1In “Performative and Passionate Utterance,” Stanley Cavell seeks to respond to an invitation that John Austin, in his How to do Things with Words, should have addressed to his heirs—the invitation to consider language not solely as a tool for description or even action, but equally as a vehicle for emotion. The Austinian theory of speech acts is thus treated as “a source of passionate utterance,”  as a provocation to dialogue and confrontation—or rather, as that which should have constituted such an appeal. Cavell’s own text, on the other hand, very much does contain one. I would like, in following his example, to try to respond in turn, and open the door that he left ajar, as if to suggest to his followers that they should come through it. On two occasions in “Passionate Utterance,” the names of Wittgenstein and Freud appear, each time closely associated with each other.  Which raises the following question: How are the Austinian and Cavellian concepts related to the thought of these two authors? To respond, we must first of all ask ourselves what the latter may themselves have in common, so as to justify invoking them side by side. Both make use of language for therapeutic ends: Wittgenstein writes to dissipate conceptual disorders, “mental cramps,” and other “diseases of the understanding;” Freud, to heal psychological disorders. The object of this study is to show that it is precisely in so far as they both raise this therapeutic voice against that of madness, that their usage of speech belongs to the performative, and more exactly to the passionate utterance. Cavell extends Austin’s theory by introducing the concept of passionate utterance as “some articulation into the region of the perlocutionary act;”  I wish to take a modest step in the same direction by extending the field of the passionate utterance in turn, by introducing into it—alongside opera and political rhetoric—the discourse of the therapist.
Therapeutic Discourse and Performative Speech
2In Freudian analysis as in Wittgensteinian grammatical analysis, the usage of discourse is therapeutic in so far as it is performative. For in both of these two cures, the psychoanalytic and the philosophical, the simple expression of a disorder suffices to provoke a deliverance from it: the description of symptoms and their causes is not, as in medicine, a prior stage to their healing, it is the method of healing. The function of discourse is not to present a theory of therapeutic procedures, preliminary to their application: it effectuates therapy. Language treats its own ills through words. The double nature of the verbal, where the illness and the remedy are one and the same, can be read, for example, in Wittgenstein’s ambiguous formula: “Philosophy is a battle against the bewitchment of our intelligence by means of our language.”  The syntax here leaves open an alternative: Is language the means of the bewitchment, or of the battle? Both, of course: Speech is at once the charm that leads thought astray, and the antidote that delivers it. Wittgenstein and Freud “do things with words,” giving them a force and effectivity. This is to say that their therapeutic usage of discourse, contrary to the usage made by medicine, is not purely neutral and constative, but effective, and therefore performative. It perfectly marries with the definition Austin gives of the latter category:
4In short, in Freud and in Wittgenstein, to say is to do, or rather to undo.
5Freud understood, in the first early stages of psychoanalysis, that the disappearance of psychic pathologies rested upon a certain mode of discourse: that by which the patient, by formulating his problems, would suddenly become conscious of them and immediately detach themselves from them. He discovered, to his surprise:
That each hysterical symptom immediately and permanently disappeared when we had succeeded in bringing clearly to light the memory of the event by which it was provoked and in arousing its accompanying affect, and when the patient had described that event in the greatest possible detail and had put the affect into words. 
7The therapeutic act rests entirely upon a saying, on a “depurative narration.”  Psychoanalysis consists, according to the name with which a patient, Anna O., was to baptize the new method, in a talking cure:  understanding by this that the cure consists entirely in the saying.
8Wittgensteinian language therapy deserves this name just as much. It proceeds from a patient description of the way in which we entangle ourselves in our own rules of grammar, a description that suffices to provoke an immediate relief, very closely similar to the abrupt feeling of illumination which, in Freudian analysis, accompanies the overcoming of repression. The therapeutic use of language is thus indeed, as much as a saying, a doing: it does not consist so much in formulating the solution to problems as in carrying out their dissolution: “The solution to the problem of life is to be seen in the vanishing of the problem.” 
9Confusions “are resolved by grasping the functioning of our language;”  simple speech suffices to make them “completely disappear.”  In consequence, just as surely as in Austin, in Wittgenstein “words are acts.”  The effectivity of discourse is a theme to which the latter will impute a particular importance. He holds that his own philosophy is not merely a “dead letter,” but that it provokes effects:
I know that my method is the correct one. My father was a businessman and I am a businessman: I want my philosophy to get something done, to get something settled. 
11Wittgenstein would also declare: “Bad philosophers are like slum landlords. It’s my job to put them out of business.”  Ineffective philosophical discourses must be cast out; only those that are sufficiently businesslike have the rights of citizenship. The former are a disease of thought, the latter are the treatment. This conviction constitutes the originality of the idea that Wittgenstein makes of his discipline: unlike his predecessors, he does not envisage it as contemplation, but as action: “Philosophy is not a theory but an activity.”  This affirmation returns constantly in his teaching practice: “What I want to teach you isn’t an opinion but a method. . . . I don’t try to make you believe something you don’t believe, but to make you do something you won’t do.”  On account of its curative efficacy, philosophy, for Wittgenstein, is a speech act. This unexpected dimension plunges into crisis the purely negative, neutral and descriptive definition that he nevertheless gives of his method: Is it really a contemplation of the workings of language, a seeing, that “leaves everything as it is,”  or rather the very forceful intervention of a doing? The receptive passivity that he often attributes to his own philosophy, refusing it the cognitive fecundity that its predecessors had claimed for their own, seems to be belied by the performative character of its remarks. In reality, the latter never leave their object intact—because they force a conversion of the gaze cast upon it, they instantaneously change our relation both to things and to ourselves.
12This ambivalence between the neutrality of observation and the power of action echoes a tension internal to Austin’s theory, and may doubtless be resolved in the same way. Curative speech, in Freud as in Wittgenstein, rests upon a correct description of the symptoms; it is thus just as much constative as performative. The two dimensions are in fact inseparable, to the point where each finds its foundation in the other: the success of the cure depends on the correctness of observation, but inversely, as we shall soon see, the therapeutic efficacity of statements, in return, endorses their truth. The two aspects can only be distinguished artificially. Is this not a contradiction, according to the Austinian definitions of the constative and performative? Things are more complex than that: Austin’s theory encounters a crisis, linked to the discovery that, in reality, every performative utterance carries an informative element and that, on the other hand, the purest of constative utterances nonetheless simultaneously constitutes an act, even if it is only that of utterance. Thus, the mixed status of Wittgensteinian or Freudian therapeutics, between cognition and action, would be an additional way to place in question the original Austinian dichotomy of statement and performance. This very particular usage of words justifies more than any other the passage to Austin’s second, no longer binary but ternary, distinction, in which each of the categories—locutionary, illocutionary, perlocutionary—is, inseparably both descriptive and effective. But under which of these forms should curative speech be subsumed?
Therapeutic Discourse and the Perlocutionary
13The Austinian triad of speech acts is articulated as follows:
Thus we distinguished the locutionary act (and within it the phonetic, the phatic, and the rhetic acts) which has a meaning; the illocutionary act which has a certain force in saying something; the perlocutionary act which is the achieving of certain effects by saying something. 
. . . how many senses there are in which to say something is to do something [locutionary act], or in saying something we do something [illocutionary act], and even by saying something we do something [perlocutionary act]. 
15Quite evidently, Freudian and Wittgensteinian speech figures in the third category: the therapeutic action that it effectuates is indeed more than the simple production of a linguistic fact, as in the locutionary act; what is more, it does not have a mere force, like the illocutionary act, but also a causal efficacy—we remarked above on the power and the spectacular immediacy of the dissolution of psychic and grammatical disorders—which is what is proper to the perlocutionary. We have at our disposal, moreover, a formal criteria to distinguish curative speech from illocutionary enunciations (for example “I name this ship . . . ,” “I take this woman for my lawfully wedded wife,” “I promise,” “I wager,” etc.): the former, unlike the latter, cannot be reduced to the explicit performative model—that is to say, a verb in the indicative present tense, in the active voice, and in the first-person singular. Abstracting from the manifest absurdity of these formulae, it is certainly not in declaring “I dissipate this philosophical confusion” or “I destroy this hysterical symptom” that the analyst of the psyche or of grammar can arrive at his aims! The talking cure thus does not produce its effects in saying something, but by what it says. The act, Cavell would say, is not “built into the verb that names it,”  it is effectuated by the speech whose designation is not the act in question. In other words, the efficacity of curative discourse, contrary to that of illocutions, does not rest on a preestablished convention. “Illocutionary acts are conventional acts; perlocutionary acts are not conventional.”  Certainly, the illuminating description, in Wittgenstein, is always the recalling of a grammatical norm, but itself does not draw its power from such a convention. Quite on the contrary, the procedure must be entirely reinvented with each application: “There is not a philosophical method, though there are indeed methods, like different therapies.”  The power of the therapeutic voice does not rest on any codification, on any “recipe” established once and for all. Liberating speech is always radically creative, violent and unexpected—an indispensable condition for its success, since it has to seize thought off guard, so as to overcome its resistances by surprise, since otherwise it may anticipate the attack. It will be objected that the words that make for the effect are not the fruit of chance. In the cure, the “happy” perlocution supposes the putting into place of the “chosen” terms, the privileged keys that alone can “open the lock,” suddenly freeing the mind from the conceptual confusion or psychic disorder under consideration. The formula likely to lift the repression is the one that is welcomed with a “yes, that’s it,” “exactly!”—the sign of a familiarity and a necessary determination of the chosen words. But this determination, precisely, is not conventional. In Freud, the words of deliverance are those that the singular narrative of the patient has charged with a symbolic weight; in Wittgenstein, it is a matter of those which correspond exactly with “what we say when,” and which help us find the well-known face of the mother tongue. Of course, this latter rests upon norms, but these norms do not dictate the way in which one might recover them: one never knows in advance which expression might suddenly rescue thought from its distress. Far from resting upon the repetition of a uniform rite, therapeutic speech is always that of an unforeseeable novelty.
Therapeutic Discourse and the “Passionate Utterance”
16The analytical discourse of Wittgenstein and Freud is therefore, among the different forms of performativity, a perfect example of perlocution. It is possible to refine further this result by referring to a new subdivision that Stanley Cavell introduces in the Austinian category of the perlocutionary, that of “passionate utterances,” whose proper effect consists in a discharge of emotion. For therapeutic perlocutionary utterances have the power to bring relief to the mind—that “peace in thoughts”  that, according to Wittgenstein, is the unique aspiration of philosophy—when, and in so far as, they constitute a passionate appeal to desire. Stanley Cavell invites us to evaluate the depth of Freud and Wittgenstein’s interest for the capacity of language to give voice to emotion:
As I read Wittgenstein, as well as Freud, [humans] become (always already) victims of expression, readable in every sound and gesture, their every word and act apt to betray their meaning . . . as if we are expression machines, and virtually never turned off (or can one say?: only rarely, and then only virtually, turning ourselves down). 
18The deepest affinity between the two authors, the characteristic that marks them out, together, in the history of ideas, would be their common sensitivity to a dimension of language often neglected by theorists—too attentive to its descriptive function—namely, its expressive capacity. Freud and Wittgenstein, breaking with the impoverished traditional theoretical schema that limits discourse to the neutral description of the real, and even going beyond the Austinian discovery of a performative aspect of speech, show that speech also has an intimate complicity with our emotions. It is up to us, now, to show that these two thinkers do not stop at “recognizing language as everywhere revealing desire,”  but capture and turn to their own, therapeutic, ends, this passional potential of speech: it is by touching the heartstrings that both of them accomplish their task as doctors of the soul. Psychoanalysis conceives pathologies as disequilibria in the psychic economy: they stem from an augmentation in the quantity of excitation—fear, shame, pain, fierce wishes—which contravene the “principle of constancy” of mental dynamics. The illness is not the affect in itself, but the deprivation of any adequate expression capable of reestablishing internal balance through a sufficient discharge of energy. The repression that affects certain overviolent emotions, impossible to assume, by the same token prohibits their natural manifestation—tears, revenge, and avowals of love, accusation, or confession—and transforms them into a psychic burden that quickly becomes the source of disorders. The repressed feeling tortures the mind all the more cruelly in that it does so in the shadows, where it has no chance of being surprised and chased out. We can thus easily understand why it is precisely only a “passionate utterance,” freeing the contained desire, that is capable of producing liberatory effects: as in Aristotelian catharsis,  we must bring the emotions onto the scene of discourse, have them before us in order to feel them frankly, with neither evasion nor denial, in order to purify the soul. Speech
brings to an end the operative force of the idea which was not abreacted in the first instance, by allowing its strangulated affect to find a way out through speech. 
20Psychoanalysis obtains results exactly in so far as it reanimates desire. The secret of its force lies entirely in the usage of “passionate utterances.”
21The same is the case in Wittgenstein’s therapeutic method. His second philosophy places affects at the heart of language, from the beginning of the Philosophische Untersuchungen onward. As Stanley Cavell remarks, the inaugural dismantling of the Augustinian conception of the verb leaves only one element intact—precisely the idea according to which words serve to speak desire.  Other passages of the book show more directly to what extent Wittgenstein is sensitive to the expressive vocation of language, to the way in which words are inhabited and molded by the emotions:
In this way, I’d like to say, the words “Oh, if only he’d come!” are charged with my longing. And words can be wrung from us—like a cry. Words can be hard to utter: those, for example, with which one renounces something, or confesses a weakness. (Words are also deeds.) 
23Wittgenstein’s insistence on the difficulty of uttering something is hardly surprising: like Freud, he is conscious of our tendency to repress desire and silence its voice. This is even his explanation for the spectacular longevity of philosophical misunderstandings: for centuries we remain prey to the same linguistic traps, because we are unconscious of the temptation to ignore the function of language, and thus cannot steel ourselves to cure an ailment that is not even recognized as such. The task of the philosopher is thus to express what is repressed, to “pass from unobvious nonsense to obvious nonsense” —to give voice to the anarchic desire that twists the rules of discourse, because so long as we refuse to hearken to it, it will find for itself modes of expression all the more effective for their being obscure and circumventory. This is why we must, according to Cavell, read the Philosophische Untersuchungen as confessions: the Wittgensteinian strategy to repair and overcome errors consisted in an admission of what one is tempted to say in such and such circumstances, in spite of the contrary prescriptions of grammar. How many remarks, in Wittgenstein, begin with the words: “I’d like to say . . . ,” “Here we should like to say . . . ,” etc.? As Cavell opines, “the voice of temptation and the voice of correction are the antagonists in Wittgenstein’s dialogues.”  This “conflict of instances,” to use the Freudian vocabulary, is the bedrock of philosophical confusions, which are born of a thwarted yet tenacious desire to subvert commonly accepted grammatical rules. If the clash between the anarchic impulse and the law brings with it the pernicious effect of a deregulation of language, it is because its outcome remains indecisive, once it is arrested within an obstinate repression. The end of hostilities and the reestablishment of the usual linguistic reference points thus supposes, as a first condition, that one lend an ear to the transgressive desire, that one confesses it (to oneself). The therapy of language consists in ceasing to ignore the forces that work language, and finally in utilizing it to give voice to desires that have been reduced to silence. Just as in Freud, the passionate utterance is the means of healing.
24The intuition of having found in the Cavellian conceptualization the correct characterization of the therapeutic usage of language demands confirmation via a test that consists in comparing analytical speech against the seven “conditions of felicity” of passionate utterances, antithetical to the criteria of the genuinely illocutionary. The first condition, and the conditions 3 and 4 that follow from it are perfectly respected: in curative discourse, just as in the passionate utterance, “there are neither conventional procedure nor effects.”  We have remarked above that therapeutic speech is only effective as long as it is creative and surprising; the predictable repetition of a code would annul its force. The second condition—that of singularizing, out of all people, the person who is the correct one-to-one—is such that, in a passionate utterance, the “‘you’ comes essentially into the picture.”  This is eminently true of the Freudian cure, which is founded on an always-unique, never-reproducible relation of transfer between two interlocutors. It is also true of Wittgensteinian therapeutic dialogue, the only legitimate heir to Platonic dialogue, for it alone inscribes thought in an alterity that is authentic, not one simulated for convenience of exposition. In Wittgenstein’s and Freud’s texts, the constant welcome of objections, and thus the assumption of the risk of a multilateral point of view, marks, at a point rarely attained in theoretical discourse, a making room for the other.  The form of an exchange is consubstantial with their thought. And it is so in virtue of an internal necessity: no therapeutic treatment can be envisaged without the “you” imparting its movement to the experience. Which leads us to condition 5b, “demanding a response,” and the sixth condition, which completes it, “here and now.” Cavell describes the emotive utterance as a direct demand to another, in a passionate effort to maintain communication, to make it so that language remains a public and common thing. This demand for an echo, a sign of reciprocity and circulation of meaning between minds, is precisely that to which Wittgenstein and Freud give voice. Madness begins where communication with the other ends. It takes root, in philosophy, in private uses of language, growing into three forms (metaphysics, skepticism, and solipsism) of theoretical autism. All the remarks in the Philosophische Untersuchungen are a full-frontal attack on this morbid temptation to isolation, and they strive to restore agreement in words, through an invitation to “play the game” of ordinary language. In the same way, Freud observes that there is no psychic disorder without a retreat into the self, a difficulty in the relation to the other. The function of the analytic conversation is thus to reestablish the continuum of meaning between the patient and others, beginning with one other, the other incarnated by the doctor. But it is here that the seventh and last condition comes in: it is always possible in principle to decline the offer of an exchange. Therapeutic speech is a tenuous miracle, as precarious and uncertain as the success of the passionate utterance. If “the ‘you’ comes essentially into the picture,” this is also because the destiny of the treatment is at its mercy. There is no obligation to mental health; refusal to be rescued from the private one-to-one with madness is an inalienable freedom. The voice of psychoanalysis has often been covered up by others, or reduced to a soliloquy. Wittgenstein’s has long been welcomed with stony silence, or, worse still, with an enthusiasm founded on a misunderstanding,  which renders only a deformed echo of his thought. What is more, the philosopher, at least according to Cavell’s reading, deliberately leaves open the door to skepticism: while striving to reestablish confidence in ordinary language, he remains aware that doubt is always in principle a possibility.  Because Freud and Wittgenstein’s therapeutic method rests upon emotive perlocution, it is, like the latter, vulnerable to refusal, and exposes them to the risk of being voices in the wilderness. This is one of the reasons why their thought is pessimistic: it is founded upon a rash wager, and its spectacular performative power is no more or less guaranteed than a lamentable failure.
25Before concluding as to the identity of therapeutic discourse and the passionate utterance, we must discuss an important difference between them. For Cavell, if “a performative utterance is an offer of participation in the order of the law,” on the other hand, “a passionate utterance is an invitation to improvisation in the disorders of desire.”  It is on this point that curative speech diverges from it. Wittgenstein and Freud work toward a lifting of repression that concerns simultaneously the infringed wishes and the repressive law, whether it is the grammatical law (Wittgenstein) or an exigency of civilization (Freud). The immoderate affect is freed, but only so it can resign itself to the moderation of convention; the voice indissociable from desire is raised only to dissolve back into the social chorus. In this, again, Wittgenstein and Freud perhaps betray their pessimism: they make of analytical discourse the scene of a violent conflict between order and disorder, desire and the law, whose outcome seems quite uncertain. We must therefore say: Therapeutic speech is an exhortation addressed to the savagery of a singular desire, in the (mad?) hope of reconciling it with the collective order of the law.
The Roots of the Performativity of Therapeutic Discourse
26We have seen that the Freudian and Wittgensteinian use of language is a performance, but we have not yet inquired as to the sources of this performative power. We would now like to show that what these two thinkers capture and turn to their own particular therapeutic ends is actually a performative dimension inherent to all language.
27Wittgenstein constantly insists on the essential interdependence of “language games” and the “forms of life” in which they are inscribed: the semantics of an utterance is never, for him, separable from its pragmatics.  The rules of a given activity determine in one and the same movement the speech that accompanies it: saying always involves a doing. The latter is, in the last instance, the foundation of the verbal, the ballast of concreteness that roots it in real life and consecrates the authenticity of its usage. The act is first in relation to discourse:
The origin and primitive form of the language game is a reaction; only from this can more complicated forms develop. Language—I want to say—is a refinement, “in the beginning was the deed.” 
29The importance accorded to praxis as a necessary element of meaning recalls Austin’s reflections on the determinative part played by the evental and material context of a speech act in the constitution of its meaning. An utterance, for Austin, can only be “happy,” and thus “make sense,” if it is concomitant with the execution of the appropriate gestures and behaviors: no speech act without a speech situation. The idea that speech is act has no other source:
Once we realize that what we have to study is not the sentence but the issuing of an utterance in a speech situation, there can hardly be any longer a possibility of not seeing that stating is performing an act. 
31The notional Austinian couplet of speech act and speech situation coincides passably well with the Wittgensteinian dyad of “language game” and “form of life.” In both cases, the establishment of a logical dependency of the first term upon the second, of verbal act upon practical environment, seeks to highlight the inextricable interlacing of every utterance with an act: “The expression ‘language game’ must here bring into relief the fact that the speaking of language is part of an activity, or of a form of life.”  The interpenetration of the meaning of utterances and the concrete context of their utterance quite obviously implies the generalization of the category of the performative to all language. The importance of this effective dimension becomes considerable, not only because of its extent—it is now pertinent for all discourse—but also and above all because of the crucial responsibility that is attributed to it. Performativity is not just one of the multiple functional virtualities of language, one of its facultative qualities; it becomes the necessary condition for any statement’s being authentic speech. In order for a phrase to have meaning, it must always be an act at the same time, part of an activity; the very definition of speech is essentially pendant to its performative function. We arrive, in Wittgenstein, at a paroxystic conception of performativity; no wonder, then, that his own—therapeutic—usage of language is of this order.
32In Freud also, the curative power of discourse finds its source in a deep and general performative ground of language, but one that, nevertheless, is envisaged in a very different way than in Wittgenstein’s perspective. The Freudian approach is ontogenetic. It refers to the very first period of psychic life, when the ego knows nothing as yet of the alterity of the outside world—this is the stage of the “oceanic feeling” described at the beginning of Civilization and its Discontents:
Originally the ego includes everything, later it separates off an external world from itself. The ego-feeling we are aware of now is thus only a shrunken vestige of a far more extensive feeling—a feeling which embraced the universe and expressed an inseparable connection of the ego with the external world. 
34The absence of any distinction between the subject and all of being implies that everything that happens in the former immediately affects the rest. The fault line that separates wishes from fulfillment, and where the possibility of frustration dwells, does not yet exist; desire—although one cannot really call it that here, since the gap of expectation, of mediation, and of risk has not begun to bear upon it—is always already satisfaction. It is enough simply to want, to have; enough to say to do. Like Kant’s god, who, simply by conceiving of objects, gives them to himself, the infant bypasses the painful step of testing against facts: the resistance and indifference of the real are unknown things to him, since he is the entire real. In Aristotelian terms, there is no room for potence, for the possible, in between the imperious expression of desire and the plenitude of the act. This archaic performativity is then conserved in the unconscious. Unlike the lucid ego, which learns little by little, through the arduous apprenticeship of frustration, of the alterity of things, and which is thus governed by the reality principle, the id remains dominated by the pleasure principle, and maintains a continuum between intention and realization:
The strangest characteristic of unconscious (repressed) processes, to which no investigator can become accustomed without the exercise of great self-discipline, is due to their entire disregard of reality-testing; they equate reality of thought with external actuality, and wishes with their fulfilment—with the event—just as happens automatically under the dominance of the ancient pleasure principle. Hence also the difficulty of distinguishing unconscious phantasies from memories which have become unconscious. 
36For the unconscious, as for the psyche as a whole in the ancient times of original narcissism, speech is intention and intention is action. Note that this explains the violence of the guilt feelings of certain neurotics: the repressed aggressive drives that are at issue have, in unconscious thought, the same status as actual murderous actions. And the remorse involved is even stronger: whereas legal justice penalizes real actions, taking no account of intentions, the psychic penalizes mere desire and its expression, since they have the status of an act. The magical efficacy of the word is also conserved in modes of thought that are still governed by the archaic feeling of osmosis with everything, such as superstition, religion, or animist beliefs—in which the “omnipotence of thought” is the rule.  It explains the taboo character of certain terms for totemic peoples but also for pious souls and obsessional neurotics, among others: to pronounce the word would be to do the deed. Thus, for the “primitive man” whose psychic life Freud examines in Totem and Taboo, to utter the name of the departed brings them back to life, with all their power to harm; to utter that of the king amounts to bringing about his presence, too strong to be borne. Just as in obsessional neurotics, certain terms that have become (according to the patient’s own word) “impossible”  are carefully neutralized with the help of a more and more complex and exhaustive avoidance strategy; for they in fact bespeak a consciousness of unconscious desires, which in turn stand for the corresponding reprehensible actions. Thus, whereas the analyst uses the performative force of speech to produce liberatory effects by exhuming the contents of repressed thoughts, inversely, all efforts of resistance on the part of neurotics consist in holding back and deflecting this dangerous power of the word, so as to keep the repressed emotions subdued. Neurotics suffer from a sort of reverential fear of the word, a distant inheritance from the omnipotent narcissistic stage where ego and world, thought and deed, word and thing, had not yet entered into the regime of separation.
Efficacity and Truth: On an Epistemological Difficulty of Analysis
37Before concluding these reflections on the performative nature of therapeutic discourse, we would like to examine a certain epistemological consequence that immediately derives from them. Austin’s research on the subtle links that unite the constative dimension and performative power in one and the same utterance may help us to hold at bay one of the classic objections to the scene of psychoanalysis, which challenges the notion that the therapeutic efficacity of the patient’s “admission” can be a criteria of its truth. This critique may equally be addressed to Wittgenstein: How can ordinary language philosophy, on the sole basis of a faith in the positive effects of its remarks, assert that it is within the true when it believes it describes “what we say when?”  Conceiving the discourse of the doctor as a performative use of words renders this genre of attacks vain and inoffensive. Firstly, if liberatory speech is a deed, strictly speaking, then the question of its truth loses all pertinence: it can no longer be evaluated in terms of truth and falsity, but only in terms of success or failure, of “felicity” and “infelicity.” Even when one seeks to judge the therapeutic usage of language according to the norm of truth—which one is perhaps authorized in doing, in so far as this usage is also a description—one must do so from an entirely different perspective than that of its detractors. To bring this to light, we must firstly examine Austin’s observation on the relations between discourse, states of affairs, and truth in the two contrary cases of the performative and constative:
We might say: in ordinary cases, for example, running, it is the fact that he is running which makes the statement that he is running true; or again, that the truth of the constative utterance “he is running” depends on his being running. Whereas in our case it is the happiness of the performative “I apologize” which makes it the fact that I am apologizing; and my success in apologizing depends on the happiness of the performative utterance “I apologize.” 
39In passing from the constative to the performative, the relation between speech and fact is inverted according to a chiasmic structure.
40In the case of constatives, the utterance is true if the corresponding fact is effectively the case—the fact has logical priority over the utterance.
41In the case of performatives, on the contrary, if the utterance is “felicitous,” then the corresponding fact is the case: the fact that I excuse myself depends, as a consequence, upon the felicity of the utterance, “Excuse me.” The utterance has logical priority over the fact.
42In other words, in performative utterances it is speech that constitutes the existence of the fact, and not the fact that founds the truth of speech. This reversal is complicated further once we leave behind ideal cases of pure description and absolute performance for real discourses, where, as we have shown earlier in this study, the constative and performative dimensions are intimately interlaced with one another. Every effective utterance also delivers a minimal informational content: for example, when I say “I promise,” the emission of the phrase institutes a new fact (my act of promising), but simultaneously describes the latter. Now, qua constative utterance, “I promise” takes its truth from the existence of a fact—the fact that I am, effectively, promising—which itself depends upon the success of this same utterance, this time qua performative. We obtain a circular structure: performative speech, in producing a fact, at the same time founds the truth of its own constative dimension. Now, again, the truth of the utterance, qua constative, rests upon the capacity of that same utterance, qua performative, to bring about the fact that makes it true. Thus it is the power of the phrase qua performance that guarantees, retroactively, its own truth qua statement.
43This study of the complex links that unite the effectiveness of the performance and the truth of the statement in the same speech act should allow us to neutralize a classic strategy used to throw discredit on the method of psychoanalysis. The following epistemological objection is often made against Freud: his conviction that rediscovered memories or disinterred unconscious thoughts do indeed correspond to the truth—to the historical event, or to that which in fact happened in the unconscious—rests on the fact that the expression of this hidden memory or repressed emotion has a therapeutic effect, the removal of symptoms. The criterion for the truth of the assertion that “this is what happened” or that “that is indeed what I desired unconsciously” thus seems to be the effectiveness of this affirmation in terms of the amelioration of the patient’s condition. This is often denounced as a vice in psychoanalytical reasoning that strips it of all credibility: the effects that the admission produce are independent of its truth value, and cannot guarantee it; to see in the former the proof of the latter bespeaks a serious category mistake. One can very well, for example, bring about positive and reassuring effects on a distressed child by telling him fables. This old critique can nevertheless be weakened, and even considered to be the result of a misinterpretation, once we regard the talking cure as a performative use of speech. To speak the hidden memory or repressed thought is above all an act: rather than a statement about the already there, the expression of what was unconscious is a creative act, an assumption of its own history through which it freely endows it with sense, rather than merely reports it. If this speech act succeeds, if it has real therapeutic effects, then we shall say, not that this is the index or the proof that the description was true; instead we shall say that this act brought about the truth of the statement. For, as Austin shows in his study of the functioning of the performative utterance, it is the latter’s success that founds, retroactively, the truth of the constative utterance that doubles it. The error underlying the classic epistemological objection consists in considering analytic remembrance as an admission; it is in reality also an act, or, in Freud’s terms, a construction.  Jacques Derrida renders this nuance most appositely in the following terms:
We are wrong, Freud tells us, to speak of translation or transcription in describing the transition of unconscious thoughts through the preconscious toward consciousness. Here again the metaphorical concept of translation (Übersetzung) or transcription (Umschrift) is dangerous, not because it refers to writing, but because it presupposes a text which would be already there, immobile: the serene presence of a statue, of a written stone or archive whose signified content might be harmlessly transported into the milieu of a different language. 
45If the theoretical model or working hypothesis that doctor and patient patiently elaborate brings about an authentic liberation, then it will give rise to a new “fact,” and consequently the description that was also contained in the creative speech act will be able to be retroactively qualified as “true.” The therapeutic effectiveness of the said lies less in the manifestation of the existence of a prior fact than in the bringing about of this fact. “The postscript which constitutes the past present as such is not satisfied, as Plato, Hegel and Proust perhaps thought, with reawakening or revealing the present past in its truth. It produces the present past.”  The Freudian method thus supposes a profound renewal of the concept of truth. In psychoanalysis, the truth of speech is no longer, as has so long been the case in Western thought, the permanent, static adequation to a preexistent thing of its description. It becomes a power of deliverance, bringing about new facts, where speech comes to be its own referent, granting it its proper place. Freud brings into being a new sense of truth, that we should like to qualify as dynamic.
46It would be tempting, in finishing, to extend to Wittgenstein these reflections on the interdependence of effectiveness and the truth in curative discourse. Can one not found the ambitious claim to faithfully grasp “what we say when” in the therapeutic, clarifying effect of the utterance? Is not speech’s capacity to bring about consequences, as in Freud, the retrospective guarantee, or rather the cause, of their veridicality? There is a passage in the Philosophische Untersuchungen, quite obviously of Freudian inspiration, that would incline us to think so:
The criteria for the truth of a confession that I thought such-and-such are not the criteria for a true description of a process. And the importance of a true confession does not reside in its being a correct and certain report of a process. It resides rather in the special circumstances whose truth is guaranteed by the special criteria of truthfulness.
(Assuming that dreams can yield important information about the dreamer, what yielded the information would be truthful accounts of the dreams. The question whether the dreamer’s memory deceives him when he reports the dream after waking cannot arise, unless indeed we introduce a completely new criterion for the report’s “agreeing” with the dream.) 
48Here Wittgenstein seems to promote the capacity of speech to bring about an effect on the life of the speaker, and to judge as secondary or as unimportant, the fidelity of what he says in relation to any preexisting fact. It would nevertheless be too hasty to entirely assimilate his thinking to Freud’s on this point. The effectiveness of Wittgensteinian discourse is certainly not that of a construction, but rather than of an admission: the tension between the confession of a new awareness and the elaboration of a new reality, between seeing and doing, is not as strong in Wittgenstein as in Freud. His therapeutic technique tends instead toward the destruction of our linguistic “house of cards” rather than toward the erection of new representations, which can all too easily become mythologies. Ultimately the ambivalence at work in the oeuvre of both authors, between the neutrality of the lucid gaze and the power of action, turns out to be entirely real in Freud, but only apparent in Wittgenstein, whose performance consists not in creating a content with a new meaning, but in modifying the point of view on that which we already have at our disposal. The performative use of language turns out very differently in the two cases: if Freudian speech does something, that of Wittgenstein seeks to undo.
S. Cavell, “Passionate and Performative Utterance,” in Contending with Stanley Cavell, ed. R. B. Goodman (Oxford and New York: Oxford University Press, 2006), 195 [“La passion,” trans. P.-E. Dauzat, in Quelle philosophie pour le XXIe siècle? L’Organon du nouveau siècle (Paris: Gallimard, 2001), 379].
Cavell, “Performative Utterance,” 195–196 [“La passion,” 377, 380].
Cavell, “Performative Utterance,” 189.
L. Wittgenstein, Philosophische Untersuchungen / Philosophical Investigations (Oxford: Blackwell, 1953), §109. Bilingual edition: German text edited by R. Rhees and G. E. M. Anscombe.
J. L. Austin, How to do Things with Words, (Oxford: Oxford University Press, 1962), 5 [Quand dire, c’est faire (Paris: Seuil, 1991), 40].
Austin, How to do Things, 12 [Quand dire, 47].
S. Freud and J. Breuer, Studien über Hysterie, (Vienna, 1895); Studies on Hysteria, trans. J. Strachey (New York: Basic Books, 2000), 6 [Études sur l’hystérie, trans. A. Berman (Paris: PUF, 1956), 4]. Emphasis modified—Freud italicizes the whole phrase.
Freud and Breuer, Studies on Hysteria, 34 [Études sur l’hystérie, 25].
Freud and Breuer, Studies on Hysteria, 30 [Études sur l’hystérie, 21].
L. Wittgenstein, Tractatus Logico-Philosophicus, published in German in the journal Annalen der Naturphilosophie (Leipzig, 1921); trans. D. F. Pears and B. F. McGuinness (London and New York: Routledge Classics, 2001) [trans. Gilles-Gaston Granger (Paris: Gallimard, 1993)], §6.521.
Wittgenstein, Philosophische Untersuchungen, §109.
Wittgenstein, Philosophische Untersuchungen, §133.
L. Wittgenstein, Vermischte Bemerkungen, ed. G. H. von Wright and H. Nyman (Oxford: Blackwell, 1977); Culture and Value, trans. P. Welch (Chicago: University of Chicago Press, 1984) [Remarques mêlées, trans. Gérard Granel (Mauvezin: TER, 1984), 63]. We find the same formula again in the Philosophische Untersuchungen, §546.
Reported by M. O’C. Drury, “Conversations with Wittgenstein,” in Recollections of Wittgenstein, ed. R. Rhees (Oxford: Oxford University Press, 1984), 110.
Reported in Drury, “Conversations with Wittgenstein,” 117.
Wittgenstein, Tractatus, §4.112.
A statement reported in C. Diamond, “Rules: Looking in the Right Place,” in Wittgenstein: Attention to Particulars: Essays in Honour of Rush Rhees, eds. D. Philips and P. Winch (Basingstoke: MacMillan, 1989), 13.
Wittgenstein, Philosophische Untersuchungen, §124.
Austin, How to do Things, 121 [Quand dire, 129].
Austin, How to do Things, 94 [Quand dire, 109].
Cavell, “Passionate Utterance,” 187 [“La passion,” 361].
Austin, How to do Things, 121 [Quand dire, 129].
Wittgenstein, Philosophische Untersuchungen, §133.
Wittgenstein, Culture and Value, 60 [Remarques mêlées ,43].
Cavell, “Passionate Utterance,” 196 [“La passion,” 380].
Cavell, “Passionate Utterance,” 196 [“La passion,” 380].
See the definition given by Aristotle in his Poetics: “a representation . . . performed by actors . . . effecting, through pity and fear, the purification of such emotions.” Aristotle, Poetics, trans. A. Kenny (Oxford: Oxford World Classics, 2013), 4.49.b28 [Aristotle, La Poétique, trans. R. Dupont-Roc and J. Lallot (Paris: Seuil, 1980), 53].
Freud and Breuer, Studies on Hysteria, [Études sur l’hystérie, 12].
See the final words of the opening quote from Augustine’s Confessions: “Ita verba in variis sententiis locis suis posita, et crebro audita, quarum rerum signa essent, paulatim colligebam, measque jam voluntates, edomito in eis signis ore, per haec enuntiabam.” Augustine, Confessions, 1.8; Wittgenstein, Philosophische Untersuchungen, §1; emphasis ours. Cavell remarks that Wittgenstein, in saying nothing with regard to this affirmation—that words serve to express my wishes—agrees with it (“Notes and Afterthoughts on the Opening of Wittgenstein’s Investigations,” in Philosophical Passages: Wittgenstein, Emerson, Austin, Derrida (Oxford: Blackwell, 1995); see especially 162).
Wittgenstein, Philosophische Untersuchungen, §546.
Wittgenstein, Philosophische Untersuchungen, §464.
S. Cavell, “The Availability of Wittgenstein’s Later Philosophy,” in Must We Mean What We Say? (Cambridge: Cambridge University Press, 1976), 71.
Cavell, “Passionate Utterance,” 193 [“La passion,” 371].
Cavell, “Passionate Utterance,” 191 [“La passion,” 370].
It is very important, in this regard, that Wittgenstein chose to place at the beginning of the Philosophical Investigations the words of someone else. The opening quotation from Saint Augustine places the book, from the start, in the order of dialogue. It even does so doubly, since Wittgenstein’s interlocutor describes his own birth in speech as an exchange with others: the opening formula “Cum ipsi (majores homines) . . .” immediately installs the child, and the reader, in the group, outside of which there is no language, no thought, and, for us, no life.
We mean the misunderstanding of the Vienna Circle, but equally the very different misunderstanding that drives the “cult of personality” around Wittgenstein today. To be treated by some as an author of a successful doctrine would no doubt have been cause for consternation, for someone who wished to teach a method applicable for everyone autonomously, rather than to transmit his personal “opinions.”
See in particular S. Cavell, Conditions Handsome and Unhandsome: The Constitution of Emersonian Perfectionism (Chicago: University of Chicago Press , 1990) [Conditions nobles et ignobles, trans. S. Laugier and C. Fournier (Combas: Éclat, 1993)]: The second paper, dedicated to the “Wittgensteinian paradox” regarding the possibility of following a rule, thinks skepticism as the necessary and indissociable companion of confidence.
Cavell, “Passionate Utterance,” 194 [“La passion,” 376–377].
See Cavell’s analyses in the title essay of Must We Mean What We Say?
Wittgenstein, Culture and Value, 46 ; Wittgenstein cites Goethe, from Faust I.
Austin, How to do Things, 138 [Quand dire, 143].
Wittgenstein, Philosophische Untersuchungen, §23.
S. Freud, Das Unbehagen in der Kultur (Vienna, 1929); Civilization and its Discontents, trans. J. Strachey (London: Norton, 2010), 15 [Malaise dans la civilization, tr. C. and J. Odier, Paris: PUF, 1971, 1983, 10].
S. Freud, “Formulations Regarding the Two Principles in Mental Functioning,” trans. J. Rivière, in Collected Papers, ed. E. Jones (New York: Basic Books, 1959), vol. 4, 13–21: 18 [“Formulations sur les deux principes du cours des événements psychiques” (1911), trans. J. Laplanche, in Résultats, idées, problèmes I, (Paris: PUF, 1984), 142]. Emphasis ours.
See S. Freud, Totem und Tabu (1912–13), G.W., vol. IX; Totem and Taboo, (New York: Empire, 2012) [Totem et Tabou, trans. S. Jankélévitch (Paris: Payot, 1984)]. See chapter 3: “Animism, Magic and the Omnipotence of Thought.”
Freud, Totem and Taboo, 26 [Totem et Tabou, 39].
This problem has been examined in S. Cavell, “Aesthetic Problems of Modern Philosophy,” in Must We Mean What We Say?; see especially pages 86–96.
Austin, How to do Things, 47 [Quand dire, 75].
See “Konstruktionen in der Analyse,” Internationale Zeitschrift für Psychoanalyse, 23(4), 1937; “Constructions in Analysis” in Standard Edition of the Complete Psychological Works of Sigmund Freud, ed. J. Strachey (London: Hogarth Press, 1964), 23:255–269 [“Constructions dans l’analyse,” trans. E. R. Hawelka, U. Huber, J. Laplanche, in Résultats, idées, problèmes II (Paris: PUF, 1985)].
J. Derrida, “Freud and the Scene of Writing,” trans. A. Bass in Writing and Difference (London and New York: Routledge, 1978), 211 [“Freud et la scène de l’écriture” in L’Écriture et la difference (Paris: Seuil, 1967), 312–13].
Derrida, “Freud,” 214 .
Wittgenstein, Philosophische Untersuchungen, vol. 2, §11:222–3. Emphasis on “circumstances” ours.