Stanley Cavell ’ s Argument of the Ordinary

My overall aim is to show that there is a serious and compelling argument in Stanley Cavell’s work for why any philosophical theorizing that fails to recognize what Cavell refers to as “our common world of background” as a condition for the sense of anything we say or do, and to acknowledge its own dependence on that background and the vulnerability implied by that dependence, runs the risk of rendering itself, thereby, ultimately unintelligible. I begin with a characterization of Cavell’s unique way of inheriting Austin and Wittgenstein – I call it “ordinary language philosophy existentialism” – as it relates to what Cavell calls “skepticism”. I then turn to Cavell’s response to Kripke in “The Argument of the Ordinary”, which is different from all other responses to Kripke’s Wittgenstein on Rules and Private Language in that Cavell’s response, while theoretically powerful, is at the same time also existentialist, in the sense that Cavell finds a way of acknowledging in his writing the fundamental fact that his writing (thinking) constitutes an instance of what he is writing (thinking) about. This unique achievement of Cavell’s response to Kripke is not additional to his argument, but essential to it: it enables him not merely to say, but to show that, and how, Kripke’s account falsifies what it purports to elucidate, and thereby to show that the theoretical question of linguistic sense is not truly separable, not even theoretically, from the broadly ethical question of how we relate to others, and how we conduct ourselves in relation to them from one moment to the next. I consider the voice of Stanley Cavell, who passed away this past June at the age of ninety-one, to be the most important Englishspeaking philosophical voice of the second half of the twentieth century, where by “important philosophical voice” I mean: a voice that young students of philosophy nowadays need to hear. I know

I consider the voice of Stanley Cavell, who passed away this past June at the age of ninety-one, to be the most important Englishspeaking philosophical voice of the second half of the twentieth century, where by "important philosophical voice" I mean: a voice that young students of philosophy nowadays need to hear. I know that such judgments have little more than merely autobiographical significance, and that the vast majority of those working within contemporary analytic philosophy will not share this judgment. The vast majority of them will not even be familiar with Cavell's work. Cavell's repeated efforts, from the very beginning of his philosophical career all the way to its end, to turn philosophy back to its origins in everyday human experience -as against its tendency to "reject the human" and seek lasting satisfaction in abstract, objectivist theorizing that attempts to bypass that experience altogether -seem as hopeless now as perhaps such efforts have always been. And surely, no philosophical argument, however sound and well-crafted, could reasonably be expected to bring an end to a basic human tendency. Nor could any philosophical argument force anyone's attention to itself. The aim of this paper is to show that there is, nonetheless, a serious and powerful argument to be found in Cavell's work for why any philosophical theorizing that fails to recognize what Cavell refers to as "our common world of background" as a condition for the sense of anything we say or do, and to acknowledge its own dependence on that background and the vulnerability implied by that dependence, runs the risk of rendering itself, thereby, ultimately unintelligible.
In presenting the argument, I will focus primarily on Cavell's "The Argument of the Ordinary", which began as the second of his three Carus lectures, delivered to the American Philosophical Association in 1988. 1 "The Argument of the Ordinary" (hereafter TAO) responds to Saul Kripke's derivation of a "skeptical argument" and then a "skeptical solution" from Wittgenstein's Philosophical Investigations, and is arguably Cavell's last sustained attempt to engage with the English-speaking "philosophy of language" and "philosophy of mind" of his time, and more specifically to save Cavell's Wittgenstein from being lost, if not through uninformed and hasty dismissals then through assimilation into the mainstream of analytic philosophy. Judged by external measures, Cavell's attempt has failed. Young students of analytic philosophy nowadays are far more likely to be familiar with Kripke's Wittgenstein on Rules and Private Language and the works it has inspired than with "The Argument of the Ordinary", or for that matter any of Cavell's other works. Indeed, I have more than once heard of courses in the philosophy of language that begin with Kripke's book, or use it as a primary text, and skip even the (relevant remarks of the) Investigations, not to mention Cavell's interpretation of it.
And this is perhaps as it should be. It is arguably impossible to truly understand the later Wittgenstein, or anyway Cavell's Wittgenstein, unless one has first found attractive, or at least experienced the attraction of, some theoretical account such as Kripke's. Finding some such account attractive, only to discover upon further reflection that it falsifies what it purports to explain and cannot ultimately be made sense of, and that no attempt at fixing it is satisfying either, for the problem lies not in this or that particular detail or theoretical commitment, but rather in the very ambition to account theoretically, objectively, and once and for all, for the sense of this or that, and more generally for our capacity to make and perceive sense -all that is essential, I believe, for an understanding of both Wittgenstein and Cavell. The problem, however, is that finding satisfaction in theoretical accounts of fundamental issues of the sorts philosophers have tended to grapple with is easier than working through such accounts to the transient recovery of what one couldn't have failed to know, and finding (transient) satisfaction in that. And young students of analytic philosophy are not likely to be encouraged nowadays to move beyond theorizing, in a Wittgensteinian or Cavellian direction. I hope this paper will provide such encouragement to some.
A significant problem in focusing on Cavell's response to Kripke's book on rule-following and private language is that so much has already been written about the latter. It seems that by now almost everyone agrees that Kripke was in one way or another wrong about linguistically-expressed meaning and its conditions, and wrong in his interpretation of Wittgenstein; and though many of those responding to Kripke have gone even farther than he did in disavowing any pretense of getting Wittgenstein right, 2 there have also been those who argue that if only Kripke had been right about Wittgenstein, he would have been right about linguistic meaning as well. 3 It might therefore seem that there couldn't be much value in the presentation and discussion of yet another critical response to Kripke. Without denying the interest and even partial validity of some of those other responses to Kripke, there is this fundamental and crucial difference between them and "The Argument of the Ordinary": that those other responses to Kripke are, essentially and exclusively, theoretical responses to what is taken to be a purely theoretical issue, or perceived difficulty. Cavell's response, by contrast, is at the same time also existential, or personal, in the sense that the thinker -here, Cavell -does not forget, and finds a way of acknowledging in her or his work, the fundamental fact that her or his own thinking (and writing, or otherwise addressing herself or himself to others and responding to others) constitutes an instance of what she or he is thinking and writing about. Moreover, this unique achievement of Cavell's response to Kripke is not additional to his argument, but essential to it: it enables him, as we will see, not merely to say, but to show that, and how, Kripke's account falsifies what it purports to elucidate, and thereby to show that the theoretical question of linguistic sense is not truly separable, not even theoretically, from the broadly ethical question of how we relate to others, and how we conduct ourselves in relation to them from one moment to the next.

Stage-Setting: Cavell's Ordinary Language Philosophy Existentialism
Though my focus will be on "The Argument of the Ordinary", I want to set the stage with a passage from a much earlier work of Cavell's. 4 It was written after Cavell, who was, as he reports, "on the road toward a proper dissertation (on the concept of human action)", attended a set of lectures given at Harvard in 1955 by J. L. Austin on "performative utterances", and a seminar that Austin taught at the same time on the subject of excuses. Cavell reports that Austin's work "knocked him off his horse". He left Harvard, took up teaching at Berkeley, began the undertaking of explicating and defending the approach and procedures he found in Austin's work and in Wittgenstein's Investigations, and planned a new dissertation on "the implications of Austin's procedures for moral philosophy"implications, as Cavell puts it, "of the sense that the human voice is being returned to moral assessments of itself". Eventually, that work in moral philosophy formed only part of Cavell's dissertation, and was published as the third part of The Claim of Reason. (Other parts develop Cavell's understanding of the Wittgensteinian notion of "criterion", and its relation to traditional forms of skepticism about "the external world" and "other minds", as well as to the philosophical tendency, and underlying motive, that Cavell came to call "skepticism" -about which I will say more below.) The passage I want to begin with is taken from that third part of The Claim of Reason.
In the passage, Cavell responds to Charles Stevenson giving what is supposed to be an example of an exchange in which one person, A, is expressing a "judgment" whose "purpose" is to persuade another person, B, to do something A wants B to do -in this case, to give a certain speech. Never mind the oddity of speaking of a judgment, moral or other, being made for a purpose. I suppose Stevenson meant to be talking about the act of expressing a moral judgment, which, if I'm right about this, underscores his conflation of what's being said with the purpose for which it is said. In any case, what most troubles Cavell is that on Stevenson's account the nature of the relationship between A and B is not relevant for an understanding of their exchange. In particular, what A says B "ought" to do, is simply what A wants B to do, not what A believes it would benefit B to do, morally or in some other respect. In other words, A's investment in, or care for, B is taken by Stevenson to be irrelevant to an understanding and assessment of A's "ought" and the utterance in which it features. Nor are the history of A and B's relationship, and the particular circumstances of the exchange, taken to be relevant for an understanding and assessment of A's "you promised" and of B's response. Here is Stevenson's example: A: You ought to give the speech, as you promised. B: That is unfortunately beyond my power. My health will not permit it.
And here is Stevenson's gloss on that example: This example deals with the consequences of a judgment's influence. A is endeavoring to influence B to give the speech. If B's reply is true, then whatever influence A's judgment may have on attitudes, it will not have the further consequence of making B speak. Realizing this, A will be likely to withdraw his judgment; he sees that it cannot have its intended effect.
[…] A may withdraw his judgment not merely because it will fail to serve its original purpose, but because it may have effects which he, in kindness, does not desire. It may lead B to be perturbed about his disability. (Stevenson 1944, 126) 5 To this, Cavell immediately responds with a list of questions; and one can feel his exasperation, but also his sense of urgency -the sense that such moments represent philosophy literally losing its mind: Does A assume that B has forgotten the promise? Doesn't take it seriously enough? Doesn't realize that what he said was legitimately taken as a promise? If so, why not tell him? If not, then why remind him of the fact? Does A not know that B is disabled? Then, when he finds out, does he "withdraw his judgment" because "he sees it cannot have its intended effect" or because he sees it would be incompetent or incoherent not to? And how does he "see" that it cannot have its intended effect? Because he sees that B is disabled? Then are we to imagine that A goes to the hospital to visit B, and after seeing both of B's legs in traction, says, "You ought to give the speech"? Or is the disability less obvious, so that A is in some doubt as to whether B's condition is as serious as he says? Then how does he "see" or "realize" that his judgment will not have its intended effect? Perhaps he sees that B is adamant; that might be a clear case of "realizing (finally, no matter how hard you try) that your judgment cannot have its intended effect". But we've forgotten that speech in our bewilderment. Was it important? Important enough so that you are willing to urge B to risk his health to give it, or go there in a wheelchair if necessary? Then B's reply "My health will not permit it" is not enough to make you "realize" that your judgment will not have its intended effect. And if the speech is that important then does B not know this? And if he does, then has he done nothing about it, having become ill? Has he, for example, not tried to find or suggest a replacement, or have the meeting rescheduled, or dictated a speech which could be read? If that would be uncalled for, then why is it so important that he give the speech? Why ought he to? But enough. The speech is not important; it doesn't exist. And neither does a moral relationship exist between these people […] (CR,285) I can think of any number of similar moments in Cavell's earlier, more youthfully combative work. And it was through such moments in Cavell that I began to see the ethical-existentialist dimension of ordinary language philosophy, and thereby came to believe that I could express myself philosophically -by which I here mean, as a professional philosopher and teacher of philosophy -and mean what I say. In my own work, I have sometimes responded in similar ways, and with something like the above tone, to what are supposed to be examples of human conversation that have been used in the works of contemporary analytic philosophers -for example, in theorizing about knowledge.
What's happening in philosophical examples such as Stevenson's? Driven by certain theoretical ambitions, and guided by certain theoretical commitments, however vaguely or partially articulated, the philosopher produces an example of a piece of human conversation, in which the protagonists utter words that it is very hard to imagine any real, competent speaker uttering; and hard not primarily because people, as a matter of empirical fact, don't talk that way, but because if anyone did utter those words under the circumstances the philosopher invites us to imagine, it would be very hard to understand them -to see what they mean, how their words are to be taken. As Cavell puts it right before presenting Stevenson's example, "it baffles imagination altogether when we try to conceive of persons saying what Stevenson gives them to say" (CR,284).
In other cases, we could imagine a context in which the words uttered by the protagonists of the philosophical example would make sense. But, in my experience, it then invariably turns out that the words thus understood do not actually support the philosopher's professed theoretical ambitions. 6 For example, in response to John Hawthorne's presentation of what he calls "the lottery puzzle" (Hawthorne 2004), which centrally concerns our "knowledge" of future contingencies, I have recently argued that no natural, genuine use of the word 'know(s)' in conversation about the future is puzzling -standing in apparent tension with what you might call our metaphysical finitude and epistemic fallibility, and calling for a sophisticated theoretical treatment -in the way Hawthorne would have us suppose (Baz 2017, 17ff.) 7 The real puzzle is what leads philosophers to give such false accounts of themselves. For, as competent employers of words such as 'ought', 'promise', and 'know', they surely must know, even if mostly non-cognitively, the conditions for putting those words to some intelligible use or another -for meaning them in some intelligible way or another -and the sorts of commitments and liabilities one would normally incur in putting those words to this or that use. Surely, Stevenson knows the relevance of the sorts of questions Cavell asks to the competent employment of 'promise', or 'ought', and to its competent assessment; and surely Hawthorne knows that the transcendence of the future, as we might call it -the familiar and inescapable fact that it always could, and occasionally does, frustrate or outstrip even some of our most reasonable expectations -stands in no real tension with our ordinary and normal use of 'know' and its cognates in conversations about the future, but rather is part of the shared background against which such conversations have the sense they have for us. What makes them forget or ignore all of that when they philosophize?
There is an ambivalence, I have found, in Western philosophy in general and in analytic philosophy in particular, toward everyday human discourse, and toward the world of experience of which speech is an inseparable part and apart from which it would not have whatever sense it has for us. On the one hand, it seems that philosophers often find it hard to go very far without using supposed "examples" of everyday speech and everyday situations, perhaps as a way of reassuring themselves and others that they are theorizing about something real and pertinent; but on the other hand, they tend to think of themselves as in the business of uncovering a metaphysical reality, or else a system of semantic rules, that underlies everyday speech and experience, and is theoretically separable from them. They cannot believe, it seems, that anything truly important or enlightening could be found in what Wittgenstein metaphorically refers to as "the grey rags and dust" of everyday situations and everyday experience, and therefore forgo altogether the work of examining them (see PI, § 52). They wish, as Wittgenstein puts it, to "penetrate", or "see through" (durchschauen), phenomena (PI, § 90)phenomena such as knowledge, understanding, intention, commitment, promises, and so on -as those come into view in the sorts of things we normally and ordinarily say in humanly significant situations; because they take our ordinary and normal use of words to be no more than an effect and indication of something else that they imagine as lying behind it; and it is that something else that their theories are supposed to capture. What the philosopher wishes to penetrate is not this or that particular phenomenon, but all phenomena -the phenomenal world altogether, if you will: he is ready to trade, in effect, the world he "converses with" or "lives through" -that is, the world he experiences and responds to, in words or otherwise, prior to any theoretical reflection -for the world he "thinks". 8 Except that anything we might sensibly call "thinking", including philosophical theorizing, is itself a mode, and modulation, of our being-in-the-world, and depends on a suitable worldly background for whatever sense it may be found to have.
Wittgensteinian criteria articulate what ordinarily and normally counts for us -that is, matters to us -in some particular way, and the conditions for things thus counting; and the Wittgensteinian elicitation of criteria, and more broadly the practice of ordinary language philosophy as I understand it and have found it practiced in Cavell, are meant to help us find our way back from, or out of, philosophical difficulties that arise, precisely, when that has been ignored, or forgotten, or taken to be inessential to philosophical enlightenment.
It is important to note, however, that not every attempt to move beyond the ground of agreement that (Wittgensteinian) criteria articulate is necessarily harmful. On something like the contrary, Cavell emphasizes that meaning our words differently from how we have hitherto meant them, "projecting" them more or less creatively into "new contexts" -and in this sense moving beyond the guidance of presently shared criteria -is actually essential to the acquisition and competent use of a natural language: we just do not know what it would be for a natural language not to allow, and sometimes require, more or less creative expansions or transformations of a word's range of application -and not only in poetry, or in scientific or religious or ideological revolutions, but also in humdrum projections such as that of 'feed' to parking meters, and 'put' to putting out the light (see CR,.
What cannot coherently be done, and what philosophers have all too often attempted, in effect, to do, is to rely on how our words are ordinarily and normally used, or meant -and so, on what may be called their "meanings" -to ensure the sense of what we are saying by means of them, while evidently attempting to use those words apart from any of the sorts of contexts in which they are ordinarily and normally used. The attempt to move beyond presently shared criteria is liable to become harmful, in other words, and more specifically to get us entangled in the sorts of traditional philosophical difficulties to which Wittgenstein responds, when it becomes what Cavell calls "the repudiation of criteria" -that is, when, rather than being motivated by some particular need for expanding or transforming the present expressive powers of our words, it is motivated by what Cavell calls "skepticism".
"Skepticism", as Cavell came to use the term, may roughly be characterized as the refusal to accept, or to acknowledge, our responsibility for the meaning, or meaningfulness, of our words, and hence for the intelligibility of our world. The age-old philosophical wish to (be able to) speak about what Kant calls "the world as it is in itself" -that is, about a world that is wholly independent of our ways of making sense of it, and hence also of the worldly-historical conditions of our doing so -thus comes to be seen, in Cavell's Wittgenstein, as the flipside of the wish not to be implicated by those ways and conditions, and partly responsible for their maintenance. 9 "We understandably do not like our concepts to be based on what matters to us", Cavell writes in "The Argument of the Ordinary", "[for that seems to imply] my being responsible for whatever stability our criteria may have, and I do not want this responsibility" (TAO, 92). That does not mean, however, that we couldn't come to accept, even embrace, that responsibility.
The philosopher to whom Cavell's Wittgenstein is responding wishes to be able to leave behind altogether the worldly conditions of making this or that sense with one's words -to transcend at once the situatedness of speech, and the countlessly many ways in which it situates us. He feels, as Cavell memorably puts it, that "he must say and think beyond these [worldly] conditions [of sense]; he wants to speak without the commitments speech exacts" (CR, 215). But he still wants to (be able to) talk about, understand, and explain promises, knowledge, justification, meaning, understanding, truth, and so forth. And then he finds himself entangled in difficulties that are rooted precisely in the expectation, or fantasy, that that should be possible.
As a basic, underlying motive -a natural aspiration to be disburdened of our endless responsibility for the meaningfulness and intelligibility of anything we say or do -what Cavell calls "skepticism" cannot be refuted (at least not by way of theoretical argumentation). All that can be done is to point out, again and again, the places where, and the ways in which, it leads us into trouble rather than to true satisfaction, and thereby to make it less attractive. More specifically, ordinary language philosophy is called for when a stretch of philosophical speech -be it the philosopher's own or one of the philosopher's protagonists' -strikes you as making no clear sense, or anyway no sense that would serve the philosopher's professed aims and commitments, and as relying for its apparent sense just on the familiarity of the philosopher's words and, as Wittgenstein has taught us to recognize, on this or that picture that we have formed for ourselves and which has been holding us captive. The ordinary language philosopher's aim would then be to show that that is what has happened, and that the philosopher's difficulties are, in a sense, self-inflicted -not inherent to the phenomena he set himself out originally to understand. But since there are no rules for ensuring or ascertaining the presence, or absence, of sense -we cannot so much as make sense of the idea of such rules, for if anyone tried to formulate such rules (and how otherwise could they be consulted or appealed to?), there'd be different ways of making sense of those formulations (or failing to), and we'd need further rules to ensure that we are doing that right... 10 -the success of the ordinary language philosopher's proposed diagnosis is always, in the end, in the hands of those to whom it is addressed. At the same time, however, since ordinary language philosophers address themselves to competent speakers, and proceed, when they proceed well, by way of the deliberate assembling of reminders of things that competent employers of the words in question could not fail to already know, their practice is not only sometimes called for, but also wellgrounded -as well-grounded as anything we do or say can sensibly be expected to be. 11 Cavell's storm of questions invites us to project ourselves imaginatively into the sorts of situations that Stevenson's example is meant to exemplify and that his theory is supposed to illuminate, and thereby to remind ourselves of what such intersubjectively significant moments are like for those who inhabit them, and of the worldly background against which such moments, and the words employed in them, mean what they mean for us. Cavell's implied claim is that if you don't know that, and how, the sorts of questions he's raising may be pertinent to the understanding and moral 10 This much we've known at least since Kant pointed it out in the Critique of Pure Reason (Kant 1998, A132/B171-173), and then worked out some of its implications in the Critique of the Power of Judgment (Kant 2000, 169; see also sections 20-22 and 38-40). I discuss this Kant-Wittgenstein-Cavell connection in Baz (2016). 11 In this way, the justification of the philosophical appeal to ordinary language parallels the justification of aesthetic claims in Kant's Critique of Judgment: both appeal to what is taken to be a ground of agreement apart from which we would not be able to communicate by means of language (see Baz 2016). assessment of the exchange Stevenson invites us to imagine, you do not know what 'promise' or 'making a promise', or 'ought', or any other morally significant term, means, and therefore cannot reasonably hope to make contact in your theorizing with the intended object(s) of your theory. And the broader lesson, as I understand it, is that if you attempt to theorize from a perspective that is imagined to be altogether outside that shared world of background, and blocks it from coming into view, then you are bound to find, with Stevenson, that human discourse is nothing but an endless series of attempts to exert literally senseless influence on one another, or for securing "agreement" on nothing we can make sense of. And this now brings us to Kripke.

Kripke's Picture
Since the argument of Kripke's Wittgenstein on Rules and Private Language has been rehearsed numerous times and will be familiar to many, at least in outline, and since what really matters is not the argument itself but rather its setup, I will mostly skip the part of formally rehearsing it. What I think would be useful to do before we turn to Cavell's response to Kripke, is to sketch the picture that informs that argument and sustains its apparent sense. For, as we shall see, and despite Kripke's express intentions, it is precisely that picture, rather than our shared criteria, that lends his argument whatever intelligibility and appeal it may be found to have. Having said that, I should add that I have found it impossible to so much as even sketch Kripke's picture without relying on philosophical jargon; and since I think each of those pieces of jargon is suspect, I will use quotation marks whenever I rely on one of them, in order to register my suspicion.
On Kripke's picture, there are literally countlessly many "meanings" words (and other signs) can "denote". Put differently, for any given word, there are literally countlessly many "meanings" it could "denote". In order to understand a stretch of human speech, you have to find out what the speaker "means by" each of her words, where that is a matter of finding out what particular "meaning" each of her words "denotes" (cf. Kripke 1982, 7).
The "meanings" words can "denote" are "rules"; and each of those "rules" determines, for any word that "denotes" it, its "correct" "use", or "application", under any set of circumstances. For the purpose of Kripke's "skeptical argument", each of the "meanings" or "rules" may be taken, and is in fact taken by Kripke, to be perfectly and finally determinate, in the sense that what it requires of a word that "denotes" it, for that word to be "used" "correctly", is not in question and not open to competing interpretations. 12 I should note, however, that it's not really clear how it -the rule itself -is to be identified, since presumably any identification of it by means of words (or other signs) would be indeterminate.
There is, importantly, no role for judgment on Kripke's story, in moving us from a "rule" to its "correct" "applications". And here it matters that Kripke's paradigm for the "rules" that constitute the possible "meanings" of words are algebraic functions that, for any proper input, determine one, and only one, output. (Not that there is no role for judgment to play in computing the output of algebraic functions, or in mathematics more generally; Wittgenstein's example of the deviant student is meant to get us to see that even here judgment plays an ineliminable role, and that it's just that here, not only is "our agreement in judgments" normally and properly taken for granted, but it may also be considered partly definitive of mathematical practice -part of what makes the mathematical mathematical, as both Wittgenstein and Cavell suggest (PI, § 240; PPF (part II of PI), § 341; and TAO, 89-91).) When Kripke turns to discuss non-mathematical words such as 'table' and 'chair' (and implies that his account applies to all other words as well), he takes it that there is no significant difference between the "meanings" those words (would normally be taken to) "denote" and the "meanings" that words such as 'plus' (are normally taken to) "denote".
According to Kripke, what's indeterminate -and indeterminate not just epistemically (i.e., from our perspective), but metaphysically (i.e., even from God's perspective) -is which of the countlessly many determinate "meanings" any of a speaker's words could "denote", it actually does "denote". Specifically, Kripke argues that neither my past "use" of the word nor my intentions with respect to it can determine what I mean by it now. My past use of the word can't fix what I mean by it now, Kripke argues, for, being finite, it is compatible with countlessly many "meanings", each overlapping with the others over all of my past "applications" of the word, but determining a different future for it, so to speak, a different continuation of the series of its "correct" "applications"; and in arguing this Kripke is clearly inspired by the mathematical fact that, as he puts it, "an indefinite number of rules (even rules stated in terms of mathematical functions as conventional as ordinary polynomials) are compatible with any […] finite initial sequence [of natural numbers]" (Kripke 1982, 18). And whatever intentions I might have with respect to the word cannot help determine what "meaning" it "denotes" either, Kripke argues, for those intentions would themselves be indeterminate -relying as they necessarily do on being formulated by other words (or signs) that are themselves indeterminate in the "meanings" they "denote". The "skeptical" upshot of all of that, as Kripke famously puts it, is that there is "no fact about me" that "constitutes" my meaning this ("meaning") rather than that ("meaning") by any of my words (cf. Kripke 1982, 9, 11, 13, and 21).
But why should we picture the meanings of our words as rules of "correct" "use" that are theoretically separable from these words' familiar use in the language we share? Wouldn't it be better -truer to our ordinary and normal use of "the meaning of a word", to the ordinary and normal criteria for "knowing the meaning of a word", and far less misleading theoretically -to follow Wittgenstein, and think of words as instruments of the language and of their meanings as a matter of their ordinary and normal uses -the contributions they normally make to significant utterances, or linguistically articulated moves, within shared practices? Why not begin with the undeniable phenomenological fact that, for us competent speakers, the meaning of 'plus', for example, may be seen in (rather than inferred from) how it is ordinarily and normally used in the mathematical and other practices in which we use it? And wouldn't it also be bettertruer to our ordinary and normal use of "how so and so means this or that word" or "what so and so means by this or that word", and far less misleading theoretically -to think of how someone means her words, or what she means by them, as a matter of how, given her and her words' history, and given her present circumstances, her words are most reasonably to be taken and responded to, which in most cases is (taken to be) obvious and does not come into question? These questions suggest an altogether different approach from Kripke's -though, as we will see, perhaps not different from the approach he meant to be following -to the elucidation of the sorts of phenomena he purported to elucidate. They suggest that the best way to dissolve Kripke's difficulties is to remind ourselves of, and become clearer about, what, as competent speakers, we already know, as opposed to engaging in theoretical construction and argumentation that proceed on the basis of questionable theoretical commitments, and run the risk, as we shall see, of losing contact altogether with what they purport to elucidate.
In any case, having identified his "skeptical problem", Kripke then offers a "skeptical solution" to it: having found that there is no fact about me that constitutes my meaning this rather than countlessly many other thats with any of my words, Kripke moves on to tell us how it is that we are nonetheless able to use language to communicate with each other as smoothly and effectively as we do. Perhaps unsurprisingly, given that the skeptical problem he has identified does not so much as even make sense apart from a questionable picture and a theoretical construction erected on its basis, Kripke's "solution" is, at bottom, rather pedestrian: as long as we "agree" nearly enough with each other in how we are "inclined" to "use" our words, the "skeptical problem" shouldn't, and evidently doesn't, bother us.
Slightly more substantive, but as we will see therefore more problematic, is Kripke's account of what, despite his "skeptical problem", "licenses" us to speak about "what so and so means by this or that word". Kripke proposes that as long as we find that another person's "use" of a word "agrees" with how we are "inclined" to "use" it, at least nearly enough, we are "entitled" to say of that other person that she means by that word what -that is, the same "meaning", or "rule", that -we take ourselves to mean by it. And when we find that a new initiate into our community of speakers "uses" her words nearly enough as the rest of us are "inclined" to "use" them, we are "licensed" to say of her that she means by those words what -that is, the "meaning", or "rule", that -we take ourselves to mean by them (cf. Kripke 1982, 90-91). 13 Except that we are not supposed to know, or to be able to tell, what we mean by those words, which means that the communal agreement Kripke envisages for us is, by his own lights, empty of content, meaningless.
conditions. 14 And this proposed mathematization of language -or more precisely, this viewing of language through a picture of mathematics as practice-independent and a-historical -means that the history of a language, and the way in which we each inherit that history, and more or less creatively draw upon it and play upon it in putting our words to use in different contexts, also do not come into view in his story.
When Kripke finds that the word 'table' -given our, or anyway his, past employment of it, and notwithstanding what "meaning" he thinks it "denotes" or has meant it to "denote" -could actually mean tabair and, if so, correctly be used to refer to anything that is a table not found at the base of the Eiffel tower or a chair found there, and that it's only "our" "inclination" not to thus "use" it that would render him "wrong" if he used it to refer to a chair in the Eiffel tower, he has not made a shocking skeptical discovery, but rather has given voice to a philosophical invention. This is what you get when you try to hold on to the notion of "meaning something with one's words" while divorcing it altogether from what Cavell calls our "natural reactions" and our agreement, however un-ensured and sometimes fragile, in how we see and respond to things, in what matters to us and how, in what makes what sense to us and howall of "the whirl of organism", 15 the Wittgensteinian "form of life", of which the Wittgensteinian elicitation of criteria seeks to remind us in the face of philosophical difficulty and confusion that result from the attempt to speak altogether outside it. On Cavell's understanding of "skepticism", as presented in the first section of this paper, it is not so much Kripke's skeptic, but Kripke himself, who is being skeptical. Whereas Cavell's skeptic is motivated by the desire to "strip criteria from ourselves", the desire not to be implicated by whatever sense our words may be found to have, Kripke's skeptic tacitly relies on those criteria, but seems "incited to complicate and build upon 14 This has been usefully stressed by Wright (1980). The basic point is instructively summarized in "Critical Notice of McGinn's Wittgenstein on Meaning": "[I]t is the basic agreement which sustains all rules and rule-governed institutions. The requirements which our rules impose on us would not be violated if there were not this basic agreement; they would not so much as exist" (Wright 2002, 127). 15 See Cavell 1969, 52 (in the essay "The Availability of Wittgenstein's Later Philosophy").
[those criteria] at will (as 'tabair' does)" (TAO, 88), with no apparent motive other than theoretical playfulness. And whereas Wittgenstein's "scene of instruction" -with the deviant student who finds it natural to go on differently from (some of) us -is meant to reveal the dependence of our criteria, hence concepts, on an agreement in natural reactions, in how we naturally perceive and respond to things, Kripke's skeptic "dissociates criteria from the realm of what Wittgenstein calls 'our natural reactions'" (TAO, 88). 16 As a result, Kripke's mathematicized concept of tabair is not the metaphysical equal of our ordinary, non-mathematicized concepts of table and chair (and found in) -as his skeptical argument presupposes -but rather is wholly parasitic on those concepts, and on a common world of background apart from which they would not be the particular concepts they are. 17 Cavell shows this by way of an illustration of the elicitation of Wittgensteinian criteria, and in showing it, pulls the rug from under the whole of Kripke's theoretical construction at one fell swoop: Our criteria for a thing's being a table -part of the grammar of the word "table" -is that this is what we call "to sit at a table". That we sit at it this way to eat and that way to write (closer in, resting an arm on it if the writing is by hand) and the other way to position a coffee cup, are all criteria for calling it a table. You can sit on a table, or on a flagpole, but not in those ways; and unlike a flagpole, a chair can be used for a table (as well as vice versa) or a tree stump serve as a table. That will not make chairs and tree stumps tables; but put some legs under it, and you can get a table. It is obviously essential, from what has been said, that the table top be a fairly smooth and quite horizontal surface supported by one or more legs generally waist high. Is this all that is essential? It is all, even a little more than, my dictionary gives. So far nothing has been 16 David Lewis proposes to respond to Kripke's "skeptical problem" by way of a distinction between "natural" and "unnatural properties", and more broadly between natural and unnatural ("perverse") ways of going on (see Lewis 1983, 375-377). The basic intuition is sound, but, as I'm about to argue, Lewis's idea that the distinction between natural and unnatural ways of going on with our words may be drawn independently of human judgment (Lewis 1983, 376) -of what we find natural, and under what conditions -is hopeless and, from Cavell's perspective, more skeptical than the "problem" it is supposed to "solve". 17 Rupert Read is therefore absolutely right when he writes, in his Cavell-inspired critique of Kripke that is very much in the spirit of the present paper, that Kripke's "skeptic can only propose the notion of 'quaddition' by borrowing our understanding of addition" (Read 2012, 85). said about scale or proportion or shape, but a disk whose radius is the length of the long side of a football stadium, supported ankle high, will not count (is it more definite that one whose diameter reaches from where I sit to just beyond the horizon, so that I cannot see the person sitting as it were directly opposite me, will surely not count?); nor will a half-inch rail the "right" height running the length of a dining room (though it, and a cross rail, may, for certain purposes, represent a table). Somewhere between the stadium possibility and the rail there will be doubtful cases. As the Kripkean skeptic properly observes […] I probably did not think of the tree stump or the stadium possibilities nor of that of the rail when I learned the term "table", nor for that matter ever thought explicitly that the support of a table must rise from the floor (not be propped from a wall, in which case it is a shelf), settling roughly (when you sit at it) at the waist [...]. I can testify that I have never thought that a shape for a table top is ruled out -even when an object is within the rough scale and proportion and height range of a table -whose edge is lined continuously with triangular projections just longer than long human arms, narrower at the base than the base of a cup, and too close to one another for a thin human being to squeeze between. You couldn't place things on it (by hand). Though I did not think of these things when I learned the word "table", I can think of them now, I can bethink myself of them, and more. I can do this because of our agreement in criteria. (TAO,(93)(94) I said this passage pulls the rug from under Kripke's theoretical construction. For obviously, there is no rule that determines all of the applications of 'table' (positive or negative) Cavell imagines, and countlessly many imaginable others; but equally obviously, that has never bothered or hindered us in our use of the word, and reveals no lack in our concept of (being a) table, and no problem. 18 As Cavell's elicitation of criteria effectively shows, our (participatory) familiarity with the ordinary and normal use of the word 'table' -and hence with the place and significance of tables in the world we share with others -is all we need to go on. That, and a sense -sufficiently, even if not always, shared with others -of what going on with the word in faithfulness to that use requires, under more or less novel circumstances, and with more or less novel interests and needs. 19 It is not hard to imagine one of us moved to challenge one of Cavell's judgments (or proposed criteria). Should anyone be moved to do so, however, she had better not appeal to her past use of 'table' or to what she has meant by 'table' (on this or that occasion or set of occasions); rather, the appeal, if competent, would be to further criteria that the challenger takes herself to share with Cavell, or more broadly to other features of the common world of background, the Wittgensteinian "form of life", that she takes herself to share with him. And it might then turn out that there is actually no real disagreement. For example, should the challenger remind Cavell that the object he imagines with the triangular projections might (turn out to) be a sacred table, or a rare and delicate one, and that the triangular projections might have been installed to protect it or prevent people from using it, it is more than likely that Cavell would happily accept that reminder. Far from undermining his Wittgensteinian "vision of language", such a reminder would further reinforce it: it would be a reminder that criteria are context sensitive (cf. PI, § 164), and that our agreement in criteria, and in the world of background they articulate and bring out, is even more intricate and complex and rule-defying than Cavell's little sketch might have suggested. It is a piece of Wittgensteinian "grammar" -a sort of necessity deeper than any imposable by rules -that if the object with the triangular projections is (to properly count as) a table, then there has to be some such explanation for the presence of the projections.
On the other hand, if it should turn out that some of us disagree with some of Cavell's judgments and that the disagreement is genuine and not resolvable by the further elicitation of criteria, that too would not undermine Cavell's Wittgensteinian vision of language. It would simply underscore the fact that nothing ensures our agreement in judgments (and criteria) -in particular, as has already been noted, no rules for the use of our words (could) ensure that agreement, for even if there were any, our agreement in how we 19 This means to echo Kant's derivation in the Critique of the Power of Judgment of our unavoidable reliance on "sensus communis" in the application of concepts, and "if cognitions are to be able to be communicated" ( § 21; see also § § 38-40). See further Baz (2016). understood them would itself presuppose that agreement -and that it may always be found to only go so far. And that -the fragility of our mutual attunement -is something that Cavell, more so than any other reader of Wittgenstein, has emphasized as no less essential to Wittgenstein's "vision of language" than the deep and pervasive attunement apart from which we would not have been able to communicate by means of language. 20 One thing that makes "The Argument of the Ordinary" a masterpiece, in my view, is that it is everywhere alive to that fragility -especially as it manifests itself in philosophy, and in the present case in Kripke's disagreements with Cavell -and acknowledges it at every turn.
But isn't all that precisely what Kripke is getting at in his "skeptical solution"? Isn't that precisely his point, when he proposes that "the success of our practices" of assigning meaning to each other's words, and presumably all of our other linguistic practices, "cannot be explained by "the fact that we all grasp the same concepts", but rather "depends on the brute fact that we agree with each other in our responses" (Kripke 1982, 109; see also 97)? No. There is a Wittgensteinian insight that Cavell has articulated more clearly and forcefully than anyone else, and which Kripke is groping toward in his book on Wittgenstein, and particularly in such passages. 21 However, Kripke's theoretical commitments and ambition lead him to misplace, and mischaracterize, the agreement in judgments (and in criteria) that Cavell's Wittgenstein emphasizes as a condition for our being able to communicate by means of language (see TAO, 67). As Cavell suggests, Kripke's "communal agreement" does not reach "the depth of confidence Wittgenstein places in our 'agreement'" (TAO, 67). Cavell's "agreement", as I've been saying, refers to agreement in what makes (what) sense to us, and how, and under what conditions. It is agreement among perceivers and responders to sense who perceive and respond to 20 "The only source of confirmation here is ourselves. An initial disagreement may be overcome […] But if the disagreement persists, there is no appeal beyond us, or if beyond us two, then not beyond some eventual us. There is such a thing as intellectual tragedy" (CR,19). 21 One philosopher on whom Cavell's insight, as encapsulated in his famous "whirl of organism" passage (cf. footnote 15 above), has had significant impact, is John McDowell. See, for example, "Noncognitivism and Rule-Following", in McDowell 1998, 198-218. each other as perceivers and responders to sense, as fellow inheritors of a common language, shared practices, and a "world of meanings", as Merleau-Ponty puts it (1996,193), who draw upon that inheritance in attempting to make sense of each other, of their world, and of themselves. Kripke's "agreement" of "inclinations", by contrast, can, by his own lights, only refer to "agreement" in what sounds we utter, or find ourselves disposed to utter, under what circumstances. 22 But if the perception and response to sense were not there from the beginning -both the beginning of a human life and the beginning of the philosopher's account -then no such "agreement" or coordinated "behavior" would bring them into being, or intelligibly into the account; 23 and if we do not find ourselves always already participating in a form of life that we share with others and in which, and against the background of which, our words and the words of others have whatever sense they have for us, then, pace Kripke (1982, 96), no "set of responses in which we agree, and the way they interweave into our activities", would constitute such a form of life for us.
Cavell says that in "stripping us of our criteria" Kripke's "matching of inclinations", which is proposed as a "solution" to Kripke's "skeptical problem", "is more skeptical than the problem it is designed to solve" (TAO, 75 and 95). What Cavell means, I take it, is that the "problem" presupposes meaning, and therefore meaningfulness, however misleadingly pictured; for otherwise, there would be nothing for the "skeptical paradox" to threaten, and no (apparent) problem. Nor would there be criteria for Kripke's "skeptic" to multiply and complicate at will. By the time we come to Kripke's 22 Early in his book, Kripke rejects the "dispositional theory" according to which an individual's meaning this rather than that "meaning" or "rule" with a word could simply consist of her being disposed (ceteris paribus) to "use" or "apply" the word in a way that accords with this rather than that "meaning" or "rule" in all possible cases -and in particular in those cases for which the two "meanings" or "rules" determine a different "correct" "application" of the word (1982,(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37). But it is also part of his story that the only way to avoid the inadequacy of the "dispositional theory" is to bring the "community" into the story, and that as long as we consider individual speakers apart from a "community", all we can legitimately say is that they are disposed to "use" their words, "with complete confidence" or with a "feeling of confidence", one way or another (1982,108). 23 This, I take it, is what Wittgenstein is getting at when he writes: "I must begin with the distinction between sense and nonsense. Nothing is possible prior to that. I can't give it a foundation" (PG, § 81, pp. 126-127).
"solution", however, meaning and meaningfulness have been completely banished from the world of speakers to a Platonic realm of eternal "rules" that are separate from the ordinary and normal use of our words. And no "agreement" in the sounds we produce or are "inclined" to produce could bring them back.
Kripke's fateful misplacement of Wittgenstein's "agreement in judgments" and "agreement in form of life" shows itself when he claims to have identified "a certain tension in Wittgenstein's philosophy" (1982, 98). The tension, as Kripke sees it, is that "on the one hand, Wittgenstein's paradox argues that there is no a priori reason why a creature could not follow [Kripke's alternative 'meanings'], and thus in this sense we ought to regard such creatures as conceivable" (Kripke 1982, 98); but on the other hand, it seems that by Wittgenstein's own lights, "we should be unable to understand 'from the inside' […] how any creature could follow [such rules]. We could describe such behavior extensionally and behavioristically, but we would be unable to find it intelligible how the creature finds it intelligible to behave in this way" (ibid.). But I find that what's unintelligible is not so much the behavior of a creature who would follow Kripke's alternative "meanings", but Kripke's scenario itself. What exactly does it invite us to imagine, or to "conceive"?
Let's try to imagine the creature whose use of 'table' accords with tabair (table outside the Eiffel Tower or chair inside it) rather than with table, and whose use of 'chair' accords with chable (chair outside the Eiffel tower or table inside it) rather than with chair. By hypothesis, that creature might have been living among us and using 'table' and 'chair' just as we do, showing herself alive to all of the context-sensitive criteria that guide our use of these words, and giving no indication whatsoever that just the location of the object matters to its being, or not being, a table, or a chair. And then one day, we find ourselves together with that creature inside the Eiffel tower, or talking about the tower, and the conversation turns to tables and chairs. And? What are we to imagine here?
Since, as has already been noted, the "concepts" of tabair and chable are parasitic on the ordinary concepts of table and chair, there are only two basic options for how we imagine the case, it seems to me. The first option is that the creature continues to identify, and to distinguish between tables and chairs just as we do, and continues to conduct herself in relation to tables and chairs just as we do. In short, there is no indication, and if we asked the creature she herself would deny, that the object's presence in the tower makes any difference to how she sees or relates to it. The only difference is that (we now find that) she uses the word 'chair' to refer to any table that is found inside the tower, and uses 'table' to refer to any chair that is found inside the tower. It's like a peculiarly geographically-limited but persistent form of malapropism that consists of replacing 'chair' with 'table' and 'table' with 'chair' whenever, and only when, the creature wishes to refer to objects found inside the Eiffel tower; except that when we try to gently correct the creature, as we normally would in response to cases of malapropism, she denies she misspoke. When we then ask her, "Why are you insisting on calling these chairs 'tables' and these tables 'chairs'?", she understands our question, for (we are presently assuming that) she continues to identify tables and chairs as the rest of us do; but all she can say is, "Oh, just because they are found in the Eiffel tower". I assume that after a while, we would give up trying to correct that creature, and would either accommodate ourselves to her peculiar way of talking, or try to avoid talking with her about tables and chairs inside the Eiffel tower (neither of which would be very hard); or we could turn our back on her and seek other partners for conversation. But whatever we would choose to do, it would not be a case of having encountered a creature whose concepts are different from ours; nor would it be a case of finding another creature's behavior, or form of life, unintelligible. It would just be a case of finding someone's peculiar way of talking wholly unmotivated, and slightly confusing and irritating. Thus imagined, that creature would be no more intelligible to herself than she would be to us, as far as her use of 'table' and 'chair' goes.
The other way of imagining what Kripke invites us to imagine, given that his alternative "concepts" of tabair and chable are parasitic on our ordinary concepts of table and chair, would be to imagine that something about the tower, or its experience, causes the creature to see (and more generally perceive) tables found inside it as chairs, and chairs found inside it as tables. Except that somehow -and unlike normal cases of seeing as, where the perceptual experience is inherently unstable and lasts only as long as we attend to the object in a particular way, and where perceiving A as B is not a case of mistaking A for B -here the perceptual experience is lasting and persistent, and the perceiver is fooled. So really, what we're trying to imagine here is more like a case of persistent and pervasive hallucination than like a normal case of seeing-as. When, inside the tower, we ask the creature to set the table, she becomes confused at first -for she sees us gesturing toward (what she sees as) a chair, and it is surrounded by (what she sees as) tables -but perhaps finally she finds a chair and attempts to set it; and when we invite everybody to come sit at the table, she grabs a table and tries to "sit at" a chair, not understanding why the rest of us grab (what she sees as) tables and sit at (what she sees as) a chair; and so on and so forth. I am not saying that such complete perceptual transposition is truly imaginable, let alone empirically possible. I'm just proposing that trying to imagine something like that is the only other way to try to conceive of Kripke's follower of tabair and chable. And what matters for my present purposes is that, once again and notwithstanding Kripke's express intentions, it seems clear that this case may not aptly be described as that of a creature whose 'table' and 'chair' have different meanings than they do for us, or embody concepts that are different from ours. And once again, however confounding that creature's condition might be, she is no more intelligible to herself than she is to us, at least as far as her use of 'table' and 'chair' goes.
And this brings us once again to "the common world of background" that is missing from Kripke's story, and to the agreement in "natural reactions" apart from which there would be no common world of background for us. Unlike the people of a foreign culture, who, as Wittgenstein reminds us (PPF, § 325), might be unintelligible to us because their ways of looking at and seeing and responding to things, and so their world, and so the meanings of their words or what you might call "their concepts", are so different from ours, 24 the presumed possessors of Kripke's altern- 24 Wittgenstein invites us to imagine that someone could find the people of another country utterly "enigmatic" and their traditions "entirely strange", even while "die Sprache des Landes beherrscht", which the English translators render as "having mastered the country's ative "concepts" would be, not unintelligible, but simply bizarre, or pathological, precisely because they must be assumed to actually possess our concepts, and therefore must be assumed to largely share our ways of looking at and seeing and responding to things.
Not that it is impossible to imagine, or to conceive of, what Kripke purportedly invites us to imagine or conceive of -namely, two words (or "the same word", but in two different dialects or idiolects) that, up to a point, seem to have the exact same use, hence meaning, but whose respective uses diverge beyond that point. On the contrary, not only is such a case easy to imagine, but there are any number of examples of actual such cases. The problem for Kripke is that those cases too reveal the inadequacy of his picture.
Take for example the English word 'feed'. You can (aptly be said to) feed a baby or a lion; but you can also (aptly be said to) feed a parking meter. 25 As long as we're talking about the baby or the lion, there is a word in Hebrew ‫)'להאכיל'(‬ that seems to translate 'feed' exactly. But that Hebrew word would never be used to refer to putting quarters in a parking meter, or to complimenting someone and thereby making their pride grow. And it's not just that Hebrew speakers happen not to use that word when talking about meters and pride; the word would not seem or sound right if used in those contexts. And yet there is surely no Kripkean "rule" that 'feed' has always "denoted", and which has determined -long before parking meters, or even cars, were invented -that putting quarters into parking meters would be a case of feeding; 26 nor is there another language". But I think the German "beherrscht" would be better translated in this case by means of "controlled", or "commanded". After all, Wittgenstein himself says that "to imagine a language is to imagine a form of life" (PI,§ 19), and that the agreement apart from which we could not communicate by means of language -that is, would not have (what would properly count as) a language -is "agreement in form of life" (PI, § 241). This seems to me to imply, what I anyway think is true, that you could not really master the language of a country whose inhabitants you found so utterly enigmatic and whose traditions you found entirely strange. 25 This example is taken from, and the following discussion is inspired by, Cavell's "Excursus on Wittgenstein's Vision of Language" (CR, 181ff). 26 And what about adding money to the new parking meters that take credit cards? Feeding or not feeding? When I ask my English-speaking students, some of them think "feeding" still fits, and others think it no longer does. I take it to be obvious that even those who find it still fits, would not have found that it does, if it weren't for the intermediary step with quarters. This brings out the crucial role of historical contingency in shaping the present meanings, or reach, of our words. "rule" that the Hebrew word has always "denoted", and which has similarly determined, even before cars and parking meters were invented, that putting quarters into parking meters would not correctly be referred to by means of that word.
The difference between the English word and the Hebrew word that makes the former fit the case of parking meters and the latter not fit it, is not a difference of underlying rules. It is rather what Wittgenstein describes as "a subtle aesthetic difference" that is a matter of "the field of a word", of "all of the widespread ramifications effected by each of the words" (PPF, § 297). As Wittgenstein notes, at first one might only be able to say "This word fits, that doesn't"; but, upon further reflection, "a great deal can be said" even about such subtle aesthetic differences (ibid.). For example, when I first asked myself why it is that the Hebrew word that would commonly translate 'feed' does not fit the case of parking meters (and other machines that can be fed), I found myself thinking that the Hebrew word suggests active participation on the part of the recipient of the action, which is lacking in the case of the meter. And then I noticed that whereas in Hebrew there is an active verb ‫)'לאכול'(‬ that would be used to refer to what the recipient of the feeding does, and that verb has the same root as the verb that refers to what the provider of the food does -which indeed suggests that both parties are actively participating in the transaction -in English, by contrast, what the recipient of the food does would either be referred to by means of the passive form of 'feed' (i.e., 'being fed'), or by means of a verb with a different root (commonly, 'eat'). The Hebrew word that would be used in the case of the meter is ‫,'להזין'‬ which in many common contexts would be used to translate (and would be translatable by) 'nourish', and which does not imply active participation on the part of the recipient. And then I thought that the English word 'nourish' would not fit the case of the meter because it suggests a certain (nourishing, benevolent) attitude on the part of the active provider that is not suggested by 'feed', 27 and which would be out of place in the case of the meter… I do not expect that everyone would accept everything I've just said in my sketchy attempt to make sense of how we use and do not use the English and Hebrew words; but I also wouldn't be surprised if many did accept it, or something along its lines. Either way, I can think of no better way of bringing out all that has gotten lost or forgotten in Kripke's account, than the contrast between the great deal that could be said even about such subtle aesthetic differences between words, not to mention all that could be said about differences that aren't so subtle, and the virtual muteness of Kripke's "agreement" in "inclinations", of "This is simply the word I am (or we are) inclined to utter here".

The Argument of the Ordinary II: Kripke's Real Skepticism
I've already mentioned Cavell's saying that Kripke's "skeptical solution" strips us of our criteria, and therefore is "more skeptical than the problem it is designed to solve" (TAO, 95); and I myself have suggested that, on Cavell's understanding of "skepticism", the real skeptic is not Kripke's skeptic, but Kripke himself. The point can also be put by saying that Kripke's "solution" strips us of our criteria in two ways. The first way it strips us of our criteria, which was the focus of the previous section, is in leaving out of view altogether the common world of background -what Wittgenstein calls "form of life" -apart from which our words would not have the sense they have for us; the agreement in "natural reactions", and more broadly in what makes (what) sense to us and how and under what conditions, without which there would have been no common world of background for us; and the (Wittgensteinian) criteria that seek to articulate and bring into view that background and recover that agreement, in the face of difficulties that result, precisely, from our losing touch with them when we "philosophize". The second way, or sense, in which Kripke's solution, and really his whole discussion, strips us of our criteria has to do with Kripke's own words -the words that are supposed to set up his "skeptical problem" and articulate his "solution" to it.
Early on in "The Argument of the Ordinary" Cavell notes "a discrepancy between Kripke's and Wittgenstein's use of the major terms Kripke cites from Wittgenstein" (TAO, 68). For Wittgenstein, none of the terms he uses in his so-called "rule-following" remarks -Cavell gives a partial list that includes 'obedience', 'following', 'interpretation', 'regularity', 'doing the same', 'ordering ', 'custom', 'technique', 'example', 'practice', 'explaining', 'understanding', 'guessing', 'intuition', 'possibility', 'intention'... -"is less or more fundamental than the concept of a rule", Cavell says, "and each of which is to be investigated grammatically (hence by way of eliciting criteria)" (TAO, 68). Kripke, by contrast, treats some of those terms -e.g., 'rule', 'meaning', 'concept', 'intention', 'inclination', 'agreement', 'community'… -as expressing what Wittgenstein would call "super-concepts" among which there is a "super-order" that somehow holds the key to an understanding of language (see PI, § 97); and he uses those words, or rather attempts to use them, apart from any of the criteria that tacitly guide and inform their ordinary, nonphilosophical use. In other words, whereas Wittgenstein seeks to lead the philosopher's words "back from their metaphysical to their everyday use" (PI, § 116), Kripke's use of his words is metaphysical through and through: what replaces our ordinary criteria in guiding Kripke's use of his words, I have suggested, is a picture, or set of inter-related pictures; and pictures cannot ensure sense, as we have already begun to see in the case of Kripke's (attempted) use of 'meaning' and 'concept' (and 'use', and 'agreement'…). One thing Cavell does in "The Argument of the Ordinary" is offer Wittgensteinian grammatical reminders that aim at showing that Kripke uses virtually all of his key expressions -"inclination" (TAO, 95), "following something blindly" (TAO, 71), "now I can go on!" (TAO, 73), and so on -or imagines them used, apart from any of their ordinary and normal criteria. As Cavell notes, even the 'this' and the 'I' in Wittgenstein's "this is simply what I do" -which Wittgenstein says he would be inclined to say when he finds he has exhausted his justifications for how he goes on, and has reached bedrock (PI, § 217) -are, on Kripke's understanding, divorced from the sort of background that would give them sense under normal circumstances. As a result, the 'this' in Kripke's understanding of "This is simply what I do" is empty, "since it counts on a criterion [of identity] that is already rejected" (TAO, 96); and it is not clear what work is done by the 'I' either, "since it seeks to represent a community that does not exist" (TAO, 96). We'll come back to this.
Similarly, I would argue that Kripke obliterates altogether in his account the distinction, and relation, between, on the one hand, "the meaning of a word' -roughly, whatever it is that the word brings with it from one use to another, and which makes it fit for being used, or meant, in certain ways but not others under suitable circumstances -and, on the other hand, "how a word is meant", or "what someone means by it", on some occasion or set of occasions. Kripke's topic is, nominally, the latter. But when someone uses a word -'inclination' for example -and we find it hard to see how he means it, or what he means by it, we do not take it that there is anything like a set of things -all as clear and determinate as addition and quaddition -that one could mean by that word, and that it's only a question of which of those he means, and whether it happens to be the very same one that we are inclined to mean with that word, or to follow in our use of it. Normally, the answer to the question of what someone means by some word -when that question has intelligibly arisen 28 -would take the form, not of a free-standing and fully determinate mathematicized concept, but of a paraphrase that aims to elucidate, well enough for present intents and purposes, how the word, as used by the speaker on some occasion or set of occasions, is to be taken (cf. PI, § 79 29 ).
The elucidatory paraphrase could be given by the speaker herself or by someone else, and in either cases would derive from an understanding of what the speaker is trying to say, or even must be trying to say given her circumstances, apparent aims, and other commitments. Cavell suggests, for example, that given Kripke's use of 'inclination', what he means by it may not aptly be paraphrased by something like "a desire under conscious check", which Cavell proposes as a rough paraphrase of what we normally mean by 28 To insist that with respect to every (competent) utterance, it always makes sense to ask the speaker "What do you mean?" is, as Wittgenstein suggests, "misleading", for "in most cases one might answer: 'Nothing at all -I say…'" (Z, § 4). 29 I'm grateful to Gisela Bengtsson for noting the relevance of that remark of Wittgenstein's to the present discussion.
'inclination', but may rather be paraphrased by something like "a generalized source of energy" (TAO,95). And this -I mean, the fact that this paraphrase roughly but usefully captures what Kripke means by 'inclination' -is not a matter of how we, or some "we", are inclined to use, or mean, that word in general, or with how Kripke normally and ordinarily uses it when he is not philosophizing; nor is it a matter of choosing between fully determinate "meanings" or "rules" one could mean by the word given its past employment. It is rather a matter of seeing how, given the word's history and Kripke's theoretical commitments and ambition, the word as he is using it in Wittgenstein on Rules and Private Language is most reasonably, and charitably, to be taken. This is one of few places where Kripke gets something right, and important, about Wittgenstein's approach to the dissolution of philosophical difficulties, but then fails altogether to be faithful to it in practice. 30 He notes correctly that for Wittgenstein, the way to become clearer about our concept of meaning would be to ask, not such questions as "In what (fact) consists someone's meaning this or that with some word?" (see Kripke 1982, 65, and 72-3; and see also Z, 16), but rather such questions as "When would we say of someone that she meant this or that with her word(s), what would be the basis of our saying it, and what would be the significance of our saying it (what commitments would we normally incur in saying it)?" (see Kripke 1982, 69, and 86-89). 31 Kripke describes the move as that of replacing the philosophical search for "truth conditions" with the search for "assertability conditions" (cf. Kripke 1982, 73, 79, and 86); and that is fine as far as it goes. But then, and partly because his picture and theoretical commitments lead him to confuse "the meaning of a word" and "what someone means by a word", Kripke goes on to make up, in effect, the practice, or set of practices, he purports to describe. He himself notes early on that "[he is] not familiar with an accepted felicitous convention to indicate the object of the verb 'to mean'" (Kripke 1982, 9, fn. 8), which is unsurprising, 30 Another place I've already discussed is his noting correctly the fundamental importance of what Wittgenstein refers to as "agreement in judgment" and "agreement in form of life", which he then goes on to misplace and recast. 31 I discuss the importance of that approach to Wittgenstein's remarks on "aspect perception", as well as what I see as its limitations, in the Appendix to Baz (2017). for, as I've noted, the ordinary and normal "object" of the verb "to mean (by some word 'x')" is either an elucidatory paraphrase, or (in the case of malapropism) another word, not a Kripkean eternal, and fully determinate "meaning", or "concept", or "rule" that is separable from the word's ordinary and normal use. The philosophically common "By 'plus' I meant quus (or plus)" or "By 'green' I meant green" (cf. Kripke 1982, 9) is a fairly recent and rather misleading philosophical invention, not an instrument of ordinary, non-philosophical discourse. 32 And this means that, by his own admission, Kripke's "skeptical solution" consists of describing the "assertability conditions" that "license" our use of an expression that has no use in ordinary, non-philosophical discourse! When Kripke invites us to give up the search for the "'truth conditions' or 'corresponding facts' in the world […] that make a statement like 'Jones, like many of us, means addition by "+'" true", and instead to "look at how such assertions are used" (Kripke 1982, 86; see also 69), he is sending us in pursuit of a philosophical chimera; for no one outside analytic philosophy uses such assertions, or expressions -certainly not in the way Kripke intends. 33 32 If the right-hand term means to refer to a concept, then the invention may be perfectly harmless (see, for example, Wikforss, 2001, 215). "By 'green' I meant green" would then just mean, roughly, "I meant what I said" (where what I said -how my words are to be taken, or understood -would still depend on the context in which I uttered them). But, in my experience, the tendency has been to take the right-hand term to refer to the word's (or concept's) extension -the set of all of its instances (see, for a representative example, Boghossian 1989) -and that seems to me deeply problematic, partly because the extensions of empirical concepts are context-dependent and judgment-dependent, which means that italicizing the right-hand term only gives us the illusion of having successfully referred to some determinate set of things, and partly because the use of many of our words -and especially our philosophically-troublesome words ('know', 'cause', 'mean'…) -is not purely representational or "descriptive", which means that looking for their extensions would be the wrong way of going about clarifying their meaning, or use. 33 Part of what has confused philosophers in this area, it seems to me, and is clearly confusing in Donald Davidson (2006), is that in the rather uncommon case of malapropism, one could correct the speaker by saying "You meant 'arrangement', not 'derangement'", where the 'meant' here means "meant to say" (I discuss the difficulty this creates for Davidson, in Baz 2017, 69 and 166-167). But that's not the common case in which the question may arise of what someone meant by a word, and where the paraphrase that would normally be offered as a response does not mean (to suggest) that the speaker misspoke. Rather, the speaker is taken to have uttered just the words she meant to utter; and the question is how those words, as uttered by her under those circumstances, are to be understood, or taken. As Cavell fully realizes and acknowledges, however, such Wittgensteinian grammatical reminders and elucidations are of limited value for someone aiming to engage with Kripke. For Kripke, as we've seen, is captivated and guided by a picture that compels him, and has compelled many of his readers -a picture of the meanings of words as eternal and fully determinate rules that are wholly independent from the ordinary and normal use of those words, and of the past employment of a word as like a finite segment of a mathematical series that fits infinitely many different polynomial functions, each determining a different continuation of the series. And it's the picture, rather than our ordinary and normal criteria, that is ultimately relied upon to ensure the sense, and truth, of Kripke's words. "What good can it do for me", Cavell asks, to insist that our grammatical criteria for the application of an ordinary word, say "inclination", tell what kind of thing an inclination is (roughly, the expression of a desire, but one under conscious check)? A view that subjects the application of a word in general to inclinations will of course subject the choice of the word "inclination" to what it calls inclinations (where roughly neither desire nor its check are in play, but instead are replaced with something like generalized sources of energy) […] What good, then, to insist that in taking a series of steps you are not inclined to take the next step within the series, unless perhaps there is a puddle just there and you are inclined to make a splash? No good. (TAO,95) There is an important sense in which Kripke is even farther away with his words into the metaphysical than the "traditional philosopher" -Descartes, or Hume, for example -to whom Another thing that might have led to the common conflation of "the meaning of a word" and "what someone means by a word (on an occasion or set of occasions)" is that we do sometimes talk, in English, about "what a word means" (as opposed to "what someone means by a word") -most commonly in talking about words in different languages, as in "'grün' means 'green'". But though we may then be said to be giving the meaning of the word 'grün', we're not doing so by giving the Kripkean "meaning", or "rule" that the word "denotes". Rather, we're giving the English word that best translates the German word -the English word, that is, that best captures what 'grün' means in German (rather than as uttered by some individual on some occasion or set of occasions). And then there is the rarer "I asked you to get green apples, and by 'green' I meant green (not greenishred)"; but the normal use of that form of expression is different from, and far more specific than, what Kripke wishes to be talking about.
Wittgenstein, and Cavell's Wittgenstein, primarily respond. The latter still purports to be telling us something, however surprising or troubling, about what we call 'knowledge', for example, or 'cause', and to proceed on the basis of what we all already know or are already committed to as competent users of those words. He therefore invites Wittgenstein's and Cavell's therapeutic reminders. In Kripke's case, the picture, and the theoretical construction erected on its basis, have taken over entirely, and are relied upon throughout to ensure the sense of his words -the clarity of what he means by them; so, as Cavell acknowledges, it is not clear what good Wittgensteinian reminders would do in responding to Kripke. He is not likely to be troubled by the fact that none of his key terms or expressions mean what they ordinarily and normally mean, as long as he feels confident that he knows what he means by them.

The Argument of the Ordinary III: The Performative Argument
But this, as I see it, is really where the heart of Cavell's argument lies, or the heart of what his argument shows; because for Cavell, Kripke is like Wittgenstein's deviant student: he reads the Investigations and then goes on to philosophize in what he takes to be its light, but in a way that, from Cavell's perspective, misunderstands virtually all of Wittgenstein's words, and gets wrong everything that is deep and important about Wittgenstein's "vision of language" and how he philosophizes. 34 And yet Cavell allows, rather charitably it seems to me, that there may be nothing internally wrong with Kripke's way of going on from Wittgenstein (TAO, 65); just as there is not necessarily anything internally wrong with how Wittgenstein's deviant student 34 In order not to further lengthen an already lengthy paper, I will not enter a detailed discussion of how Wittgenstein's remarks work, and how different it is from how Kripke proceeds. I will only note that Cavell is surely right when he remarks that "Wittgenstein takes the ideas Kripke is explicating and organizing to be more various and entangled and specific than Kripke seems to me to give him credit for" (TAO,67). Also useful in bringing out Kripke's failure to appreciate the level at which Wittgenstein's remarks on rulefollowing are meant to operate is Goldfarb (1985). understands "add 2". 35 And in either case, to respond by saying "This is simply what I do" would likely be an empty gesture, or worse, partly because in such situations, as Cavell notes, "the common world of background against which I define the this that I do is not available" (TAO, 72; see also 71), and partly because, in the absence of that common world, and especially in the former case where the fundamental disagreement is among peers, it is not clear with what authority one could say this (see TAO,72). Nor would it do to try to prove Kripke wrong in how he goes on from Wittgenstein by appealing to how the members of "our" community -and which community would that be?! -are inclined to mean, or understand, Wittgenstein's words.
This means that Kripke's account of linguistic communication and its conditions has misrepresented its purported subject matterof which Cavell's philosophical encounter with Kripke is surely an instance -beyond recognition. But it also means that the question of how Cavell ought to have responded to Kripke is a difficult question, just as the question of how one ought to respond to a student who finds it natural to go on differently from how "we" go on is a difficult question. It is the question of what to do when one finds one's most basic ways of looking at and seeing and responding to things, and hence one's world, confronted by another's, and thrown into question; 36 so it is ultimately a broadly ethical question, not a purely theoretical one. And this is precisely what Kripke misses altogether, and covers up, in his depiction of what Cavell calls Wittgenstein's "scene of instruction". Not only would saying "This is simply what I (we) do" be empty in the face of someone for whom it is natural to go on with their words -to develop, so to speak, the linguistic inheritance (we thought) they share with us -very 35 I say that Cavell is being charitable to Kripke here, because, as the discussion in section 3 aims to show, it is actually not clear that Kripke's way of going on from Wittgenstein ultimately makes sense, not even purely internally. 36 The confrontation need not be so radical, or dramatic. Others can disagree with us in how they go on with their words in less fundamental ways. Such moments are common; and they too cannot really be understood from Kripke's perspective. An important feature of Kripke's (mis)reading of Wittgenstein's "scene of instruction" is the conflation of the student who "doesn't get it" and does not know how to go on, and the person -Wittgenstein's student, for example, but possibly a peer (TAO, 73) -who gets it and goes on, but differently from us. differently from us, but there is also the question of our relationship with that other, and how we choose to develop it. Saying "This is simply what I do" -as an assertion of brute power, for example, or in exasperation -would itself be a significant step, or move, the taking of a position, in a world shared with others, and in relation to some particular other. As with anything we may do or say, we will become responsible for it; and as with everything else we do or say, its meaning or significance is not up to us to choose or decide, and could always be found to transcend our express intentions.
"Whether and how we accommodate ourselves to our impotence" in the face of someone who disagrees with us fundamentally in how he goes on, Cavell writes, "will probably depend on our relationship with this person, if there is any" (TAO,(88)(89). Reaching bedrock, he writes, only means that one (finds one) "cannot keep going straight" (TAO, 82). And then he adds: But does this not leave me room, perhaps ground, for choice over whether to take this stumbling rock as a rejection, from which I recoil, or as a discovery, say of the other, to which I yield? I mean, if I discover resistance I might shift my ground, or take a new approach, or blast my way through, or exclude the site and this block from my plans altogether. However I take it, the scene with its spade is going to remain for me […] one of cultivation, or constraint. (TAO,82) In my experience, even those in contemporary mainstream analytic philosophy who recognize the ethical dimension of what they theorize about -in the present case, the transmission, acquisition, and use of language -and recognize its importance, would be likely to insist that the theorist may, and indeed should, set that whole dimension aside for the purposes of his theorizing. One of the most important lessons I have learned from Cavell is that the pursuit of philosophical theories that leave me out -both in the sense that their sense is not supposed to implicate me, and in the sense that I am not expected to be able to recognize myself, and my experience, in their portrayal of us -would lead me nowhere, or anyway nowhere I wish to go. In the present case, if you try to divorce what may sensibly be called 'meaning' from meaningfulness, hence from what we each find meaningful and worth noting and saying, and hence from what Cavell calls 'voice', whose utter absence from Kripke's account he notes (TAO,; if, specifically, you portray the acquisition of linguistic competence as a matter of coming to talk as others do, or in a way others accept, rather than as a matter of coming to speak for yourself and becoming responsible, not only for the truth or practical utility of your words but also, before everything else, for their sense; if, in other words, you attempt to separate what someone means with her words from what she cares about and how, and against what worldly background, and hence from your understanding of that person and how she relates to you; then you are bound to find, with Kripke, that a speaker could be meaning indefinitely many things with her words, or mean nothing at all. You might then turn to your community of fellow speakers, hoping that it would decide for you how to take her words, and how to go on more generally. But it would be unrecognizable. In addressing himself to the American Philosophical Association of the late Nineteen Eighties, Cavell knew he was likely to be misunderstood, and dismissed. He knew his style of philosophizing went against the grain of how, for the most part, philosophy was being done in the English-speaking philosophy departments of that time (see Cavell 1990, xi). In some respects, that would have been even more true today. But it was partly this knowledge of the possibility of isolation -intellectual, spiritual, or other -that moved Cavell to respond to Kripke's proposed understanding of what's involved in making sense with our words. As he puts it in the final lines of "The Argument of the Ordinary": "At any time I may find myself isolated. A moral I derive from the Investigations […] is accordingly: I am not to give myself explanations that divide me from myself, that take sides against myself, that would exact my consent, not attract it. That would cede my voice to my isolation. Then I might never be found" (TAO, 100).
In the face of Kripke's foreignness, and the foreignness (for Cavell) of his portrayal of linguistic communication and its conditions, and being fully aware of the foreignness of his own approach to philosophizing to Kripke and any number of others in the mainstream of analytic philosophy of that time, Cavell could have chosen to disengage, as so many nowadays, on every side of any number of philosophical (and other) divides, choose to do. He could have chosen to address himself only to those who already shared his general approach and philosophical sensibilities; and that would not necessarily have been an insensible choice. But at least that time, Cavell chose to engage with Kripke, drawing on, and trying to recover, what he thought he shared with Kripke and with his audience of analytic philosophers. Most importantly, he presented himself, in his refusal to disengage -his refusal to give up on communicating with those who in certain respects might be more alien to him than the members of some isolated tribe. In doing that, he offered the most powerful refutation I can think of, not only of Kripke's account, but also, and notwithstanding Cavell's express disavowal of any such ambition (see TAO,and 83), of Kripke's skepticism. 37