20th WCP: How Private Must an Objectionably Private Language Be?

Some philosophers, taking their cue from Philosophical Investigations (PI) 243 - 315, suppose that a private language is objectionable only when its terms refer to Cartesian mental events. In this "strong" sense private languages are very private indeed. Others (notably Kripke, 1982) have focused on PI 201 and the surrounding remarks about rule following, and have explicated the notion of an objectionably private language as (roughly) that of a language used by just one isolated individual unsupported at any time by any source of external or community correction and approval. I think of this as a "weaker" sense of 'private language.'

In sec. 1 I attempt to defend the "Kripke - Wittgenstein" (henceforth 'KW') version of the private language argument against some objections proffered by Simon Blackburn. KW takes languages which are private in the weaker sense to be objectionable, and claims that the later discussion (PI 243 - 315) deals with a "special case" falling under the more general discussion of rule following in earlier sections. In section II I briefly consider some possible objections from Wittgenstein himself to my defense of Kripke.

"This was our paradox: no course of action could be determined by a rule, because every course of action can be made out to accord with the rule"(PI 201). According to KW the reason any course of action could accord with the rule is that there is no fact about an individual to which he can point in justifying going on one way (in the application of a word, continuation of a number series, etc.) rather than another. For example, what justifies me in answering '125' rather than '5' when queried '68 + 57?' This can be thought of as a skeptical question, asked by myself about myself, in which past training and intentions are related to present behavior. Perhaps the rule I follow in the application of '+' is "bent" in the following way; I give the answer to any query 'x + y?' in the usual way so long as x and y are < 57, and otherwise I give '5,' so that to answer '125' for the problem above would be to deviate from the rule, it would be to fail to go on in the "same" way. Kripke calls this a "skeptical" paradox. He tries to show that all of our usual intuitions about the madness of such a "skeptical" suggestion are unsupportable. It turns out that only the reactions of a community of users serve to "rule out" such mad responses as '5.'

There is only one of KW's skeptical arguments which Blackburn wishes to dispute. KW considers the possibility that the fact about me which makes '125' rather than '5' the correct answer is a dispositional fact. In learning to add (use '+') I acquired a complex disposition, which included the disposition to answer '125' etc. So if I answer '5' my present behavior does not accord with my past training and resultant dispositions. So my meaning some function _ by '+' just is my acquiring such dispositions. But KW attacks this idea on several fronts. The one which I find most worthy of consideration, because of its clear reliance on the normativeness in the notion of meaning, is this; my dispositions may include dispositions to make mistakes. The mistakes could conform to the "bent rule". So dispositions cannot be what distinguishes the bent rule follower from the straight rule follower, and thus cannot be the fact about me which guarantees that I mean some function _ by '+'. Blackburn, contrariwise, thinks that extended dispositions could distinguish the bent rule follower from the straight rule follower. ". . .Consider, for instance, the bricklayer told to add bricks to a stack two at a time. If this means to him 'add 2 up to 1000, and then 4', his reaction to the foreman may be quite different. Perhaps he cannot carry four bricks at a time" (Blackburn, 1993, p. 221). Or suppose someone uses 'round' to mean 'squound', where something is 'squound' if it is round before Jan. 1, 1984 and square thereafter. Such a person would have to show some behavioral tendencies distinct from the tendencies of straight users of 'round.' Thus a person who meant squound by 'round' might in 1983 show concern that next years autos might not roll so well, a tendency to hoard tires, etc. That would show that he did not really think of himself as going on in the same way in his bent use (cf. 1984, 81-2).

I do not think Blackburn's arguments succeed. How a worker is disposed to respond to a foreman who commands '+2' may depend upon factors which are completely independent of whether he understands '+' straight or bent (e.g. how tired he is). With respect to the 'squond' case, imagine someone, S, following the rules of checkers straight, and his opponent has one move left to crown a piece. S might now be worried, concerned about the new consequence which turns up at this point of a move according to the straight rule. Or he might not. What is such a fact supposed to show about what rule or rules S is following? If S didn't show any fears would that show that he did not take the rule in the straight way? Blackburn claims that if such bent rule followers did not show "undue fear lest next year's motor cars will have round wheels" we can "know them not to interpret the words in the bent ways defined."(82). But what we fear, expect, look forward to or shun, can vary from person to person even where everyone is following the same straight rules. Thus by themselves such facts show nothing about what rules we are or are not following. Blackburn believes that his arguments show that "What the learner cannot do is both take the initial samples and explanations in the bent way, and show no awareness of the bend"( 1984, 81). What his arguments might show at best is that where there is a bend in a rule the user is bound to be aware of something being different. But his arguments do not show that what the user must be aware of is the bend. They do not show that he must be aware of it in such a way that he is no longer entitled to think of himself as going on in the same way. A rule provides one source for the idea that what I am doing is the "same." This point does not depend upon some notion of "relative identity." (Cf. Kripke, 1981, 18,19). But it should be clear that my thinking I am, say, making the "same" move does not entail my thinking that nothing is different about the move in question (it might be, for example, the same move I made yesterday).

Now Blackburn thinks that dispositions must supply a standard for distinguishing bent rule from straight rule following individuals since he thinks that we cannot show that individuals have a problem in this regard which communities do not have. A community might be following a bent rule. That within the community so far there has been agreement in applications shows nothing about whether or not it is following a straight rule. And all the same consequences which ensue for the individual regarding the possibility of meaning anything at all by a sign would ensue for the community also.

Just why does Blackburn think that the skeptical paradox leveled against the individual can also be brought against the community? On KW's view, members of a community "see each other as" or "dignify one another as" rule followers. They have justification conditions for this "seeing as," for the individual's practice must accord with the community's in order for him to be so seen. But Blackburn objects, ". . . we do not know what a community would be lacking if its members failed to see each other this way, or if they continually saw each other in the light of potential bent rule followers."(1993, 222)

There seems to me to be an obvious reply to the last sentence quoted. What "it" would lack would be "being a community." Blackburn assumes that we can identify 'it' even where such "seeing as" is missing. I do not see how. Almost as if to acknowledge this last point, Blackburn goes on to entertain the possibility that in a community ". . .the mutual support itself supplies the standard of correctness."(1993, 223). There might be an analogy to an orchestra in which attunement with others was itself the standard of correctness, there being no score to follow. Yet according to Blackburn this will not work for judgement, for even if my entire community starts saying that 57 + 68 = 5, that would not, Blackburn insists, make me wrong in claiming that it is 125. Here, as elsewhere, it is difficult to know what Blackburn has in mind. I suppose that in the situation envisaged someone might just continue to insist "I am right, everyone else is wrong" but such a reaction would be, it seems to me, very odd. More likely a person in the situation described would be thrown into utter confusion and wonder just what he had misunderstood or how he had gotten so completely off the track. The conclusion to draw here is precisely not the "disastrous conclusion" against which Blackburn warns ( ibid) namely that what the majority says and believes constitutes a "truth condition" for any claim. The thought that that is the conclusion to draw derives from a failure to observe that "It is what human beings say that is true or false, and they agree in the language they use. That is not agreement in opinions but in form of life" (Wittgenstein, 1953, 241) . If my opinions diverge wildly from everyone else's then I am likely to think that the problem is not the opinions but something else. It can't be that what is at stake is merely a mistake, a blunder. "For a blunder, that's too big."(Wittgenstein, 1967, 62).

Blackburn is however not easily dissuaded. He suggests that we think of the matter in this way: "The members of a community stand to each other as the momentary time-slices of an individual do. So just as the original sceptic queries what it is for one person-time to be faithful to a rule adopted by a previous person-time, so the public sceptic queries what it is for one person to be faithful to the same rule as that adopted by another "(1993, 224). I believe that we cannot think of the matter this way. There is an important difference in the relations between person-slices, on the one hand, and persons, on the other. It does not make sense to think of person-slices as correcting each other, holding each other responsible, and so forth. I do not mean that a person cannot correct herself. I mean that when she does she employs community standards. She does not correct herself "by herself," so to speak, and could not. Whatever she might do by herself could not count as correcting, anymore than Sancho Panza's blows administered to his own bottom could count as punishment or a penalty. Or so it seems. Whether it seems so also to Wittgenstein will be taken up further below.

This last point brings us to the notion of a practice, which Blackburn ultimately, and correctly, sees as being central to this discussion. "Correcting" belongs to practices if anything does. Blackburn relies upon a general preliminary sense for what such a thing might be, in order to proceed to the central question, namely, "What kind of thing does a practice have to be if it is to block the skeptical paradox?"( 1993, 226). I believe that Blackburn's failure to consider the role of constitutive rules in practices leads to his misconstrual of the following example (itself borrowed from Dummett). Blackburn asks us to suppose that a born Crusoe might find a Rubick's cube washed up on the shore and over the years evolve a technique (a private practice) for solving the cube. Blackburn thinks it is "easy to go through the thought experiment of coming across such an individual."(1984, 84. cf. 1993, 226-27). I think what he should say is that it is easy to imagine such an individual, in somewhat the same way in which it is easy to imagine a bear speaking English. But if you really think about it (the term "thought experiment" is quite honorific here) you may not get far. Crusoe discovers a "solution."? And where did he get the concept of a solution? How is he to know what would count as a solution? Where would he even get the idea of something counting as something? Crusoe is sitting on the beach. The cube washes up. He picks it up, fools with it. For years perhaps. Then suddenly it pops into his mind: "perhaps there is a solution to this." This is ludicrous. Or, nothing pops into his mind (in particular no English sentences ) but he simply starts manipulating the cube, making little marks in the sand, looking back at them now and then, and eventually, he has it! What? The solution! I personally can neither imagine, conceive, nor make sense of this. In particular I do not see how anything that Crusoe did could be considered by him to be mistaken or incorrect, except in the sense of failing to get him from A to B in the way that some other "move" did. But there could not be anything "correct" about getting to B itself, (though doing so might be pleasing, satisfying, conducive to survival, etc.) We can imagine a born Crusoe finding the solution to the Rubicks cube only by forgetting who a born Crusoe is, perhaps, by forgetting that he is a born Crusoe.

Blackburn supposes that Crusoe ". . .dignifies himself . . .as having a principle of application. . .In doing so he allows the possibility of mistake. . .It is a component of his attitude that a particular judgement might turn out better regarded as mistaken."(1993, 228) What perhaps makes this seem plausible is the reliance on the notions of "attitude" and "allowing" which Blackburn brings to bear on the case. Yet there is something very fishy in the notion that something might turn out to be a mistake because someone has the attitude that it would be better regarded as a mistake. Better by what standard of superiority? Blackburn suggests the following; the isolated individual's judgements are playing a role in a project "of ordering the expectation of recurrence of sensation with an aim at prediction, explanation, systematization, or simple maximizing of desirable sensation"(ibid.). I have allowed that Crusoe might find certain ways of arranging the cube more pleasing than others, and it is not impossible to imagine a story in which its being arranged one way rather than another might have survival value. None of this amounts to showing that arranging it that way is "correct." That something is pleasing to me at this moment cannot make it "correct." If something quite different should please me a moment from now, then what more can I say or think than "whatever seems to please me does please me. There is no difference." And that arbitrariness will transfer to any notion of a "mistake" which might be imported into this context, so that "whatever seems right to me will be right," i.e. the notion of right, and mistake, will be demolished, as Wittgenstein insists (PI 201).

There is reason to question whether Wittgenstein himself would find the replies to Blackburn sketched in the previous section satisfactory. In PI 243 he claims that "a human being can encourage himself, give himself orders, obey, blame and punish himself," and that "we could imagine human beings who spoke only in monologue." Hacker contends that "such reflexive speech acts are not parasitic on non reflexive ones"(1993, p.16). If they are not, then what Wittgenstein here claims we can imagine does indeed seem to be the very thing I have claimed (in response to Blackburn) we cannot imagine. But as Pears points out (1988, p.339) the phrase 'their language' in PI 243 allows for the possibility that they speak the same language, perhaps as a result of an eavesdropping which contains covert external correction. Yet both Pears and Hacker and many other commentators hold that the decisive thing about an objectionably private language is that it be private in the sense that its terms refer to essentially private Cartesian cogitationes. KW, on the other hand, argues in effect that any language which is not subject to external correction, whether private in the Cartesian sense or not, could not be intelligible to anyone, including the one who uses it.

How far apart are these positions? After all, Wittgenstein certainly aims to overthrow the notion that words could acquire significance by being attached to such cogitationes. So clearly on his view no one understands a private language, or otherwise put, such a "language" is no language at all. This obvious fact lends some plausibility to Kripke's view that the "sensation language" discussed in PI 243 and following is just a special case falling within a more general discussion of what is involved in following a rule. The issues are too complex to be discussed in detail here but I will make just one suggestion. The apparently dogmatic claim that nothing a born Crusoe did could count as (in)correct, for himself or anyone else, depends upon correctness being specified relative to a constitutive rule. I do not deny that Crusoe could think something incorrect in the sense of failing to conform to something like a rule "in the summary sense." Now I think it is Wittgenstein's view (though he wouldn't have put it just this way) that linguistic practices crucially involve constitutive rules, and such rules essentially regulate relations within public practices.

I believe these considerations, when more completely spelled out, will support the claim that objectionably private languages do not have to be private by virtue of being tied to Cartesian mental content.