versión On-line ISSN 1851-9636
Anal. filos. v.25 n.1 Ciudad Autónoma de Buenos Aires mayo 2005
"A Queer Sort Of Misunderstanding": Wittgenstein On The Model Of One's Own
My comment on Juan Larreta's paper ("Conceiving of someone else's pain on the model of one's own") has two sections. In the first one, I will single out several points in the debate and review his argument. In the second section, I will examine the main points of his position and suggest reasons why it could be questioned.
I. The case against private language that Norman Malcolm's (1954) believes has found in §302 of Philosophical Investigations (PI) takes objection with the assumption that once I have felt what pain in me is like, I can transfer the idea to the world outside myself(the privacy thesis (PT) expressed in &283).1 Malcolm presents two versions of the argument. The first one reaches the conclusion that it is impossible without contradiction to speak about other people's sensations and therefore, that the notion of private language leads to solipsism. The second one comes to the conclusion that sentences that attribute sensation terms to other people, even if not contradictory, are unintelligible. The first version reads:
(1) If I were to learn what pain is from perceiving my own pain, then I should, necessarily, have learned that pain is something that exists only when I feel it. For the pain that serves as my paradigm of pain (i.e., my own) has the property of existing only when I feel it. That property is essential, not accidental; [...] So if I obtain my conception of pain from pain that I experience, then it will be part of my conception of pain that I am the only being that can experience it. For me it will be a contradiction to speak of another's pain. This strict solipsism is the necessary outcome of the notion of private language (Malcolm, p. 189-90, my emphasis).
The second (which is "more correct" according to Malcolm) reads:
(2) If the privacy sensation thesis is true, then [...] "he cannot have any idea of what would count for or against the occurrence of sensations that he does not feel. His conclusion should be, not that it is a contradiction, but that it is unintelligible to speak of sensations of others" (ibid, p. 190, my emphasis).2
Saul Kripke (1982) rejects Malcolm's first version of the argument by pointing out that it does not do justice to §302. In this fragment, in fact, Wittgenstein makes no special mention of essential or accidental properties. He simply finds it difficult to imagine "pain which I do not feel on the model of a pain which I do feel" (Kripke, p. 116). In order to show that Malcolm's first exegesis is inadmissible, Kripke chooses a natural class term (duck) which performs a similar function to proper names. His argument is that nothing stops us from imagining another duck, phenomenically identical to the one we know, though with a different genetic code, even if we had learned the concept of duck from a given paradigm such as Central Park ducks. In other words, nothing hinders the conceptual extrapolation of the term even if one shares the ideas that are set out in Naming and Necessity (1981).
Unlike Malcolm, Kripke sees that Wittgenstein raises a second sceptic paradox in §302 (Kripke, pp. 133-134). Here the problem is no longer how to extend a given concept to a new case (as in the previous "plus" and "quus" paradox). It is rather the special difficulty presented by expressions like "I am in pain". The difficulty lies in that the subject seems unable to have an idea of a "self" or of "having" applicable even to his own case, "let alone a generic concept of a 'self ' that in addition to 'me' includes 'others'" (Kripke, p. 124)3. If so, the subject would not be able to understand what it means to say that other minds have sensations as opposed to saying that he himself had them there. Kripke believes that these issues (not those that Malcolm indicates in his second exegesis) were those that led Wittgenstein to vacillate between solipsistic tension and a (sceptical) solution closer to conductism.
In any case, in the two versions of Malcolm's so-called "external" argument (which as he points out, is not a reductio), he criticises the aforementioned (PT) assumption that once I learn/know what pain in me is like, I can transfer the idea to other people.
Following Kripke, Juan4 makes an interesting observation. If we accept a theory of personal identity in the line of Parfit (former and successive selves) where the notion of "psychological continuity" is subject to degrees and levels, we must also admit two things: (1) that our past pains are not strictly speaking "ours" but rather belong to our former selves, so that nothing prevents a private language from meaningfully attributing pains to other person stages; (2) that the (contingent) relation ofpsychological continuity degrees allows us to think of extremely weak links between my present self and my successive or former selves, and even to conceive a zero-degree continuity nexus. If this possibility is taken into account, to extrapolate the concept of pain to a self whose link to our present self were null would be similar to meaningfully extrapolate the concept of pain to that which other people may feel. The cases of fusions and fissions presented reinforce points (1) and (2) in the sense that they admit recollection from the"inside" of sensory experiences of our former ones with the important nuance that, as noted, it is related to other people. In short, what Juan wishes to show, if I understand him correctly, is that the conclusion of the external argument (to conceive other people's experiences is unintelligible) rests ion the non-problematic character of conceiving our own past or future experiences (an idea that is linked to a strong identity theory). And that if we admit, on the other hand, a weak identity theory, self-ascriptions and attributions of sensations to other people need not be considered in a different category. It would follow, by modus tollens, that the argument against private language (in its external version) is inconclusive.
II. I will now discuss Juan's proposal focusing on three of his assertions: (i) Wittgenstein's most famous private language argument rests on the doctrine of the necessary privacy of subjective experience. (ii) Such doctrine takes for granted that conceiving one's own past experiences is not problematic. (iii) §350 shows that Wittgenstein accepts (ii) as being implicit in (i). Let's take up the first one.
Aprivate language, as defined by Wittgenstein, is a language that logically cannot be understood by other people (§243). In my opinion, what this paragraph is meant to show is that such a language would not be intelligible even to me. Let's recall the thrust of §258. The ceremony with which I believe I am granting a name "S" to my present sensation (the internal ostensive definition that institutes its meaning) might succeed if every time I used it in the future, I referred to that sensation and not to another. For this to happen, the connection between the name and the named must be the correct one in every instance I apply the word. Now, given that I lack an independent identity criterion that can ensure the connection (since "whatever is going to seem right to me is right"), I am unable to fix the appropriate use of the word, i.e., I am unable to speak meaningfully of "my sensorial experience" whether it be present, past or future. Thus the ceremony through which I allegedly learnt/acquired knowledge of the meaning of the sensation at hand failed, because what would allow the distinction between the correct and incorrect use of a word - a vital condition for its meaningful use - could not be established. And the words that make up a language are not mere sounds that I may use at will.5
"The quid is that an essentially private object can not justify the use of a word neither for other people nor for himself. The private object not only does not playa part in the public game, but it cannot playa part in our private game either" ("Notes for the Philosophic Lecture", p. 450).
Now, if this is so, even assuming that I think I could internally capture the experiences of my former selves, this would not, I believe, change things. The subject - product of fusion - could not establish through an internal ostension that his words (those referred to his immediate private sensations, incomprehensible to anybody else, §243) become from that very instant intelligible to him. And by definition, he could not even understand his former self, the private linguist with whom he has zerodegree continuity.
It is true, as Juan points out, that he does not deal with this argument. I nonetheless believe the issue matters because what we have said, if I am not wrong, conceptually precedes the point emphasised by Juan. The idea is the following. Whichever personal identity theory is judged best, to be able to state that it is possible to extrapolate a given concept, it is necessary to have the concept that needs to be extrapolated in advance. And this is precisely what the private linguist cannot have. Thus I think Juan's objection is not independent (as he would like it to be) from what is frequently considered "the most famous argument" against private language, that is the argument following §243.6
According to Juan's reasons (notes 2 and 5), the external argument's implausible conclusion rests on two things. According to §302, (i) we can meaningfully attribute our own pain to bodies other than our own ("The Pope's head hurts me"), whereas we cannot meaningfully attribute "pain" to other people ("The Pope has a headache"). On the other hand, according to the analogy presented in §350, (ii) we can conceive our own past experiences with no difficulty at all. My understanding is that Juan's interpretation of both paragraphs is not consistent with what Wittgenstein intends to convey. So that I will try to justify my claim in the remaining lines of these notes.
Let's start with §350. The private linguist's idea (the PT's supporter) is that meaningful attribution of other people's pains is simply equivalent to say that the other person is undergoing the same that I so often have. An analogous case may be thought to take place when we understand what "5 o'clock on the sun" means. What we understand is that it's the same time there as it is here when it is 5 o'clock. Wittgenstein points out the error in this reply. The problem is not with the grammar of the notion of "sameness", which is taken for granted in every language game. The difficulty lies in that in order to say that someone is undergoing the same pain as I, I must understand in advance what it means that he is in pain and I am in pain. That is why the example of the sun displays an equivalent error: talking about "the same" time here and there lacks meaningfulness because we do not understand the respective sentences in advance (we only believe we do). The error's source lies in that we overlook the fact that the expression "5 o'clock" is only meaningful with respect to the sun's position since we measure time precisely with regard to the sun. We may recall a similar case in §279 when someone allegedly says he can estimate his height by laying his hand... on top of his head! For these reasons, I believe that §350 cannot be quoted to support point (iii) as Juan would like.
Let us now consider §302. In the first place, it is far from obvious that the paragraph encloses a variation on the private language argument. Personally, I do not manage to recognise any argument in it. Neither do I believe that it alludes to a special difficulty, as Kripke thinks. In my opinion, it rather shows up, as is often the case, a misunderstanding or confusion. Let us see why.
The very idea of "my own experience" with its immediacy and intimacy is a permanent target of Wittgenstein's critiques. He attacks both its epistemic (privileged access, direct knowledge) and its ontological sides ("the real thing is this").7 He believes that the privacy thesis, thus understood, leads to all sorts of philosophical puzzles. I quote Wittgenstein:
"It is said that we do not know other minds, and we know our own. But in a sort of way - it is rubbish. But why do people say it? It is based in a queer sort of misunderstanding. [...] It is like saying: in tennis there are no goals. Therefore there is no use for 'I wonder if I have got a pain', 'I am pretty sure that I have a pain' or 'I know that I have a pain'. You can not score a goal in tennis. Is that because it is difficult?" (Wittgenstein's Lectures on Philosophical Psychology 1946-47 (LPP), p. 227, my italics).
In paragraphs preceding §302, the speaker indicates that an insistence on the link between the concept of pain and pain behaviour tends to do without the inner sensation, as if it didn't exist. He thinks that imagining someone's pain obviously is not imagining a given behaviour (gestures/exclamations). He thinks it is equally obvious that we sometimes imagine someone's suffering even if nothing in her behaviour betrays it. These simple observations suffice to take notice that what "really matters" is the inner sensation. The rest is at least secondary (&&281-283).
Wittgenstein closely links the privacy thesis with such concerns. In this context, he examines the concept of imagination. Why "it is not easy" to imagine someone else's pain on the model of one's own? (§302). Let us try to. First option: I imagine his sensation by putting my sensation in its place. But then, it is utterly contradictory to say that I can imagine someone's sensation while I omit it. Hence the irony when saying... "transition is none too easy to do". Surely not because it is problematic but because it is absurd.
Second option: I can replace "I'm in pain" with a yelp. Can I imagine that my yelp can replace "He's in pain"? Suppose that I started crying out as I pointed to a part of his body. According to Wittgenstein, this behaviour would not be understood as saying, "He's in pain", but as "I have pain there - as if his body were an extension of mine" (LPP, p. 52). And again, I could not meaningfully say that I imagine his pain. In short, the image construed in the imagination does not playa role in the meaning or the understanding of terms related to other people's sensations.8
What both cases overlook is that the concept of sensation cannot be separated, (grammatically speaking), from the person having the sensation. 9 Because of this, when the aforementioned link is violated, something totally out of place is said. But how have we come this far? The PT introduces a gap (metaphysical and epistemic) between the psychological field (the phenomenical, the immediate, the internal) and the behavioural field (the observable, the indirect, the external). In this context, Wittgenstein insists that the gap itself is the product of a misunderstanding. The key to psychological concepts does not lie in "the inner" but rather in the links (conceptual, grammatical) established during learning, concept acquisition and mainly, in the complexities of the human situation: "what takes place inside is only meaningful in the flow of life".10
Then again, in the sentence "Only I know my pain", to know simply alludes to the fact that I do not doubt at all - I am never wrong - regarding whether something hurts me or not. But the point that is interesting to stress here is that in a context where by definition doubt and error are out of place should also be one where justification and knowledge are beside the question. If this is so, in the "I am in pain" language-game (avowals in general), there is no room for the epistemic, just like there is no room for goals in a tennis game. However, the misunderstanding of claiming that these expressions (cogito-like) constitute a genuine epistemic base leads to say things like "we do not know other minds but we do know our own". Also, by association, it invites the perennial philosophical puzzles (idealism, solipsism, scepticism regarding other minds) that are Wittgenstein's subject of criticism in the Investigations and in &302 concerning the issue we are debating. His main point, as I understand him, is to make evident a false idealization, a nonsense or a chimera which basically emerges when we speak in a way that is valid in one game of language in another game to which it actually does not belong to.11
1 "What gives us so much as the idea that living beings, things, can feel? Is it that my education has led me to it by drawing my attention to feelings in myself, and now I can transfer the idea to objects outside myself? That I recognize that there is something there (in me) which I can call 'pain' without getting into conflict with the way other people use this word? -I do not transfer my idea to stones, plants, etc." (PI: &283)
2 According to Malcolm the PT's defender implicitly rejects the notion of criterion: "A proponent of the privacy of sensations rejects circumstances and behavior as a criterion of others, this being essential to his viewpoint. He does not need (and could not have) a criterion for the existence of pain that he feels. But surely he will need a criterion for the existence of pain that he does not feel. Yet he cannot have one and still hold to the privacy of sensation" (Malcolm, p. 190, note 1).
3 The difficulty seems to be special because of the following: eventually I could make up the concept of duck by abstracting from those particular ducks. But given the weight of the Hume-Lichtenberg line in Wittgenstein, what could I abstract from to make up the concept of pain experienced by someone else if I wish to eliminate 'myself'?: "I have no concept of a 'self ' nor of 'having' to enable me to make the appropriate abstraction from the original paradigm" (Kripke, p. 125).
4 I do not fully understand Juan's position in the debate. Note 2 seems to indicate that his interpretation is in line with Kripke's sceptical interpretation but then the example of the Central Park ducks does not help to show that "Wittgenstein's" external argument, as Juan says, is not justified. It rather shows that Malcolm misinterpreted §302. On the other hand, if Kripke interpreted the paragraph correctly, and if Juan follows his line of thought, then the problem is that the subject could not avail itself of the concepts of a "self ' or of "having" from his own case, as the traditional discussion on other minds assumes.
5 Later, in "Knowledge of Other Minds", when Malcolm (1958) observes that the assumption (PT) is the "fundamental error" of all arguments by analogy (Mill, Russell, Price, Hampshire), he quotes and comments §258 as a palpable proof that the assumption "destroys itself'. He adds that in acknowledging the key role of the notion of criterion, the "problem" of other minds is immediately destroyed. Although I do not personally believe that Wittgenstein attempts to "solve" the problem (if any) with the notion of criterion, Malcolm appears to do so and his is the argumentative line chosen by Jonathan Dancy (1985). Therefore, I cannot quite see how the notion of criterion in the external argument can be eliminated, as Kripke and Juan wish it should.
6 I am not dealing with Kripke's peculiar interpretation by which "the real" PLA (the sceptical paradox) ends in §201 and what follows §243 is a corollary to its solution; for two reasons: 1) because Kripke himself admits that he does not reconstruct "Wittgenstein's argument" as it is but the argument as it struck him (Preface, viii); and 2) because he notoriously omits the second part of §201 where it is said that the paradox is the result of a "misunderstanding" and the reason for it.
7 Cf'. PI: §§245; 246; 247; 264; 274; 293; 294; 298; 398.
8 Cf. PI: &&390-396, 449.
9 "What would a pain that nobody had be like? A pain that did not belong to anybody at all?" (Philosophical Observations, VI, p. 65).
10 Zettel: &&532-548; (LW, II, &44. "Wouldn't it be ridiculous if a lawyer said in court that the witness cannot know if someone was angry because anger is internal?" (§85). "If I lie to him and he guesses it from my face and tells me so - do I still have the feeling that my internal self is in no way accessible to him but is hidden? Don't I rather feel that he sees right through me? It is only in certain cases that the internal is hidden from me and, by the way, it is not hidden because it is internal" (§33). "Mental is not a metaphysical, but a logical, epithet for me" (§63). "Indeed, logic explains the expression 'the internal', rendering it comprehensible" (§62). (Last writings on thephilosophy of psychology (LW, vol. II: The inner and the outer).
11 LW, vol. II, &44.
1. Dancy, Jonathan (1985),An Introduction to Contemporary Epistemology, Oxford, Blackkwell Publishers. [ Links ]
2. Kripke, Saul (1981), Naming and Necessity, Oxford, Basil Blackwell. [ Links ]
3. Kripke, Saul (1982), Wittgenstein on Rules and Private Language, Cambridge, Massachusetts, Harvard University Press. [ Links ]
4. Malcolm, Norman (1954), "Wittgenstein's Philosophical Investigations", in Ludwig Wittgenstein: The Man and His Philosophy, ed. H. T. Fann, New York, Dell Publishing, 1967. [ Links ]
5. Malcolm, Norman (1958), "Knowledge of Other Minds", in The Nature of Mind, ed. David M. Rosenthal, New York & Oxford, Oxford University Press, 1991. [ Links ]
6. Wittgenstein, Ludwig (1953), Philosophical Investigations, Oxford UK & Cambridge USA, Blackwell Publishers. [ Links ]
7. Wittgenstein, Ludwig (1941), "Notes for the Philosophical Lecture", ed. D. Stern, in Philosophical Ocassions, ed. J. Klagge and A. Nordmann, Indianapolis, Hackett Pub., 1993, pp. 447-458. [ Links ]
8. Wittgenstein, Ludwig (1946-1947), Wittgenstein's Lectures on Philosophical Psychology, ed. P. T. Geach, Chicago, The University of Chicago Press, 1988. [ Links ]
9. Wittgenstein, Ludwig (1967), Zettel, ed. G. E. M. Anscombe & G. H. von Wright, Berkeley and Los Angeles, California, University of California Press. [ Links ]
10. Wittgenstein, Ludwig (1948-1949), Last Writings on the Philosophy of Psychology, ed. G. H. von Wright & H. Nyman, Oxford, Blackwell Publishers, 1982. [ Links ]
11. Wittgenstein, Ludwig (1949-1951), Last Writings on the Philosophy of Psychology, ed. G. H. von Wright & H. Nyman, Oxford, Blackwell Publishers, 1992. [ Links ]