This challenge to the descriptive fallacy is even more significant for Austin, for he is not content to simply divide statements into constatives (which would be liable to empirical validation) and performatives (which could be successful or unsuccessful—a distinction that would, by itself, leave intact the Vienna Circle’s logical positivism); rather, he understands all statements as containing a performative dimension. More precisely, Austin’s central claim is that every manifestation of thought in speech is definitely a form of action. The very organization of the argument in How to Do Things with Words manifests this strict synonymity. Indeed, before “Lecture VIII,” Austin divided statements into constative and performative, as if it were a matter of two distinct classes that are not unified in a single statement. But this bifurcation12 gave rise to so many aporias that he was led to ask whether, rather than a division between performatives and constatives, each statement could be simultaneously a locutionary, illocutionary, and perlocutionary act. A locutionary act says something; an illocutionary act does something by saying it (performatives of the sort “I promise you that . . .” or “I swear that . . .”); finally, a perlocutionary act aims to cause some effect in another (such as “to threaten,” “to intimidate,” “to persuade,” etc.). With this distinction, Austin constructed a general theory of action that was also, at the same time and in the same way, a general theory of speaking in its multiple aspects. From the eighth lecture forward, Austin—who was so suspicious of philosophers’ generalizations—effected a generalization of speaking as an act, even talking of a “more general theory”13 that would establish a typology, for example, of “explicit” performatives (verdictives, exercitives, commissives, behabitives, and expositives). The very notion of truth is thus profoundly altered with respect to logical positivism.
But how, precisely, should Austin’s new approach to the notion of truth be understood? In fact, two conceptions could come out of it. First, one could insist that the notion of a performative as the realization of an act is not related to the notion of truth. Performatives, as Austin said, are successful or unsuccessful (“felicity” or “infelicity”14), not true or false. Consequently, if we insist that the dimension of success or failure is tied to limited circumstances, it is possible to say that Austin rejects any notion of truth by linking it to a given context in which the empirical, singular individual is plunged more or less by chance. With the notion of a “successful” or “failed” act, and by insisting on the context’s (by definition) contingency,15 truth is boiled down to a question of what works or doesn’t, as in classical utilitarianism. In this sense, the theory of “speech acts”16 rejects any notion of truth and any idea of intrinsic rules governing the elaboration of universal propositions, quite simply because the act succeeds or fails at the whim of external circumstances. Thus interpreted, Austin can give rise to a radical skepticism, which is later exemplified by Barry Stroud.17 Alongside this first conception of truth that Austin may have held, we can propose a second, quite different, interpretation: even though he defined an utterance in terms of its success or failure, Austin would nevertheless not have renounced all definitions of truth, nor would he have embraced utilitarianism. Indeed, Austin could also mean (and the eleventh lecture of How to Do Things with Words leads us to think so) that the truth is defined as what is appropriate. Thus understood, Austin’s proposed theory of truth is quite different from the Vienna Circle’s representationalism, and his works give rise to a new philosophy. This is, by the way, what he himself seems to say when he writes, “Now we said that there was one further thing obviously requiring to be done, which is a matter of prolonged fieldwork. We said long ago that we needed a list of ‘explicit performative verbs’; but in light of the more general theory we now see that what we need is a list of illocutionary forces of an utterance.”18 His reflection is not only therapeutic, negative, and critical but also pioneering, creative, and productive of new tasks for thought. But this positivity could also be understood in two senses: a return to a certain form of positivism or an entirely new version of classical skepticism. These two paradoxical sides of Austin’s legacy have been taken up by two very different thinkers, John Searle and Stanley Cavell. Let’s examine these two diametrically opposed legacies.
Searle as a “Positivist” Version of Austin
Starting explicitly from Austin’s claim that an expression’s meaning cannot be reduced to its simple propositional content (to what is said) but also encompasses the saying, Searle attempts to inventory the inevitable implications of speaking (“pragmatic implications”). Every successful use of a discursive act supposes that a certain number of conditions have been satisfied; hence, the task of a philosophy of language ought to be to discover them and lay them out in some sort of typology.19 There are a variety of conditions for a discursive act—for example, the “sincerity condition” inherent, for Searle, in an utterance’s illocutionary force. This implicit condition means quite simply that if I assert that p then eo ipso I express the belief that p, if I promise that x then eo ipso I express the intention to do x, etc. To assert, declare, warn, promise, etc., has the sincerity condition that the speaker believes that p; to advise has the sincerity condition that “S believes A will benefit H,”20 etc. It follows that there is a necessary connection between my speech and the conditions that are implied for my speech to be successful. Succinctly put, the speaker cannot, without contradiction or paradox, perform a discursive act and deny its implications. This is why the relation between the speech act and its conditions is one of necessity. And these implications (like sincerity) are always posited as pragmatic constraints. These implications cannot be denied without a “pragmatic contradiction.” So the sincerity conditions are thus said to be inherent in the speech act, since they are immanent in the act of “making” a promise, issuing a warning, and even of making an assertion.
This necessary implication is what makes it possible to formalize the theory of speech acts, which Searle sets out to do with Daniel Vanderveken in Foundations of Illocutionary Logic.21 They break down illocutionary force into its basic components capable of formal combinations. In so doing, they offer a certain number of universal laws that necessarily govern the successful accomplishment of a speech act. Every illocutionary act has the form F(P), where F is the illocutionary force of the utterance and P the propositional content of the statement. Having made this first formalization, they combine the necessary components of illocutionary force (F) in the following formula: F = <πμκθεψη>, where π is the illocutionary point, μ is the degree of strength of the illocutionary point, κ is the mode of achievement of the illocutionary point, θ is the propositional content condition, ε is the preparatory conditions, ψ is the sincerity condition, and η is the degree of intensity of this sincerity condition. To show the necessary implications of a speech act and to reduce it to a number of primitive components that could be combined and formalized—this is indeed the goal of Foundations.22
Searle is interested here in the universal rules that govern the structure of the act. His project thus consists of systematizing Austin’s intuition by attempting to construct a vast typology of the set of implications that constitute speech acts like promising, threatening, pleading, etc. But to inventory all such speech acts and to find the universal rules that govern their successful function, to formalize them and to deduce their intrinsic and formal components—these are all efforts that manifestly revive the motivating demand of the first analytic philosophers. As it happens, this is what the formula F(P) indicates, as it articulates less a critique of the Vienna Circle’s claims as the temptation to enlarge its field of investigation. Austin’s project is indeed brought back alongside a certain classically scientistic demand—to make speech acts into a formalized science, following the model of logic.
But, one might object, isn’t this concern for a system of acts, and for an inventory of its implications understood as universal rules, precisely the opposite of the way of thinking that Austin encour
aged? Isn’t Searle, with this systematization, rather distanced from Austin? This is Michael Soubbotnik’s view, echoing other authors like Sandra Laugier or Stanley Cavell:
Searle is not content to effect a technical (even considerable) reconstruction which would transform a brilliant “discovery” into a field of knowledge that was, if not mature, at least adolescent. I wouldn’t even hesitate to say that despite the use of concepts like “performative,” “illocutionary act,” “force,” etc., and certain seeming points of convergence on the question of truth, Searle and Austin quite simply do not speak the same philosophical language.23
It is not within the scope of my project here to take a position on the nature of the inheritance passed on by Austin’s oeuvre, whose undeniable laconic diversity allows multiple possibilities. My concern is simply to show that numerous suggestions in Austin’s oeuvre foreshadow the “possibility” of Searle’s philosophy. To be convinced of this we need only remember that Austin spoke of his hope for a vast “science of language.” Searle does not betray Austin but rather exhibits one potentiality. This potentiality consists, at least in Foundations and Speech Acts, of reviving a certain form of philosophical practice proper to the Vienna Circle, in which it is a matter of constructing vast typologies that indicate the correct use of language. Briefly, once it is acknowledged that Austin’s great innovation was to have shown that statements are not merely concerned with facts but also with actions, then it is a matter for Searle of showing that these different forms of action (to promise, assert, threaten, warn, etc.) are capable of being formalized in the classic form F(P). Foundations of Illocutionary Logic thus constitutes the positivist interpretation of Austin’s project—a reading that, I have insisted, is a legitimate interpretation of an oeuvre that speaks quite clearly of his hope for the dissolution of philosophy within a discipline “sure of its object and of its method.”
Cavell as a “Skeptical” Version of Austin
Opposed to this reading of Austin is Stanley Cavell’s interpretation, which clearly presents itself as skeptical in comparison with the systematizations of the early Searle. For Cavell, Austin is the philosopher of the ordinary, and not the philosopher of “speech acts.” Austin wanted to precisely analyze the multiple—that is, infinite—occurrences of the terms we use, not construct a system of universal rules governing all language. We ought to pay attention, therefore, to our use of language. This way of reading Austin, which Cavell embodies, consists in developing a moderated skepticism, which is inscribed, as I read him, in the system of reflective judgments of Kant’s Critique of Judgment, and thus gives his own statements only the status of plausibility. It is not a matter of saying, with Hegel, that “the I is a we,” but rather of stating that “everything happens as if” and we “act as if” the “I” could be a “we,” “everything happens as if” our statements could be, if not universal, at least virtually shared by others. This reading of Cavell’s project, starting from the “as if” or reflective judgment of the Critique of Judgment, seems to me to be corroborated first of all by the constant references to the author of the “allgemeine Stimme,”24 who, in Cavell’s texts,25 is offered as the answer to the question of language’s claim to speak the truth, or even to an individual’s claim to speak in the name of everyone. As it is not possible to say that this claim is entirely empty and illusory without asking about this proposition’s own status—isn’t it, too, empty and illusory? Barry Stroud’s radical skepticism falls into this sort of abyss—as it is not a question, for all that, of renewing the foundational demand that so-called classical philosophy expresses, Cavell’s chosen plan is intermediary. Our judgments (statements) are no longer constitutive but regulative (“everything happens as if”), a crystallization of the finitude of human speech. Moreover, this interpretation of Cavell’s project as an extension of Kant’s argument in the Critique of Judgment seems to me all the more plausible since Cavell repeats the Kantian argument against metaphysics, namely, that of “radical finitude.” In his eyes, classical metaphysics cannot accept the absence of a “guarantee” (a foundation) because it denies finitude and “misfortune”—always interpreted as the “failure inherent in the act of speaking.” His return to ordinary language, to what he banally, that is, trivially, tells us of the world, constitutes a recognition of finitude as radical and inescapable. This explains, in my view, Cavell’s recourse to autobiography and to the individual “I.”26 This recourse allows Cavell to avoid sinking into a totally self-refuting argumentation (“I say that nothing is universal or valid, and at the same time and in the same way, I claim my own proposition is universal and valid”). With this use of autobiography transformed into a form of philosophical writing, it is a matter of asserting the fragility, precariousness, and fallibility of subjective speech while also “calling for” a “we” and infinitely hoping for it as a possible validation.
Having justified this reading of Cavell framed in terms of the Critique of Judgment, I must now show that even if, as I have said, Cavell does not immediately sink into the aporias of Barry Stroud’s radical skepticism, his measured skepticism or “critique” is nevertheless not exempt from a certain number of pitfalls that closely resemble the logical pathology of self-refutation. This tendency to pragmatic paradoxes can be found at several levels.
First of all, to deny every generalization, as Cavell does, taking up one of Austin’s claims, is to risk falling into an infinity of speech situations. Indeed, to attempt to define a statement’s pertinence only as measured by its context is to risk confronting the irreducible incommensurability of contexts. Briefly, the new “pitch of philosophy” would be less the proposal of a new writing, “skeptically consistent” because it draws its source from the “everything happens as if” of the Critique of Judgment, but rather a polygraph test of situations so infinite, singular, and atomized that no generalization, discussion, or argument could corroborate it. In sum, the study of usage, in Austin, is part of an inventory of idiomatic expressions; and autobiography, in Cavell, would be the most refined example of what Wittgenstein called private language.
Moreover, if the ordinary is the only judge of our claims (as Cavell claims, following Austin), how are we to explain the very possibility of analysis of ordinary language? Indeed, from the perspective of the ordinary, the status of the “philosophy” or the “analysis” of the ordinary remains problematic. How are we to understand the possibility of Austin’s reflection on different usages, a reflection that is not part of ordinary language but of its analysis? What is the truth of the philosophy of ordinary language? The question here is indeed to determine whether one can define the truth (as being immanent in ordinary language) without asking, on the other hand, a question about the truth of what one says about ordinary language. If we do not wish for ordinary language philosophy to fall into self-contradiction (which would consist in claiming one’s analysis of ordinary language to be true while rejecting that ordinary language is true), then we are forced to ask, as do both Austin and Cavell, the question of the pragmatic consistency of their practice of ordinary language analysis. It follows that, even though Austin and Cavell refuse the term “philosopher,” they cannot avoid, insofar as they are theoreticians of ordinary language, an interrogation of their own analyses’ status. Shouldn’t we therefore examine this use—that is, the philosophical use of language—rather than contradictorily denying its relevance? The categories of “performatives” such as “verdictives, exercitives, etc.” are not ordinary uses of language and are concepts just as obscure, for the neophyte reader, as the Hegelian categories of “Aufhebung,” “Wirklichkeit,” and “für sich” and “für uns.” How are we to classify them, as Austin does, while maintaining the irrelevance or the noxiousness of nonordinary language? Doesn’t ordinary language at some point have to reflect on itself, go back over itself, understand itself as speech or language? One could retort that it is precisely because there have been deviant uses over the course of the centuries that we must go back to th
e ordinary. But how then are we to explain that ordinary language could have produced these monsters of metaphysical usage if it was initially the very locus of truth? After all, isn’t this return to the ordinary, as Austin’s very practice attests, effected by the implementation of a nonordinary language? And doesn’t this practice raise, ultimately, the question of philosophical language, of its uses and its own rules? The analysis of ordinary language or “linguistic phenomenology,” in its claim to say something, must also reflect itself as analysis, otherwise it falls into self-refutation in the same way that radical skepticism does.
Finally, the Cavellian conception of a truth placed under the sign of the plausibility of an “everything happens as if” engenders, again, the question of the pragmatic consistency of such an affirmation. Indeed, to say that finitude is radical and that this is why we can claim the plausibility of “everything happens as if” amounts to claiming the truth of what one says. What indeed must we presuppose to say that the truth cannot be attained because finitude is truly too radical if not the unavoidable truth of this very proposition? This pragmatic paradox becomes established in what I have shown to be Cavell’s implicit model, namely, the Critique of Judgment’s model of reflective judgments; indeed, as I had the occasion to note in my studies of Kant and the post-Kantians,27 a paradox that the post-Kantian skeptics (Aenesidemus,28 Salomon Maimon) were superbly able to reveal is hidden at the heart of the Critique: the thematization of what is said in the Critique of Judgment (for example, “the living world comes under reflective judgments”) presupposes a point of view in which this thesis is assumed to be valid. The division between determinant and reflective judgments must be given by the philosopher writing the Critique as a clearly nonreflective judgment; in other words, the point of view from which the Critique of Judgment is written must not come under “everything happens as if” without leading us to a series of absurd statements, such as “everything happens as if the Critique were true, everything happens as if the distinction between reflective and determinant judgments were relevant,” or even, facetiously, “everything happens as if humanity had ended.” To put it in a formula, in such a configuration, humanity has ended to the extent that there is no legitimate point of view from which the proposition “humanity has ended” can be uttered. These three pitfalls of Cavellian skepticism show how much his discourse, like many others, is ultimately marked by the logical pathology of self-refutation.
The Death of Philosophy Page 8
