) is particularly useful with regard to the prominent place given to equity in the exercises (78-85). He cites Latro's words (Controv. 9.4.9) in lege … nihil excipitur; sed multa, quamvis non excipiantur, intelleguntur et scriptum legis angustum, interpretatio diffusa est as the keynote of the spirit of the schools and emphasizes the various applications of natural law with illustrations from the declamations (84). On the term thema, see above n. 65.Google Scholar

74 See Bonner 54-7. The sententiae, in the broader sense of ‘considered opinions,’ tended to be expressed more and more as sententiae in the narrower stylistic sense of brief, wittily balanced antitheses or ‘points’ (traits). The same tendency can be observed in the literary debates of courtly love, in the conversational courtesy books, and the novelle .Google Scholar

75 For the same reason, Bornecque comments, that ‘les personnages sont toujours de simples abstractions, père, fils, riche, pauvre, pirate, sans la moindre caractéristique; à l'élève de mettre les couloirs’ (Les déclamations, 86). Quintilian says that in the schools as distinct from the forum ‘the facts of the case are definite (certa) and limited in number (pauca) and are moreover set out before we begin to declaim: the Greeks call them themes, which Cicero [Top. 21] translates by propositions’ (7.1.4).Google Scholar

76 Bardon points out that the terminology was by no means consistent. Yet, when concerned with equity, the quaestio still 'fût réservé au développement moral se limitant à un cas précis, tandis que tractatio se rapporterait au développement de portée générale (71). Though occasionally tractatio could treat equity ‘sous ses deux formes, générale et particulière,’ the term is never found in the treatment of ius (74). The concern with divisio in the Oresteian issues of the vir fortis sine manibus goes back to Aristotle: ‘The example of the fallacy of division (ἐκ διαιρέσεωσ) in the Orestes of Theodectes: “It is just that a woman who has killed her husband” should be put to death, and that the son should avenge the father; and this in fact is what has been done. But if they are combined, perhaps the act ceases to be just’ (Rhet. 2.24.3).Google Scholar

77 The general meaning of figura which Quintilian gives (9.2.75) is very close to color (see n. 83), and both resemble terms used to discuss such figurative devices as allegory in the Renaissance. Galileo, in criticizing Tasso's narrative incoherence, compares his allegory to a carefully determined oblique point of observation necessary for bringing certain types of illusionistic paintings (called ‘anamorphoses’) into focus. This oblique perspective turns such a painting, when seen normally from in front, into a confusion of lines and colors. So allegory compels the ‘narrative, originally plainly visible and viewed directly’ to ‘adapt itself to an allegorical meaning seen obliquely, and implied’ (trans. Panofsky, E., Galileo as a Critic of the Arts [The Hague 1954] 13). The obliquity is reminiscent of Lactantius' view, discussed below, that a poeticus color is a real incident obliqua figuratione obscuration (Div. inst. 1.11.30); see also Isidor Etym. 8.7.10. Galileo also compares the fragmentary nature of Tasso's stanzas to a tarsia picture made up of a composite of little varicolored pieces of wood (op. cit. 17). Both passages suggest the notorious incoherence of declamatory structures and points of view as a series of verbal lumina and interpretive colores. Galileo further compares the Gerusalemme Liberata to a ‘study of some little man with a taste for curios’ — bric-a-brac like dried insects, little antique figures, and sketches by Bandinelli or Parmigianino: the ‘jumbled Kunst- und Wunderkammern so typical of the Mannerist age.’ In contrast, he praises the Orlando Furioso by comparing it to ‘a regal gallery adorned with a hundred classical statues’ and ‘countless complete historical pictures’ (op. cit. 18-9). Mannerist curiositas, as declamatory lumina and colores, is associated with an enclosed proximity of the cabinet, in contrast to the more open and extensive gallery, which perfectly realizes the Horatian distinction between the painting to be viewed from close up and that to be seen from a distance (see ‘The Meaning of Horace's Ut Pictura Poesis, JWCI 36 [1973] 1-34). Galileo includes the intricate allegory which strains the reader's glossatorial powers among the other Hellenistic forms of figurative ingenuity.Google Scholar

78 For other examples of sententiae which also serve as colores , see Bornecque, , Les déclamations, 53.Google Scholar

79 Fortunatianus uses this theme to illustrate one form of iniectio qualitatum which occurs cum qualitas qualitati supernascitur: Rapta raptoris mortem aut indotatas nuptias optet. Rapuit et profugit: dedit earn pater marito alii, reversum raptorem vult pater producere ad magistratum: ille contra dicit. Quo modo hic supernascitur qualitas qualitati? quoniam, quae antea rapta erat, hodie iam nupta est (Halm 95). This theme with variations was common in later collections of declamations and occurs in the gesta romanorum. The process of increasing qualification is fundamental to the building of narrative incident. See n. 11.Google Scholar

80 For the complicated textual history of this passage see Else, (251-63). The metaphorical importance of χύδην for Aristotle's analogy with painting is perhaps greater than his recent commentators have suggested. Derived from χέειν (= to flow) and usually rendered ‘at random,’ ‘confusedly,’ ‘without order,’ the word also has a technical literary meaning of prose composition as distinct from measured verse (μέτρα), as in Aristotle's Rhetoric (1409b7), which corresponds to oratio soluta as opposed to oratio numerosa or metrica. (With regard to Aristotelian usage, N. A. Greenberg comments that ‘if meter is impressed upon a logos, the logos will become poiema,’ ‘The Use of Poiema and Poiesis,’ HSCP 65 [1961] 268). In addition, there is a strong connotation, among the usages of χύδην, of a whole whose parts are as yet unstructured but which are, nevertheless, capable of σύνοψισ (cf. Plato 537C, 858B), This is the basic meaning of Socrates‘ use of χύδην in criticizing the arrangement of the parts of Lysias’ speech (264 B-E). It is not that they ‘are thrown out helter-skelter,’ as H. N. Fowler renders the phrasing (LCL), that is, that Lysias has used no art, but rather that he has used poor art. There is no rhetorical necessity (ανάγκην λογογραφικήν) for the order of his topics; he has arranged them as they have occurred to him. The parts of a discourse, on the contrary, should be as clearly organized as those of a living being (ζῷον). Plato, furthermore, distinguishes the freely flowing prose of dialogue in his Laws from poetic composition (ἐν ποιήμασιν) by the word χύδην (811DE). Such passages strongly suggest a specific literary meaning for the term which brings out precisely Aristotle's distinction in the Poetics. However essential the arguments, verbal expressions, and characterizing speeches may be, considered by themselves they are only potential dramatic materials awaiting realization in the coherent shaping of the incidents. In this sense, they resemble the ‘bricks’ which must be collected one by one (χύδην) in constructing the edifice of Plato's laws (858B). If laws are simply heaped up indiscriminately, for which Aristotle also uses χύδην (1324b5), no direction will be apparent in the legislation. Once outlined, the plot may ‘receive’ the other dramatic materials as the design on white may receive the ‘flow’ of colors. In this sense, the common term of the analogy between poetry and painting may be χύδην itself which has a distinct meaning for each of the arts. Free flowing speech (prose) in contrast to measured speech (poetry), the informal association of topics in dialogue in contrast to logically ordered discourse: both contrasts illustrate a distinction between the disparate materials of drama and their principle of coherence. So, also the materials of painting, the colors, will flow in an unsatisfying profusion if there is no cohering principle of outline. The verb upon which χύδην depends is ἐναλείφω, to anoint or dye. In an Aristotelian passage on the growth of embryos (De gen. an. 2.6; 743b20-5), cited as a gloss on the verb as early as the sixteenth century (Petri Victorii Convn. in prim. lib. Arist. de arte poet. [Flor. 1560] 71 and V. Madii et B. Lombardi in Arist. lib. de poet, communes explanationes [Venet. 1550] 114), it seems to mean ‘to paint within lines’: ‘All the parts are first marked out in their outlines and acquire later on their colour and softness or hardness, exactly as if Nature were a painter producing a work of art, for painters, too, first sketch in the animal with lines and only after that put in the colours’ (trans. Pratt, A., Works 5 [Oxford 1912]). I. Vahlen viewed the analogy in the Poetics as two stages in a process such as that described above. Bywater, paraphrasing Vahlen in his own commentary, says that the process is ‘(1) λενκογραφεῖν εικόνα, and (2) ἐναλείφειν φαρμάκοισ, and serves to illustrate the order of procedure in Tragedy, where the μῦΘοσ is said to come first, as the ἀρχὴ τὴσ τραγωδίασ, and the ἤΘη second’ (Aristotle On the Art of Poetry [Oxford 1909] 171). Though Bywater feels his analysis is unnecessarily subtle, Vahlen's account corresponds exactly to the general process of composition I have been describing. Though he does not point out the technical literary meaning of the word, he comments on the vis vocabuli χύδην, quae consilium et necessitatem excludit, and cites the Phaedrus (264B) and De Gen. An. (743b23) — Aristotelis de arte poetica liber (Lipsiae 1885) 123. Vettori, as well as Maggi, presumably also saw the two stages in Nature's ‘painting’ an embryo as applicable to two stages in ‘painting’ a play: id est animantem illam, quam prius lineis tantum ductis descripserint, postea coloribus illinunt (loc. cit).Google Scholar

81 ‘How to Study Poetry’ (16), Moralia 1.83. This distinction may be indebted to Aristotle's earlier discussion of mimesis (1447b9-23) in which he excludes Empedocles as Plutarch himself does in this passage. The relation of Plutarch's distinctions to the dichotomy of the true and the fabulous narrative is discussed above (48). Petronius likewise contrasts the poet free to go per ambages of mythical allusion and exalted prophecy to the historian bound to give a scrupulous account of things that really happened as if on oath before witnesses (Saty. 118). See n. 45.Google Scholar

82 See Lowinsky, E. E., ‘The Musical Avant-garde of the Renaissance or: The Peril and Profit of Foresight’ in Art, Science, and History in the Renaissance, ed. Singleton, C. S. (Baltimore 1967) 113–16.Google Scholar

83 ‘A myth (μῦΘοσ) aims at being a false tale, resembling a true one (λόγοσ ψενδὴσ ἐοικῶσ ἀληΘινφῷ); wherefore it is far removed from actual events, if a tale is but a picture and an image of actuality, and a myth is but a picture and image of a tale (λόγοσ μὲν ἔργον, καὶ λόγον δὲ μῦΘοσ εἰκὼν καὶ εἰδωλόν ἐστι). And thus those who write of imaginative exploits lag as far behind historians as persons who tell of deeds come short of those that do them’ (Moralia 4.507-9). The concept of color resembles that of figura in Quintilian which implies a different meaning from the one expressed and occurs frequently in figuratae controversiae (9.1.14, 9.2.65). Figurae may serve as a psychological counterpart of the use Quintilian assigns to hypothesis (5.10.96): ‘Some things, again, which cannot be proved, may, on the other hand, be suggested by the employment of some figure. For at times such hidden shafts will stick, and the fact that they are not noticed will prevent their being drawn out, whereas if the same point were stated openly, it would be denied by our opponents and would have to be proved’ (9.2.75).Google Scholar

84 L. Caeli Firmiani Lactanti Opera Omnia , ed. Brandt, S. and Laubmann, G., CSEL 19 (Vindobonae 1890-97) 3–4. For a brief account of literary studies and the early Christian writers, see Curtius, E., op. cit. 446-67, esp. 454-5 on Lactantius and his influence.Google Scholar

85 One cannot more fittingly stress the importance of these passages than by recalling that they include the principal theoretical citation of Petrarch's coronal oration (trans. Wilkins, E. H., Studies in the Life and Works of Petrarch [Cambridge, Mass. 1955] 306) as well as the fact that Boccaccio relies heavily on the Institutes in his Genealogie deorum gentium. (For the relation of Petrarch's Privilegium to his oration, their sources, and the precedents for, and circumstances of, the ceremony itself, see Wilkins' admirable ‘The Coronation of Petrarch,’ Speculum 18 [1943] 155-97.) The alternative phrasing in Lactantius' epitome of his Institutes (11-2) is worth consulting as a gloss on color and figura in Quintilian's sense.Google Scholar

86 Compare Macrobius' account of the type of fabulae properly used by philosophers: argumentum quidem fundatur veri soliditate sed haec ipsa veritas per quaedam composita et ficta profertur, et hoc iam vocatur narratio fabulosa, non fabula, ut sunt cerimoniarum sacra, ut Hesiodi et Orphei quae de deorum progenie actuve narrantur, ut mystica Pythagoreorum sensa referuntur (Commentarii in somnium Scipionis 1.1-2, ed. Willis, I. [Lipsiae 1963] 5–6). Boccaccio clearly categorizes the types of fabulae on the basis of their narrative similarity to factual truth. The first type superficially lacks all verisimilitude; the second in superficie non numquam veritati fabulosa conmiscet; the third potius hystorie quam fabule similis est; and the fourth is an old wives' tale which nil penitus in superficie nec in abscondito veritatis habet (Genealogie deorum gentilium libri , ed. Romano, V. [Bari 1951] 706-7). The justification of poetry by associating it with history becomes more pronounced in Christian defenses of fiction. (In the final Seniles, for example, Petrarch tries to give Boccaccio's fabellae greater respectability by treating them as historiae — cf. Branca, V., Boccaccio Medievale [Firenze 1964] 102-4, 121-2 nn. 3-6.) The contribution of the Christian euhemeristic, humanistic, and even exegetical traditions to an already strong Hellenistic emphasis upon history in rhetorical and poetic theory, is most important for the origins and later developments of neoclassicism. As Quintilian proclaims, tanto robustior quanto verior (2.4.2). In the context of law, to judge an action one must know what actually happened; history, res verae gestae, becomes the handmaiden of justice; res fictae — in the guise of a false testimony (mendacium) — of injustice. The strong influence of Horace's rhetorical attitudes toward poetry upon Renaissance neoclassicism may best be seen in the opening chapters of B. Weinberg's A History of Literary Criticism in the Italian Renaissance (Chicago 1961). For the possible legal influence on the concept of the poet as liar, see App. A. Google Scholar

87 See St. Gregory's prooemium to his Super eantica canticorum expositio (Migne, PL 79.473): Sic est enim Scriptura sacra in verbis et sensibus, sicut pictura in coloribus et rebus; et nimis stultus est qui sic picturae coloribus inhaeret, ut res quae pictae sunt ignoret. Nos enim, si verba quae exterius dicuntur amplectimur, et sensus ignoramus, quasi ignorantes res quae depictae sunt, solos colores tenemus. Littera occidit, sicut scriptum est, spiritus autem vivificat (II Cor. iii. 6); sic enim littera cooperit spiritum, sicut palea tegit frumentum; sed jumentorum est, paleis vesci; hominum, frumentis. This passage, with the rest of the prooemium, is later borrowed word for word by Richard of St. Victor (PL 196.406A). The exegetical use of the comparison of Scripture with painting influences the medieval scholia, as well as Renaissance commentaries, on Horace's analogy of ut pictura poesis (A.P. 361-5). See Botschuyver's, H. J. Scholia in Horatium (Amstelodami 1935) 1.447, where superficies verborum and empty sonoros versus are compared to what by this time was considered the inferior picture to be viewed from a distance and the serious sensus et materies to the picture considered to be superior and to be scrutinized from nearby.Google Scholar

88 For the early exercises see Reichel, G., Quaestiones progymnasmaticae (Lipsiae 1909); Marrou, H. I., Histoire de Véducation dans l'antiquité (Paris 1960) 238 ff. and passim; Haarhoff, T. J., Schools of Gaul (Johannesburg 1958) 68-78; for their broader context of legal debate, see Matthes, D., ‘Hermagoras von Temnos 1904-55,’ Lustrum 3 (1958) 58-214 and Bonner, S. F., op. cit. 1-26; for the transmission of the exercises in the Middle Ages and Renaissance, see Clark, D. L., ‘The Rise and Fall of Progymnasmata in Sixteenth and Seventeenth Century Grammar Schools,’ Speech Monographs 19 (1952) 259-63 and Baldwin, C. S., Medieval Rhetoric and Poetic (New York 1928) with an English translation of Hermogenes (23-38); for Priscian's Latin version of Hermogenes, see Halm 551-60, and for Salutati's MS of Priscian's exercises with other minor rhetorical manuals, see Ullman, B. L., The Humanism of Coluccio Salutati (Padua 1963) 154; and for the textual transmission of Theon's exercises, see Lana, I., I progimnasmi di Elio Teone (Torino 1959). For Aphthonius, I have used Rhetores Graeci , ed. Spengel, L. (Lipsiae 1854) 2.21-56 and ‘The Progymnasmata of Aphthonius,’ trans. Nadeau, R., Speech Monographs 19 (1952) 264-85.Google Scholar

89 Though Suetonius is not chronologically explicit, he describes exercises preceding the development of the controversiae and suasoriae which are identical to the progymnasmata. Besides explaining figuras, casus, and apologos (i.e. Aesopian tales — Quintilian 5.11.20) in speeches, teachers would compose (as models) narrationes, (briefly or at length), encomia or vituperations, moral essays on the expedient or inexpedient, questionings or defenses of ‘the credibility (fidem) of myths (fabulis), an exercise which the Greeks call “destructive” and “constructive” criticism.’ These exercises are succeeded by debates upon historical narratives (perhaps representing, like suasoriae, famous persons in well-known situations) or upon actual events of recent time (more similar to controversiae). He gives two examples, easily adaptible to novelle, which he implies might illustrate either fictitious or legal materials (‘fictas aut iudiciales’) — Suetonius, Rhet. 1, trans. Rolfe, J. C., LCL, 2 vols. (London 1959). When the rhetoricians took up debates, the earlier exercises seem to have fallen to the grammarians and to have remained with them.Google Scholar

90 Aristotle also observes that ‘we ought not to discuss subjects the demonstration of which is too ready to hand (σύνεγγνσ) or too remote (λίαν πόρρω); for the former raise no difficulty, while the latter involve difficulties which are outside the scope of dialectical training (κατὰ γνμναστικήν)’. This passage from the Topics (105a-9) is quoted by John of Salisbury, along with the Elder Seneca (Controv. 1. pr. 21), in his criticism of over ly loquacious logicians (Metal. 2.8).Google Scholar

91 See AHF 22–6, esp. nn. 23-8, and above 9-14.Google Scholar

92 See AHF n. 6. Isidore describes the distinction between sententia and chria in precisely the same terms as earlier writers and as he himself distinguishes thesis from hypothesis (2.15): De Sententia. Sententia est dictum inpersonale, ut (Ter. Andr. 68): Obsequium amicos, veritas odium parit. Huic si persona fuerit adiecta, chria erit, ita: ‘offendit Achilles Agamemnonem vera dicendo,’ ‘Metrophanes promeruit gratiam Mithridatis obsequendo.’ Nam inter chrian et sententiam hoc interest, quod sententia sine persona profertur, chria sine persona numquam dicitur. Vnde si sententiae persona adiciatur, fit chria; si detrahatur, fit sententia (Etym. 2.11, ed. Lindsay, W. M. [Oxford 1962]).Google Scholar

93 See Reichel, , 36-7, 107-8, and Quintilian 3.5.8-9. The necessity for the hypothesis to introduce not just a particular person but one who is well known to all is an important characteristic of the exemplum in its development from the hypothesis which I have traced in AHF 65-71 — to whose bibliography should be added the citations to Branca, V. and Battaglia, S. in n. 106. Exemplary power is gained from the ‘historically’ recognizable. No statement of this principle is clearer than the one that Dante places in the mouth of Cacciaguida (3.17.136-42). The confinement of the rhetorical hypothesis within the limits of history, however, had often to be relaxed in law as in literature by the use of personae fictae in the interests of a greater power of generalization. See AHF n. 23 to which should be added L. Fuller's Legal Fictions (Stanford 1967), which has an extensive bibliography on the legal implications of H. Vaihinger's Philosophy of ‘As If’ (94-6), and Barfield, O., ‘Poetic Diction and Legal Fiction,’ Essays Presented to Charles Williams (London 1947) 106–27; see also this essay nn. 49, 94, and 127. The persistent demands of exempla, historical or fictional, upon an awareness of a continuing human experience emphasized the advantages of verisimilitude and, indirectly, encouraged the transmission of neoclassical literary attitudes.Google Scholar

94 It is interesting how the medieval arguments over the real existence of universals crossed and recrossed this middle area. John of Salisbury comments that terms which are considered universal cannot be defined by logic or analyzed by grammar too strictly. Relative pronouns themselves which are used to qualify universals grammatically are suspended between the generic and the specific much as an indefinite agent is postulated in the proposal of a law: ‘In the saying: “Wise and happy is the man who has recognized goodness, and has faithfully conformed his actions to this,” the relative words “who” and “this,” even though they do not designate the specific person [and act], are nevertheless in a way limited, and freed of their indefiniteness, by specification as to how they are to be recognized’ (The Metalogicon of John of Salisbury 2.20, trans. McGarry, D. D. [Berkeley 1962] 124 f.). Such pronouns, as well as more definite figurative expressions like ‘A tree both bore the cause of our death, and that of our life,’ do not descend to the particular thing but retain their generic freedom. Cf. n. 127. It is just when such a statement as ‘a good man will be one who …’ becomes a premise in an enthymeme, as Michel observes of Aristotle and Cicero (op. cit. 205-6), that the thesis enters the rhetorical argument.Google Scholar

95 Donaldson, E. T. rightly sees Andreas' scholastic argument on both sides of the Ovidian question as diversion for clerical, as well as courtly, ingenuity (‘The Myth of Courtly Love,’ Ventures: Magazine of the Yale Graduate School [1965] 16–23). The form of the De amore, however, should be placed in the still broader declamatory tradition which had already contributed so much to the medieval literature of débat (see Walther, H., Das Streitgedicht in der lateinischen Literatur des Mittelalters [Munchen 1920] esp. 17-27, 126-35, and Curtius, E., op. cit., 154 f. for the continuing influence of the declamationes attributed to Quintilian). For the early influence of declamatory presentation upon the literary habits of Augustan society, see my paper on ‘The Meaning of Horace's Ut Pictura Poesis’ previously cited. Horace's analogy, in distinguishing a presentation before a friendly audience from one before an impersonal iudicis argutum … acumen (see below), has distant implications for all scholastic debates. For Andreas, I have used the Latin text of E. Trojel as it is reprinted by Battaglia, S. ( Trattato d'Amore [Roma 1947]) and the translation of Parry, J. J. (The Art of Courtly Love [New York 1941]). For dating and biographical information, see Rajna, P., ‘Tre studi per la storia del Libro di Andrea Cappellano,’ Studi di Filologia Romania 5 (1891) 193-272.Google Scholar

96 This view of the moral efficacy of fiction, independent of its preceptorial or allegorical intentions, I have sketched in greater detail in Appendix A. Nearly all subsequent uses of the Stesichorean palinode appear to derive from the Phaedrus (Cf. Ep. 3.319E; Philostratus, Life of Apollonius 6.11). As a prayer for the withdrawal of what had been granted to a previous prayer imprudently made, the term ‘palinode’ occurs in the second Alcibiades (148B). The likelihood of such imprudence is recognized in the Pythagorean prohibition, reported by Diogenes Laertius (8.9), against our ever praying for ourselves on the grounds that we cannot predict what will turn out well for us. A similar difficulty faces the maker of laws (Theat. 179A), and with respect to their hymns, poets must be carefully censored lest, in hope of gaining a blessing, they inadvertently, like Midas, ask for a curse (Laws 801). St. Augustine introduces his Retractationes with an echo of this ancient legal censorship: 'ut opuscula mea sive in libris sive in epistulis sive in tractatibus cum quadam iudiciaria severitate recenseam et, quod me offendit, velut censorio stilo denotem (ed. Kröll, P., CSEL 36 [Vindobonae 1902] 7). For uses of Παλινῳδία by Christian writers for the ‘Stesichorean recantation,’ for ‘repentance’ or simply for ‘changing one's mind’ (palinodiam canere), see Lampe's, G. W. H. Patristic Greek Lexicon. Although there is little chance that much detail of the Phaedrus could have been known to Andreas, Plato's dialogue bears a striking resemblance to the De amore in theme and structure. Both works concern the arts of love and of discourse. The relation between Socrates and Phaedrus is close to that between Andreas and Walter: Andreas gives in to Walter and writes an ars amoris of which he claims to disapprove. Socrates gives in to Phaedrus and rewrites Lysias‘ praise of the non-lover's superiority in the art of love whose thesis he later reverses in his palinode. Lysias’ pragmatic emphasis upon the non-lover's ability to manipulate the beloved expediently reflects the intention behind Andreas‘ Ovidian ars amoris. Socrates’ recasting of the speech, though it purports to take the same side, functions, in effect, as Ovid's remedium amoris in that its undisguised banality clearly exposes the vicious assumptions which Lysias' poor composition had obscured. Similarly, the banality of the anti-feminist argument in Andreas' reprobatio, some of which is no less secular in its terms than his preceding Ovidian arguments, might equally emphasize for Walter as a young courtier the triviality of the assumption behind such arguments. As Socrates' palinode in praise of the Platonic Eros offers an escape from the secular context in totally new terms, so Andreas' palinodial conclusion urges Walter to accept the broader Christian perspective of divine love. Although for Socrates the contemplative garden of the Muses is turned inside out (since the active life of the ‘outside world’ is within the enclosed city, 227-30), both speakers, having reaffirmed that all artists must one day be judged as men, lead their young listeners back out of the scholastic locus amoenus. Much closer in time to Andreas' relation to Walter is St. Augustine's to his student Licentius in their ‘retreat’ at Cassiciacum, see n. 99.Google Scholar

97 If one allows for the arbitrary inclusion of nearly all valuable activities in his concept of ‘play’ and the exclusion of clearly pejorative uses of the word, Huizinga's general description of the ‘rules of the game’ in Ch. 1 and the materials in Chs. 6 and 9 are useful for showing how the division of the artistic and prudential faculties manifests itself in various forms of discourse (Homo Ludens [Boston 1967]). C. S. Singleton mentions the applicability of Huizinga's study, which itself refers to the amatory debates as a form of ‘play’ (125), to a definition of courtly love (‘Dante: Within Courtly Love and Beyond’ in The Meaning of Courtly Love , ed. Newman, F. X. [Albany 1968] 47-8). Abbé Norbert De Paepe, ‘Amor und verus amor bei Andreas Capellanus. Versuch einer Lösung des reprobatio-Problems,’ Mélanges offerts à René Crozet (Poitiers 1966) 921-7, sees the courtly conversations also as a kind of play (Spiel). The impossibility, he feels, of this more refined love's existing in the real world outside of the fictional debates forces Andreas, when the fiction can no longer be sustained, to accept physical love as the only possibility, to judge it, and to reject it in the reprobatio .Google Scholar

98 Since Boccaccio (Comento Lez. 16) and Petrarch in his many references to the controversiae in his Epistolae Familiares do not distinguish the works of the younger from the elder Seneca, and similarly since such MSS as that of Salutati (cf. Ullman, B. L., op. cit., 171), as indeed many complete Renaissance editions, contain work of both men without distinguishing between them, it is difficult to know how widely the father's declamations were known in the earlier Middle Ages. Fifteen of the ‘controversial’ plots are represented in the gesta romanorum; E. Norden refers to verse paraphrases of the declamations of Seneca and Quintilian (Die antike Kunstprosa [Leipzig 1898] 897–8); Ullman, B. L. points out that the declamations attributed to Quintilian, at least, were represented in the florilegia which probably originated in northern France in the 12th century (CP 27 [1932] 35-7); to the references to Quintilian in n. 95 may be added St. Jerome's citation from Decl. 13.2 quoted in the Historia Calamitatum of Abelard (transl. Muckle, J. T. [Toronto 1954] 52.) Gilbert de la Porrée refers directly to Seneca's declamations (PL 64.1255BC), and John of Salisbury, in his criticism of trivial subtleties of school logicians (Metal. 2.8) quotes Seneca's comment on the excessive subtlety of the rhetors (Controv. 1.pr.21) and later reminds Becket (Metal. 4.prol.) of Seneca's words on the pleasures of reminiscence (Controv. l.pr.l). John also compares the mock battle of military training (militie imaginarie) with the early education of a logician (Metal. 3.10) in a manner reminiscent of Seneca's contrast of the training in both the gladiatorial and rhetorical schools to the actual combat in the arena and forum. Digladiari is an early metaphor for verbal disputes (Acad. 1. frag. 1, LCL 457) and is retained in humanistic criticism of scholastic debate. As in Andreas' second dialogue, the miles amoris of Ovid's amores is compared to the real soldier with the connotations, similar to Seneca's, of leisure (otium) and escape as opposed to trial and responsibility (cf. Robertson, D. W., A Preface to Chaucer [Princeton 1969] 408-10). Given the relationship between their patrons, and John's frequent embassies, Andreas might easily have known the Metalogicon (1159) with its direct reference to the primo declamationum of Seneca, and may even have met John himself.Google Scholar

99 As Andreas finds Walter already in love and agrees to teach him its art so that he may proceed more cautiously in the future (Praef.), so St. Augustine finds Licentius so in love with poetry (fingendis versibus vacavit, quorum amore ita perculsus est) that he feels responsible to persuade him to give philosophy the larger share of his affection (Contra Acad. 3.1). The interesting parallel between the two texts lies in the way Andreas and St. Augustine go about discouraging their students. They both wish them to learn to practice their respective artes, amatoria and poetica, with the greatest skill, although neither teacher approves of the arts themselves. Andreas teaches the plenamque amoris doctrinam, not so that Walter should become a lover, but that eius doctrina refectus et mulierum edoctus, he become better able to resist the temptation (362). Not to have written the treatise would be to have given in to inertia (416). St. Augustine, likewise, finding Licentius eagerly composing verses, tells him: Opto quidern … tibi ut istam poeticam quam concupisti, complectaris aliquando: non quod me nimis delectet ista perfectio; sed quod video te tantum exarsisse, ut nisi fastidio evadere ab hoc amore non possis; quod evenire post perfectionem facile solet (Answer to Skeptics, trans. Kavanagh, D. J. [New York 1943] 128, 140). Only through perfecting an art will one either grow weary of it or become impatient with it through being difficult to please. In this context, a teacher may, with complete sincerity, give the best possible instruction in an activity of which he disapproves. With regard to the word fastidium it is interesting that the first reason given why Boccaccio's brigata should return to Florence is ‘aceiò che per troppa lunga consuetudine alcuna cosa che in fastidio si convertisse nascer non ne potesse’ (10. Concl. 6). Storytelling, as other arts, if continued too long (so, perhaps, as to be no longer a challenge), may become irksome, and to avoid its becoming so, one returns to the active life of the city. This perhaps is the same fastidium that Augustine hopes will return Licentius' attention to philosophy.Google Scholar

100 It is interesting to see in how many of the 221 stories, whether among the fifteen derived from the controversiae or not, this structure is apparent in the gesta Romanorum as printed by W. Dick in Erlanger Beiträge zur englischen Philogogie 7 (Leipzig 1890) 2.1-273. A law, a proclamation, or a contract is initially given which the subsequent incidents violate in such a way that its application is ambiguous and some equitable adjudication is required. The structure is syllogistic in that the law forms the major premise while the act falling under it forms the minor premise. The brief adaptation of the raptor duarum (3) opens with a formula repeated twenty-four times in various stories: such and such a king regnavit, qui statuit pro lege, quod. … The legal thesis once established, the theme is completed by the circumstances to be judged, the hypothesis, which are introduced by some variation of accidit casus, quod. … As Bornecque observes (Les déclamations 84-6), the tales arrive at a ‘conclusion’ whereas the controversiae remain ‘undetermined.’ In the case of the raptor duarum, the judgment in the tale is that offered by Seneca's lenient rhetor who argues that inter pares sententias mitior vincat (1.5.3). The rhetorical origins and transmission of this structure in medieval narratives, including that used in the setting up of such questions as those debated in the De amore and the Filocolo, are revealed in the school illustration of a dispute, transcribed from Cicero (De inv. 2.153-4), in a Carolingian treatise. Nam scripti controversia est ea, quae ex scripta lege nascitur, hoc modo: Lex: qui in adversa tempesiate navem reliquerint, omnia amittant, et eorum sint onera et navis qui remanserint in ea.' A narrated incident then follows which is not quite covered by the law and must be adjudicated (The Rhetoric of Alcuin & Charlemagne 9, trans. Howell, W. S. [Princeton 1941] 76).Google Scholar

101 The structure of the questione d'amore is foreshadowed in Plutarch's Amatorius. The participants retreat from the noisy town of Thespiae to the ‘garden’ of Mount Helicon, where a series of general questions about love arise from a narrative incident involving a wealthy young widow's pursuit of a handsome but poor adolescent. One such general thesis asserts that ‘To choose a woman for her wealth rather than for her character or birth would be ignoble and base; but if character and good breeding are added, it would be ridiculous to shun her’ (Moralia 754A, trans. Helmbold, W. C., LCL [London 1961]).Google Scholar

102 The greater luridness may simply result from the fact that the declamations deal with criminal, rather than social, quaestiones. It is, however, very difficult to gauge the attitudes of any earlier age toward matters we may regard as sensational. When Boccaccio, for instance, comments atlength upon Dante's reference to‘ Seneca morale’ (1.4.141), he lists as praiseworthy books the most famous of the moral essays, and continues: ‘Ma sopra tutti fu quello delle Pistole a Lucillo, nel quale senza alcun dubbio, ciò che scriver si può a persuadere di virtuosamente vivere, in quel si contiene: e quello ancora che si chiama le Declamazioni’ ( Il Comento di Giovanni Boccaccio sopra La Commedia Lez. 16, ed. Milanesi, G., 2 vols. [Firenze 1863] 1.397 f.). Subsequent references will be to this edition and cited in the text. The declamations, along with the Epistulae Morales, are recommended especially for their moral instruction in contrast to the Apocolocyntosis Divi Claudii, which is ‘molto più poetico che morale.’ However more delicate we may consider Andreas' courtly refinements than the mores of the controversiae, he, nevertheless, suggests that force is not out of place with country girls (1.11).Google Scholar

103 As in Seneca (cf. Controv. 2.7) traveling spouses often return home to similar complications in the declamations. Among the shorter exercises attributed to Quintilian occurs the following (n. 347): Adulterum cum adultera liceat occidere. Uxor peregrinantis mariti mortem rumore cognovit. Heres inventa, nupsit adolescenti cuidam, & domum in dotem dedit. Supervenit maritus nocte: utrumque occidit. Reus est caedis. Variations upon the ‘spouse surprised’ or the ‘double seduction’ occur again and again, often qualified by having a rich man in competition with a poor man. In fact, the conflict of interests between rich and poor pervades the controversiae almost as much as the considerations of social rank dominate the dialogues of the De Amore. Among those attributed to Quintilian involving distinctions of wealth or sexual questions or both, with occasional travel, see nos. 247, 251, 252, 257, 259, 262, 270, 276, 280, 286, 301, 335, 343, 344, 347, 363, 379, as well as a number assigned to Calpurnius Flaccus (M. Fabii Quintiliani … Declamationes XIX majores et … CXLV minores … curante Petro Burmanno (Lugd. Bat. 1720]).Google Scholar

104 The shaded leisure necessary for having disputations about love is clearly described: Quadam ergo die, dum sub mirae altitudinis et extensae nimis latitudinis umbra pini sederemus et amoris essemus penitus otio mancipati eiusque suavi et acerrimo disputationis conflictu studeremus investigare mandata … (176).Google Scholar

105 ‘Io mi sono un povero pellegrino d'amore, il quale vo cercando una mia donna a me con sottile inganno levata da' miei parenti; e questi gentili uomini i qualicon meco vedete, per loro cortesia nel mio pellegrinaggio mi fanno compagnia’ (Il Filocolo, ed. Battaglia, S. [Bari 1938] 297). Subsequent references will be to this edition and cited in the text.Google Scholar

106 Branca, , Boccaccio Medievale (Firenze 1964) 109. See n. 100. Branca's study, especially Ch. 4 ‘Le nuove dimensioni narrative’ (originally presented at the University of Padua in 1954), is particularly helpful for the exemplary structure of the novelle. See as well, Battaglia, S., ‘Dall’ esempio alla novella' (Filologia Romanza 7 [1960] 21-84) and ‘Carattere paradigmatico e qualità realistiche dell' esempio medievale,’ both included in Giovanni Boccaccio e la riforma della narrativa (Napoli 1969) 1-81. Important for the origins of Boccaccio's questions, several of which are treated also by Andreas, is Rajna, P., ‘L'episodio delle questioni d'amore nel Filocolo del Boccaccio’ (Romania 31 [1902] 28-81). It is interesting how the method of speaking in utramque partem is reflected in the names for divisions of poems in some of the source material in the sense of taking turns: partimen, joc partit, jeu parti, partito, giuochi partiti (52 f.). Rajna comments on the close relation between the questioni and the novelle: ‘nel Filocolo questioni d'amore che prendono volentieri aspetto e possono aver anche contenuto di novelle, nel Decamerón novelle che possono dar materia a questioni’ (34). Questions 10 and 12, for instance, contain narrative possibilities ingenious enough to yield novelle. The eighth question, to look back on the preceding stage of development, is identical to those of Capellanus: ‘quale di due donne deggia più tosto da un giovane essere amata, piacendo equalmente a lui amendue, o quella di loro che è di nobile sangue, e di parenti possente, e copiosa d'avere molto più che il giovane, o l'altra la quale non è nobile né ricca, né di parenti abbondevole quanto il giovane’ (343).Google Scholar

107 ‘E da ora, s'io nel mio parlare troppo mi distendessi, a voi e appresso agli altri circustanti dimando perdono, però che quello ch'io intendo di proporre interamente dare non si potrebbe a intendere, se a quello una novella, che non fia forse brieve, non precedesse’ (311). This general process corresponds to the rhetorical revelation of general truths within the particular circumstances of a case. When to a universal denunciation of vice, Quintilian says, one adds a name, it becomes an accusation. No general truth is so universal, however, that it can fit any case without some link to the specific controversy, lest it appear tacked on. Moral commonplaces must seem to arise from the very nature of the circumstances and persons involved (2.4.22-32). On the other hand, remove those persons and the definite question returns to a general disquisition (De part orat 106).Google Scholar

108 The question of what should be sought and avoided, so fundamental to the questioni d'amore and to the novelle, almost inevitably involves the comparison of what is more (or less) honestum, aequum or utile and the means to attain (or avoid) it. Comparatio was regarded in the rhetorical treatises primarily as a complication of the qualitative issue (De Part Orat 66; De Orat 3.111-17). See n. 40. Plutarch's ‘comparisons,’ appended to certain of the Lives, based on the actions just narrated, resemble the iudicia of the love debates whose questions demand a ‘fiction’ of detail. Russell, D. A. calls them ‘model answers for a rhetorical exercise: you have heard the two stories, what points of similarity and difference can you see?’ He cites Quintilian on encomium and invective and on comparison (2.4.21): ‘“Which is the better man and which the worse?” This gives double the amount of material to handle and deals not only with the nature of virtues and vices but with the degree (modus) in which they are present. Plutarch's sunkriseis are specimens of this kind of work’ (op. cit. 110).Google Scholar

109 All references will be to Il Decamerone , ed. Branca, V., 2 vols. (Firenze 1960) and will be cited in the text.Google Scholar

110 See above 52-58. As I suggested in AHF n. 86, fiction is a form of logical presentation which can avoid the paradox of an infinite regress.Google Scholar

111 The definition of a question to be discussed in utramque partem, which underlies Boccaccio's story 10.5 and to some degree controls its structure, seems equally to be the objective of Chaucer's Franklin who concludes his analogous tale by saying: ‘Lordynges, this question, thanne, wolde I aske now, / Which was the mooste fre, as thynketh yow? / Now telleth me, er that ye ferther wende. / I kan namoore; my tale is at an ende’ (The Complete Works of Geoffrey Chaucer, ed. Robinson, F. N. [Cambridge, Mass. 1933] 174). Robinson cites other instances of rhetorical questions asked at the end of stories or episodes, but the one he cites from the Canterbury Tales (1.1347 ff.) solicits no discussion from the pilgrims themselves (774, 831).Google Scholar

112 On several occasions Boccaccio reasserts the poet's immunity from doctrinal criticism and from the logical severity of disputation by distinguishing the schoolroom from the garden — both of which are secure from the practical exigencies of the forum: Phylosophorum insuper est in ginnasiis disputare; poetarum in solitudinibus canere‘ (Gen. deor. gent. 14.17, cf. 14.4, 14.5, 14,7, 14.11 and Osgood's n. 1, 14.19). Defending the length of the Decameron, he says he has written for those at leisure, for 'le cose brievi si convengon molto meglio agli studianti, li quali non passare ma per utilmente adoperare il tempo faticano. … E oltre a questo, per ciò che né ad Atene né a Bologna o a Parigi alcuna di voi non va a studiare, più distesamente parlar mi si conviene che a quegli che hanno negli studi gl'ingegni assottigliati’ (Concl. 21). And so, also, the ‘strettezza della intenzion,’ which threatened the tales of the tenth day, ‘molto più si conviene nelle scuole tra gli studianti che tra noi’ (10.6.3). Similarly, although the variety of errors and of (unspecified) works, clearly including doctrinal matters, permits no certainty, the phrasing of the ‘condemnation of 1277,’ which includes Andreas‘ De amore, allows for the possibility that certain objections may have been raised primarily because the students’ questions were indecorous when debated in the schools where they were out of place: nonnulli Parisius studentes in artibus proprie facultatis limites excedentes quosdam manifestos et execrabiles errores … quasi dubitabiles in scolis tractare et disputare presumunt (Chartularium Universitatis Parisiensis, ed. Denifle, H. [Paris 1899; repr. Bruxelles 1964] 1.543). The generic context of these distinctions goes back at least to Plato's comparison of the philosopher with the rhetorician (Theaet. 172C-75), which is reflected in Boccaccio's comparison of the poet with the lawyer (Gen. deor. gent. 14.4). Being free from the timekeeper of the forum, the philosopher may converse in peace at his leisure, pausing and continuing his argument, or leaving it for a new topic or illustration at will, not caring how long or short his discourse is so long as he fully presents the truth (172D-E). Plato compares this type of leisure explicitly to that appropriate to telling stories (Rep. 376D). As Orlando says, ‘there's no clock in the forest’ (3.2.303). The specific context pertains to a distinction between types of philosophical discourse itself once one enjoys the shaded leisure of contemplative activities. Opposed to the specialization of eristic disputation proper to the schools is the liberal philosophical discussion about the general nature of virtues and vices proper to their novelistic illustration in the garden. The more liberal the philosophical intention, the closer to the poet's. An exact parallel, furthermore, exists between types of rhetorical discourse. Long speeches on questions of justice and injustice belong to the liberal intentions of deliberative oratory, while disputation among private persons over contracts is broken up into bits of questions and responses (Soph. 225A-C). The most detailed discussion of these types of rhetorical discourse is in Aristotle's Rhetoric (3.12). By Horace's time Aristotle's distinctions have been applied to poetry and underlie his famous analogy of poetry with painting. As in philosophy, the most liberal intention of rhetoric is nearest the poet's. The most liberal terms of both disciplines, regarding technique as well as intention, merge and become almost indistinguishable in literary theory. In writing brief moral essays (genus exercitationum) Cicero says he transposes materials treated in the schools disputatiously (in scholis Θετικῶσ) to his own rhetorical style of discourse (ad nostrum hoc oratorium transfero dicendi genus). If questions such as ‘What then is good?’ are discussed too dispassionately (lentius), they will be tiresome: they need to be ‘illuminated’ (illustranda) by the life and actions of eminent men, for ‘wordy discussion of them seems to be excessive subtlety’ (Parad. Stoic. 5, 10). This is precisely Boccaccio's view of the function of narrative. See n. 116.Google Scholar

113 Analogous to the determination by the leggi of what materials, or arguments, were to be included in the novelle of a particular day, Quintilian remarks (7.1.14) that even ‘in scholastic themes … the laws are sometimes stated merely with a view to connecting the arguments of the case (in scholasticis ponuntur ad coniungendam modo actae rei seriem).’ With respect to the leisure shared by the literary artist and the rhetor of the schools in their truant retreat from the active life of the forum, two anecdotes stand out in the long tradition, already fully suggested in Plato (Phaedrus 278E), which culminates, for antiquity at least, in the Dialogue of Tacitus. In explaining why he fails as a declaimer though he excels in the forum, Cassius Severus comments that he is used to addressing a judge, not an audience seeking entertainment; to responding to an adversary in debate, not to himself; and to avoiding superfluous words as much as those which would prejudice his case (Controv. 3.pr.12). In the same spirit Quintilian relates how the poet Accius, when asked why — given his great skill in repartee in his tragedies — he did not become a lawyer, ‘replied that in his plays the characters said what he himself wanted them to say, whereas in the courts his adversaries would probably say just what he least wanted them to say’ (5.13.43). The artistic construction of colores could not be made ‘against the clock.’ As in the declamations and Lactantius' Divine Institutes, the verb colorare, meaning a construction (often deceitful) put upon motives or actions, continues to be used by Andreas (1.9) and Boccaccio (Decam. 1.3.7; 8.7.10; 9.1.7).Google Scholar

114 When, for instance, specific leges or regulae are ‘unexpressed,’ the ‘natural laws,’ expressing the universal intentions of justice which the writer shares with his readers, must take their place. Fiction, released from the prescriptions of law or preceptive philosophy as well as from the stipulations of history, operates upon the same principles as equity. See nn, 41 and 43.Google Scholar

115 Quoted from Elizabethan Critical Essays, ed. Smith, G. G., 2 vols. (Oxford 1950) 1.168 f. Subsequent references to Sidney will be to this edition and cited in the text. Though part of the scientific and psychological materials of my essay to follow, certain topics discussed by Fisher, N. W. and Unguru, S. in ‘Experimental Science and Mathematics in Roger Bacon's Thought’ (Traditio 27 [1971] 353-78) might be mentioned here. Bacon's attempt to describe quality in quantitative terms (major vero pars praedicamenti qualitatis continet passiones et proprietates quantitatum 373 n. 90) is brought back into a balanced relation of quantitative to qualitative measurement congenial to fiction by a strongly qualitative strain in his conception of quantity itself. First, quantity and mathematical demonstration are closely associated with sensory perception (sed quantitas est maxime sensibilis 373 n. 89) and with visual exemplification which renders the demonstration apprehensible and persuasive (exemplum sensibile, et experientiam sensibilem figurando et numerando, ut omnia ad sensum manifestentur: propter quod non potest esse dubitatio in ea‘ 360 n. 32). Such insistence on visual exemplification bears an interesting relation to St. Thomas’ emphasis upon the value of sensory images in Scripture (S.T. I I, qu. 1, ar. 9) and to St. Augustine's upon the power of emblematic figures to move the emotions (Ep. 55.11.21). All these views reflect the rhetorician's faith in enargeia to move an audience. Second, the proper observer does not record isolated phenomena at random but is directed by a prior intuition about which questions will reveal causes for which effects. Bare facts alone, like Sidney's ‘bare Was’ which yields ‘no cause’ unless it be ‘poeticall,’ will, for Bacon, produce only the bare truth without the cause (‘nudam veritatem sine causa’ 365 n. 53). So, in law, Cicero says (Parad. Stoic. 24), if you simply give the bare facts, their qualitative ‘reality’ necessary for judgement is apt to be lost (nuda ista si ponas, iudicari qualia sint non facile possunt). Bacon's psychological emphasis upon persuasive demonstration and upon the observer's seeing individual experiments in the light of his total experience gives the measurement of quantities a qualitative shading. Analogously for law, Quintilian could see the quantitative status definitiuus ultimately reduced to qualitative considerations (7.4.15-6). The similarity, furthermore, of Bacon's terms to Cicero's distinction between divination and science is striking. Divination is based on random coincidence and, never moving beyond pure conjecture can never establish causes for effects (De div. 1.12-6, 24 f., 35, 86, 127; 2.146 f.). It is pure empiricism and can only be thought of as an art at all sub specie aeternitatis — when enough time shall have elapsed for sufficient recurrence to be observable. Among the three non-legal types of narrative, fabula would correspond to divination, res gestae to history. In the humanistic defenses of literature, such as Boccaccio's and Sidney's, it is plausibly presented causation which most distinguishes poetry, in subject matter and structure, from old wives' tales, a phrase which St. Augustine himself had used for ancient divinations (C.D. 4.30).Google Scholar

116 The need for relaxation and the value of entertainment, proclaimed by Plato (Laws 653C-58A) and Cicero (Deorat. 2.19-24), is stressed as well by moralists and theologians in the Middle Ages. The terms of their arguments retain something of the ancient defense of the schools of declamation despite the ambivalence of later ancient attitudes toward the rhetor's, or the poet's, retreat from the active life of the forum in Quintilian (10.3.22-7) and Tacitus (Dial. 9, 12-3). Quintilian, for instance, comments that although the schools present but a false semblance of forensic practice (in falsa rerum imagine) and that a young man should not remain too long in their shade (umbra) lest he later fear real dangers (vera discrimina) as he might the full sun (velui quendam solem), declamatory exercises may indeed nourish eloquence with a refreshing variety and relief from the contentionum asperitate fatigata. For if one only treats the cotidiana pugna of legal actions, his mind will become dull and stiff (10.5.14-8). I have discussed the antithesis between the ‘shade’ of the rhetorical schools and the ‘sun’ of the forum in JWCI 36 (1963). These comments of Quintilian, as well as those in 11.3.26-7 and 12.6.5-6, should be added to my discussion there, along with Horace's own reference to his ‘pilgrimage’ to Greece where the shaded leisure of academic gardens is contrasted with the hot exertion of the active life (Ep. 2.2.41-8). After reading Homer in Rome, he seeks paulo plus artis in Athens to learn to distinguish the straight (rectum) from the crooked (curvo) and to search for the truth in the groves of the Academy (inter silvas Academi quaerere verum). But dura tempora snatch him from the loco grato, and the heat (aestus) of civil war thrusts him unprepared (rudem) into active service (in arma). St. Augustine seems to have associated the etymology of the word academia itself with the idea of retired seclusion: sed ab hoc jam litigioso tribunali secedamus in aliquem locum, ubi nobis nulla turba molesta sit; atque utinam in ipsam scholam Platonis, quae nomen ex eo dicitur accepisse, quod a populo sit secreta (Contra Acad. 3.9.18). His editor, Kavanagh, D. J., suggests that he regarded the word as ‘composed of ἔκασ = afar off, and δῆμοσ = the populace,’ and points out that Suidas thought of the original Greek form as Ἑκαδημία (255-6).Google Scholar

117 All references to the Vita are cited from Vita di Dante e difesa della poesia, ed. Muscetta, C. (Roma 1963) 33–9. For the relation of the illustrative and hortatory exemplum of rhetoric to the heuristic and eschatological exemplar of philosophy and theology, see AHF 65-71 and below n. 127. In addition to the bibliography there, see the essays of V. Branca and S. Battaglia cited above. Battaglia particularly stresses the ethical neutrality of the exemplum as an illustration for either good or bad actions. He cites Festus' exemplum est quod sequamur aut vitemus and Forcellini's definition of an example as something proposed for imitation tum in bonam, tum in malam partem (10 n. 1). The Academic treatment in utramque partem of the ancient topics of moral philosophy could be easily adapted to exemplary narrative by casting them in the form of what to seek and what to avoid. Aristotle associates the form with deliberative rhetoric (1358b22-9), whose principal virtue is prudence through which happiness (εὐδαιμονίαν) may be attained (1366b20-2). The attainment of happiness governs what things are to be pursued or avoided in oratory (1360b2-14) as it does the selection of actions in the drama, since men are κατὰ δὲ τὰσ πράξεισ εὐδαίμονεσ (1450a20). A passage in the Rhetoric discussed earlier in relation to equity (n. 33) shows how the rhetorical tradition might transmit this form of representing ethical choice for literary purposes (1417a16-35). Here (as later in 1418b1-3) Aristotle insists that the orator draw attention ot his own prudence and virtue rather than to his skill in argumentation (ἀπὸ διανοίασ). If an expression revealing prudence seems unprepared for, he should add a reason (αἰτίαν) for it as Sophocles does in Antigone. The same thing would hold true for adding motivations through episodes to any dramatic hypothesis (cf. AHF 43-5). Of the endless considerations of what to pursue or avoid, the following is a representative selection: Tim. 87B; Top. 104b1 ff; Rhet. ad Her. 3.4; De inv. 2.158-9; Top. 84; De orat. 2.67; De off. 1.153; Hor. Serm. 1.1; Juv. Sat 10; Per. Sat. 4; Sen. De tran. 9, Ep. 22, 31, 32, 60, 95; Inst. orat. 3.6.56; Diog. Laer. (Epicurus) 10.30, 117, 132; Isidore Etym. 2.4.3.Google Scholar

118 La Divina Commedia 1.5.118-20, ed. Vandelli, G. (Milano 1949).Google Scholar

119 Il Comento Lez. 20: ‘Col quale come ella poi si giungesse, mai non udii dire, se non quello che l'autore ne scrive, il che possibile è che così fosse. Ma io credo quello essere piuttosto fizione formata sopra quello che era possibile ad essere avvenuto, chè io non credo che l'autore sapesse che così fosse’ (op. cit. 1.477-9).Google Scholar

120 Comentum super Dantis Aligherij Comoediam, ed. Vernon, G. W. and Lacaita, J. P. (Florentiae 1887) 1.213: ‘Dicit ergo: ma dimme a che e come, idest ad quid et qualiter amor concedette, idest permisit tibi, che conoscessi i dubiosi desiri, quasi dicat: quomodo potuisti cognoscere quod Paulus amaret te concupiscenter, quod erat dubium propter vinculum affinitatis!’ For Cicero, although the treatment of how (quemadmodum) something came about might enrich the narratio (De oral. 2.328), it lent its greatest support to the confirmatio, especially when one asked about the modus of an action, that is quemadmodum et quo animo factum sit (De Inv. 1.41).Google Scholar

121 Quoted from the commentary of D. W. Lucas on Aristotle's use of ‘universal’ in Poetics 1451b8 (Oxford 1968).Google Scholar

122 In the medieval considerations of optics, a similar ratiocinative power is accorded to the senses. Alhazen, Witelo, and Roger Bacon distinguish simple perception from apperception, the latter becoming a kind of argumentum — a syllogistic process so rapid that we are unaware that it is taking place (cf. De Bruyne 2.244 ff.). In the process from perception to apperception the individual particular becomes specific. St. Thomas, also, accounts for our sensory enjoyment of proportion by granting a rational potentiality to the senses: sensus ratio quaedam est et omnis virtus cognoscitiva (S.T. 1.1 qu. 5, art. 4, ad. 1). Cf. Sir David Ross: ‘The passage from particulars to universals … is made possible by the fact that perception itself has an element of the universal; we perceive a particular thing, it is true, but what we perceive in it is characters which it shares with other things’ (Aristotle [London 1966] 55).”Google Scholar

123 The influence of Roman legal terminology upon St. Paul's vocabulary and, through his epistles, upon the language of Scriptural exegesis, and, through exegetical interpretation of both sacred and secular texts, upon the transmission of literary theory from antiquity would be a fruitful subject for a separate study. Certain correspondences, furthermore, between the four ‘rational’ status and the four ‘levels’ of exegesis and between the subsidiary ‘legal’ status and the semantic challenges to the exegete are suggestive. For example: one must arrive at a ‘position’ to ‘see’ the meaning of the case and of the text; there may be several status in a single case as there may be several ‘points of observations’ for a single text; and in neither the case nor the text need all the status nor all the meanings be involved or brought out. In the Middle Ages, the invocation of the author's intention (intentio, sententia) — corresponding to the lawmaker's intention (voluntas, sententia) — in the face of possible misinterpretation at a later date of the literal meaning of the fiction (figmentum) — corresponding to the written law (scriptum) — may often refer to a legal analogy instead of an allegorical involutum. Juan Ruiz's defense of his work is a case in point: ‘Lo primero, que quiera bien entender a bien juzgar la mi entención por que lo fiz,’ e la sentencia de lo que y dize, e non al son feo de la palabras; e segund derecho, las palabras sirven a la intención, e non la intención a las palabras' (Libro de Buen Amor, ed. Willis, R. S. [Princeton 1972] 10). In objecting to lawyers as critics of poetry, Boccaccio points out, as Cicero says of the jurisconsults (Pro Mur. 27-9), that they offer only a strict interpretation of the law without understanding its intentions (sola scriptorum valent memoria non ex ingenio, sed ex Uteris legum latorum iura reddentes). They limit themselves to particular trivialities, such as nunquid ardens femina solvi posset a frigido viro, taking no note of important — though remote — manifestations of nature (circa excelsa aut semota nature). To become totally involved in statutes that vary relative to place and time is to ignore the stabilis and fixa scientia, eternis fundata atque solidata principiis, of the laws of nature which concern the poet (Gen. deor. gent. 14.4). See the notes to this chapter in C. G. Osgood's edition, particularly n. 12, which locate literary terminology firmly in ethics and the most liberal concerns of legal science (Boccaccio on Poetry [New York 1956] 148-53). In a certain sense, the entire philological enterprise behind the Genealogie is ‘exegetical’: a reconstruing of the old Law, the scriptum of the pagan past, in terms of the new Law, the voluntas of the Christian present. This is possible in literature because fiction, having no one way per figmentum vera referendi (Macrob. Comm. 1.2.10), can be seen through like a velamentum so that the poets may appear as the eruditos viros they really were. Having perished from not being understood (falsa opinione perisse), they are now redivivi and restored as reipublice insignes, through whom ad altiores sensus etiam ingenia legentium excitantur (15.1). This hermeneutic spirit combines easily with Cicero's expression of literary culture in the Pro Archia which concludes Boccaccio's fourteenth book. See nn. 43 and 128.Google Scholar

124 Gen. deor. gent. 14.9. The passage is strikingly similar to Plutarch's didactic observation: ‘Philosophers, … for admonition and instruction, use examples (παραδείγμασι) taken from known facts; but the poets accomplish the same result by inventing actions of their own imagination, and by recounting mythical tales’ (Mor. 20C). Boccaccio twice states that what the syllogism is to philosophy, fiction is to poetry (ibid. 14.9, 14.17). The second passage clearly implies the syllogistic coherence of the poet's invention.Google Scholar

125 Letters of St. Augustine , trans. Cunningham, J. G., A Select Library of the Nicene and Post-Nicene Fathers of the Christian Church , ed. Schaff, P. (New York 1892) 1.230; Latin is cited from PL 33.77. For the possible association of the Holy Ghost (as the ‘author’ of the Scriptures) with the Ciceronian conception of the Platonic Ideas as permanent principles of intelligibility constituting the animum mundi (Acad. 1.24-33), see above 21 f. Google Scholar

126 Il Comento Lez. 19: ‘In queste parole intende l'autore d'ammaestratci, che noi non dobbiamo con la meditazione semplicemente visitar le pene de’ damnati; ma visitandole e conoscendole, e conoscendo noi di quelle medesime per le nostre colpe esser degni, non di loro, che dalla giustizia son puniti, ma di noi medesimi dobbiamo aver pietà, … E usa l'autore di monstrare di sentire alcuna passione, quando maggiore, e quando minore in ciascun luogo: e quasi dove alcun peccato si punisce del quale esso conosca se medesimo peccatore (op. cit. 1. 474). Such emotional effects Plato had recognized and distrusted in mimetic representation, for few ‘are capable of reflecting that to enter into another's feelings must have an effect on our own: the emotions of pity our sympathy has strengthened will not be easy to restrain when we are suffering ourselves’ (Rep. 606B, trans. Cornford, F. M.). Diogenes Laertius comments (3.80) that Plato often intermingles myths ‘with his works in order to deter men from wickedness, by reminding them how little they know of what awaits them after death’ (Lives of Eminent Philosophers, trans. Hicks, R. D., LCL , 2 vols. [London 1925]). See A. H. F. n. 78. Quintilian sees the power to enlist the feelings of a judge as the spiritus and animus of oratory (6.2.5-7). He describes the force of visual images to arouse the emotions in a passage (6.2.24-36) which forms the basis and source of Hamlet's reflections on his being moved as ‘in a fiction, in a dream of passion’ by the players (2.2.576-98). This power to move the listener by vividly describing or presenting either one's own emotions or those of others is a rhetorical commonplace of which Dante, as indeed Boccaccio (cf. Decam. 5.8), takes full advantage. Walter Map, in a passage rarely cited by literary historians, brings out the ethical psychology underlying the medieval justification of secular narrative. He distinguishes, furthermore, the emotions of this world, under the power of the artist, from those of the next. To turn from the first to the second constitutes a kind of palinode. Nam historia, que veritate nititur, et fabula, que ficta contexit, et bonos fine florenti beant, ut ametur benignitas, et fedo malos dampnant interitu, volentes inuisam reddere maliciam; sibique succedunt inuicem in scripturis tum aduersitas prosperitati, tum e conuerso muiacione frequenti, quatinus utraque semper habita pre oculis neutri fiat propter alteram obliuio, sed se medico temperamento moderentur, ne unquam modum superet eleuacio uel fractura, scilicet ut contemplacione futurorum nec sit a spe uacua meditacio, nec a metu libera, futurorum dico temporalium, quia caritas perfecta que celestis est foras mittit timorem (De nugis curialium 1.31, ed. James, M. R. [Oxford 1914] 62).Google Scholar

127 On this paradox see AHF 51-55. Though primarily important for the study immediately to follow on the philosophical transmission of literary theory, a chapter of John of Salisbury's Metalogicon on the nature of universals (2.20) bears equally upon the rhetorical distinctions in relation to fictions. Aristotelian universals, expressed in such concepts as ‘genericness’ (genus) and ‘specificness’ (species), are for John simply notiones (ἐννοίασ) or phantasiae (εἰκονόφανασ), imagines, and umbrae of things which actually exist. Take away the existent particular objects and such monstra will vanish like dreams or like the images of the objects reflected in the pure mirror of the mind (tanquam in speculo, native puritatis ipsius anime). John calls these mental forms exemplaria and the real objects of which they are the images (imagines) exempla. Herein lies the early association of exemplar with the imago, giving the image the value of universality, and the exemplum with the rhetorical illustration, giving the example the value of particular demonstration. These associations foreshadow Sidney's attribution of an ideality to the poetic image and an illustrative validity to the historical example. That both the ideality and the exemplification (significatio and illustratio) come, in general practice, to be expressed in the single term exemplum shows the increasing imposition of rhetoric and history upon poetic theory for their didactic value. John points out, however, that universalia, though without existence in themselves, may be considered ‘substantial’ (substantialia), that is, necessary prerequisites and hence ‘causes,’ with respect to our knowledge of particular things (ad causam cognitionis). Things may be prior to concepts (i.e. mental representations of them) existentially but concepts are prior to things cognitively. Starting with given things or deeds, we proceed inductively from individuals to species to genera and, then, ‘confirm’ our knowledge by returning deductively (with genus and species now functioning as prior conditions, or causes, of individuals) to the initial things or deeds, which may now be regarded as ‘interpreted.’ As in the qualitative status of a legal case, John maintains that generic terms denote not ‘what’ but ‘what kind of what’ (quale quid) a thing or act is. Such denotation is perforce general, and John, as an Academic, will not demand the degree of accuracy expected of exact sciences which leads others into contentious debate. In his Creation God distinguished things by the general categories of number, weight, and measure (Sap. 11.21). Numerus differentiates things, pondus gives them generic value (ad generis dignitatem), and mensura provides quantitative delimitation. Quoting St. Augustine (De Gen. ad lit. 4.3-5), who describes weight as omnem rem ad stabilitatem trahens, John associates pondus with the qualitative status, which Augustine himself placed under the protection of the Holy Spirit. (The identification of pondus with love is metaphorically stated in the City of God [11.28] and in the Confessions [13.9] — and developed by Dante [3.1.136-41] — and with voluntas as a unifying power negotiating between sensory images in the mind and the magnitudes memory assigns them [De Trin. 11.10-11].) Universale, on the other hand, are figuralia: genera et species non omnino quid sunt sed quale quid quodammodo concipiuntur: et quasi quedam sunt figmenta rationis seipsam in rerum inquisitione et doctrina subtilius exercentis.' Fictions, now in the guise of universals, are accorded validity, once again, in the investigation of quality. Civil law, in turn, will have, John says, its own kind of fictions (ius ciuile sua figmenta nouit), as indeed will all branches of learning (cf. AHF n. 23 and above n. 93). The conclusion is that genera and species, which are exemplars of particular things (exemplaria singulorum), as instruments of learning, may be regarded as a monstruosa … figmentorum speculatio which may dispense with the consideration of individual things (usque ad uentilationem singularium). It is only through the abstracting power of fictions, John suggests, that one may free oneself from the unique properties of a particular substance. This freedom of the ‘universal’ is similar to that of poetry from the particularities of history and that of rhetoric from the particular circumstances of a legal controversy. Joannis Saresberiensis Episcopi Carnotensis Metalogicon ed. Webb, C. C. I. (Oxonii 1929) 99–113.Google Scholar

128 The importance of the relation between Scripture and secular poetry for the Humanists generally may be seen in Trinkaus, C., In Our Image and Likeness (London 1970) 555–774, esp. in his treatment of theologica poetica (683-721). See also Curtius, E., op. cit., 214-27, and Singleton, C. S., Dante Studies I (Cambridge, Mass. 1954) 84-98. The technical analogies between Scriptural allegory and secular exemplification are satisfactorily brought out by De Bruyne 2.302-70. When Boccaccio movingly laments that the difficulties of composing the Genealogie will be non aliter quam si per vastum litus ingentis naufragii fragmenta colligerem (1.pr.), he says that it is only by revealing the hidden significance of these disparate tales that a principle of order will emerge by which to arrange them (porro, princeps eximie, uti componendo membra deveniam, sic sensus absconditos sub duro cortice enucleando procedam, non tamen ad unguem iuxta intentionem fingentium fecisse promittam). Functioning as a thesis, such an interpretation will reveal not only the voluntas of the ancient authors but offers a principle of formal coherence long obscured beneath the misunderstood litterae. So in the human community, beyond the private world of humanistic letters, the equity of natural law will supply the mortar for the fragmentary laws of society. See n. 123.Google Scholar