8. Language, Norms, and the Force of Reason
The last several chapters have constituted a detailed examination of the concepts and values of “language,” “meaning,” “practice,” and “use,” “rule,” “regularity” and “institution” in the dialectic of analytic philosophy in the twentieth century. At each stage, I have examined the relationship of these concepts with the notion of a language as a total logical, grammatical, or practical structure, and with the ambiguities inherent in an appeal to language that constantly tends to figure it as a structure of signs, while subsequently finding just this structure to be inadequate to account for its own institution, extent, limits, or ultimate guiding principles. In the repeatedly enacted dialectic that I have explored, the attempt to describe or theorize the logical form or structure of language in terms of acorpus of analytic rules, principles, or norms has, I have argued, repeatedly been contested by those moments of presence, genesis or institution that resist being included in the structural system of language as simplyanother element or another moment (see chapter 1). The dialectic has repeated itself consistently, unfolding each time out of the inherent dynamic of the analytic tradition’s founding and originally determinative recourse to language. Language, with almost every resort that the analytic tradition has made of it, then appears ambiguously as an objectively present structure or system, accessible in principle to the schematic resources of a theoretical description of its structure or form; and then again, in its moments of founding principles, limits, or ultimate nature, as something radically transcendent to, mysterious or problematic for any such accounting.
There are few themes more pervasive in the discourse of analytic philosophy of language today than the invocation ofordinary lived practices as the ultimate source of linguistic meaning and intersubjective intelligibility.[^379] The appeal to practices figures, in the recent literature, most centrally in projects that attempt to explain the meaningfulness of language as grounded in essentiallypublic andsocial practices of communication, deliberation, evaluation and criticism. In many of its versions, it seeks as well to account for the “normative ” dimension of language - in other words, for distinctions between correctness and incorrectness in linguistic usage - by reference to the existence or regularity of socially learned and inculcated standards, rules, or norms. But as I shall argue, this appeal to practices, in most of its formulations, is simply another version of the characteristic and repeated attempt to comprehend language as a total structure, and the force of reason as the force of its rules in application to a human life. In this final historically focused chapter, I shall consider three recent linguistically oriented projects that consider the longstanding question of the force of reason in relation to the forms of our access to the language we speak. Despite superficial similarities, these projects diverge widely, I shall argue, in the ways they construe the force of the better reason as operating to determine thought and action; these divergences mark some of the different contemporary possibilities for taking up the analytic tradition’s ongoing critique of linguistic reason, or continuing it in the space of a broader history of critical thought.
In his recent textMaking it Explicit , Robert Brandom outlines a complex, far-ranging and innovative project of semantic and pragmatic analysis. One of his overriding aims is to make the practical foundation of reason and our practices of reasoning intelligible in a new way by showing how the norms that he sees as governing them can besocially instituted and maintained. One of the most urgent aims of the“ normative pragmatics” that Brandom develops is to provide an alternative to the “representationalist” view that construes propositional contents as fixed and determined in themselves, independent of their characteristic roles in inferential and communicative practices. Drawing on readings of Kant, Frege, and Wittgenstein, Brandom argues that the norms of reasoning and the contents of concepts are in fact wholly determined by these practices.[^380] Thus, instead of seeing conceptual norms, in the first instance, as rules, laws, or commandments representedexplicitly in our description of them, we ought to see them as typicallyimplicit in our actual social practices of making and attributing judgments and our practical attitudes of treating the judgments that others make as legitimate or illegitimate.[^381]
Following the suggestion of some of Frege’s polemics against a psychologistic treatment of logic, Brandom distinguishes sharply between the merely causal consequences of linguistic performances and the distinctivenormative significance that these exercises take on when they are understood as involvingreasons and aiming at thetruth . For Brandom, the first sort of significance is describable from a naturalistic perspective, whereas the second sort is not. In particular, the normativity of reasoning comes into view whenever performances are legitimately assessable ascorrect orincorrect .[^382] Norms of reasoning do not, like natural laws, specify whatwill happen, but rather what ought to happen: which inferences, for example, it iscorrect to draw from some set of premises or assumptions. This liability to assessments of correctness does not, Brandom argues, adhere to events described purely naturalistically, where what is at issue can be, at best, the regularity or normalcy of a performance, but there is no legitimate application of the concept of correctness. It is, moreover, distinctive of the peculiar “force” of normative rules in reasoning that we are bound, not by these rules directly, but by ourconceptions of them. Normative rules, as opposed to causal ones, have force in determining how we ought to reason only for beings capable of conceiving of them as having this force, andas so conceived. Brandom argues that this demarcates the realm of normativity from the realm of facts and phenomena accessible to explanation in purely naturalistic terms, thereby marking us as the particular kinds of beings we are, responsive not only to natural, but also to rational, force. It is only because our acting on normative rules is dependent upon our recognition or conception of them, our accepting or grasping them, implicitly or explicitly, that we are “denizens of the realm of freedom,” rational agents, at all.[^383]
This description of the basis of normativity has its roots in Kant, and has more recently played a central role in a variety of analytic projects that have discussed our “responsiveness to reasons” or the possibility of characterizing our social and linguistic “reasoning practices” as involving commitments to “norms” in a fundamental way.[^384] For these projects, normative entities such as standards or rules are to be distinguished from non-normative ones in that their force in determining what we do depends on ourrecognition of them as such. By contrast with natural laws or regularities, they are not binding “in themselves,” but only insofar as we can recognize them as binding, or (equivalently) recognizeourselves as bound by them. Their force is not, then, that of the natural laws that compel the movements of bodies, but a categorically distinct kind of rational force that depends on our recognition of it as binding, a recognition that, we may further suppose, is experienced and negotiated primarily in linguistic and social practices of justification, explanation, and evaluation. Its paradigm is the “game of giving and asking for reasons” in which we offer, accept or reject not only particular claims to truth, but also more general criteria for their evaluation and criticism of specific linguistic performances, and so gain clarity about (what we will then take to be) the standards or claims of reason in relation to our world-directed attitudes.
Although Brandom, following Kant, thinks of reasons as (at least potentially) having the form ofrules governing possibilities of correct or incorrect linguistic performance, it is also one of his overriding goals to argue against an unrestricted “regulism” according to which what makes a performance correct or incorrect is always its relation to anexplicit rule or principle.[^385] Though performances can sometimes be evaluated by reference to explicit rules that they violate or comport with, the more usual case is that they are simply treated as correct or incorrectin practice , without any explicit reference to rules or principles. Ordinary attitudes of treating or taking a performance as correct or incorrect, shown in actual behaviors of praise, censure, approval or disapproval are sufficient, according to Brandom, to establish the normative status of a particular performance in a social context. This kind of normativity is always social, since it depends on the interaction of the performer of an action and those who are in a position to evaluate it, but the evaluation need not wait on the formulation of any explicit principle or rule that underlies it. In this sense, norms are typicallyimplicit in practice before they become explicit in a stated principle. The possibility of identifying such implicit norms or “proprieties” of practice is in fact essential to Brandom’s case for the essentiallysocial character of normativity. Such proprieties are “normative statuses - the status a performance has as correct or incorrect according to a rule or practice.” (p. 628).
Brandom’s argument against regulism, and his defense of the implicit/explicit distinction, relies heavily on what he takes to be the main point of the rule-following considerations of Wittgenstein’sPhilosophical Investigations . For Wittgenstein as Brandom reads him, the notorious paradox ofPI 201 poses a general, and insurmountable, problem for the view that all proprieties of practice, all evaluations of practices as correct or incorrect, are dependent onexplicitly represented rules. For the application of an explicit rule, in a particular case, is itself something that is amenable to evaluation as correct or incorrect. As Brandom puts it, “applying a rule in particular circumstances is itself essentially something that can be done correctly or incorrectly.”[^386] But, then,if all proprieties depended on explicitly represented rules, it would be necessary in each case to determine the application by resorting to another explicitly represented rule, leading to a bottomless infinite regress. The application of each rule, in each case, would depend on the specification of afurther rule; since rules cannot interpret themselves, the process of interpretation would be endless. The solution to the paradox, according to Brandom, lies in recognizing the fact that at bottom, the determination of the correctness or incorrectness of a performance is irreduciblypractical :
The question of the autonomy of the intellectualist conception of norms, presupposed by the claim that rules are the form of the normative, is the question of whether the normative can be understood as ‘rules all the way down,’ or whether rulish properties depend on some more primitive sort of practical propriety. Wittgenstein argues that the latter is the case. Rules do not apply themselves; they determine correctnesses of performance only in the context of practices of distinguishing correct from incorrect applications of the rules. To construe these practical proprieties of application as themselves rule-governed is to embark on a regress. Sooner or later the theorist will have to acknowledge the existence of practical distinctions between what is appropriate and what not, admitting appropriatenesses according to practice as well as according to rules or explicit principles.[^387]
For Brandom, then, the solution to the regress paradox that makes trouble for the regulist position that norms must beexplicit rules is the recognition of a more primitive level of normativityimplicit in practice.Given primitive proprieties of practice that suffice by themselves to determine individual performances as correct and incorrect, even without an explicit rule, the regress is blocked. For these primitive proprieties, being already ingrained in our practice, are already effective in determining possibilities of evaluation and criticism, whether or not they are ever explicitly formulated. There is no need, in particular cases, for a further explicit formulation of the principles that govern these attitudes, though they remain effective in demarcating correct from incorrect performances in practice.
The interpretive grounds for attributing the full extent of this argument to Wittgenstein are in fact obscure. Wittgenstein himself never suggests a general distinction between ‘implicit’ and ‘explicit’ norms, and indeed only rarely uses the term “normative,” or any of its variants, at all. Indeed, its only appearance in thePhilosophical Investigations comes in the context of Wittgenstein’s criticism of his earlier view of language as a calculus or a game, a view which he held in writing theTractatus :
- F. P. Ramsey once emphasized in conversation with me that logic was a ‘normative science’. I do not know exactly what he had in mind, but it was doubtless closely related to what only dawned on me later: namely, that in philosophy we oftencompare the use of words with games and calculi which have fixed rules, but cannot say that someone who is using a languagemust be playing such a game.[^388]
When the later Wittgenstein does refer to a philosophical operation, comparable to Brandom’s “explicitation”, of showing or bringing to light structures thought tounderlie our practical determinations of correctness and incorrectness, he seems to be quite skeptical of the possibility or helpfulness of any such procedure.[^389] In any case, in the sections immediately following his invocation of Wittgenstein’s argument, Brandom provides a detailed account of the way in which, according to his theory, the “normative statuses” of various reasoning practices are instituted. According to Brandom, the normative status of any performance is always dependent on more basic evaluative “attitudes” of taking or treating that performance as correct or incorrect. The evaluative attitude of treating a performance as correct or incorrect may itself be adopted in various ways. Most directly, evaluative attitudes are connected tosanctions : one treats a performance as incorrect by punishing it, and treats it as correct by rewarding it.[^390] However, Brandom argues that treating a performance as correct or incorrect by negatively or positively sanctioning it must go beyond simply treating it asregular orirregular according to the standards of the community. For genuine normativity, it is essential that positive and negative sanctions adhere to actions that arecorrectly taken to be correct or incorrect, rather than simply to those that arein fact taken to be correct or incorrect.[^391] Since the attitudes that institute proprieties are in this way themselves normative, there may be no way to reduce normativity, so described, to anynon-normative basis in regularities of practice; the basis of social practices of reasoning may be, as Brandom puts it, “norms all the way down.”
The sanctions that institute and stand behind actual attitudes of approval and disapproval may be as simple as corporal punishment - for instance hitting offenders with sticks - or they may extend to more complex, fully social attitudes and actions, such as extending or restricting permissions or rights. Turning to the particular way in which the force of reason works in intersubjective communicative situations, moreover, it is important to Brandom that the source of sanctions is not, first and foremost, the community itself, but rather its individual members in concrete interlocution. The primary social relation, in which normative evaluations first become possible, is not between the individual and the community but between two individuals. Appreciating this, Brandom argues, is essential to avoiding an “I-we” account of intentionality, one that misleadingly sees the community as itself a source of evaluative attitudes. By affirming that evaluative attitudes, sanctions, and normative statuses occur, first and foremost, in particular communicative situations of interpretation, we can instead, he suggests, uphold a more realistic “I-thou” model of intentionality and rational significance.[^392] As Brandom points out, even given community-wide agreement, it still ought to be conceivable that the community is wrong. And this is conceivable, given a social model of reasoning, only if determinations of truth and objectivity are themselves evident, first and foremost, in concrete evaluative attitudes taken in concrete episodes of interlocution, before anything like a “community standard” appears. As Brandom points out, moreover, this makes normative statusesperspectival in an important sense: the determination of the concrete commitments of social actors is always made from a particular interpretive perspective.
The implicit/explicit distinction further facilitates the shift away from an abstract, universalizing account of rationality by treating normative standards for correct and incorrect reasoning as implicit in concrete communicative interactions, and demonstrated in the practical attitudes of the participants toward each others’ performances, even if neither party could explicitly formulate them. Nevertheless, despite this significant element of perspectivalism, Brandom thinks of the institution of normativity and indeed the contents of concepts themselves as explicable by reference to the total structure of linguistic and attributional practices that a community is interpretable as engaging in.[^393] This commitment to structuralist explicability in terms of social practices is most evident in the context of the other half of Brandom’s project inMaking it Explicit , the “inferentialist semantics” that is to complement his “normative pragmatics.” According to inferentialism, concepts are determined as having the contents that they do only by their occupying the particular positions that they do in complex networks of propositional inference and deduction. For the inferential semanticist, conceptual contents are therefore defined by the complex network of formal and material inference rules that govern “moves” from one claim to another, and from explicit claims to intentional actions, in the language as a whole. These rules may of course be largely implicit in practice rather than explicitly formulated; it is the job of rational reflection, in fact, to act as the “organ of linguistic self-consciousness,” bringing what is “implicit in practice” to explicit expression. For Brandom, in particular, the “inferential norms” that govern the use of expressions in everyday practice are conceived as conveying upon these expressions the content that they have. Later, explicitation makes these norms clear and thus displays this content in the context of a general inferentialist description of the structure of content in a language as a whole.[^394]
As we have seen, Brandom construes Wittgenstein’s argument as challenging the notion that the determination of correctness or incorrectness in concrete interlocution could ever rest wholly onexplicit norms or principles, and so as arguing that it must rest instead on norms or proprietiesimplicit in practice. This leaves out, however, the possibility of construing Wittgenstein as issuing a more radical challenge, one directed against the very attempt to portray normative judgments as depending on structures intelligible as norms or proprieties (no matter how “primitive”) at all. issues As I shall argue, though, it is just such a challenge that Wittgenstein can indeed be read as issuing; and reading him this way helps to show what is taken for granted in Brandom’s account.
At PI 202, Wittgenstein says, apparently in an attempt to resolve the “paradox” of PI 201, that “‘obeying a rule’ is a practice;” a few paragraphs earlier, he says that “to obey a rule, to make a report, to give an order, to play a game of chess, arecustoms (uses, institutions).” (PI 199). These remarks, as well as Wittgenstein’s scattered references to “forms of life,” have encouraged interpreters in taking him, as Brandom does, to be accounting for the ordinary possibility of following explicitly formulated and consciously recognized rules by reference to more primitive or basic norms, implicit in the “practices” we share. For some commentators, these norms are to be taken as regularities instituted by some form of convention or social agreement; according to others, they depend on the “natural” or biological regularities of human behavior.[^395] But another, quite different reading of the significance of these remarks becomes possible when we consider the specific way in which Wittgensteinsituates his “appeal” to practices (if such it be) within a broader consideration of the basis and limits of philosophical explanation of the “uses of words” themselves. In seeking to explain how it is that it ispossible to follow a rule correctly (to go on as we do), we may easily and naturally be tempted to advert to a basis for our practices in a more fundamental or primitive fact of agreement:
“So you are saying that human agreement decides what is true and what is false?” - It is what human beings say that is true and false; and they agree in the language they use. That is not agreement in opinions but in form of life.
If language is to be a means of communication there must be agreement not only in definitions but also (queer as this may sound) in judgments. This seems to abolish [aufzuheben ] logic, but does not do so…[^396]
In response to the interlocutor’s attempt to reduce truth and falsity to such a fact of “human agreement,” Wittgenstein responds by adverting to the even more basic fact of “agreement” in thelanguage that we use. He explicitly distinguishes this kind of agreement from agreement on “opinions” or beliefs, and further calls it agreement “in form of life.” This kind of “agreement,” which is not an agreement on opinions or beliefs, and so is not a matter of sharing interpretations of facts, holding common creeds, or the like, is nevertheless pervasivelyshown in ordinary life, in what we call “following a rule,” “responding to an order,” “going on in the same way,” and so on. When these ordinary phenomena occur, the “basis” of their possibility is not ordinarily called into question; we take it for granted, by and large, that others who share our language will go on as we do. But when it does come into question, we will respond by appealing to the fact of agreement that is (perhaps only now) perspicuous as having existed all along, the ground of our sharing a language or a way of life. Such agreement can have the force of reminding our interlocutor of her commitments to such shared routes of significance and patterns of judgment, of what she already knows but may have forgotten, of what actions or decisions we may perceive her life, as she has lived it before us, as committing her to as she goes on to live it.
Despite differences in emphasis, Brandom can take most of these points in stride; indeed, it is almost impossible, once we have taken Wittgenstein to betheorizing meaning in terms of something like “practices,” to resist interpreting the fact of “agreement in judgments” that he cites as the fact of our sharing a broad range ofspecific anddescribable “linguistic practices,” including centrally, but not limited to, our practices of judging and evaluating the linguistic performances of others. As is well known, though, Wittgenstein has deeply seated animadversions, in general, against conceptions of the task of philosophy as consisting inexplanation or theory at all. Even more decisive in the present context is his specific critical sense of the relevant force of appeals to “agreement,” whether in the course of ordinary discourse or philosophical explanation.[^397] The “agreement” that we share in sharing a language, he makes clear in the passages where he more closely considers the ordinary sites and implications of our appeal to it, is notitself to be understood or explained, in general, in terms of any morebasic set of facts or phenomena, not even more basic normative attitudes or proprieties of practice.
This becomes clear, especially, in Wittgenstein’s consideration of what is called going on in the “same” way, for instance in completing a series of numbers or using a word in new cases. Where an interlocutor’s performances are recognized as deviant with respect to the standard we take ourselves to be committed to, we will ordinarily criticize them as failing to go on in the “same” way we do. But such criticism, Wittgenstein makes clear, is not itself based on any criterion or standard of “sameness” more basic than the fact of agreement itself. Theuses of the words “agreement,” “same” and “rule” are indeed, he says, deeply interwoven, both in the teaching of practices and their criticism.[^398] But this interweaving is not such as to confer priority on any one of the notions they express, in relation to the others. The appeal that we may be tempted to make, in response to a recognizably deviant performance, to the “sameness” of a way of applying a rule or a way of going on in the completion of a series or the use of a word, isalso , irreducibly, an appeal to the simple fact of ouracting the way that we do; it does not adduce deeper explanatory grounds for this fact, but simply gives expression to it.[^399] In particular, the appeal to “sameness” in explaining what we do, or criticizing the performance of an interlocutor who fails to do this, does not itself adduce any grounds for our agreement in practice that are deeper than that agreement itself. Our appeal to it cannot, therefore,in general sustain a retrospective description of those explicit standards and rules that we are later in a position to present ourselves as agreeing in, as having beenimplicitly present in our practices all along.
Of course, we maymake this appeal, in ordinary life as well as in special contexts of philosophical theorizing. Where an interlocutor or a learner fails to go on in the expected way, to do what we do, to follow the rule that he seemed to understand in the way that (as we understand) itmust be followed, we may appeal to the fact of that which he shares with us, the understanding that he already manifested in his performances before, the regularities or proprieties that he already showed in the previous instances of his practice. But as Wittgenstein’s rule-following paradox itself brings to the fore, any such appeal mayitself be variously understood or understood (as we shall say) wrongly. It may then yield a performance that is, by our lights, deviant; we may then criticize the performance, censure its performer, repeat our demand for him to recognize what he shares with us. Wittgenstein’s paradox, in its most general form, calls upon us to account for our making, and enforcing, this demand by reference to some formulable item or basis that we can cite as underlying it. Its critical upshot is that no such item or describable basis can do so. We are, in practice, thrown back upon repeating the demand itself, and nothing can guarantee its success inany case.
In the passages in which he considers most closely what is involved in our evaluation of certain responses, as opposed to others, as being “normal” or “natural,” Wittgenstein emphasizes the specificity of the surroundings in which such evaluations themselves “normally” occur, and against the backdrop of which they function. For instance, at PI 143, he considers the various possibilities of a learner’s response in a language-game that involves writing down “series of signs according to a certain formation rule”:
143…. And here we may imagine, e.g., that he does copy the figures independently, but not in the right order: he writes sometimes one sometimes another at random. And then communication stops atthat point. - Or again, he makes ‘mistakes ’ in the order. - The difference between this and the first case will of course be one of frequency. - Or he makes a systematic mistake;. . Here we shall almost be tempted to say that he has understoodwrong .
Notice, however, that there is no sharp distinction between a random mistake and a systematic one. That is, between what you are inclined to call “random” and what “systematic.”
Perhaps it is possible to wean him from the systematic mistake (as from a bad habit). Or perhaps one accepts his way of copying and tries to teach him ours as an offshoot, a variant of his. - And here too our pupil’s capacity to learn may come to an end.
In ordinary cases of learning, the pupil can be brought, relatively easily and by means of the relevant training, to do what we do, to go on in the right way; but Wittgenstein’s point here is that nothingguarantees that this must always be possible. The pupil’s capacity to learn may always come to an end; and when it does, there may benothing more to which we can appeal to ensure her future agreement. And this possibility of breakdown, the possibility of my being unable to find grounds for demanding agreement, or of my grounds failing to appeal to the other, affects in an essential way anything we should call an explanation of my justification for my following the rule as I do:
- “How am I able to obey a rule?” - if this is not a question about causes, then it is about the justification for my following the rule in the way I do.
If I have exhausted the justifications I have reached bedrock, and my spade is turned. Then I am inclined to say: “This is simply what I do.”[^400]
The point here is closely related to the earlier one about therole of appeals to agreement, judgments of sameness, and applications of “identity” in the ordinary cases in which we evaluate and criticize linguistic performances. It is that explicitly cited grounds for these appeals and judgmentsmay fail to motivate in any case, and thatwhen they do so fail, nothing need necessarily ensure that the learner will indeed go on in the right way. Here, any appeals to normsimplicitly shared in practice will be just as idle, and just as little capable ofensuring agreement, as the appeal toexplicitly stated norms or agreements that it replaces.
Thus, far from demanding, as Brandom takes it, that norms made explicit in reflection be construed as having a basis in inexplicit but nevertheless normative proprieties of practice, Wittgenstein’s rule-following paradox, read in this general way, poses a much deeper-seated threat to the project of a socially based inferentialist semantics than Brandom can see. For it effectively challenges the thought that the motivating force of reasons in ordinary conversation can ultimately be explained, in the course of a general explanation of the possibility of communication, by reference to anything “implicit in practices” at all. The appeal to commonalities of response or underlying agreements may fail in any case; and in each case where it does fail, an appeal to the existence of a shared practice can do no better. If this possibility of failure is indeed always present, indeed, then there are in generalno such things as univocal standards or norms that are silently present, determining correctness and incorrectness even where there is, as yet, no explicit standard. It is anessential feature of our ordinary discourse that it may always, and sometimes in fact does, “bottom out” in the “bedrock” of which Wittgenstein speaks, that the chain of reasons may always come to an end. Butwhen bedrock is reached, in the interlocutor’s continued (as we may put it, “stubborn”) failure to see what I see, her refusal to find grounds in what I say for going on as I do, as I say shemust , my further appeal to proprieties implicit in the practices we share, to common routes of judgment or standards for evaluation that we must share insofar as we share a language at all, goes and can go no further than my appeal to this bare fact of our sharing a language itself. Here, one might say, there is no longer anyquestion of accounting for the incorrectness of the interlocutor’s performance. For the normal surroundings of commonality that provide so much as thepossibility of accounting for the performance as correct or incorrect have failed. I will then be inclined, as Wittgenstein says, simply to repeat the fact ofmy practice, of my grounds and of my ways of going on. Since I can no longer see these ways as determining, or necessitating, the performance of the other, I can in the end only point again to it, repeating my appeal to the legitimacy ofmy way only, this time, by demonstratively indicating its bare existence.
The bedrock of which Wittgenstein speakscan always be reached, in the order of practice. But it is one of the implications of the rule-following paradox that itis reached, in the order of explanation,whenever we try to give a general account, in terms of more primitive (implicit or explicit) underlying “norms,” proprieties, or standards, of what we suppose to be the practical “basis” of the fact of our using language at all. Most directly, of course, the paradox bears against the picture that takes our linguistic action to be everywhere determined by underlying and describable rules. Given this picture, it shows that since “any course of action can be made out to accord with the rule,” no course of action can actually be presented asdetermined by it.[^401] Once we realize the generality of this paradox and the extent of the problem it represents, the picture of our ordinary practices as governed everywhere by describable rules is visible as a “mythological” description of our practice. No such description, in particular, canaccount for the fact that we go on in the way we do, since no appeal to rules can adduce grounds more basic than this fact itself for supposing that itmust hold.[^402] But if the appeal to explicit rules cannot explain what is involved in our practice, than neither can, for similar reasons, the appeal to inexplicit proprieties of practice that Brandom makes. Within the course of a general attempt toexplain the possibility of communication or account for the possibility of criticizing the performances of others, the mythology of implicit proprieties of practice is in fact little different from the mythology of rules that Wittgenstein most directly opposes. If, as I have argued, the fact of agreement is not explicable in terms of anything more basic of itself, to talk of a standard or a norm here,even an implicit one , is to commit a grammatical confusion; it is to presuppose a metaphysical picture of our binding to a linguistic, grammatical or “pragmatic” structure that cannot survive Wittgenstein’s staging of the self-undermining fantasy of constraint upon which it relies.
Thus, in Wittgenstein’s consideration of what is involved in “following a rule,” the ordinary and hardly eliminable possibility of communication breaking down inany case can be seen to pose a pervasive general problem for any accounting that, like Brandom’s, seeks toexplain our usual agreement in ways of going on by reference to more primitive features of practice, even those that are not yet “explicit” in reflection or judgment. The explanatory project founders, in particular, at the point of bedrock, where the simple fact of my action is no longer explicable in terms of anything more basic than it itself. Here, there are no longer facts or norms (even implicit ones) that I can appeal to inexplaining my action, since, as Wittgenstein puts it, there is no longer the specific kind of doubt that such an explanation could answer.[^403]
The problem here is not, it is important to note, that there is anything wrong or suspect about Brandom’s claim that even the most “basic” or foundational human behaviorscan be characterized in terms that are “normative” in Brandom’s sense.[^404] At PI 289, for instance, Wittgenstein says with reference to an immediate, first-person expression of pain, that to “use a word without justification does not mean to use it without right.” That is, even at the point of bedrock, where no further justificatory explanation is possible or useful, we maystill describe performances as legitimate or illegitimate, and even (in many cases) “correct” or “incorrect” according to some standard. This possibility of “normative” description at this level is not at issue between Wittgenstein (as I am reading him) and Brandom; whatis at issue, however, is the possibility, essential to Brandom’s account, of making it the basis of asubsequent general explicitation of the “norms implicit in practice” all along. Of course we may sometimesdescribe our existing practice as having involved, all along, some set of distinctive commitments; such descriptions will be useful, in general, only where there is some specific reason for doubt about those commitments or their bearing on the particular case, and may, again, always themselves be accepted or rejected. But in emphasizing the standing possibility that such descriptions fail, that we find ourselves at bedrock, without any possibility of further appeal, Wittgenstein challenges the notion of rules that sees them as always already silently determining our uses of words, throughout a language as a whole.[^405] If, as I have argued, in the “bedrock” situation, appeals to implicit norms fare no better, than Wittgenstein’s paradox is just as fatal for Brandom’s inferentialism as it is for the mythological picture of rules that it aims to replace.
Again, this critical claim is not, it is important to note, based on some version of the argument that if it is possible, inany case, for me to fail to find my ground with another, it must be impossible to find ultimately workable grounds or standards for agreement inevery case. Such an argument, though perhaps resembling Descartes’ argument for perceptual skepticism on certain reconstructions of it, would be a bad one, trading on what might seem to be a peculiarly philosophical tendency to absolutize the imperfections of our abilities to know or our liabilities to respond. It is, indeed, no part of Wittgenstein’s claim to suggest on philosophical grounds that our capability to understand one another, or to find grounds for understanding where they at first seem to be lacking, goes any less far (or farther) than it in fact does. The fact that itdoes go as far as it does, indeed, can be seen as a remarkable one, and all the more so, in view of how little we can say, in a general sense, to explain it. Wittgenstein’s claim is, rather, that, whatever this fact may be taken to involve, the justificatory or explanatory appeal to it cannot be either discharged or shored up by an appeal to factsor norms more basic than it itself. Our appeals to the fact of agreement, in the actuality of everyday conversation as well as in philosophical explanation, can in the end only retrace themselves, ceaselessly gesturing at the fact which is presupposed to, but never wholly explained by, all of our reference to rules, norms, or practices, whether explicit or implicit: the omnipresent but scarcely comprehended fact of our sharing (what is called) “a language” at all.
Again, seeing the way in which Wittgenstein’s rule-following paradox gives us reason to doubt the ultimate coherence of any general account of language-use as depending in primitive proprieties implicit in practice need not prevent us from acknowledging the existence, in many actual cases, of just the process of reflectiveexplicitation that Brandom describes so carefully. Doubtless, this processdoes go on, and indeeddoes play an essential role in a wide variety of human institutions and “linguistic” practices. The present point is just that it is ultimately incoherent to make it, as Brandom does, the basis of an explanation of the possibility of meaningful language itself. A good example of the actual process of explicitating norms, indeed, is thejuridical practice of the articulation and reflective determination that Brandom cites as a model, wherein laws and standards of justice are articulated by reflection on past precedent and what can be seen to have been implicit in their previous application to particular cases. And in many, probably most, of the vast range of cases in which something like the articulation or explicitation of binding standards of practice or judgment does go on within a particular practice, it will be possible for its participants to see the standards or rules thus articulated as grounded in (what they will now be able to see as) proprieties or standards of judgment and evaluation that were (at least as they willnow see it)implicit in their practice all along. But it is one thing to say (what is surely true) that such a distinction between what is implicit in practice and what is later to be seen as explicit in rules plays an important role in our pursuit and description of a wide variety of human practices; it is quite another to appeal to the implicit/explicit distinction, in the general way Brandom does, as providing the basis for a generalexplanation of the possibility of human communication, and the contentfulness of its concepts, overall. In so doing, as we have seen, Brandom misplaces the specific and uniquely perilous situation of appeals to standards, and hence of the possibility of (what we may or may not be able to recognize as) “explicitation,” in our claims and demands on one another.
The difference between Brandom and Wittgenstein on this point has important consequences for the broader question of the specificforce of reason , the basis and nature of the claim of the “better” reason over our actual decisions and acts of judgment. For consider how implicit and explicit norms are pragmaticallyenforced , according to Brandom’s social pragmatist picture. As we saw, for Brandom the enforcement of norms, and hence the institution of normativity, always depends on the practice of imposing positive or negativesanctions for correct or incorrect behavior. The stake of reasoning, what underwrites the force of the obligations we undertake in committing ourselves to particular claims, is always dependent on the threat or promise of the sanctions imposed by our peers, including the determination of whether we are entitled to membership in the community at all. Where the underlying threat is not immediately present, the force I take it to have may depend, to be sure, on myrecognition of it as applying to me, or (what is equivalent) my recognition ofmyself as subject to its force.[^406] Thus, to evaluate a performance as according or failing to accord with some explicit standard is always, for Brandom, to assess its liability to be rewarded or punished; it is this liability to sanction that underlies the possibility of specific performances being assessed at all. The liability to sanctions and rewards is seen as already existing, even before it is explicitly articulated in formulable standards; it is by reference to it, according to Brandom, that appeals to such explicitly articulated standards have the force that they do.
But as Wittgenstein’s consideration of rule-following, by contrast, brings out clearly, the “articulation” of standards to criticize specific performances isitself the operation of a fundamental claim of force. It is so, most of all, inasmuch as such explicitation effectivelyconstitutes a standard of criticism, and sointroduces determinate possibilities of criticism and sanction, punishment and reward, that did not exist before. (For instance, the laws or standards that prohibitextorting money from a corporation , and thus make it punishable to do so, do not exist prior to the determinate forms of social life and institution that give them sense). Of course, it is an integral part of the force of this kind of explicitly formulated normative claim that it can present the standard it “formulates” as having already existed, operating silently as a determinate but “implicit” component of the practice that we already accepted. Brandom’s picture, in seeing the articulation of norms as always dependent on such implicit proprieties of practice, consents uncritically to this claim, both in general and in the manifold specific cases where it plays a role in the determination of our perceptions of rightness and the pursuit of our projects.
If understood in the way I am recommending, though, Wittgenstein’s rule-following considerations, by contrast, yield grounds for demystifying the theoretician’s claim to retrospectively recognize “proprieties” held to have been “implicit” in practice all along. These grounds are thus also grounds for interrogating critically the claim of force or power that the demand for this kind of “recognition” involves. From this perspective, the fact that “we” (but who is included in this “we”?) can constitute standards of judgment and then apply them to new cases in a way that is (largely) recognizable as “uniform,” is, again, bound to appear remarkable; but as it is not founded in any determinate or describable fact of our agreement on beliefs or contents of judgment, it is not, also, founded in the commonality of threats we all fear or rewards we all seek. When a performance is recognized as deviant, an appeal to “what we all do” or to the rule or regularity implicit, in any case, in our practicescan have the effect of bringing the performer back in line; it will have this effect, in particular, whenever the performercan recognize herself as having been committed, all along, to the standard we thereby articulate. But this recognition will be shown, if at all, only in the complexity of what she then goes on to do; and it may, again,always be refused.
The claim to articulate binding standards of rationality, regularities of practice, or rules of use conceived as having always already (if implicitly) guided possibilities of significant expression in the practice of a language, is in any case always grounded in a claim of mastery, a claim on the part of the critic to be able to oversee, and thus articulate, the relevant possibilities.[^407] The basis of this claim, as it is operative in our actual discourse, is not, in general, any actual or even promised application of real sanctions or rewards, but themystified and evenimaginary picture of language that is also the core of the metaphysical picture of rules that Wittgenstein most directly criticizes.[^408]
The picture figures deeply in ordinary as well as philosophical practices of criticism; its methodological basis is the ordinary ambition to gain insight into the abstract expressive possibilities of the structure of language as a whole, and to portray them at some level of abstraction from the variety of actual performances they are seen as determining. Brandom, as we have seen, shares this ambition with others who have theorized language as grounded in “social practices.” And although he takes pains to avoid an implausible regulism or any simple attribution of norms to the standards of “communal practice,” his picture nevertheless replicates the fundamental instability of the earlier, less sophisticated social-practice structuralisms whose explanatory ambition it shares. These projects have in common that they take for granted both the accessibility of the basis of linguistic meaningfulness to theoretical description, and the utility of some coherent concept of linguistic “use” or “practice” in explaining it. But Wittgenstein’s rule-following paradox, in challenging the structuralist to account for the gap between rules and their application, poses a fundamental problem for this configuration of commitments. It does so, most of all, by exposing the open problem of theapplication of a word to a new case of its use. The paradox of rule-following shows that this problem will always be open, as long as we picture language itself as a structure intelligible to theoretical description. No matter how complete this description is, no matter how much it adduces in terms of the proclivities of our practices or the commitments said to be inherent in them, it will still leave the open gap between the structure of language and the life of its use. The attempt to cross this gap with “implicit” proprieties of practice is, from this perspective, as futile as the earlier one to cross it with explicit, symbolically formulated rules. In each case the standard that is designed to explain the use of the word fails to do so, since ititself can be used in various ways. With Wittgenstein’s posing of the paradox, the authority of the structuralist picture is undermined in that it is shown up as inadequate, and indeed futile, for its explanatory purpose. It is thereby exposed to immanent critique at the point of the claim of power that it, in the guise of neutral explanation, recurrently exerts.
From his first published works on Austin and Wittgenstein, Stanley Cavell’s writing is marked by his profound critical engagement with the methods of ordinary language philosophy.[^409] In the articles “Must we Mean What we Say?” and “Knowing and Acknowledging,” for instance, Cavell takes up the question of the relationship of these methods to the traditional problem of skepticism, a question that will occupy him as well throughout the complex argument ofThe Claim of Reason .[^410] Here, Cavell develops the methods of reflection on ordinary language pioneered by Austin, Ryle, and Wittgenstein not in order to provide a direct or indirectrefutation of skepticism, but rather to articulate the unique position from which this reflection can engage in a dialogue with skepticism. For Cavell, the special resources available to this reflection arise most directly from the form of its most typical question, the question of “what we should say when. .” in a variety of different circumstances. The appeal of this question is not to factual or statistical knowledge about normal patterns of speech behavior, but rather, in each case, to what the speakerherself will say in a new case. In a remarkable way, according to Cavell, the procedures of ordinary language philosophy appeal to a kind of knowledge we ourselves possess simply in virtue of being speakers of a natural language, a kind of knowledge that essentially involves our capacity to project our reasons into new situations.
Like Brandom’s own argument for the implicit/explicit distinction, Cavell’s appreciation of the distinctive methods of the ordinary language philosopher rests heavily on an interpretation of the implications of Wittgenstein’s “rule-following considerations.” For both philosophers, it is also significant that the possession of a language, and hence of a social existence, depends upon the ability to project a familiar concept into a new context in ways that our peers will deem appropriate. But whereas Brandom’s inferentialism understands both this ability and the ability to determine appropriateness as governed by describable norms, ordinarily implicit if not explicit in practice, Cavell’s appeal to ordinary language philosophy figures our ability to project conceptsitself as the object of the philosopher’s appeal. In practicing ordinary language philosophy, the philosopher does not seek todescribe the norms governing discourse, but directly engages the interlocutor’sown ability to make judgments of correctness and incorrectness.
It is significant for this appeal that the projection of words into new contexts is, as Cavell puts it inThe Claim of Reason , characterized by both “‘outer variance’ and ‘inner constancy’”.[^411] That is, the meaning of a word can (in some sense) be thesame , regardless of the social, pragmatic, or semantic context in which we use it. But contexts are heterogeneous and diverse. Despite our intuitive sense that words have more or less stable meanings, the question of whether a wordcan appropriately be used in a new context is never completely determined, at least in advance ofour determination of this:
We learn and teach words in certain contexts, and then we are expected, and expect others, to be able to project them into further contexts. Nothing insures that this projection will take place (in particular, not the grasping of universals or the grasping of books of rules), just as nothing insures that we will make, and understand, the same projection. That on the whole we do is a matter of our sharing routes of interest and feeling, modes of response, senses of humor and significance and of fulfillment, of what is outrageous, of what is similar to what else, what a rebuke, what forgiveness, of when an utterance is an assertion, when an appeal, when an explanation - all the whirl of organism Wittgenstein calls “forms of life.”[^412]
According to Cavell, then, the ongoing projection of words into new contexts is neither arbitrarynor “determined” by rules or norms. Rather, on the level of the methodological practice of ordinary language philosophy, the question of the application of an old word in a new case is not any longer a question of degrees of determinacy or arbitrariness with respect toany standard, but involves an appeal that must be madeprior to the grasping of any standard. Any advance delimitation of the range of contexts in which a wordcan appropriately be used would destroy some of its fertile and constitutive ambiguity, and hence some of its sense. But it is the task of ordinary language philosophy, or of a practice of ordinary interlocution informed by it, to negotiate the determination of appropriateness again and again, in each case appealing to the interlocutor’s own senses of propriety, significance, and relevance. As Cavell puts it,nothing insures that the “right” projection will take place; recognizing that there is no standard or principle whose formulation must convince means recognizing that there is no substitute, in the practice of ordinary language philosophy, for the ever-renewed appeal to what Cavell calls the “projective imagination.” The openness of this appeal, its ability to engage the imaginative work of language itself, would be lost if we took it, as Brandom does, that it always amounts to the appeal to what could then later be presented as norms implicit in practice. That this appeal must be renewed inevery new case, and that its application in each case is, to some extent at least, an exercise of the imagination, serves to mark it off from any comprehensive attempt to theorize the norms of language and reason once and for all.
The specific way in which Wittgenstein, according to Cavell, resists an interpretation of linguistic practice as essentially rule-bound comes out more clearly if we consider the concrete practices of reasoning in which the giving of rules and justifications ordinarily takes place. These practices essentially involve, as well, thedetermination of whether rules have indeed been followed. For Brandom, as we saw, this determination amounted to the application of critical “score-keeping” practices whereby interlocutors evaluate one another; such evaluation was, according to Brandom, “essentially something that can be done correctly or incorrectly.” But as Cavell points out, there is an importantdifference between the practice of following a rule itself and the practice of determining whether a rule has been correctly followed:
For Wittgenstein, ‘following a rule’ is just as much a ‘practice’ as ‘playing a game’ is (PI, 199). Now what are its rules? In the sense in which ‘playing chess’ has rules, ‘obeying a rule’ has none (except, perhaps, in a special code or calculus which sets us some order of precedence in the application of various rules); and yet it can be done or not done. And whether or not it is done is not a matter of rules (or of opinion or feeling or wishes or intentions). It is a matter of what Wittgenstein, in theBlue Book , refers to as ‘conventions’ (p. 24), and in theInvestigations describes as ‘forms of life.’ (e.g., PI, 23). That is always the ultimate appeal for Wittgenstein - not rules[^413]
In other words, though it may be the case that determining correctness or incorrectness is itself something that can be done correctly or incorrectly, it is significant that the practice of making this determination in each case is notitself , in general, something that is governed by determinate rules. In this sense, for Wittgenstein as Cavell interprets him, the practice of giving and asking for reasons isnot one of “norms all the way down.”[^414] For the determination of the correctness or incorrectness of a performance may indeed advert to a rule, implicit or explicit; but the determination, in a particular case, of how to understand what the rule itself requirescannot be made by introducing another rule, or indeed anything like a rule at all. Rather than seeing reasoning practices as embodyingimplicit rules, capable of subsequent explication in analysis, therefore, the ordinary language philosopher can only gesture toward what Wittgenstein calls “bedrock.” And when this bedrock is reached - when there are no more reasons to give - the ordinary language philosopher’s method does not seek to render it explicit or explicable, to summarize it in a set of principles or norms or a corpus of rules.[^415] Rather, the ordinary language philosopher must simplyappeal to it, mutely, insisting upon what we must share if we can share a world at all. At the same time, this appeal can itself always fail, breaking down into mutual incomprehension, exhibiting the claim of reason as something weaker, more limited and less assured in its operation, than any explicitation of norms and principles can express.[^416]
For Cavell, the normativity of concepts is not, then, constituted by the explicitor even implicit structure of norms presupposed in discourse; for even where such principles are presupposed, their application in any particular case isitself a matter that must be settled, ineach case, by the exercise of the interlocutors’ own ability to project concepts into new contexts. That there is no substitute for this appeal, both in actual practices of reasoning and in the forms of philosophy that are best suited to demonstrate what is involved in them, is, according to Cavell, the most important implication of the ordinary language philosopher’s consideration of reasoning. InThe Claim of Reason , Cavell further develops his account of the ordinary language philosopher’s appeal to this inexplicit and inexplicable ground of human attunement by considering Wittgenstein’s own distinctive way of using the concept of “criteria.” For Cavell, “criteria” are what competent speakers of a language share, what they agree in, if they share a language at all; but criteria can always fail us, and agreement in them is never to be assured by a standard conceived as determinate in advance. In this special sense, to recognize oneself in another on any particular occasion can be described as “agreeing” in criteria; but it is important that this agreement, shown in particular cases, is not reducible to agreementon any general set of explicit or implicit principles. And it is essential to our way of sharing criteria, of being mutually attuned, that we can also turn out not to share them, tofail to be attuned:
Our ability to communicate with him depends upon his “natural understanding”, his “natural reaction”, to our directions and our gestures. It depends upon our mutual attunement in judgments. It is astonishing how far this takes us in understanding one another, but it has its limits; and these are not merely, one may say, the limits of knowledge but the limits of experience. And when these limits are reached, when our attunements are dissonant, I cannot get below them to firmer ground. The power I felt in my breath as my words flew to their effect now vanishes into thin air. For not only does he not receive me, because his natural reactions are not mine; but my own understanding is found to go no further than my own natural reactions bear it. I am thrown back upon myself; I as it were turn my palms outward, as if to exhibit the kind of creature I am, and declare my ground occupied, only mine, ceding yours.[^417]
There is in principle no way, according to Wittgenstein and Cavell, to foreclose this possibility of refusal, no ultimate authority to appeal to when the attempt to find oneself in the other fails. It follows that what is at stake in reasoning, in accepting or refusing an interlocutor’s explicit justificationsor what is simply implicit in his ways of life, is neversimply a matter of compliance or failure to comply with intelligible normative principles. Where disagreements arise, rules may be cited, and the introduction of explicit normative principlesmay suffice to convince one or another party to the dispute. The introduction of explicit rules is itself, for the ordinary language philosopher, an integral part of the variety of practices that we call reasoning, deliberating, arguing, and convincing. But the citation of an explicit principle, even if it is offeredas normative for the kind of language-game that we are involved in, or as constitutive for rationality itself, may always itself fail to convince. And when this happens we are, as Cavell suggests, “thrown back upon ourselves” in a peculiar sense, left with nothing more to say, left to occupy our own ground silently, capable of appealing, in the end, only to ourselves.
This staking ofourselves in reasoning, figured in the ordinary language philosopher’s methodological appeal to our own sense of the projection of our words and in her recognition of the ongoing possibility for appeals to rules tofail to convince, distinguishes the ordinary language philosopher’s conception of these practices from other conceptions current within the analytic tradition. The standards or norms implicit in these ordinary practices may always be described in terms of rules, and the introduction of explicit rules will in fact in many cases help us to see what was involved in our practices all along, and thus show us the extent of our obligation to them. But if the introduction of explicit rules mayalways fail to convince, then there is an important sense in which this description by means of explicit forms of rules mustalways fail to portray its object. The theoretical adumbration of rules meant to describe the grammar of ordinary language practices can go only as far as the ordinary explicitation of rules within these practices itself goes; and there can be no hope that the introduction of any set of rules could suffice to eliminateall disagreements. What is made possible by the explicitation of any particular standard is then, at best, the appeal of one interlocutor to another (“see it this way!”)within the practice of reasoning, an appeal that might always be taken up, or might be refused. Such an appeal may be an appeal to an explicit or explicitable standard of judgment, but it may also be an appeal toways of judging, routes of significance, ways of seeing what is significant in a new case or worthy of our attention in an old one. It may appeal as much to our powers of imagination as our capacities of judgment; what is at stake in it is as muchhow we shall think aswhat we shall do.
For the ordinary language philosopher, the peril of deviant reasoning is not so much sanctioning asalienation , the possibility that I may find myself (that any of us may find ourselves) at bedrock, unable to find words to justify myself to another, unable to find or articulate the ground of our mutuality. The threat of this alienation is not, at least in most cases, that I mayactually be ostracized or forcefully excluded from the community; it is, rather, that I will not be able to find myself within it, will not be able to identify with its modes of action or its determinations of significance. And where theI find these modes and determinations lacking, where I cannot find a grounding for their assumptions in myself, the place of the alienation they threaten can also be the opening of the possibility of their critique. Either way, what is staked is not so much our freedom from negative sanctions or even our membership in a community, but the very possibility of community itself, of the unthought ground of mutuality that enables me to receive the other, and to be received by her, at all. In taking up or failing to take up the other’s words and reasons as words and reasons that can be ours as well, we will find or miss the ground of our mutuality, the extent to which we can share reasons, the extent to which we find ourselves capable or desirous of community with the other.[^418]
The method of ordinary language philosophy, as Cavell reconstructs it, is thus practically unique in refusing to see theforce of reason as dependent on theenforcement of norms, or indeed as amounting to anyauthority more distinct or elevated than that of the mere and never-ensured possibility of our relation to one another.[^419] It is for this reason that Cavell’s appeal to the methods of ordinary language philosophy, in constant dialogue with the threat of skepticism, culminates in his recognition of the need to refigure the traditional problematic of skepticism as one ofacknowledgment rather than knowledge. The skeptic figures the problem inherent in skepticism as a problem of inadequate knowledge, as if recognizing our human situation meant recognizing that there is something that wecannot know of the object before us or the person who speaks to us. The appeal to ordinary language does not, according to Cavell, block this conclusion directly, but rather interrogates its ground in the kind of projection of the ordinary uses of terms that it demands. This projection, evident in, for instance, the skeptic’s question whether we know of the existence of thewhole of an object before us, whether we can reallyknow (and not only assume, infer or guess) that our interlocutor is not simply an automaton, is, according to Cavell, neither fully “ordinary” nor completely “extraordinary”. Instead of simply rejecting or ruling out the skeptic’s appeal, Cavell interrogates the movement of its desire, revealing it as coeval with the desire to develop a totalizing analysis that would speak to the human epistemological condition outside anyparticular context. But this desire to project our words “outside language-games,” to find a place to speak outside the practical contexts and concerns that alone give speech its ordinary surrounding, is not itself simply to be rejected, for it is inherent in the projective character of our language itself, in our tendency to project terms ever again into new and unanticipated contexts.[^420] Through the ordinary language philosopher’s own appeal to the projective imagination, though, it becomes clear that what is at stake in it is notsimply an inadequacy of knowledge. To work through my skepticism is tolive it, to stake myself, in the concrete discursive recognition of another, on the possibility that thereis a context of reasoning, desiring, and suffering that we can share.
The skeptic’s worry, which can masquerade as a theoretical one about the possible adequacy of knowledge, then stands revealed on the level of the real anxiety from which it arises, the anxiety of alienation or isolation, of failing to find myself with another, of being “thrown back upon myself” in solipsism. That this anxiety is always possible, for Cavell, means that traditional skepticism manifests (though darkly) something like a disappointment with the human condition as such, with the fate of having to seek recognition, finding and losing it ever again, outside the possibility of any conclusive refutation of our need for it.
Cavell’s use of the methods of ordinary language philosophy culminates by showing that we can see the stake of reasoning as the need for acknowledgment, of the way in which we live or fail to live the mutuality of our words. In this way, Cavell’s investigation of skepticism offers to orient its problematic away from a question of the completeness of knowledge, and toward the question of our ability to acknowledge one another. This breaks with the totalizing impulse of the structuralist understanding of language, offering instead to re-articulate the source of this impulse at the level of our need, or desire, for mutual understanding, agreement, or attunement, our need or desire to find a context of interests and reasons that we can share, a world in which we can live together. But to see how this alternative ethics arise from, and in turn requires, both an alternative conception of philosophical practice and a renewal of reflection about the nature of language itself, it is helpful to turn to the work of Levinas, a philosopher who is not in any sense a part of the analytic tradition, but whose work on language and ethics nevertheless may bear some significance for our understanding of how that tradition might, today, be received.
From his first philosophical work, Levinas’ thought is marked by the attempt to understand the foundations of our understanding of one another outside theclosure of a totalizing system of metaphysics, phenomenology, or ontology. These comprehensive approaches of these projects, Levinas argues, will always fail to adequately respect the ethical implications of our human relationships with one another by failing to acknowledge the respect in whichdifference oralterity figures in these relationships, a way that is, according to Levinas, more basic than any theoretical accounting for it. Recognition of the primacy of alterity, Levinas argues inTotality and Infinity , calls for an ethics that is at the same time “first philosophy.”[^421] This ethics, according to Levinas, would recognize that the ethical claim of one upon another is in fact prior to those claims of ethical or metaphysical theory that would portray it as a form of relation between two already constituted terms. In this respect, for Levinas as for Cavell, the fundamental ethical imperative is the demand for acknowledgment, a demand whose satisfaction cannot be guaranteed by the systematic inscription of any set of norms, rules, or principles that could be known, but must be experienced in the experience of the possibility of my relationship to another. The question of my relationship to the other, for Levinas as for Cavell, is not first and foremost a question of knowledge (as much as it may seem to be within the traditional projects of philosophy), but rather a question of whether I can rise to the stringent exigency of an “ethical” demand, of a claim of the other upon me, that is never simply a dictate of comprehension.
For Levinas as for Cavell, the ethical demand of the other begins where I am tempted to say that my knowledge of her must be incomplete, where it is no longer possible to comprehend our relationship as that between two terms in a system of relations governed by theoretical principles or rules. If society as such is founded upon the regular or contractual relationship of autonomous subjects, fidelity to the ethical relationship itself demands an acknowledgment that comes before this contract. According to Levinas, it requires, instead, a recognition of the way in which the possibility of the relationship to the other, a relationship marked by “infinite” distance, itself marks the very form of our subjectivity. For according to Levinas, there is no responsibility outside the possibility of this genuinely constitutive relation to alterity, to the other as other, irreducibly singular and unique.
Levinas’ own ethics of alterity takes shape, most determinatively, against the backdrop of his critical rejection of Heidegger’s project of ontology, a project in which Levinas detects a repetition of the totalizing gesture of philosophy as such. This gesture, according to Levinas, aims to eliminate alterity and the ethical relationship by reducing it to the univocity of a monological description, in this case the description of the closure and totality of being. His rejection of Heidegger’s ontological project culminates in the dense and ellipticalOtherwise than Being or Beyond Essence , where Levinas again insists upon a form of subjectivity that is defined by the possibility of my putting myself in the place of the other, outside any possibility of a theoretical comprehension of her situation.[^422] It is only in this form of substitution, Levinas suggests, that the concreteness of the ethical relationship can appear in its full strangeness and difference, a concern that unsettles the subject to its core, a concern for alterity that is in principle uncapturable as a concern for anything “in being” itself.
For Heidegger as Levinas reads him, the univocity of being meant that language as such must be the language of being, the speaking of being with one voice in the primordiality oflogos .[^423] Rejecting this univocity, Levinas returns to language to find in it the possibility of ethics as anotherwise than being , a form of relationality and difference that cannot be reduced to the totalization of a single voice. Levinas’ consideration of this primordiality yields one of the most suggestive distinctions ofOtherwise than Being : the distinction between thesaying (as the original form of the ethicalappeal of one to another) and thesaid (of propositions, demonstration, and knowledge) in which it will always already be fixed:
From the amphibology of being and entities in the said we must go back to the saying which signifies prior to essence, prior to identification, on the hither side of the amphibology. Saying states and thematizes the said, but signifies it to the other, a neighbor, with a signification that has to be distinguished from that borne by words in the said. This signification to the other occurs in proximity. Proximity is quite distinct from every other relationship, and has to be conceived as responsibility for the other; it might be called humanity, or subjectivity, or self. Beings and entities weigh heavily by virtue of the saying that gives them light.[^424]
According to Levinas, the possibility of ethics - indeed, the possibility of subjectivity itself - depends on the possibility of a return to this paradoxical saying before the said, to a linguistic relation that is grounded in an exposure to the other. He treats this exposure as a kind of “signification” that is prior to the fixture of what is said in the form of propositions or contents; it is “prior to all objectification” and to any giving or exchanging of signs.[^425] For Levinas, the possibility of any social relation, any intersubjective agreement of principles or judgments, any “game” of reasoning together in debate, discussion, argument, or conversation, depends on this more primordial saying. The appeal of the one to the other, for Levinas, is the voicing of a demand that cannot be captured in the objectivity of a set of rational contents, of a totality of propositions bearing rational relations to one another.
Were the critical and reflective methods of ordinary language philosophy, descendents of the envisioning of language that first began the analytic tradition as such, to take up this Levinasian discourse of the saying and the said, the distinction would necessarily be subject to far-ranging and difficult critical questions whose scope can only, at best, be indicated here.[^426] To talk of the primacy of the saying over the said, of is greater “originality” and of the more basic and “grave” responsibility that stems from it, is at best to gesture toward the same ineffable ground that Wittgenstein calls bedrock, the ground of mutuality that itself, in the ambit of any general theoretical attempt to elucidate it, stands revealed as groundless. As we have seen, both the critical upshot of Wittgenstein’s rule-following considerations and the practice of ordinary language philosophy depend on the possibility of a methodological acknowledgment of this groundlessness, of the ultimate baselessness of the fact of our agreement. But the methodological acknowledgment of this groundlessness within a practice of philosophical reflection or linguistic criticism demands as well our recognition that, at this point of bedrock, “my spade is turned,” that the movement of articulation here fails in the very saying.
The most significant legacy of these alternative considerations of linguistic reason and reasoning, then, is not an alternative account or even a single alternative practice. It is, rather, the opening of a set ofquestions about the nature of language and its relationship to what we treat as the ordinary forms of social life. Within the ambit of these questions, it must be asked whether it is even so much as possible to grasp the “structure of language” as the basis for an explanatory account of these ordinary forms, or of the role of what was once grasped as reason in determining and controlling them. The effect of posing such questions can be, as well, to re-open the question of the basis of rational force, of the ground for what we take to be the claims of reason in application to the pursuit of our lives. To ask them is also to interrogate more closely the relationship between claims for the force of reason and the real systems of power and violence with which they have sometimes made common cause; it is to re-open the ancient question of the relationship of the force of language to that which binds a community together, ensures its regular life, or seeks to conserve or protect its integrity against internal or external enemies. These questions, as we have seen, are recurrently be re-opened by the historical trajectory of structuralism, even as it tries incessantly to foreclose them; that they bear a deep significance for any future thinking of the political, of ethics, and of the claims of rational reflection in today’s world, seems beyond doubt. Their opening, in the text of analytic philosophy’s sustained consideration of intersubjective meaning and interpretation, ought to reveal as well the way in which the tradition’s sustained inquiry into language deepens and radicalizes them.