H. P. GRICE E J. L. SPERANZA: LA CONVERSAZIONE -- I VERBALI: PEZZAROSSA

 G.: Pezza-Rossa’s title is already too ambitious for comfort: Sopra la sola confutazione possibile dello scetticismo. S.: The sole possible refutation. Italians do like the article when they mean to end history. G.: Quite. It is one thing to refute the sceptic; another to announce that there is only one admissible corpse. S.: Yet you have your own objection to the sceptic. G.: I do, but I have never claimed monopolistic burial rights. S.: Woozley thought otherwise. G.: Woozley thinks many things sotto voce, and not all of them deserve publication. S.: Still, the Malcolm line is serious enough. G.: Very serious. If the sceptic says “I know there is cheese on the table” is self-contradictory or absurd, yet admits it is an ordinary expression, he flirts with the impossible. S.: Because Malcolm says an ordinary expression cannot be self-contradictory. G.: More carefully, he says that an expression which would never be used to describe any situation cannot at once be ordinary in the relevant sense. S.: And the sceptic must concede ordinary use. G.: Yes. “I know that there is cheese on the table” is not a private code-word, nor a grammatical hallucination. It belongs to the furniture of discourse. S.: Malcolm’s point then is that the sceptic cannot both admit the furniture and deny that there could be any room in which it is properly used. G.: Exactly. One cannot sensibly say that an expression is ordinary and at the same time logically outlawed from any correct descriptive employment whatever. S.: Unless one plays games with “ordinary use.” G.: Quite. And I did try to give the sceptic that escape route. S.: The “cauliflower” society. G.: Yes. Suppose everyone commonly called roses “cauliflowers,” yet, upon reflection, insisted that these applications were wrong. One then asks whether frequent use suffices for correctness, or whether a whole people may need correction. S.: A nice nightmare for lexicographers. G.: And a more respectable one than most sceptical triumphs. S.: But now you want to go further than Malcolm. G.: I do. My point is that another possible confutazione trades not merely on ordinary use, but on utterer’s meaning itself. S.: That is the asterisk-p line. G.: Precisely. Let *p be an absurd proposition. S.: Give me one. G.: Let *p be: pirots karulise and do not karulise elatically. S.: Already unbearable. G.: It should be. That is the point. If a proposition is absurd in the relevant sense, it cannot be the proper target of a reasonable communicative intention. S.: “Reasonable” being the loaded word. G.: Entirely loaded, and usefully so. S.: Then formulate. G.: If U means by x that p, in the nonnatural sense, U intends A to come to entertain, or perhaps to believe, that p, at least by recognising U’s intention and the grounds for it. S.: Through reason, not mere causation. G.: Exactly. I do not merely want A to produce a reflex. I want A to arrive at p by a recognisable rational route. S.: So if p itself is absurd, the route collapses. G.: That is my suspicion. If *p is not merely false but absurd, then no account of utterer’s meaning that makes reason central can easily permit U to meanNN that *p. S.: Unless U is performing a joke, a reductio, a parody, a linguistic pathology. G.: Quite so. But then what U means is not *p as such, but something else by way of *p. S.: Such as “see how absurd this is.” G.: Precisely. The absurd proposition may be mentioned, brandished, or exhibited, but not straightforwardly meant as what the addressee is to come reasonably to accept. S.: Then how does this hurt the sceptic. G.: Because the sceptic, in many of his grander moods, aims to make us entertain propositions about knowledge that threaten the very reason-governed practice by which he addresses us. S.: More slowly. G.: Very well. If the sceptic says that no one ever knows any empirical proposition, he is not merely proposing an odd thesis about cheese, tables, and doubt. He is threatening to deprive a large stretch of ordinary linguistic practice of rational legitimacy. S.: Which Malcolm says already makes trouble for “ordinary expression.” G.: Yes. But I want to add that the sceptic must also count on our taking his own thesis seriously, that is, as something he means us to understand through reason. S.: Not by hypnosis. G.: Exactly. He must intend us to recognise grounds, infer conclusions, see incompatibilities, feel the pressure of his q against our “I know p.” S.: So he relies on reason-giving discourse while undermining the very credentials of reason-governed knowledge claims. G.: Just so. S.: But that still sounds like a pragmatic transcendental argument, not yet an argument from utterer’s meaning. G.: It becomes one when we ask what it is for the sceptic to mean what he says. S.: Go on. G.: If U meansNN by x that p, then U must intend A to come to entertain p via a rational appreciation of U’s communicative move. S.: Yes. G.: But that demands that p be, at minimum, a reasonable candidate for rational uptake. S.: Not necessarily true, but not absurd. G.: Exactly. One may intelligibly mean something false. One may mislead. One may err. But one cannot straightforwardly meanNN that *p where *p is so absurd that no reasonable addressee could be expected to adopt it by reason. S.: You are building a reasonability constraint into meaning. G.: I am. Or rather drawing it out of the very role of recognition and rational uptake in nonnatural meaning. S.: Then the sceptic’s thesis might fail, not because it is unpopular, but because it cannot be the object of the kind of uptake he needs. G.: That is the shape of it. S.: Give me the formal skeleton. G.: Gladly. Let M(U,x,A,p) abbreviate: U meansNN by x, to A, that p. S.: Good. G.: Then, roughly: M(U,x,A,p) requires that U intend A to entertain p by recognising U’s intention that A entertain p, and by taking x as a reason-bearing move in the circumstances. S.: A Gricean mouthful. G.: All decent theories are. S.: Continue. G.: Add a reasonability condition R(p): p must be such that it is a reasonable candidate for rational entertainment or belief in the given exchange. S.: Not certainty, but reasonability. G.: Exactly. Then for absurd *p, not R(*p). S.: Therefore not M(U,x,A,*p), at least not literally and directly. G.: Correct. Unless the true p is something like “*p is absurd,” in which case the content meant is no longer *p itself. S.: So if the sceptic’s own thesis collapses into absurdity of the relevant kind, he cannot mean it in the very sense required for philosophical assertion. G.: Precisely. The fatal objection would be that he tries to occupy the illocutionary posture of a reason-giver while offering a content unfit for reason-governed uptake. S.: This begins to look like your stronger answer to the sceptic. G.: Stronger, or at least differently targeted. Malcolm attacks the sceptic’s relation to ordinary expressions. I attack his relation to the conditions of meaningNN itself. S.: Yet you will need to show that the sceptical proposition is absurd in your strong sense, not merely distressing. G.: Of course. One must not promote mere discomfort to contradiction. S.: Then where exactly is the absurdity. G.: In the sceptic’s demand that we treat as unintelligible or systematically incorrect a whole range of reason-governed empirical claims, while still expecting us to take his own meta-claim as a serious, reason-directed contribution to inquiry. S.: So the content is parasitic on the very practice it seeks to globally disqualify. G.: Exactly. The sceptic depends on our capacities for recognising evidence, incompatibility, correction, and rational warrant, but then tells us these cannot underwrite knowledge in any empirical case. S.: Many would say that is only surprising, not absurd. G.: True. One must be careful. Not every parasite is a contradiction. S.: Then perhaps your argument works better if one tightens the notion of “knowledge” in the sceptic’s mouth. G.: Yes. If the sceptic insists that “I know p” is always self-contradictory or logically absurd when p is empirical, then he is not merely revising a standard; he is denying the possibility of a central ordinary practice while exploiting that very practice’s rational machinery. S.: Then *p here might be not simply “no one knows anything empirical,” but “the ordinary use of ‘I know p’ for empirical p is both genuine ordinary use and logically absurd.” G.: Excellent. That is much closer. S.: And that may indeed be too much for meaningNN to carry. G.: I think so. Because to meanNN that *p one must intend the addressee to recognise, through reason, the force of a proposition whose very content destabilises the rational standing of the practices relied on in the exchange. S.: The sceptic saws off the branch and invites us to admire the carpentry. G.: Very good. Keep that. S.: Thank you. G.: Then the relation between scepticism and meaning runs both ways. S.: Explain the reciprocity. G.: On the one hand, the response to the sceptic feeds into the theory of meaning because it shows that not any old content can be the object of reasonable communicative intention. S.: So meaningNN is constrained by reasonability of content. G.: Yes. Not every concatenation, nor every absurd pseudo-thesis, is apt for genuine meant content in the strict sense. S.: On the other hand. G.: On the other hand, the theory of meaning feeds back onto the response to the sceptic because it displays the sceptic’s dependence on reason-governed uptake. S.: He must count on us as rational addressees. G.: Precisely. He cannot merely emit noises and hope to produce despair causally. He argues, therefore he presupposes the very commerce of reasons he affects to distrust. S.: Then your objection to causal theories of meaning becomes relevant. G.: Entirely. If meaning were merely a tendency to produce attitudes, the sceptic could perhaps aim simply to induce unease, hesitation, or suspension by whatever means. S.: Like the tail-coat case. G.: Yes. But meaningNN is not secured by mere causal tendency. It requires intention plus rational recognisability. S.: So the addressee’s attitude must be achieved via reason and not merely caused. G.: Exactly. And that brings the requirement of reasonability of p to the centre. S.: Not *p. G.: Not *p. If the only way to induce *p is by confusion, intimidation, semantic vertigo, or sheer philosophical fatigue, then U has not meaningNNly brought A to entertain *p in the relevant way. S.: He has only broken the furniture. G.: Often the sceptic does little else. S.: Yet some sceptical arguments are subtle and not absurd. G.: Certainly. One must not abolish scepticism by bad manners. I am not saying every sceptical challenge is itself *p. S.: Only the strongest global sceptic who says the ordinary empirical “I know” is inherently absurd while still speaking as a reason-giver. G.: Yes. That stronger sceptic invites the fatal objection. S.: Then Pezza-Rossa’s “sola confutazione possibile” may be one route, but not the only one. G.: Exactly. His climate wants a decisive philosophical proof against scepticism. Mine allows a different pressure-point: the sceptic’s dependence on the conditions of meaningful, reason-directed utterance. S.: Which is less a single sword-thrust than a constriction of the breathing apparatus. G.: A pleasingly medical metaphor. S.: Oxford has its uses. G.: Occasionally. S.: Let us formalise once more, more soberly. G.: Very well.  M(U,x,A,p) requires that U intend A to entertain p by recognition of U’s intention and of x as reason-bearing. Therefore M(U,x,A,p) requires p to be fit for rational uptake in the exchange, call this R(p). For absurd *p, not R(*p). Therefore, absent a change of target content, not M(U,x,A,*p). If the sceptic’s thesis is of the form *p, he cannot straightforwardly meanNN it as a serious philosophical claim. But his whole performance presupposes he is so meaning it. Therefore his position collapses at the level of communicative act as well as content.  S.: Nicely brutal. G.: Only moderately. S.: Someone will object that rational uptake need not end in belief; entertainment is enough. G.: Fine. Let R(p) be suitability for rational entertainment rather than acceptance. The point remains. Some contents are unfit even for serious entertainment as live philosophical deliverances, except under a metalevel description. S.: Such as jokes, reductios, examples, nonsense tests. G.: Exactly. One may present *p without meaningNN that *p. One may mean that it is nonsense, absurd, revealing, or instructive. S.: So when you say “pirots karulise and do not karulise elatically,” you do not meanNN that pirots karulise and do not karulise elatically. G.: Mercifully no. S.: You mean that this is an absurd proposition, or rather an absurd-looking one, fit to test the boundaries. G.: Precisely. It is a specimen, not a creed. S.: Then the sceptic may reply that his proposition is not like that at all, because it is grammatically and conceptually well formed. G.: He may. Then the dispute shifts to whether the sceptical content is truly absurd, or only revisionary and uncomfortable. S.: And there the cauliflower society returns. G.: Indeed. We would need to ask whether our actual linguistic behaviour with “know” resembles the rose/cauliflower case: widespread use, but corrigible upon sufficiently clear reflection. S.: If yes, the sceptic may yet survive. G.: In some diminished form, perhaps. He might then force us into saying not that ordinary “I know p” is flatly self-contradictory, but that its correctness is unstable or indeterminate. S.: Your option two or three. G.: Exactly. But the grand sceptic who wants definite incorrectness everywhere while retaining full philosophical seriousness of his own utterance is in deeper trouble. S.: Then your “fatal objection” is really targeted at the maximal sceptic. G.: Precisely. Philosophers often overgeneralise the prey. S.: And Pezza-Rossa. G.: Pezza-Rossa belongs to a climate that wants scepticism killed with one principled blow, perhaps Rosminian in spirit, perhaps more civic-rational in ambition. S.: While you are content with several objections, some semantic, some pragmatic, some ordinary-language, some about meaningNN. G.: Quite. I have no desire to deny colleagues their favourite anti-sceptical weapon provided they do not insist it is the only possible one. S.: How many times, after all, can one kill the same sceptic. G.: As often as he reappears, unfortunately. S.: Then the charming result is that the theory of meaning and the anti-sceptical strategy become mutually supporting. G.: Yes. MeaningNN needs reasonable, reason-governed uptake; scepticism, to be intelligible as a serious position, must inhabit that very space; but radical scepticism about empirical knowledge threatens to undermine it; therefore the sceptic’s own act of meaning becomes suspect. S.: Meaning polices scepticism; scepticism reveals the commitments of meaning. G.: Admirably compressed. S.: Thank you. G.: Keep that too. S.: You are being unusually benevolent. G.: Only because Pezza-Rossa forced me into generosity by his title. S.: Then the final line. G.: Very well. The sceptic may doubt cheese, tables, bodies, futures, and all the rest; but if he wants us to understand him by reason, he cannot ask us to recognise as seriously meant a content that is itself unfit for reasonable uptake. S.: Not *p. G.: Not *p. Absurdity is not a content fit for meantNN belief; it is at best a specimen for diagnosis. S.: So if the sceptic must traffic in *p, he ceases to be a philosopher and becomes an exhibit. G.: Dry enough? S.: Sufficiently Mantuan, with one eye on St John’s and the other on the cheese.