Double-Take. Reading De Man and Derrida Writing on Tropes

1. De Man's late essay "Anthropomorphism and Trope in the Lyric," delivered in a series of lectures at Cornell in the spring of 1983,¹ begins with an argument which proceeds as a reading of the first third of a sentence in "On Truth and Lie in an Extramoral Sense," the words affirming that truth is a mobile army of tropes.² It's a sentence famous or notorious enough so that, as one might say, "it hardly needs translation." Perhaps indeed it needs re-translation into a foreign tongue, or so one might be tempted to say by the disorienting usage of the word "translated" in the opening paragraph of De Man's essay. De Man quotes from Nietzsche in German and then alludes to his quotation as "translated" even though he hasn't translated it from German to English but left it in the original: 

   Was ist also Wahrheit? Ein bewegliches Heer von Metaphern, Metonymien, Anthropomorphismen…” Even when thus translated before it has been allowed to run one third of its course, Nietzsche’s sentence considerably complicates the assimilation of truth to trope which it proclaims. (Rhetoric 239)

   I want to call attention to a way in which  "Anthropomorphism and Trope in the Lyric" and two essays of Derrida frame what Nietzsche lists in his breakdown of "truth" into rhetorical terms, a "mobile army of metaphors, metonymies, anthropomorphisms" (Nietzsche 3:314). I'll be moving back and forth between the De Man and Derrida essays, drawing analogies, paying attention especially to a certain passage in "Anthropomorphism and Trope in the Lyric" and a comparable passage in moment in Derrida's "White Mythology: Metaphor in the Text of Philosophy." In each essay there is an abrupt slowing down of the reading--or you could call it a drawn-out double-take.

2. There is an odd moment in the preceding essay in Margins of Philosophy, "The Supplement of Copula: Philosophy Before Linguistics"—a stressed elision, which occurs in reference to a passage in "On Truth and Lie in an Extramoral Sense." Nietzsche's essay serves up an argument analogous to certain familiar arguments of and about philosophical language, but "with," Derrida writes, "an entirely other aim" (178). "White Mythology: Metaphor in the Text of Philosophy" can be read as elaborating on that elision. And the outcomes of Derrida's reading of Aristotle on metaphor can find a language, I would argue, in Nietzsche's essay's odd list of tropes as it is read by De Man.³

3. In the text of Aristotle, as read in "White Mythology," thinking about metaphor takes place as a thinking through metaphor. Statements of truth come in the form of metaphor and in reflections about metaphor. Reading Aristotle, Derrida will almost not find anything but what can be thought of by means of analogies, the metaphorical linkage, by analogy, of one idea with another. But the reading brings us to a knot amid those woven together threads, and something in the texture of Derrida's essay changes.

4. What had "White Mythology" and "Anthropomorphism and Trope in the Lyric" been saying?

5. Each essay has asserted or demonstrated that the tie between truth and metaphor is so close as to be a coinciding. "Anthropomorphism and Trope in the Lyric" sets out with a quotation from "On Truth and Lie in an Extramoral Sense”; De Man has restated it, blandly, "At this point, to say that truth is a trope is to say that truth is the possibility of stating a proposition." Truth is a trope, the trope truth is the possibility of stating a proposition" (Rhetoric 241). A metaphor says A says B. "White Mythology" has been reading Aristotle and showing that statements of truth come in the form of metaphor and in reflections about metaphor, and that in these inaugural texts, what one finds are metaphors which are philosophemes: which can be stated, and also offered in example, like truth, light. In "The Supplement of Copula: Philosophy before Linguistics," where Derrida is concerned particularly with Benveniste, Saussure, and again Nietzsche, "truth" is a "trope" means something a little different: the reversibility into each other of the two ostensibly opposing sides or phases of an argument or what Derrida calls a "philosophical scheme" (179). Here is an instance: someone challenges the authority of philosophical discourse by pointing to the fact that it's determined by linguistic constraints, constraints such as abstract ideas consisting in metaphors or dead metaphors. But just that point is simultaneously a quintessential philosophical gesture—the determination to get to the true truth that metaphors ultimately cannot conceal; the meaning, always there, and essentially unaffected by its transportation from one sign, or analogy, to another. One or another version of that thesis, and its denial, or its reversal, form a "schema" or "philosophical scheme," and it's as if they formed a Mobius strip. "White Mythology" and "The Supplement of Copula" trace the contours of several such reversible philosophical arguments. One of the subjects explicit in the latter essay is "the truth" as "the possibility of stating a proposition" (Rhetoric 241), and how it may be related to the putting into question of the inevitability of "is" or "to be." "Truth is a trope" in these two different kinds of instance--the reversible schemas and the unavoidable metaphors—and these tropes are—although in two different ranges of reference—what "truth" is.

6. When Nietzsche resorts to the philosophical schema whereby philosophy or truth is language or metaphor, or rather to that phase of the schema, he does so "with an entirely other aim," Derrida affirms, than those that accompany this schema in other texts (for instance the aim to put philosophy in its place, "after" "linguistics"). In first undercutting the assumption that Nietzsche's sentence about tropes has a polemical thrust, De Man reads similarly.⁴ In the case of Nietzsche, for De Man as for Derrida, something else is going on. "The definition of truth as a collection [...] of tropes" is "a purely structural definition, devoid of any normative emphasis": a definition which "implies that truth is relational, that it is an articulation of a subject (for example 'truth') and a predicate (for example 'an army of tropes') allowing for an answer to a definitional question (such as 'what is truth?') that is not purely tautological." Truth is a trope is a proposition. Nietzsche's sentence is "a statement with no critical thrust" when it is read as the assertion given in its first ten words (Rhetoric 241).

7. But Nietzsche's list has, by dint of its next item, "considerable critical power." It is power coming not first, as one might have anticipated, from the reference to "a mobile army." Rather—double-take—Nietzsche's sentence, Nietzsche's list, has critical power via the word anthropomorphisms.  The sentence-fragment takes on critical power due to an anomaly in the list of rhetorical terms. De Man states this point with a gravity that slows us down and an abruptness which emphasizes the incongruity.

   But "anthropomorphism" is not just a trope but an identification on the level of substance. […] The apparent enumeration is in fact a foreclosure which acquires, by the same token, considerable rhetorical power. (241-2)

   The presence of "anthropomorphism" in the list of tropes is an anomaly on which De Man's reading of the sentence turns, and the reading which had started starts again.

8. The double-take in "White Mythology" happens like this: We are moving along in the metaphors in many texts of Aristotle threaded together by Derrida, metaphors interwoven with Aristotle's pursuing of metaphors to say the truth about metaphor. Derrida has talked about the "tropic movement" of these figures, the mobility back and forth among such tropes as the eye, light, sources of light, gold, value, sources of value, generation, regeneration, and truth. He has observed that the type of metaphor Aristotle's text brings out is analogy, metaphor as the comparison of two and hence four different elements:  two similar meanings each of which has (or is) its own metaphor. The two meanings being compared are like each other, and they also are alike in their internal construction. Each meaning consists in its word and its word's meaning. The meaning's word essentially belongs to its meaning. They may even be impossible to detach from one another, which is the case with those metaphors in Aristotle's text. But, and this is important, even where the concept and its signification merge, they do so in the context of a discussion or conceptualization of metaphor, that is, of two things and their comparison. Thus they are conceived (and conceived only) as essentially comparable yet distinguishable from one another. In other words, these are analogical structures, metaphors structured and stabilized as analogies. An analogy is a strong and suggestive unit. It is always implying, or could imply, another one, and it constitutes a possible proposition. An analogy forms connections with other possible propositions. That (we were seeing), in Derrida's Aristotle, is what truth is.

9. True metaphors generate other true ways of speaking. This would naturally be the case, since the single external and singular referent, and at the same time the generator of all these figures and analogies, is the sun—which produces rays, light, heat, life, true life, and so forth. The sun-figure's generativity is crucial to the integrity or wholeness of meaning. It's crucial to the system of truth and trope being one, to the spreading out from the basic philosophemes, or metaphors, of all the analogies, and all the truth statements, that it will be possible to find or to make.

10. "White Mythology" analyses a "tropic movement" in and among several senses of value. Derrida argues that it is not possible, by using as a tool the concept "metaphor” or "analogy” or "value," to ground a science, or a conceptual structure able to house a knowledge of its own language—be it a metaphorology (a systematic knowledge of metaphors) or a science of linguistics (of language as a system of values). Theorisations on this order have their boundaries erased in advance. For the metaphor network, the concept of value, or the analogy theoretically deployed by any of these, drops into what Derrida calls "a wider discourse of figuration" (243).⁵

11. Then far into the essay comes an odd moment. We seem to hear Derrida saying, "when has anyone ever seen the sun sowing?" ... Double-take... What Derrida has written is this: "Where has it ever been seen that there is the same relation between the sun and its rays as between sowing and seeds?" (243). Derrida's interrogative sentence doesn't just pose a rhetorical question. It mimes an incredulous query, a query which voices at the same time the presumption that somewhere, it might have been possible to see that the sun sows.

12. There have been all those analogies, an extending and continuous fabric of light—analogies, figures all like or comparable to one or another quality of the sun; all those metaphor-philosophemes we couldn't do without (like "light" and like "like.")  But here there's some trouble. Derrida has just quoted these sentences from Aristotle:

    It may be that some of the terms thus related have no special name of their own, but for all that they will be metaphorically described in just the same way. Thus, to cast forth seed-corn is called "sowing" [speirein]; but to cast forth its flame, as said of the sun, has no special name [to de ten phloga apo tou heliou anonymon]. How is this anonymity to be supplemented? This nameless act, however, stands in just the same relation [homoios ekhei] to its object, sunlight, as sowing to the seed-corn. Hence the expression in the poet, "sowing around a god-centered flame [speiron theokistan phloga]." (Poetics 1457b25-30) (Derrida, Margins, 242-3)

    That "'expression in the poet'" is a catachresis standing in for another figure.

13. It's more than an oddity: in there amid all those respectable figures (analogies) comes up their generator—a condition of possibility.

14. As Derrida has noted, Aristotle needs the figure of the ray-sowing sun, needs the true sun to be able to generate all those figures. The question arises whether the sun—the single unchallengeable and singular referent in this system of figures—ought to be able to be seen to do what it is said to do. If we cannot "see" the sun sow, it is because this generative "figure" and singular referent comes up like a proper name among all the figures (indeed an army of metaphors and metonymies) Derrida has been identifying in Aristotle's writings on metaphor. One can validly juxtapose here De Man's conclusion about the occurrence of the word anthropomorphism in Nietzsche's list: "The apparent enumeration is in fact a foreclosure" (241).

15. As telling as the term catachresis would be the word setzen. Setzen, "to posit," is introduced in De Man's second Nietzsche essay ("The Rhetoric of Persuasion" Allegories of Reading 120-4). The term setzen reappears in "Shelley Disfigured," an essay which, reading Shelley's broken off poem "The Triumph of Life,” concerns what is not quite lyric, like "Anthropomorphism and Trope" (which turns in the second part to a reading of Baudelaire's “Correspondances"). In borrowing from "the poet"—in settling on this assertion "the sun sows its rays"—Aristotle's text is positing something: "the metaphor of metaphor [...] an ellipsis of ellipsis" (243). It's positing what for the text's conditions of possibility to exist would have to be able to be; metaphorically, or without statement: would have to be able to be put in a comprehensible metaphor. That's why this one--a sowersun--isn't. It marks instead the positing of meaning. An anthropomorphism, writes De Man,

    Takes one entity for another and thus implies the constitution of specific entities prior to their confusion, the taking of something for something else that can then be assumed to be given. Anthropomorphism freezes the infinite chain of tropological transformations and propositions into one single assertion or essence which, as such, excludes all others. It is no longer a proposition but a proper name, as when the metamorphosis in Ovid’s stories culminates and halts in the singleness of a proper name, Narcissus or Daphne or whatever. (Rhetoric 241)

    Not a proposition and not a metaphor, “the sun sows its rays” is an anthropomorphism, that extra term in Nietzsche's hence impossible list. Instead of a simply comprehensible metaphor, we have what is in effect a proper name—Sowersun, for instance; one that entails not just light (all our analogies and metaphors), but the "freezing" of propositions.  The metaphor of metaphor slides (writes Derrida) into having to be called an "ellipsis of ellipsis"; subjective and objective genitive (the two senses of "of") slide into each other. Metaphor disappears into "its bottomless overdeterminability" (243).

16. As De Man's essay calls our attention to the fact that "anthropomorphism is not just a trope but an identification at the level of substance" (241), it acts out an intensified wariness. Something similar has happened when "The Supplement of Copula: Philosophy Before Linguistics" alludes to Nietzsche on metaphor. Derrida quotes a passage which says: it could only be by "forgetfulness" that anyone could believe that it would be possible to say what is true by means of our words, for they represent the imposition by which "sensory excitations" are accorded the status of "objective judgments" (178). Derrida quotes from Nietzsche in order to show how inescapable the self-inverting scheme is whereby no matter how violent the reminder to the philosopher of the limits placed on him by his language, philosophy always reappropriates this critique (which consists in a version of the quintessential philosophical thesis) (177). But on every occasion on which he points to the fact that this "law of reappropriation by philosophy" comes into play in "On Truth and Lie ...," Derrida's text suggests that that is not all that is going on, and that what is going forward in the passages of that Nietzsche’s essay is something yet to be explicitly understood. One such reference to Nietzsche occurs in Derrida's opening analysis in a reading that principally unravels the arguments of Benveniste's Problems in General Linguistics regarding philosophy's origination in empirically definable "facts of language."  "The Supplement of Copula: Philosophy Before Linguistics" begins by talking about the line of thought that makes use of the notion of a "discourse," and in which the concept of a certain thing termed a "discourse" replaces other kinds of attention to texts that exist across an uneven, fractured history. "Philosophical discourse" is held to have become possible thanks to a specific linguistic situation and to be determined and constrained by features of the "language" in which it was framed. This thesis is a version, Derrida observes, of the recurrent philosophical move consisting in the pushing aside of the mere language of a text to get at its meaning or significance; even Nietzsche is drawn into the same collapsible construction, for instance in the paragraph in "On Truth and Lie " challenging the validity of existent words. It is in this context that Derrida writes that Nietzsche "must resort to an analogous argument [but] with an entirely other aim" (178).

17. It's again a text of Nietzsche that gives De Man the word "posit," which as De Man begins quoting and reading the text in question, loses its innocuous, inconspicuous character. I quote from the 1887 passage in The Will to Power (which is quoted in "The Rhetoric of Persuasion"):

    If, according to Aristotle, the law of contradiction is the most certain of all principles, if it is the ultimate ground upon which every demonstrative proof rests, if the principle of every axiom lies in it; then one should consider all the more rigorously what presuppositions [Voraussetzungen] already lie at the bottom of it. Either it asserts something about actual entities, as if one already knew this from some other source; namely that opposite attributes cannot be ascribed to them [können]. Or the proposition means: opposite attributes should not be ascribed to them [sollen]. In that case, logic would be an imperative, not to know the true [erkennen] but to posit [setzen] and arrange a world that should be true for us. (Allegories 120)

    De Man comments, “What has and will be shown, within the confines of this particular fragment [the passage from which I quote above], is the possibility of unwarranted substitutions leading to ontological claims based on misinterpreted systems of relationship (such as, for instance, substituting identity for signification)" (123). In De Man's use of the term, "positing" has a peculiar impact on meaning. Again, from "Shelley Disfigured": "language posits, and language means (since it articulates), but language cannot posit meaning" (Rhetoric 117). If meaning is posited, can it be "meaning" in the sense that it is possible to know—have words or alignments that say—what is true? To that question "On Truth and Lie in an Extramoral Sense" is saying no, in the paragraph Derrida refers to; but the necessity of positing is being introduced, is being inscribed. Inscribing positing is, so I'd put it, the "other aim" of Nietzsche alluded to in "The Supplement of Copula." A supplement of to be: Derrida signals that we should register the fact that what's being pressured, in this passage about words, is the value of the "is."⁶

18. "On Truth and Lie in an Extramoral Sense" indeed “complicates the assimilation of truth to trope that it proclaims" (Rhetoric 239). De Man next asserts the congruence of Nietzsche and Kant: Nietzsche's essay and the Third Critique are alike, a denial or undoing of the certainty of meanings and of sensory objects, and the intent to recover a controlled discourse. Despite their considerable difference in tone, De Man writes, that tonal difference "cannot conceal the congruity of the two projects, their common stake in the recovery of controlled discourse on the far side of even the sharpest denials of sense-certainties" (240). Amid these lucid intents, however, in both texts there come into play patterns that cannot be assimilated to the main one. "What interests us primarily in the poetic and philosophical versions of this transaction," De Man writes, "is not, at this point, the critical schemes that deny certainty considered in themselves, but their disruption by patterns that cannot be assimilated to these schemes" (240). Not the exploration of truth being trope will be De Man's topic, but rather a "disruption," one in the same area where a disruption is registered by "White Mythology."

    The disruption arrives as "anthropomorphism" is juxtaposed with "trope." Far from being the same, tropes such as metaphor (or metonymy) and anthropomorphisms are mutually exclusive. The apparent enumeration is in fact a foreclosure which acquires, by the same token, considerable critical power. Truth is now defined by two incompatible assertions: either Truth is a set of propositions or truth is a proper name. Yet, on the other hand, it is clear that the tendency to move from tropes to systems of interpretation such as anthropomorphisms is built into the very notion of trope. One reads Nietzche’s sentence without any sense of disruption, for although a trope is in no way the same as an anthropomorphism, it is nevertheless the case that an anthropomorphism is structured like a trope: it is easy enough to cross the barrier from trope to name, but impossible, once this barrier has been crossed, to return from it to the starting point in truth. Truth is a trop; a trope generates a norm of value: this value (or ideology) is not longer true. It is true that tropes are the producers of ideologies that are no longer true. Hence the army metaphor. (Rhetoric 241-2)

    Meaning is foreclosed for the statement "Truth is tropes," the in itself undisturbing identification which "White Mythology" locates in "metaphor in the text of philosophy," as it is extended in the sentence of Nietzsche. The simultaneous assertion that truth is trope and truth is anthropomorphism implicitly disqualifies both assertions, since the first amounts to the statement that truth is a series of propositions (or metaphors), the second that it is an entirely different sort of list, that of names posited as being the proper names of entities. "Anthropomorphisms" means the reference to entities the identity of which is fixed, whether through a metamorphosis over the course of a history, or through a definitional metaphor. Giving the name "anthropomorphisms" to what "truth is," Nietzsche's sentence refers to the taking as given of beings and things thereby being posited as existing in a certain way—posited in the mode of entities on the way to being named not by means of nouns and verbs, but by means of proper names. The figure of figuration and of generation is the sun, introduced in the non-figure "the sun sows its rays." The "figure" of a sowing sun is the opaque proper name of a sower of figures such as light, value, exchange, gold; is the name, say, "Sowing-sun" or "Sunsower." Derrida's listing of metaphors in the text of philosophy pulls up at a point at which it is true that in Aristotle's text "Truth is now defined by two incompatible assertions: either truth is a set of propositions or truth is a proper name" (241).

19. The passage just quoted from "Anthropomorphism and Trope" should be set alongside "Rhetoric of Persuasion (Nietzsche)," the last of three chapters on Nietzsche in Allegories of Reading—in particular the section of the chapter which considers how the passage from a cognitive to a performative rhetoric is "irreversible," but is also interrupted, since there is no passage forward (or backward) to the possibility of knowing that language is in a particular instance doing something, that it is able to act or "perform." Language does not operate in the mode of knowing or constatation of truths, but in the mode of persuasion, of rhetoric as a type of power rather than a mode of knowing (the knowledge conveyed by metaphors, figures, or propositions). In the course of the breakdown Nietzsche's text carries out, De Man is saying, we have arrived at a performative model of language, but by no means does this imply that we could revert, or proceed, to knowing what has occurred, or to knowing whether a performative came into play (Rhetoric 130).⁷

20. The indispensable metaphor-philosophemes Derrida has tracked in Aristotle appear (in both Aristotle's text and Derrida's) in the context of a conceptualization of metaphor or figure. That context affords them the structure of trope, of metaphor, of analogy, and, implicitly, of propositions. But suppose the conceptual distinction within the category "metaphor" were annulled, and those figures stayed around? Are such figures tropes? Or would "anthropomorphisms" be the appropriate term for all? The identification of metaphor with truth "in the text of philosophy," inescapable, is what "Anthropomorphism and Trope in the Lyric" terms "an identification on the level of substance [...] the taking of something for something else that can then be assumed to be given." It situates the "discourse of figuration," in short, in a mythology. What seem our metaphor-philosophemes might as well be a list of "figures": a list of names. De Man's emphasis on Nietzsche's list of tropes—and its sudden incongruity at the word "anthropomorphism"—allows us to describe what happens with the sequence of exemplary, then incongruous, metaphors Derrida follows in Aristotle. Thus "White Mythology" jolts at the catachresis "sowing sun" and registers it as the proper name of the generative figure of generative figures in Aristotle. We are brought to a special kind of non-figure: an ostensible proposition which consists in the "proper name" which is only an indicator, which indicates, but does not communicate or express. And here, the referent of that indicating is only itself: indicating the necessity and the impossibility of it being a true name generating true figures. The list of tropes in Nietzsche's sentence and the list of figures in Aristotle would be, so the context of Derrida's and De Man's essays suggests, a list of proper names gone opaque, without a referent other than themselves.

21. The “figure” by which the action of a human being is ascribed to the single and only irreplaceable visible thing is an anthropomorphism of a peculiar kind, the source of visibility. The effect is to make the listed figures into the reiteration of one, like names of a God. Repeated enough times, the figure of a generative sun, in the words in which Derrida finds it in white mythology, becomes I am saying a stammered name (such as SowerSun). An idea we think we know about—the indistinguishability of an object and its inscription in a system of interpretation—would freeze or have frozen, in fact, thought.

22. That available illusion is being repulsed in some later sentences in De Man's argument. Some words of Baudelaire's "Correspondances" serve in these lines, which have an unheimlich atmosphere not dispelled by one's fitting them into a reading, nor referring them to Baudelaire's text. De Man writes:

    And if man (l’homme) is at home among "regards familiers" within that Nature, then his language of tropes and analogies is of little use to them [sic.]. In this realm, transfer tickets are of no avail. Within the confines of a system of transportation—or of language as a system of communication—one can transfer from one vehicle to another, but one cannot transfer from being like a vehicle to being like a temple, or a ground. (Rhetoric 251-2)

23. Proper names, like the verbal units which are proper names for things, recede. Explaining what is at stake in Derrida's readings in Glas, in "Hypogram and Inscription" De Man discusses Saussure's parenthetical and left-off work published by Starobinski under the title Les mots sous les mots. Saussure had been studying Latin inscriptions seeking to determine whether certain patterns of letters were or were not anagrams. De Man describes Saussure as finally letting this line of thought trail away once it became sufficiently clear that there was going to be no way to determine the question one way or the other. The letters, across and among which Saussure could trace the names of certain Latin authors, were indubitably there. "Randomly" or "by intention" of the inscribers? That alternative does not suffice to describe what Saussure was looking at, De Man writes. In the not-quite-for-sure anagrams, Saussure was seeing the indeterminably significative status of what had been supposed to be units of meaning or of legibility; the dismemberment of words (Resistance 37). The mobility of metaphor brings up, not at a central, nor at a proper, name, but at an opaque figure such as that of the "cup without wine" which is the final trope in Derrida's list—a figure placed in the decontextualizing or overcontextualizing locale of a parenthesis (Derrida 243). A possibility being implied and analysed by De Man and Derrida is that words might turn to names and names to unreadable inscriptions. Such would be the implication of drawing the two readings close: of that spooky slide from Aristotle to Nietzsche.

24. Yet the signature "Nietzsche," far from freezing up the sentences in Derrida's and De Man's essays, allowed for readings that undo the yoking of tropes to what would freeze them into "an identification on the level of substance," that of a god's substance or a Nature's substance or man's. My own reading was prompted by my sense of a rhetorical effect. De Man and Derrida exact from us a double-take, a surprise, or deepening uneasiness, at what appears as the added term in a list (of examples of true metaphors; of names of tropes). The readings in "Anthropomorphism and Trope in the Lyric" and "White Mythology" aim at a kind of disconnecting. Derrida drops philosophemes into "a wider discourse of figuration." "The Supplement of Copula" pulls to pieces Benveniste's conception of a determined and determining language of philosophy (namely Greek); aims to pry apart the fused grammar and lexicon of the words akin to "be." De Man's readings of Nietzsche scatter a dismembered sentence; their aim, to underdetermine the significance of tropes.