On parole, Blanchot, Saussure, Paulhan

Nous l'avons détruit, nous avons libéré l'étoile-sans rayon désormais ; il roule obscur, l'astre du désastre, disparu, comme il le souhaitait, dans la tombe sans nom de son renom. — Maurice Blanchot, L'écriture du désastre

Now that Maurice Blanchot has also died, we can do nothing with him but take him at his word. We have no choice but to take the words he has given us to think without him. And among all the words Maurice Blanchot has given us to think, is there one from now on more pressing and in need of unfolding than that of disaster? He gave us his word, disaster, and in so doing stepped aside, leaving us to take up the space liberated by a fallen star. Disappeared, into the ruined space of disaster, catastrophe, calamity, such a word as his also marks the contours of a certain tomb, though one whose proper name, as either origin or end, is always obscured by its being repeatedly on the move! By what light would it now be possible to track the word that has been given in this way only through its incessant disappearance and errancy, through its tall into a sepulchral space of separation and erasure of its own rays? And just what kind of authority would it take for us to grant such a disastrous parole in the first place?

Nothing, it seems, could be farther removed from the dark constellation of reflections enshrouding Blanchot's writing of the disaster than the sunny atmosphere of clarity in which Saussure's Course in General Linguistics makes its way from one pedagogical point to another in laying bare the hidden principles of language and the way it functions. And yet, as is well known, it is precisely Saussure's most original founding gesture that threatens to turn the entire semiological enterprise into a sterile science of dead-end fragments and simplifications. For, it should always be recalled, Saussure's Course, like any text for that matter, is able to begin only by giving us his word, his own unmistakable parole, though in such a way that it seems to be immediately and irremediably retracted from us. That luminous and now tamous word Saussure first gives us, of course, the one that will always have allowed his text to get underway as a systematic Course, is la langue. But have we ever adequately stopped to consider how la langue itself appears only to withdraw in a mode of unthinkable obscurity? Before Saussure's own text executes this tectonic distinction within the French language between la langue and la parole, la langue (as a system of purely formal conventions), which can subsequently be defined and then known only in its strict opposition to la parole (as a particular execution of these conventions), does not and cannot exist as such. It can only be la parole of Saussure's own text, the active and idiosyncratic fact of the Course as a concrete instance of linguistic materiality, that releases la langue as a genuine force of semiological analysis and knowledge.

La langue is therefore not simply a formal system whose empirical existence Saussure was able merely to observe or deduce; it is also, as the neologism that it will become only in the wake of Saussure's Course testifies, a material event that his text has to inaugurate in its own right. Louis Hjelmslev's attempt to resolve what he called the fundamental "ambiguity" lying at the heart of Saussure's distinction between langue and parole gives some indication of the radicality of this event. In Hjelmslev's stratified vocabulary, what Saussure actually does in the Course is to execute an individual "act" — of parole -that retrospectively necessitates an alteration of all "usage" - in this particular case, of la langue - which then and only then can be regarded as the "norm" for every possible "schema" - of intelligibility. (2) For Hielmslev, of course, usage is everything, since the norm is a merely fictive law projected onto it, and every individual act must conform to the existing limits fixed by it. But in the case of Saussure's distinction, the use of the word "langue" to refer precisely to the systematic ele- ments of language by their pure opposition to acts of parole constitutes a "change of definition" that, Helmslev had noted, "would involve a change of language" and thus the appearance of" another language from the one we know" (82). Whenever the available range of usage is ex- ceeded in this way, Helmslev adds, "the description of usage itself would itself have to be modified" 88). Saussure's improvisation therefore leaves a material trace on the entire French langue, and beyond it, on the description and knowledge of language as such. We can from now on understand what la langue is, in contradistinction to la parole, only because Saussure's Course implemented this particular parole of la langue in the first place. And it did this when it made la langue into the only possible object of linguistics by virtue of its difference from la parole.

In the more familiar terms of the Course as it is ordinarily ex-pounded, la langue is a semiotic system that, like all social institutions, is based on certain determinate patterns. These abstract and formalized conventions, unlike the concrete and heterogeneous instances of la parole, can be delimited and then studied in themselves to produce a mode of scientific knowledge. This is why Saussure places such insistence on a further characteristic of the distinction: la langue will constitute only that which is absolutely essential in language's capacity to signify, whereas la parole will comprise all those factors and functions of signification that remain in whole or part entirely contingent. In other words, in order to learn anything epistemologically reliable about language - and since language is the "most important" of the innumerable semiotic systems that make up social life in general, in order to learn anything worth knowing at all — it is necessary once and forever to separate language from itself, to sever the generality of la langue from its intimate ties to the immanent particularity of la parole.

"These can only be," Saussure assures us in the firm tone of a practicing surgeon, "one solution to all these problems: it is necessary right from the start to place ourselves squarely in the domain of la langue.... By separating la langue from la parole, we also separate what is social from what is individual, and what is essential from what is merely contingent." (3)

In an ironic gesture that has not gone wholly unnoticed, even if it has not often been remarked as such, Saussure thus asks us to take his word for it (to believe him sur parole as it were) that we can only learn to know what language is in its most systematic and knowable state once we have excised all instances of la parole from our exami- nation of la langue! And, of course, this originary acte de parole - the one Saussure gives us in the very name of la langue, moreover, though only in order to cut all others away from it — is at the source of the rich and inconclusive history of Saussure's own reception.

For, of all the Saussurean legacies to twentieth-century thought, none has been more subjected to investigation, judgment, and sentencing than this founding bifurcation between la langue and la parole. Between language as social system and language as individual practice, as we have just seen, Saussure seemed to find an absolute incompatibility, and in the face of this paralyzing alternative for any rigorous attempt to produce knowledge about language, he resolutely chose the systematic coherence of la langue as the only proper object of study for linguistics. There can be no doubt, moreover, that the dominant reception of Saussure still considers such a priority given to la langue over la parole to characterize in its essence what goes by the name of "structuralism," a wholly dépassé, if interesting moment in the much larger scheme of linguistic, literary, and cultural history. This emphasis on la langue-language as a system, but language construed as such as well — is still widely regarded to be Saussure's fundamental fault, defect, and in some sense, an infraction against the larger and more encompassing laws of both language and history. For thus having introduced into the study of language the possibility of a split between system and act, state and evolution, signification and intention, Saussure's text has been placed within a theoretical space of detention and rehabilitation. By pretending to consider the exclusion of la parole from consideration of la langue as a simple délit or offense that might and should have been avoided, and that can now be remedied and legitimated by a new acte de parole instantiated by their own text, nearly all readings of Saussure, implicitly or explicitly, lay claim to a highly suspect and potentially unreliable authority of la parole to know and to police the frontiers of just such knowledge. If Saussure insists upon the necessary separation of the systematic and formal nature of la langue from the unruly but coercive power of la parole, he does so precisely because he remains attentive to the original asymmetry between the powerless knowledge offered by la langue and the unknowable power that is always unleashed in la parole (4). What has most often been lost in the reception of Saussure is thus an awareness of the precise ways in which his own text serves to resist underwriting just such claims to legitimacy and authority. Indeed, in the very section of Saussure's Course that assigns to la langue the preeminent place it will have to assume from now on in any reliable science of language, the material facticity of la parole is also disclosed as the only truly historical means by which la langue will ever have had to accede to this position: "la langue is necessary for la parole to be intelligible and produce its effects; but la parole is necessary for la langue to occur at all [pour que la langue s'établisse]; historically, the fact of la parole always comes first" (18-19). This "circle," as Jacques Derrida has referred to it, would not be hermeneutic in any classic sense, and therefore it would not easily lend itself to an operation of reattachment of beginning to end. (5)

The fact is, la parole, by having to speak first, and first of all by having to speak for and thus herald the intelligibility of la langue, serves to disjoin la langue from itself from the start, no matter how much it would claim otherwise. la parole, just as we have seen in the writing of the Course itself, must come first in order for la langue to happen historically. But in order for this parole to function effectively, that is to say, to achieve intelligibility as what it always will have been only subsequently, it must at the same time already be disguised as precisely what it is not yet, which is to say the systematic coherence of la langue. The parole that announces and thus lays down the law of la langue's intelligibility also and by the same token transgresses it simply by virtue of preceding it. This gaping enigma, which one could perhaps now ven- ture to call the paroling of la langue, is originary with respect to all the subjective freedoms and responsibilities that can eventually be granted to la langue by its parole. In Saussure, at least, such a parole is not and cannot be granted by any determinate authority, such as philosophy, theology, or linguistics, least of all by one claiming a specific privilege or legitimacy for its own discourse. It is rather the cleft authority upon which any and all intentionality can eventually come into limited and thus provisional possession of its own langue. By drawing attention to the way Saussure's text enacts trom its very beginning this kind of paroling of la langue, we uncover a historicity to language that is unconditional, though not at all simple in nature, It is always a discrete act of parole that grants material reality and thus gives genuine authority to a langue that would otherwise remain an empty and isolated possibility - but it is always la langue that, through the power of its parole, then promises to be governed in its behavior by semantic laws that have not yet been formulated, much less tested tor consistency or coherence of meaning. La langue is therefore poised equally, though in a most volatile and unpredictable mode, between a prior act of parole that grants it the freedom to occur historically, and a future moment of critical reckoning in which it will have to account for its borrowed potential to become and remain a unitied system of intelligibility.

But who is really going to take any of that seriously? The paroling ofla langue? Mere word play, calembour, paronomasis. What is little more, after all, than a rather sorry pun works by allowing Saussure's found ing distinction for structural linguistics, langue/parole, to become contaminated by its purely accidential relation to idiomatic English expressions taken from the legal and penal systems, "to be granted parole," "to parole," or "to be paroled," and so on. The question, in its very trivolousness, brings us back to our starting point, specifically to Blanchot and his text, The Writing of the Disaster (6). Asking, examining, and intervening in some of the most troubling and important questions of our time, Blanchot's text returns over and over again to consider the way the French word, désastre, prompts us toward a richly suggestive but ultimately unreliable thinking of the philosophical relation between concealment and disclosure, or truth and error, The French word for disaster hides, though it can also be seen to shelter within itselt a root that idiomatic usage can no longer or not yet see or hear clearly: astre, a word for celestial bodies of light capable of providing us with "illumination." Blanchot's text outlines in this way a question about whether and how we could learn to read the writing of the disaster, l'écriture du désastre, as constituting a kind of writing granted to us by disaster. Can the word, disaster, that is, also teach us something about the way disaster as such could one day become a mode of disclosing the truth, helping us to rediscover an originary light or star whose path has been up until now lost or hidden within the disasters of history? The most frivolous, the play of the letter, can always turn into the most serious — and this actually happens here when the concealment and disclosure at play in the French word, désastre, are then turned by Blanchot's text toward an encounter with Heidegger's questioning of the (Greek) word for truth: aletheia. From a mere pun-un- covering and thus revealing the light, astre, concealed within the word, désastre - can eventually be wrested an urgent question for our own historical consciousness, if we only stop to listen to its call to us from within language. What is the truth of the disaster, and can this truth, no matter how dark and dangerous it has become, be made to shine anew? Blanchot's response is remarkable, even astonishing, tor its tenacious refusal either to legitimate or to dismiss the philosophical relation of concealment and disclosure, of forgetting and remembering that Heidegger finds at play in the word aletheia. Rather, Blanchot says at one point, a writer like Heidegger, who returns to the root of certain words that are considered fundamental in order to develop variations upon other ideas and words, "makes 'true' the concept that there is in the root a power that is at work and that also incites to work" (107). For Heidegger, truth as aletheia, or a mode of unconcealment, has been concealed in truth as veritas, or a mode of adequation and correspondence. This slight shift in meaning can actually be discovered, argues Heidegger, at the heart of the Greek word, aletheia, by heeding its formation from the alpha privative, aletheia. Truth, now read as aletheia or unconcealment, is to be understood as a mode of wresting or even robbing what is true from a counter movement of concealment, lethe, that is there from the beginning.? But, suggests Blanchot's text, what is philosophically true in this demonstration is not what Heidegger says about the Greek word for truth, for that belongs to a mode of truth that could never be demonstrated by philosophy alone with respect to aletheia. What is true and beyond question here is rather that Heidegger puts the word for truth to work, and in so doing, he does indeed manage to make the idea that it contains an effective power of philosophical explanation and understanding "work," or succeed, despite whether it is genuinely true or not. It will therefore not suffice either simply to take Heidegger at his word here, Blanchot implies, or merely to dismiss what he has to say about the truth as it is hidden and disclosed by the word aletheia. Rather, philosophically and historically speaking, one has to begin to account for the very precise way that Heidegger's writing of the truth in this word achieves the power to "work" — in both senses.

Now that particular understanding of the power at work in words whose play is put to work in this way is not something that Blanchot could have read as such in Heidegger's own text. Rather, it is an under- standing of the mysterious relation between language and truth, work and play, that he owes in large part to Jean Paulhan, in particular to a little and little-known text called Alain, ou la preuve par l'étymologie [Alain, or Proof by Etymology].® For it is in this text that Paulhan goes about the delicate task of exploring the "nature and play," that is, the laws according to which etymology works (280). Etymology concerns the semiotic relation between words and their meanings, though it does so by privileging the way any given word can be traced back to a form and a meaning that precede the ones it has today. As Paulhan is quick to point out, moreover, the word etymology doubles as its own best advertisement, since its roots can be traced back to the Greek, etumos logos, or true meaning (265). The "proof" by etymology to which Paulhan refers in his title, then, is to be understood as an epistemolog- ical one: etymology undertakes, and then claims to demonstrate the true meaning of words by reconstituting a knowledge of their origins. The "proof by etymology" is thus always the proof of etymology, where the root of "true meaning," the etumon, is both the means and the end, the vehicle as well as the destination. The problem is, however, the status of this "truth," or of the proof or authentic meaning that is reached by recovering the word's etumon, is not itself epistemologically univocal or sound. Paulhan's text thus reaches back behind etymology to ask how the authenticity of a word's ultimate "truth" can itself be authenticated.

On the one hand, Paulhan reminds us, the demonstration of true meaning derived from etymology is mere play, like an ordinary play on words (jeu de mots), since the only genuine "science" of etymology, made possible by the developments in modern linguistics over the last century or so, leaves no doubt that etymological explanations are at best either redundant or inconclusive, and at worst, downright misleading (275-77).

On the other hand, though, it remains a fact, however contrary it may run to the overwhelming evidence of linguistics, that writers and philosophers alike, indeed all ordinary language speakers, continuously rely on learned as well as folk or literary etymologies as the most authoritative means of establishing and justifying the truth of their own discourses (267-68, 278). The enigma that Paulhan thus discloses is a subtle but most provocative one. For Paulhan's text on etymological proof is not just a demystification of its epistemological pretensions, and this is what sets it apart from so many other critical texts on the subject. The real interest of this text, and one that Blanchot's reading of Heidegger pursues and amplifies, is the way in which it underscores that proof by etymology - the search for and validation of the "true" meaning of words by means of nonscientific procedures of knowledge production-continues everywhere unabated despite the demonstrated inadequacy of its premises as well as its conclusions. (9) How and why, Paulhan's little text dares to ask, does proof by etymology retain a special kind of efficacity in determining "true meaning," a practical efficacity that seems able to resist the most rigorous linguistic and epistemological critique?

Paulhan's answer is surprising only for its simplicity and obviousness. How the search for "true" meaning by etymological exploration is able to bypass the findings of linguistic and epistemological analysis is obvious: it belongs to an entire series of devices that have little or nothing to do with the kind of truth reached by scientific methods, including those of both modern linguistics and classical logic (280, 300).

Rather than obeying the laws of a science of linguistics or a logic of phi. losophy, "etymological proof," says Paulhan, is always literary, no matter how learned it may appear. 10 In fact, in order for etymological proof to work at all, Paulhan argues, it must be literary through and through, for only then could it exceed the merely verifiable history of a word's signification—which may be epistemologically sound but which remains of limited explanatory value with respect to the purely accidental stages of its own true meaning (275-77). Etymological proof, on the other hand, is a "rhetorical ploy" (tour de rhétorique), and as such it always goes beyond the truth of the observable laws of linguistic evolution in order to explain and therefore motivate in its every detail the ultimate meaning of any given word as well as the entire edifice of corollary arguments that can then be built with it (280-82). Etymology, then, like "poetry" for Mallarmé, is a fiction, but it is a "supreme" fiction in that it serves to repair the original defect of all language, which is to permit an infinite diversity of meaning without being able to offer a philosophically coherent explanation and thus justification for its own arbitrary capacity to appear true. Etymology dispenses with the methodological constraints of both linguistics and philosophy, and thus it allows even the most random features and func- tions of language to appear fully motivated and meaningful within the context of their own fictive constructions. It is decidedly not a form of verifiable knowledge, "une science" or "une connaissance," says Paulhan, but rather the fulfillment of a mere desire to know, "un souhait," which could never be otherwise satisfied by language. Why such "etymological" fictions, which by this point in Paulhan's text have been extended to include effects of paronomasia, ellipsis, antithesis, allusion, and all the other devices and wiles of a generalized rhetoric, manage to resist the laws of scientific observation and classical logic that always and everywhere challenge their epistemological legitimacy is simple: they possess an undeniable force that both linguistics and philosophy, far from effectively contesting, actually require for carrying out their own programs. Paulhan says this force is itself something of a fiction, since it consists in an etymological "projection," the sudden and unexpected " discovery" of a meaningful design, or explanation underlying the structure of language. It is the projection that allows for and makes the discovery in language, but it is the discovery that then serves retrospectively to "prove" that indeterminate or even arbitrary connections between words, parts of words, and the philosophical systems they eventually allow to be constructed upon them, can all be traced back to the single and over-arching purpose that has now been projected upon them. Such a "projection," Paulhan adds, which appears all the more "violent and irrevocable" for its requiring more time and more work to discover and then refine, ends up joining "the inevitability of a natural phenomenon" with "an intellectual voluptuousness composed simultaneously of pleasure and power" (284-85).

That such a fiction of true meaning tends eventually to "impose itself" with the "stability, permanence, and self-certainty" of proof, concludes Paulhan, does nothing to change the fact that it results from "fallacious intentions" and "chimerical origins." The entire process is itself a "trap," Paulhan had already warned, but a necessary one that always works (282-83). By reinforcing the intellectual pleasure we cannot help but feel at our own sense of intellectual power, the etymological projection, and even "rhetoric in general," thus "dupes" us into granting its discoveries a "legitimacy" that can only be provisional at best, since we always lend it without any firm basis in knowledge - the actual source of the intellectual pleasure and power we take from language remaining a "secret" necessarily kept from us (285-86). Should the structure of this formulation, if not the very terms, bear an uncanny resemblance to the asymmetrical and dysfunctional relation in Saussure between the virtual coherence of la langue and the coercively discontinuous acts of la parole, this would not be purely by accident. Indeed, Paulhan's "etymology" does not merely resemble Saussure's parole, it is nothing but an instance of parole through and through. That is to say, for both Saussure and Paulhan, the "true meaning," or etymology, of any given parole always entails the usurped authority to speak power to truth.

In Paulhan's text, this becomes clear when la parole is examined as a specific case of etymological proof in its own right. For, as Paulhan suggests, la parole is never just one example of etymology among others. la parole belongs to the most privileged examples of etymology conceivable, since it doesn't just provide an illustration of etymology at work. Rather, thanks to the example of its own etymology, la parole actually serves to reinforce the circular proof of all etymology as an effective ploy for establishing "true meaning" in general (265). "What happens if we consider parole?," Paulhan asks. "The term comes from the Greek, parabole, which designates Christ's word, or truth par excellence" (266). What could be further removed, at first glance, from Saussure's understanding of la parole as always being too individual, momentary, and willful in its operation to provide reliable access to true knowledge, than this retracing of la parole back to a divine origin in the word of God, or truth incarnate? Etymology, or the true meaning of a word, seems to find its perfect complement and justification in this etymology of la parole, which uncovers the word as the ultimate foundation for truth. The true meaning of the word parole would teach us in this way that words are themselves the best way to reach truth. In the beginning was the Word, and etymology is the path that allows us to retrieve its true meaning. But Paulhan's entire argument is aimed against taking at face value the sta- tus of just such an etymological "proof" of the "divine origin" of all words. And so, when he returns to the example of the particular proof of la parole, he carefully shows how the privileging of la parole as an instance of authority that could somehow be traced back to the word of God, and thus to the ultimate ground for true meaning, is itself predicated on a rhetorical sleight of hand. The etymology of la parole is not really a proof at all, but rather the construction of a "myth," whose own authority is entirely arbitrary and therefore unwarranted. As Paulhan had argued in his 1938 text on "the power of words," the myth of this authority is based on an illusion, but the regularity of our belief in the illusion ends up by endowing the myth with a power of "natural law." (11) When it comes to retracing the true meaning of la parole, there is no reliable means of justifying any one over all the other possibilities at hand: "C'est au choix," says at last Paulhan in mock frustration — "take your pick" (278-79). And if la parole eventually gains the power to impose a single truth on us as though it were the word of God, this is not, Paulhan says, because it has been granted to us by divine authority. On the contrary, it is we who learn to bestow upon all its discrete relations of proximity and mere contingency a unified network of meanings that can eventually point to a much more serious, and even sacred design at work. (12) Like all etymology, moreover, parole is a figure for the way language in general functions as a projection of epistemological authority that works, not because it is based on true knowledge, but rather because it lends us back the power we constantly but without justification attribute to it: the power to speak, and thus to know, the ultimate truth. Paulhan's etymology thus brings us back to the "circle" Saussure's course had already struggled to identify and then escape: in order for speaking subjects to have a meaningful parole, they must obey the formal laws of la langue; but in order for la langue to come into being, it must first be given parole by the entire community it can then be said to govern. La parole, then, is etymology insofar as it serves to motivate, and ultimately to legitimate, though only in the mode of a rhetorical fiction, an entire series of permutations between language and meaning that in themselves are devoid of explanatory value with respect to the origin and end of their own true meaning. Whereas Saussure, once he was compelled to notice this unruly element in the power of parole, had sought to control its disruptiveness by cordoning it off from the systematic study of the "constitution and laws" governing all signs (semiology), Paulhan takes an entirely different tack. (13) Indeed, as we have just seen, the etymology of parole-and the parole of etymology for that matter-entails an infraction against the epistemological laws of both linguistics and philosophy by claiming to possess a form of "true" knowledge that cannot be legitimated by either linguistic or philosophical methods alone. However, to the extent that neither the discourse of linguistics nor that of philosophy, nor any meaningful discourse as such, could ever simply do without recourse to the very same words (paroles) and true meanings (etymologies) that they also serve to call into question, the ensuing "trap" offered by paroles and their etymologies becomes itself inevitable, and therefore a new kind of "law" in its own right. This strange "conclusion" to Paulhan's argument becomes readable in the two rambling chapters he appends to the body of Alain, ou la preuve par 'étymologie. The chapters themselves are called "Notes and Observations," and each takes the form of a letter responding to critical reactions to an earlier text by Paulhan, called Petite préface à toute critique [A Short Preface to All Criticism]. (14) What unites both responses is the unexpected though recurrent claim by Paulhan that his own work remains governed by the possibility of attaining a critical point of view on the rhetorical operations of language that would itself be epistemologically legitimate, despite precisely those disruptions of philosophical and linguistic laws that seem always to be enacted by the literary object, or any object partaking of literary elements, which is to say all language. "My whole point," Paulhan writes in the first response, "is in fact to use literature as the basis for sketching out a system of knowledge that would be precise and rigorous, in short, scientific" (289). But is such knowledge possible whenever language is involved, which is to say always? If, as Paulhan argues in the rest of this text as well as in his others, the "critical methods" available to us for the study of language are themselves always the result of an "illusion of language" - such as those afflicting and disqualifying both terrorist and rhetorical methods of analysis - then, Paulhan imagines himself being asked, "wouldn't such an enterprise have to be merely chimerical?"

Paulhan's answer offers what is perhaps the most surprising twist of the entire essay. For, says Paulhan, the possibility that every critical method taking language as its object of study-whether it be literary, political, or philosophical in nature—proceeds at bottom from an "illusion of language" in no way counters the need to continue the undertaking of " critical reflection" with ever more precision and rigor. (15) "That would be just one more reason," says Paulhan, "for pushing the inquiry even further" (290). Far from leading to a dead-end, or to a relativistic approach where no discrimination whatsoever is possible among the competing "critical values" that necessarily emanate from different " critical methods," Paulhan's dogged pursuit of the traps, confusions, and illusions hampering the formation of any truly reliable system of knowledge about language only reveals the problem to be one that is more imperative and engaging to address (302). In a chapter interrogatively entitled "Whether there are laws of expression," Paulhan returns yet again to a question that is ever-present in his texts, which he does not hesitate in this context to qualify as "a simple case of the most important problem concerning mankind" (297). The specific challenge addressed here is the formation of a critical method appropriate to the study of literature, but Paulhan also says this is merely one instance of the much larger question of whether there are "laws" governing the exact way that words are related to meaning, or language is related to thought, and finally, that mind is related to body, and spirit to flesh. This question, in all its universal generality, from the functioning of language to that of literature, and on to the extralinguistic domains of philosophy and politics, remains an insurmountable stumbling block for metaphysicians, critics, poets, and linguists alike, Paulhan says, to the precise degree that such regional discourses all fail to discern the "laws" governing how thought can be directed by language, or inversely, how language can be directed by thought. But if the only dependable law governing our kowledge of the way language works is one that consists in the regularity with which critical method falls prey to illusions, traps, and errors in attempting to respond to this question, then where exactly does that leave us with respect to the kind of knowledge that Paulhan's own text, in the name of truth, both demands and proposes? When the law becomes itself a rule of error, then how can the critical enterprise of discovering truth lead to anything but radical tailure? As early as 1936, in the first version of Les fleurs de Tarbes, Paulhan had spoken of just such a "law of failure" to describe the regularity of error that upsets critical method from within. (16) And yet, this law of failure, even in the 1936 text, was not the final word on the matter, but rather a fresh opportunity to find a way out, a way to "hope" for a "solution" that, by subjecting its own critical method to methodical scrutiny, would not simply repeat the same errors of Rhetoricians and Terrorists alike. What has become, we may now ask, of this law of failure and the paradoxical hope of finding a methodological solution by the time Paulhan writes Alain, ou la preuve par l'étymologie? For etymological discoveries, as Paulhan amply demonstrates, belong indeed to all those "illusions of language" that serve as proof, alas, not for the establishment of "true meaning, " but only for the necessary failure of every rhetorical device to provide a firm means for securing our knowledge. However, as is typical of Paulhan's texts, especially of their endings, it is precisely at the moment when all seems lost that a new way of considering the entire problem is introduced into the argument. Blanchot, in the first text he devotes to Paulhan, draws attention to this signature move by Paulhan, and also points out that it can occur only when the "writer... accepts the law by means of which he enters into an obscure zone where there is neither pathway nor marking. (17) Approaching Paulhan's conclusion about language's trap of "permanent illusion," and describing the ensuing space of methodological uncertainty that is opened up by it, Blanchot says at this point that "we need to take one more step, though without believing it will take us very far." This last step, or pas, to which Blanchot refers with respect to Les fleurs de Tarbes, is also taken at the far end of Alain, ou la preuve par l'étymologie, after Paulhan has tried every resource of critical method, only to find them all wanting. Still, he says, "there remains one hope" (296). What is this one remaining hope espoir-according to Paulhan? As Blanchot suggests, it will have to be a step beyond hope, which will actually obscure hope by stripping it of every recognizable feature. In other words, this last hope can be no more hope; in fact, as it turns out, the obscure hope Paulhan holds out to us is only the shadowy hope of despair, l'espoir du désespoir, and therefore only another step back into language and all its problems and obscurities. In a gesture that anticipates Blanchot's reinscription of Heidegger's a-letheia in The Writing of the Disaster, though with his own off-hand and understated manner, Paulhan, at the end of the text on etymology, thus gives us a new word and a new etymology for "hope," by turning the word "despair" into the only hope we will ever have to reach a true knowledge of our own language and all its predicaments. "There remains one hope," Paulhan writes with respect to discovering the semantic laws that govern language (and that language governs), "or, rather, perhaps it could be said — a desperate solution Il reste un espoir. Ou plutôt dira-t-on peut-être,une solution de désespoir" (298-99). The tinniness of the translation here is, as always, highly overdetermined. It serves to reveal, by contrast to the way Paulhan makes "espoir" in the first position resonate in the second place and actually turn into its own negation, "désespoir," something more significant at play than mere assonance in Paulhan's text. For how could "despair" ever be "said" to be a solution for anything, much less for this all-important problem of determining the laws that govern language and meaning? Despair, properly speaking, is not and could never be a "true" solution, since by itself genuine despair would merely aggravate the original problem, whatever it was. But isn't that the point of the whole argument after all? " A desperate solution," like everything else in language, moreover, is already and forever a pat phrase whose "true meaning" - neither quite total despair, for one continues in such cases to act toward a goal, nor yet wholly a solution, since desperate solutions more often than not end nonetheless in miserable failure - could never be traced back and derived from its roots in any reliable way.

And yet, tracing despair (désespoir) back to its etymological root in negated hope (dés-espoir) is exactly what Paulhan does in order to explain and remotivate the "desperate solution" he manages in fact to offer at the end of this text. To some degree, it must even be granted that Paulhan succeeds in this attempt, since a solution for which "hope" has been all but lost in "despair," a "solution" that can from now on function only by hoping against hope, as it were, would actually offer a decent translation for the current signification of "desperate solution, whose own meaning with respect to the relative success or failure of its undertaking remains largely open. Despairing of discovering a positive law that would ever permit direct access to a "true knowledge" of language by somehow avoiding all of its traps or illusions, Paulhan therefore proposes instead a methodological solution that would be ruled first and foremost by the recurrent fact of language's potential to deprive us of anything but illusory truths.

"Yes," Paulhan acknowledges with respect to the arguments and proofs advanced by any given critical method, "[they] are based on illusory assumptions - but the illusions that serve to establish such assumptions are themselves facts that are perfectly real" (298). What is true, as Blanchot, following directly in the steps of Paulhan, had pointed out with respect to Heidegger's "truth" (aletheia), is therefore not the philosophical knowledge that a study of the relation between language and meaning, or between language, meaning, and world would disclose and then allow us to reach. What is true and perfectly real is the way that words necessarily produce an illusion of knowledge as soon as we examine with methodological rigor the way that they actually work. And since, by not stopping to make such an examination we merely allow the play of such illusions to function all the more coercively on our thought, we have no choice but to ask these questions about truth and reality with obsessive regularity if we are to think at all. There can be no hope of ever escaping illusory judgments about thought and language, for the reality is, we always make them willy-nilly inside language, and thus without ultimate rhyme or reason - whence the despair. But since the absence of rhyme and the nullity of reason are themselves perfectly constant and regular in our judgments, we could at least begin to examine critically the laws of their functioning, and thus our own willingness to conform blindly to their play - hence the hope.

Such, at least, would be one desperate attempt at rendering, and thus translating, Paulhan's next step into the unknown: "And, of course, Rhetoricians and Terrorists judge all the wrong way [jugent à tort et à travers] —but it is by conforming to such constant errors [suivant des torts si constants], and such regular [confusions des travers si réguliers], that, for want of a critical method, they themselves turn into laws [ils se changent eux-mêmes en lois]" (298). While he thus despairs of making any critical judgment that would not itself be "à tort et à travers," another commonplace within the French language, and that means perpetually in error and therefore ultimately "groundless," Paulhan emphatically does not despair of the one hope remaining to all users of language - Rhetoricians and Terrorists alike — which is to subject the inevitability of that very law of error to renewed critical analysis. (18) And so, in the same sentence that has to use a mere tic of language, "à tort et à travers," to qualify all judgments about language and thought as being ultimately "groundless" — à tort et à travers — he also breaks that very cliché apart into its discrete elements, "des torts si constants," and "des travers si réguliers," and begins in this way to subject the rule governing the conventionality of its own language to a new force of critical reflection and dispersal.

For, Paulhan goes on to suggest, it would be precisely for want of performing such an analysis — "à défaut d'être savants" — that the users of any given language always risk becoming merely used by it, and thus turn themselves into the groundless and mechanical laws they should be subjecting to interrogation: ils se changent eux-mêmes en lois. Of course, nothing can ever guarantee that Paulhan's own "desperate solution," uncovering over and over again the way that language always works to conceal and thus to deprive us of any definitive truth or security, will not itself turn into a merely repetitive tic, error, or confusion. Like every true parole, moreover, this one last glimmer of hope has to be granted or received according to a law that could never prevent it from disappearing all over again in genuine despair or even disaster. But, at least in the case of Paulhan and Blanchot, and thanks to the writing of the disaster and the etymology of despair, the very particular hope and light held out by such a parole would, in addition to conforming to the letter of this law, also serve to warn us against taking the source of its own authority for granted once and for all.

---

1. At least one of the sources of Blanchot's word, disaster, is to be found in an obscure corner of Jean Paulhan's text, Alain, or la preuve par l'étymologie: "L'écart de désastre à catastrophe et à calamité n'a plus rien à voir avec les étoiles, le bouleversement, la paille (What separates disaster from catastrophe as well as from calamity no longer has anything whatsoever to do with the stars, or overturning, or straw)." The following essay, as will become clear, is one attempt to situate with more precision the role that this text by Paulhan could play in a reading of all such words whose original light has disappeared. The specific reference to disaster can be found in the edition of Paulhan's Œuvres complètes (Paris: Cercle du livre précieux, 1966-1970), vol. 3, 277. Further references in the text.
2. Louis Helmslev, "Langue et parole," in Essais linguistiques (Paris: Editions de Minuit, 1971). Further reterences in the text.
3. Ferdinand de Saussure, Course in General Linguistics, trans. Wade Baskins (New York: McGraw Hill, 1959), 9, 14; translation modified, further references in the text.
4. This unknowable power is first of all and most importantly the power of change: "... It is in la parole that the germ of all change is found..." (Course, 98, tr. modified). As such, and as we shall see, it is also the most unforeseeable and unstoppable power of what can actually occur as history.
5. Jacques Derrida, "Sémiologie et grammatologie," in Positions (Paris: Éditions de Minuit, 1972), 39-40.
6. The Writing of the Disaster, trans. Ann Smock, (Lincoln: University of Nebraska Press, 1995). Translations occasionally modified.
7. Heidegger's exegesis of the word aletheia is complex and differentiated, and it evolves over the course of some thirty years or more. Starting out as a mere case of etymological derivation, it ends up coming much closer to a free-standing paronomasis. Two concise but thorough considerations of what is at stake in Heidegger's insistent thinking of aletheia as a-letheia, are: Robert Bernasconi, "Aletheia and the Concealment of Concealing," in his The Question of Language in Heidegger's History of Being (Atlantic High- lands, NJ: Humanities Press, 1985), 15-27, and John Sallis, "At the Threshold of Metaphysics," in his Delimitations (Bloomington: Indiana University Press, 1995), 170-85.
8. Blanchot refers explicitly to this text by Paulhan when he considers Heidegger's "etymology" of aletheia (93-97), and Paulhan includes a reference to Heidegger in his comments on the use of etymology by philosophy (267).
9. In this respect, and most aptly in the context formed with Heidegger and Blanchot, Paulhan's considerations on language and truth could be fruitfully compared to those of Nietzsche. A recent dissertation by Anna-Louise Milne documents with care and perception Paulhan's interest in Nietzsche and its traces in several of his texts. See, especially, "Dionysus and Apollo," in Jean Paulhan's Commonplace: The Genealogy of a Critical Concept (PhD, Columbia University, 1999), 256-70.
10. Such would be the "mystery" of literary language in general and poetry in particular: a capacity to work according to a law lying beyond and interfering with the reach of epistemological laws. Blanchot, in an essay tellingly entitled "Mystery in Literature," says that Paulhan aims at "finding a law whose legitimacy would be grounded in mystery, which is to say, one that could be a law in conformity with that which escapes the law" ("Mystery in Literature," in La part du feu, Paris, Gallimard, 1949), 59. Michael Syrotinski cites this passage in his excellent study of the relation between Blanchot and Paulhan. In many respects, the following argument is merely following up on directions suggested by Syrotinski's reading ("Blanchot Reading Paulhan," in Defying Gravity [Albany: State University of New York Press], 77-104).
11. See Paulhan, "Letter to the Directors of the Nouveaux Cahiers: On the Power of Words," in Les fleurs de Tarbes, ed. Jean-Claude Zylberstein (Paris: Gallimard, 1990), 235. It is in this context that a comparison between Paulhan and Barthes would be relevant, since Barthes's own treatment of "myth" stems from a highly combustible mix of Saussurean semiology and German ideology critique. As Barthes will clearly state in his Preface to the 1970 edition of Mythologies, there can be no effective political analysis that does not recognize its dependence on semiological principles, and no consistent semiology that bypasses its own political dimensions and implications. The main difference between Paulhan and Barthes is that Barthes, at least at this stage, is more utopian than Paulhan in his belief in the possibility of cleansing his own discourse of a rhetorical, epistemological, and political authority whose own legitimacy remains suspect and in need of critical analysis in its turn. See Barthes, Mythologies, trans. Annette Lavers (New York: Noonday Press, 1972), 9.
12. This is exactly what Saussure had argued about the necessary relation between parole and langue: in order for the signifying functions of la langue to become established as the seat of intelligibility, la parole, in all its randomness, had to come first. "And the same goes for parole. Yes, our word comes from parabole, Christ's teaching. But before parabole? Actually, the Greek parabolè only means confrontation, comparison; it is a simple term of rhetoric, like metaphor or antithesis. And before rhetoric? Parabolè only means proximity, contiguity: two things happen at the same time, two things find themselves placed next to each other. There is nothing very serious or sacred about that ..." (2.79). And as Saussure had also said, it is out of such arbitrary relations of proximity and contiguity that the signifying "articulations" of la langue will have to appear.
13. Saussure tends to use etymological examples precisely when he needs to argue for the necessary exclusion of parole from la langue, or of diachronic from synchronic linguistics. Some of the more memorable examples occur in the section on "onomatopoeia," itself a sub-category of paronomasis (69). The fact that he will attempt to make etymological "proof" indiscriminately illustrate both the arbitrary and the motivated nature of parole indicates a certain lack of control over his own examples of etymology. Parole, like etymology and paronomasis, is a fiction capable of motivating every relation, though without itself ever escaping a radically arbitrary element. One of the most overdetermined examples of etymology, paronomasia, and parole to be found in Saussure's Course (175), is picked up by Paulhan in his text on literature, knowledge, and politics, De la paille et du grain (5:316). It is of course the example of Alsace: Choucroute is an attempt to grant parole to a foreign and intruding langue, Sauerkraut, by re-endowing the sound of its German letters with French meaning. Saussure calls this an example of those "crude attempts to explain refractory words by relating them to something known" (173).
14. The Petite préface à toute critique dates from 1951. The two critics to whom Paulhan responds at the end of his text on etymology are Maurice Nadeau and Aimé Patri. The original edition of Alain, ou la preuve par l'étymologie (Paris: Editions de Minuit, 1953) contains several paragraphs at the beginning of "Notes and Observations" that are not reproduced in the Oeuvres complètes. They are written under the title, "I have to admit it," and concern Paulhan's feeling that critics have missed the point of his earlier text - which he promises to make clearer in the responses that follow.
15. Paulhan's position here is not without analogy to that of Paul de Man as it is articulated in "The Resistance to Theory." This is not surprising to the extent that de Man, like Paulhan, is addressing the problem of critical method, or "theory," from a post-Saussurean point of view that emphasizes those factors and functions of language — and de Man will refer both to etymological proof ("Cratylism") and paronomasis in this context - that cross between and eventually cross up a neat division between rhetoric and epistemology, la langue and la parole, literature and philosophy, fiction and referentiality. Like Paulhan in Alain, or la preuve par l'étymologie, de Man will examine the rhetorical element in etymology, or paronomasis, as a "trope that operates on the level of the signifier and contains no responsible pronouncement on the nature of the world - despite its powerful potential to create the opposite illusion." The power at work is one of illusion, and it necessarily operates on every referential discourse of knowledge as well as of power, See, "The Resistance to Theory," The Resistance to Theory (Minneapolis: University of Minnesota Press, 1986), esp. 9-10.
16. Paulhan, Les fleurs de Tarbes, ed. Jean-Claude Zylberstein (Paris: Gallimard 1990). The pages in question here and following, which treat the problem of critical method in terms of fatlure, hope, solution, and method, are 247-50.
17. Maurice Blanchot, "Comment la littérature est-elle possible?," in Faux Pas (Paris: Gallimard, 1943, 1971), 100. Further references are from the same page. For a deceptively powerful and beautiful meditation on this move in both Paulhan and Blanchot, see Ann Smock's book, What is there to say! (Lincoln: University of Nebraska Press, 2003), especially the section entitled "Getting Across," 30-44.
18. Such is the privative "law" of all language: it forever separates us from any direct access to light (astre) and hope (espoir) by making it always necessary to pass first through disaster and despair (l'écriture du désastre, une solution de désespoir). Another way to put this is found in Paulhan's "conclusion" to The Flowers of Tarbes. Since the illusions of language are constant and permanent, we are forever condemned to a critical method of dis-illusionment: "Where illusion creates the thing, disillusion ruins it" (3:80). But since the "dis-illusion" here must itself work by means of an etymological play present in all language, this necessarily delusive search for truth is also necessarily endless. See also, "Letter to the Directors of the Nouveaux Cahiers: On the Power of Language" for a slightly modified version of this predicament in the search for all truth (235).