ISSA Proceedings 2002 – Topos In Rhetorical Argumentation: From Enthymeme To Figure
James J. Murphy some ten years ago raised the question, “[W]hat is the relation between Topos and . . . Figura?” (Murphy, 1990, 240), a question he understood as one of an historical development yet but dimly known, and whose answer would require long and difficult scholarship. No doubt he was right, but one might equally well ask a related, equally important, and perhaps more manageable question: “What are the intrinsic (structural) relations between topos and figura?”, relations which are presupposed by the historical developments which Murphy rightly insisted need to be investigated.
It is the latter question that this paper proposes to engage – I say engage, not answer, because a full answer relies also on historical developments, though not as profoundly as the answer to Murphy’s question above. This paper investigates in a preliminary way the historic relations between topos and figure, and, using recent developments in the theory of topos, argues that the figures are enthymemes constructed from particular topoi. The paper proceeds in three steps:
1. What are the (historically constituted) intrinsic relations between topos and figure?
2. What does recent scholarship say about the relation between topos and enthymeme?
3. To what extent does the notion of the figure as an enthymeme constructed from a topos explicate the extrinsic relationship between topos and figure?
To initiate an answer to the first question, consider the following example. In his Rhetoric, Aristotle considers one of the general topoi, rational correspondence, more commonly understood as similarity or proportion, A:B::C:D (1399a34 ff.). In the Poetics, the figure metaphor is explicated as a proportion (1457b). Indeed, to use one of Aristotle’s not-so-excellent metaphors from the Rhetoric, “the arrow flies” (1411b35), the proportion
(the motion of the arrow):(the motion of the bird):: X: flight
explicates precisely the metaphor in question. In other words, from the beginnings of written rhetorics, there is the notion that the figure (metaphor) is based on a topos (similarity). Furthermore, and more to the point of this paper, that the figure (metaphor) is an enthymeme using a topos (similarity) is implicit in the explanation:
If the bird flies, and if the motion of the bird is similar to the motion of the [shot] arrow, then the arrow flies.
The only part of the enthymeme that is expressed in the metaphor is of course the conclusion.
Four tropes (of the usual set of about eight) have been identified as master tropes: metaphor, metonymy, synecdoche, and irony (Burke, 1969, 503 ff.). Whatever Burke’s reasons for choosing them – and we will return to these reasons in due time, these four do occupy a prominent place in the history of relations between topos and figure, a place that is so prominent that it might well represent the larger proposition that this paper is arguing. I anticipate – that is, I am already arguing figuratively [using the figure prolepsis], something that I have not yet shown I can do. But then, practice, as rhetors have reminded us, precedes theory (utens before docens). To return to my point – of these four master tropes, it is a commonplace that metaphor occupies the premier position, and it is that figure’s relation to similarity from Aristotle till today that I shall focus on.
Quite as Murphy long ago noted, the Roman rhetoricians present full-blown lists of the figures: the anonymous writer in Rhetorica ad herrenium, Cicero in De oratore, Quintilian in Institutio oratoria. As in Aristotle, the link between, say, metaphor and the topos similarity is made, though more obliquely than in the Rhetoric. Thus, whereas Aristotle clearly names the topos proportion in his treatment of the figure metaphor, the Romans, who use the words “resemblance” and “similarity” (e.g., Cicero, 1942, §§155 and 157) do not note that these are topoi, which in any case are not treated in their rhetorics but in their writings on the topics. Still – the association is there.
By the Middle Ages, if we take the early and late examples of St. Augustine (Robertson, 1958) and Geoffrey de Vinsauf (Gallo, 1971) as representative, we find a vastly diminished rhetoric of the figure and a nearly complete abandonment of the topoi altogether (not just in relation to figure). Regarding metaphor, e.g., de Vinsauf says it involves “transposing” a word from its literal meaning (Gallo, 1971, lines 770 ff.), but gives no theoretical definition; the topos normally associated with metaphor, comparison, is discussed separately under “Methods of amplification” (lines 241-263). Thus, Curtius, in his study of the literature of the Middle Ages, rightly treats the topics in Chapter 5 (1991, 79-105) and metaphor in Chapter 7 (1991, 128-144) as unrelated and separate, without drawing the Aristotelian connection between them. We might note that this is not very surprising, given that the Roman rhetorics were well-known in the Middle Ages but Aristotle’s was not.
A shift occurs in the Renaissance, however. Thomas Wilson’s The arte of rhetorique (1553), one of the first Ciceronian ones in English, says that metaphor is “an alteration of words from the proper and naturall meaninge, to that which is not proper, and yet agreeth therunto, by some lykness that appeareth to be in it” (1962, 194); and his “coloures of rhetorique” (212-214) include the topos similitudo (as well as most of the Aristotelian topoi), while his “places” refer to the places of logic (18 and 37).
It is well-known that Peacham’s stylistic rhetoric, The garden of eloquence (1593), includes most of the classical topics in its lists of the figures. Less well-noted, perhaps, is the fact that Peacham’s book begins with a remarkable table (Table 1):
What caused Peacham to use this device I do not know, though I observe that Ramism was already known in England, and that it may have influenced Peacham as it most certainly did Fenner and Fraunce in their rhetorics of the same time. Ramism, with its binary structuration (the core of its infamous “method”) and its transfer of all matters dialectical from rhetoric to logic. Indeed, Fenner, an avowed Ramist, “diagrams” all four master tropes under the heading of comparison (a topos), using “the method” (1966, n.p.) (Table 2):
The significance of the English Renaissance’s understanding of the figures was recognized only much later, however. Sisiter Miriam Joseph’s Rhetoric in Shakespeare’s time (1947) is one of the first extended studies to argue conclusively what the Ramistic diagrams show implicitly, that the figures are based entirely on the topoi. For Sister Miriam, metaphor is an application of the topic similarity (Miriam Joseph, 1962, 328), synecdoche an application of the topic division (315). Another study contemporary with Sister Miriam’s is Rosamund Tuve’s Elizabethan metaphorical imagery (1947). Tuve’s Part II, “Logical Functions of Imagery,” argues – independently of Sister Miriam – that the topoi ground the figures. For example, the basis of metaphor is the predicament quality (and the predicaments are among the logical topoi), division (a general topos) is the basis of synecdoche. Nearly twenty years later, Rosalie Colie, though taking a somewhat different point of view, essentially argues that at least some of the figures are based on the topoi in her Paradoxia Epidemica (1966). Paradox is, among other things, a rhetorical topos, and as such leads to conceits like the courtly lovers’ predicament (Colie, 1966, 89 ff.), the infinite as a figure for God (145), the problematization of non/being (303), and so on.
The Aristotelian association fades again during the Enlightenment. The rhetors after Locke (1690) were much too busy rescuing their subject from his new philosophy, sometimes by using that philosophy against itself. Yet from Bacon’s (1620) “idols of the marketplace” (1993, 1273) – by which he meant abuses in public discourse or rehtoric – to Campbell’s “tropes conducive to vivacity” (1776, 299) – including metaphor which “represents things intelligible by things sensible” (304) – is a straight line that leads directly through Locke’s view of rhetoric as an instrument of deceit.
How refreshing, then, to find at the very end of the Victorian era’s feeling (following hard on the Enlightenment’s reason) the astonishing rhetorical sophistication of a Nietzsche. Blair’s translation of Nietzsche’s lecture notes shows that he argued forcefully that language is inherently not accidentally figurative: “all words are tropes in themselves, and from the beginning” (1983, 107). A sentence like “The grass is green,” for instance is a metaphor because grass (a plant) is not literally green (a colour). Similarly, any name that substitutes distinguishing function for description is synecdochic (the present-day computer, e.g.), and so on. This argument is much more radical than mine in this paper, for where I claim only that the figures are based on the topoi, Nietzsche claims that all language is topical.
Completing this preliminary and admittedly reduced survey of topos and figura, the Modern Age is replete with support for the thesis in question. Richards’s metaphor as tenor/vehicle (1936, 97-100); Perelman’s rhetorical figures within argumentation (1969, 167-179); Saussure’s sign = signified / signifier (1966, 66-67) which leads directly to Group Mu’s “general rhetoric [of the figure]” (1981); and Burke’s motivation for identifying the four master tropes – metaphor=perspective, metonymy=reduction, synecdoche=representation, irony=dialectic (1969, 503 ff.). [To these might be added Eco’s comments on Aristotle’s notion of metaphor, and Genette’s figure as a gap with the sign (in metaphor, this gap is called “resemblance”).] In short, the historically constituted relation between the topos (viz. similarity) and the figure (viz. metaphor) is that the latter is squarely based on, derived from, and constituted by the former.
To move on to the second stage of my argument, it has very recently been argued that the Aristotelian topoi and enthymemes are related as follows: T, a binary relation between linguistic terms, is a topos exactly when “If P(x) and if T(x,y), then P(y)” is an acceptable enthymeme (Dyck 2002). This rather simple statement is of course a reduction of the fuller argument which deals with Aristotle’s twenty-eight general topoi and therefore with rather more complicated enthymemes also. Evidently, if this argument is correct, then the relations between rhetoric and logic may have to be rethought: for, since implication is a (logical) topos and has a form identical to the above, it follows that the (rhetorical) enthymeme is a generalization (weakening) of the (dialectical) syllogism.
But such esoteric byways are not my interest here and now. Consider the following topoi and the enthymemes associated with them:
1. S = similarity: If P(x) and T(x,y), then P(y).
2. C = contiguity: If P(x) and C(x,y), then P(y).
3. R = representatation: If P(x) and R(x,y), then P(y).
It is immediately clear that the assertions “P(y),” under the given conditions, are the figures metaphor, metonymy, and synecdoche, respectively. I have already dealt with the first case, in which the topos is similarity and the figure is metaphor, and the second case is handled in the same way: contiguity or “is associated with” leads to metonymy. The third case involves the topos pars pro toto, and we might examine the textbook example of synecdoche, where a ship is represented by its sail, or R(sail,ship). To assert “Ten sails are on the water” is to assert the conclusion of the following enthymeme:
If ten ships are on the water, and if R(sail,ship), then ten sails are on the water.
The third part of my argument is, in other words, obvious, and in the case of metaphor may be put in strictly Aristotelian terms using “Dionysus’s shield” as an example (Poetics 1457b20):
If Ares’s shield (object) is a shield (function), and if shield:Ares :: cup:Dionysus, then Dionysus’s cup (object) is a shield (function).
[If Shield(Ares’s shield) and if S(Ares’s shield, Dionysus’s cup), then Shield(Dionysus’s cup).]
It would be inexcusably poor rhetoric if I did not present a telling example of my over-all claim, namely, an example showing how argument by figure works. For this I need a familiar poem, and I know of none more familiar or more excellent than Shakespeare’s 116th Sonnet.
Let me not to the marriage of true minds
Admit impediments. Love is not love
Which alters when it alteration finds
Or bends with the remover to remove.
O no, it is an ever-fixed mark
That looks on tempests and is never shaken;
It is the star to every wandering bark
Whose worth’s unknown although his height be taken.
Love’s not Time’s fool, though rosy lips and cheeks
Within his bending sickle’s compass come:
Love changes not with his brief hours and weeks
But bears it out even to the edge of doom.
If this be error and upon me proved,
I never writ nor no man ever loved.
“116” is a Shakespearean sonnet: (3×4) + 2 = 14. This disposition (arrangement), or subgeneric structure, imposes severe restrictions on the reader as well as on the writer. Three quatrains rhymed abab and a closing heroic couplet suggest a triadic approach to the topic love followed by a summative or concluding statement. And the sonnet bears this out: quatrain 1 = what love is not; quatrain 2 = what love is; quatrain 3 = a mix of what love is/not; the couplet = an “if -, then – ” summary.
But what is one doing when one says, of a word or an idea, that it is not such-and-such, but is such-and-such? One does not have to be a rhetor to recognize this is a definition, but one would probably have to know some rhetoric to know that definition is a topos. Like any educated man in the Renaissance, Shakespeare knew this, and he adapted the topos of definition to the disposition of the sonnet, beginning negatively in the first quatrain, moving to the positive in the second, and mixing the two in the third. Shakespeare was inventing this particular poem by using a topos to generate an argument for a given disposition.
The closing couplet of this sonnet has earned enormous attention because it supposedly is difficult to understand. But for rhetoric the couplet is an instance of a figure called syllogismus in Renaissance rhetorics. Syllogismus is an apparently valid syllogism: if [my definition proves false, then [I never wrote and no man ever loved]. A bit of logical analysis reveals that all this syllogism asserts is that the definition of love given in the poem, call it L, must be true: “if not-L, then N” is logically equivalent to “L or N”; and since N = “I never writ nor no man ever loved” is evidently false, we see that L must be true for the syllogism to be valid. In other words, this syllogismus is a petitio principii, begging the very question it pretends to answer – Is this definition really true? A slightly different analysis of this couplet may be given, for the enthymeme is constructed from the topos implication. But of course the topos implication, namely, “if -, then -,” is exactly the topos syllogismus.
What then does the couplet add to the rest of the poem, if it does no more than assert what is required to be proved? It can at best measure the poet’s conviction that his definition must be right; it may at worst suggest his underlying fear that he might be wrong: it is, in short, a cry of near-desperation – if love is not what I believe it to be, how could I ever have written or any man ever have loved?
The poem therefore presents a larger enthymematic argument based the topoi definition and implication. The full extent of Shakespeare’s genius in constructing this poem has not yet been broached, however: for the details of the argument are presented using another topos, similarity, in a series of brilliant metaphors. These metaphors, utterly characteristic of Shakespeare’s style, are also part of the invention of his argument. Here are some of the ones employed in generating the positive aspects of his definition of love:
[Love] [is] “The marriage of true minds” (line 1)
[Love] “is an ever-fixed mark” (line 5)
[Love] “is the star to every wandering bark” (line 7)
[Love] [is] [a grim, Time-like reaper] (line 10)
(The last metaphor derives from the negative personification “Love’s not Time’s fool” (line 9) and the synecdoche “rosy lips and cheeks” (standing for youth) which come “within his bending sickle’s compass” (line 10).)
The poem’s full matter (content, the Renaissance res ) is inseparable from its manner (form, the Renaissance verba). Its definition and concept of love may be stated as love is that mental or spiritual relationship which is perfect, steadfast, trustworthy, independent of circumstance, and eternal. This love, in other words, is a highly idealized (in the sense of Plato’s ideas) love, not a love as it most probably exists in the world. Few if any, even in the Renaissance as today, likely experience such love.
I conclude with one small observation. I have argued synecdochically that the figure is an enthymeme derived from an appropriate topos – in other words, doing something while I was arguing that it could be done (utens before docens). But such circularity underlies also the very claim I am making. Long before enthymemes and topoi were understood or even articulated as such, far back (in other words) in an imagined place at an imagined time, a prehistoric female dropped her infant into the soft grasses: Ah – dam, she grunted, and figuratively gave birth to the enthymeme and to my argument.
Aristotle. (1984a). Rhetoric. In Julian Barnes (Ed.), The complete works of Aristotle, Revised Oxford Translation, Vol. 2,. Princeton: Princeton University Press.
Aristotle. (1984b). Poetics. In Julian Barnes (Ed.), The complete works of Aristotle, Revised Oxford Translation, Vol. 2. Princeton: Princeton University Press.
Bacon, F. From Novum Organum. In M. H. Abrams (Ed.), The Norton anthology of English literature, Sixth Edition, Vol. 1 (pp. 1271-1276). New York, London: WW Norton & Company.
Blair, C. (1983). Nietzsche’s lecture notes on rhetoric: a translation. Philosophy and rhetoric 16, 94-127.
Burke, K. (1969). A grammar of motives, Second Edition. Berkeley: University of California Press.
Campbell, G. (1963). The philosophy of rhetoric (1776). L. F. Bitzer (Ed.). Carbondale : Southern Illinois University Press.
Cicero. (1942). De oratore, Vol. III. Loeb. Cambridge, MA: Harvard University Press.
Colie, R. (1966). Paradoxia epidemica. Princeton, N.J., Princeton University Press.
Curtius, E. R. (1991). European literature and the latin middle ages. Reprint. Princeton: Princeton University Press.
Dyck, E. F. (2002). Topos and enthymeme. Rhetorica 20.2, 105-117.
Fenner, D. (1966). The artes of logike and rhetorike (1584). In R. D. Pepper (Ed.), Four Tudor books on education. Gainesville, Florida: Scholars Facsimiles & Reprints.
Fraunce, A. (1950). The arcadian rhetorike. E. Seaton (Ed.). Oxford: Basil Blackwell.
Gallo, E. (1971). The “poetria nova” and its sources in early rhetorical doctrine. Paris: Mouton.
Group Mu. (1981). A general rhetoric. Transl. P. B. burrell and E. M. Slotkin. Baltimore: Johns Hopkins Press.
Locke; J. (1979). An essay concerning human understanding (1690). P. H. Nidditch (Ed.). Oxford : Clarendon Press.
MacIlmaine, Roland. (1969). The Logike of the Moste Excellent Philosopher P. Ramus Martyr (1574). C. M. Dunne (Ed.). Northridge, CA: San Fernando Valley State College Renaissance Editions.
Miriam Joseph, Sister. (1962). Rhetoric in Shakespeare’s time (1947). New York: Harcourt, Brace & World.
Murphy, J. J. (1990). Topos and figura: historical cause and effect? In G. L. Bursill-Hall, Sten Ebbesen & Conrad Koerner (Eds.), De ortu grammaticae (pp. 239-253). Amsterdam/Philadelphia: John Benjamins.
Peacham, H. (1954). The garden of eloquence (1593). Gainesville, Florida: Scholars Facsimiles & Reprints.
Perelman, C. (1969). The new rhetoric. Notre Dame, Ind.: University of Notre Dame Press.
Richards, I. A. (1936). The philosophy of rhetoric. New York, London: Oxford University Press.
Robertson, D. W. Jr. (1958). Augustine: on christian doctrine. Upper Saddle River, N. J.: Prentice Hall.
Saussure, F. de. (1959). Course in general linguistics. C. Bally, A. Sechehaye, and A. Reidlinger (Eds.). Transl. W. Baskin. New York: Philosophical Library.
Tuve, R. (1961). Elizabethan metaphysical imagery: renaissance poetic amd twentieth-century critics (1947). Chicago: University of Chicago Press.
Wilson, Thomas. (1962). The arte of rhetorique (1553). R. H. Bowers (Ed.). Gainesville, Florida: Scholars Facsimiles & Reprints.