“Whose house is that?”
Wittgenstein on the Self
“Think of a picture of a landscape, an imaginary landscape with a house in it. – Someone asks ‘whose house is that?’ – The answer, by the way, might be ‘It belongs to the farmer who is sitting on the bench in front of it.’” (PI, 398) Wittgenstein tells this story in the midst of a discussion on the self, the I, or better: on the ways we use the word “I”. The gist of this discussion appears to be that there is nothing to which the word “I” refers, that there is no self. But what is the point of the jocular story in these considerations? Does it suggest that it is not I who imagines the landscape, house, and farmer? Does the farmer stand for the self even though there is no such thing? What does it mean that he can own but not enter the imagined house? What does the story tell us about the I and/or the “I”? There is a lengthy answer to all this.
Wittgenstein reflected on questions concerning the mind, mental states, processes, and acts throughout his life and in that context he came regularly back to the I, self, soul, or subject as he called it more or less indiscriminately. The first time he talks about the mind – in the middle of a finicky passage on the things that make up the world – he announces unexpectedly: ”There really is only one world soul, which I for preference call my soul…” (NB, p. 49) and a few weeks before his death, thirty-six years later, he writes in another notebook: “But it is still false to say:… I is a different person from L.W.” (LWPP II, p. 88). No wonder that somewhere on that road he came to think that “The I, the I is what is deeply mysterious.” (NB, p. 80)
The scope and complexity of his writings on the mind establish that these were not by-products of other, more immediate concerns – such as an interest in language. Wittgenstein said in the 1930s that philosophy needed to deal only “with those points about language which have led, or are likely to lead, to definite philosophical puzzles or errors.”
Everything else he considered “not the philosopher’s business.” (G.E. Moore, “Wittgenstein’s Lectures in 1930-33,” Philosophical Papers, Allen & Unwin, London 1959, p. 324.) The same attitude is reflected in the Tractatus. where Wittgenstein writes that he is concerned with “the problems of philosophy” and that language is of interest to him because these problems arise out of a misunderstanding of “the logic of our language.” (TLP, p. 3) In the Philosophical Investigations he writes in a similar vein that “the work of the philosopher consists in assembling reminders for a particular purpose.” (PI, 127) The philosopher, we might spell this out, assembles reminders about how language is used for the philosophical purpose of uncovering “plain nonsense” and “bumps that the understanding has got by running its head up against the limits of language.” (PI, 119) Again, it is the philosophical puzzle that motivates the concern with language.
This does not mean that he had no genuine interest in language, but that it was always motivated by philosophical considerations. Philosophical questions concerning language he saw, in turn, linked to questions concerning the mind. To trace Wittgenstein’s discussion of the self means, therefore, to trace the complex web of connections between questions of mind and language.
As one reads Wittgenstein, one must keep an eye on the changing landscape of his thought. But it is just as important to be alert to the continuities in his thinking and such continuities are particularly visible in what he says about the mind. One unifying theme in this area is his enduring hostility to the idea of an individuated substantive self. Insofar as the belief in such a self is most easily associated with Descartes, we can call Wittgenstein’s position an anti-Cartesianism.
How much did Wittgenstein actually know about Descartes? How much had he read? There are no reliable estimates concerning either question. He did discuss Descartes in connection with the “Cogito”, but insisted that it was unnecessary to reconstruct his thinking, instead “one must do this for oneself.” I take the “this” here to refer to a rethinking of Cogito that will determine what made Descartes’s view plausible to him but not necessarily to us. (O.K. Bouwsma, Conversations 1949-1951, Indianapolis, Hackett 1986, pp. 12-14.)
This attitude is already evident in the Tractatus where he asserts that “there is no such thing as the subject that thinks or entertains ideas.” (TLP, 5.631) That he intended this to be an anti-Cartesian remark is made clear in the Blue Book where he writes, first, that our language creates the illusion that the word “I” refers to “something bodiless, which, however, has its seat in our body,” and then concludes: “In fact this seems to be the real ego, the one of which it was said, ‘Cogito, ergo sum’.” (BB, p. 69) He returns to the theme once again in the Philosophical Investigations where he writes: “I” is not the name of a person, nor “here” of a place, and “this” is not a name. But they are connected with names. Names are explained by means of them. It is also true that it is characteristic of physics not to use these words. (PI, 410)
In order to appreciate how deeply anti-Cartesianism infuses Wittgenstein’s thinking about the mind, we must look at a notoriously obscure passage in the Tractatus where this idea is first vented (TLP, 5.54-5.55) – a passage that begins with the claim that “propositions occur in other propositions only as bases of truth-operations” (5.54) and proceeds from there at hazardous speed to the conclusion that “there is no such thing as the soul.” (5.5422) The argument is so dense, that it requires some spelling out to make it transparent.
The first requirement for that is to pay close attention to the truth-functionality thesis with which the passage starts. The thesis says that the relations between a complex proposition made up out of simpler ones and those simpler propositions themselves are inevitably truthfunctional in character. But the claim appears to run into immediate counter-examples. Are there not contexts in which the truth and falsity of the complex proposition is independent of the truth or falsity of the component ones? Are modal propositions like “It is possible that p” or “it is necessary that either p or q” not such counter-examples? Wittgenstein sees himself forced to admit that “at first sight it looks as if it were also possible for one proposition to occur in another in a different way.” (5.541) He points to what he calls “certain forms of proposition in psychology,” i.e., propositions of the form “A believes that p” and of the form “A has the thought p” where “A” is taken to stand for some subject or self and “p” for a proposition believed or entertained by A. By focusing on these propositions he has forged a link between his discussion of the logical structure of complex propositions and the question how we are to understand the relation between a judging, believing subject and the contents of his judgments or beliefs. It remains to be seen whether that linkage is accidental.
Wittgenstein’s counter-examples are, clearly, open to three different responses. The most straightforward one would be to abandon or to restrict the scope of the truth-functionality thesis. A second response would be to maintain that the apparent counter-examples are not bona-fide propositions at all. A third, and perhaps the most problematic response, would be to argue that the apparent counter-examples do not, in fact, conflict with the truth-functionality thesis. That would amount to saying that they do not really have the logical form which they seem to possess.
It is this third response which Wittgenstein makes his own, as he indicates right from the start when he writes that “at first sight it looks as if” the apparent counter-examples were exceptions to the truth-functionality thesis. His response requires, first of all, an appeal to the distinction between the apparent and the real form of the proposition which Wittgenstein had made earlier in the Tractatus when he had written: “Language disguises thought. So much so, that from the outward form of the clothing it is impossible to infer the form of the thought beneath it.” (4.002) In discussing his apparent counter-examples to the truth-functionality thesis he contrasts accordingly what they look like “if they are superficially considered” with their supposedly real form. (5.541 and 5.542) Superficially considered it looks in both cases, “as if the proposition p stood in some kind of relation to an object A.” (5.541) That relation would not be truth-functional in character and, hence, on the superficial account these propositions would indeed be counter-examples to the truth-functionality thesis.
At this point Wittgenstein introduces a new thought. It is that “modern theory of knowledge” actually takes the superficial appearance of these propositions to display their real logical form. (5.541) Modern theory of knowledge is thus tainted by a pervasive philosophical flaw since “most of the propositions and questions of philosophers arise from our failure to understand the logic of our language.” (4.003) In case we are in any doubt, Wittgenstein goes on to mention “Russell, Moore, etc.” as representatives of the modern theory of knowledge he is attacking. It may surprise us to find him accusing Russell here of a failure to distinguish between the superficial appearance of a proposition and its real logical form, since elsewhere in the Tractatus he credits him with having introduced that distinction into philosophy. “It was Russell who performed the service of showing that the apparent logical form of a proposition need not be its real one.” (4.0031) It has long been understood, however, that Wittgenstein’s criticism at 5.541 is directed against a specific Russellian text, a book on the theory of knowledge which Russell was writing while Wittgenstein was in Cambridge and which he never completed as a result of Wittgenstein’s criticism. ( David Pears, “The Relation between Wittgenstein’s Picture Theory of Propositions and Russell’s Theories of Judgment”, in C.G. Luckhardt, Wittgenstein. Sources and Perspectives, Ithaca N.Y.: Cornell Univ. Pr. 1979, pp. 190-212.)
In that text Russell had argued for a broadly Cartesian conception of the self as a necessary presupposition for a coherent theory of meaning. ( I call Russell’s conception “broadly Cartesian” since I do not want to imply that he adopted the Descartes’s complete picture of the self. Russell did hold that (1) there is a real self which is (2) capable of thought and perception, and (3) logically simple.) He had reached that conclusion in the context of a question that had long preoccupied him, the question of the unity of the proposition. Already in the Principles of Mathematics of 1903 he had pointed out that there is surely a difference between a proposition and a list of its components. The first, he had said, exhibited a peculiar kind of unity which was absent from the second. But how was this unity to be explained? Russell’s initial response had been to adopt an essentially Fregean solution and to argue that within every proposition there occurs one component which ties the elements of the proposition together. But by 1903 he had already convinced himself that the conception of propositions as real unities was problematic, since it seemed to lead to a type of logical antinomy. By 1911 he had, therefore, come around to thinking that the solution lay in the assuming that the apparent unity of a proposition was in each case due to a thinking subject holding the elements of the proposition together in its thinking. Every proposition “p” was thus really of the form “A judges/thinks that p”. The unity of the proposition was grounded in the unity of the thinking subject which stood in a relation of thinking or believing to the proposition. A broadly Cartesian conception of the mind, thus, appeared to be a pre-requisite for a satisfactory theory of meaning.
It was precisely this theory which Wittgenstein attacked in sections 5.54ff of the Tractatus. Once we are aware of this, we can see that his choice of the apparent counter-examples to the truth-functionality thesis was motivated by the idea that he would, thus, be able to defeat, at one stroke, both a certain theory of meaning and an associated theory of the mind. His argument can therefore be understood to be illustrating the claim (made in the preface of the Tractatus) that misunderstandings of the logic of our language give rise to philosophical problems and as a result generate philosophical theories. In the passage beginning with 5.54 Wittgenstein is clearly showing how Russell’s misconceived account of the unity of the proposition generates the need for his misguided conception of the self.
Against Russell, Wittgenstein declares at this point simply that “it is clear, however, that ‘A believes that p’ and ‘A says p’ are of the form ‘”p” says p’.” (5.542) The remark must remain mysterious until we are clear about the form of the proposition “‘p’ says p”. It evidently speaks of a relation between the proposition “p” and the situation p; but what kind of relation is here under discussion? Wittgenstein characterizes it elsewhere in the Tractatus in the words: “This proposition represents such and such situation.” (4.031) He also says that in order for the proposition “p” to succeed in this task of representation, it must have some structure to it. “It is only in so far as a proposition is logically articulated that it is a picture of a situation.” (4.032) In fact, the proposition and the situation must, in some way, be equivalent to each other. “In a proposition there must be exactly as many distinguishable parts as in the situation that it represents. The two must possess the same logical (mathematical) multiplicity.” (4.04)
These words contain an implicit critique of Russell’s view of propositions. They say that a proposition has a unity in itself which does not depend on a thinking or judging subject. A proposition is essentially articulate and as such is “not a blend of words.” (3.141) In the proposition there exists, rather, a “nexus” between the signs. “Only in the nexus of a proposition does a name have meaning.” (3.3) The articulated proposition is for that reason itself a fact – i.e., something in which the elements hang together like links in a chain. (3.14) We can see then that the sentence “‘p’ says p” expresses for Wittgenstein a relation between a fact (i.e., the proposition) and a situation that has the same logical multiplicity. As he says in the passage under discussion: the sentence “does not involve a correlation of a fact with an object, but rather the correlation of facts by means of the correlation of their objects.” (5.542)
When Wittgenstein says that “A believes that p” and “A says p” are really of the form “‘p’ says p” he is, in effect, telling us that believing and saying (as well as thinking to which his earlier example referred) are or involve representation relations. In order for A to believe that p or to say p, A must be able to represent the content of the proposition “p” to itself. But this means that A or something in A (that which does the representing) must have the same logical multiplicity as p; A or something in A must, as a result, be itself a fact. It cannot be a simple object. The idea that a Cartesian self can represent situations for itself must, therefore, be abandoned as logically incoherent. Two things are apparent as far as Wittgenstein is concerned. The first is that the unity of the proposition cannot be sought in the Cartesian subject that Russell had postulated. Russell’s account can, in fact, not explain why only certain combinations of words form propositions. “A correct explanation of the form of the proposition ‘A makes the judgment p’, must show that it is impossible for a judgment to be a piece of nonsense. (Russell’s theory does not satisfy that requirement.)” (5.5421) Since for Russell the unity of the proposition is brought about by a subject holding its components together in consciousness, the elements “drinks,” “eats,” “merry,” should under appropriate conditions form the content of a meaningful judgment, which they clearly do not.
Wittgenstein applies this account immediately to the phenomenon of Gestalt perception. With respect to the well-known Necker cube he asserts that “there are two possible ways of seeing the figure” because “we really see two different facts.” (TLP, 5.5423) I take the remark to mean that the shifting perception is not to be explained by saying that the subject perceives the cube in two different ways; the explanation is rather that there are two “ways of seeing”, two perceptual representations, verbally expressed as (1) the a points of the cube are in front of the b points and (2) the b points are in front of the a points.
These representations picture two different facts, two actually different positions of the cube in space, and they stand to these facts in the relation of a logical isomorphism. The perceiving subject, thus, drops out of the account of Gestalt perception. The second thing that becomes apparent to Wittgenstein at this point is that the subject cannot be conceived of in Cartesian terms as both simple and representing (i.e. thinking, believing, judging etc.). These two characteristics, which the classical modern tradition from Descartes through Leibniz to Russell has taken to be compatible, are, in fact, not so. And with this observation he cuts through the Gordian knot of the modern conception of the subject. The idea that a simple self could also be a representing self is, indeed, absurd. That conclusion might lead one to postulate, instead, a complex representing subject. But Wittgenstein is certain that this is an equally unviable notion. Laconically he writes: “A composite soul would no longer be a soul.” (5.5421) While he does not tell us from where he takes the force of this conviction, its radical consequences within the structure of the Tractatus are obvious.
For Wittgenstein sees himself now justified to conclude “that there is no such thing as the soul.” This can hardly be called a proof, since the premise that the soul cannot be composite is not argued for. And it is, in fact, not clear that Wittgenstein means to be giving a proof. When he says: “This shows too that there is no such thing as the soul” we need not take the word “shows” to mean “proves”; we might take the sentence instead to be saying that this makes once again evident that there is no such thing as the soul. Such a reading fits the fact that he ascribes belief in a soul to “the superficial psychology of the present day.” And this “psychology” – which we might want to interpret literally here as a theory of the psyche, i.e. of the soul or the subject – had already been dismissed earlier as “no more closely related to philosophy than any other natural science,” (4.1121) i.e., as irrelevant to philosophy. The context makes clear that Wittgenstein associates “the superficial psychology of the present day” with Russell’s theory of knowledge. Theory of knowledge as a whole is dismissed as “the philosophy of psychology.” (TLP, 4.1121)
By the time Wittgenstein was writing the Tractatus anti-Cartesianism was, of course, a well-rehearsed idea and objections to the Cartesian conception of the mind were familiar. Hume and Kant, Schopenhauer and Nietzsche, as well as Mach and Freud come to mind as exponents of one or another form of anti-Cartesianism. In Beyond Good and Evil, for instance, Nietzsche argues against what he calls “soul atomism,” a view he characterizes as “the belief which regards the soul as something indestructible, eternal, indivisible, as a monad, as an atomon,” adding, “this belief ought to be expelled from science.” (Friedrich Nietzsche, Beyond Good and Evil, sect. 12.) Shortly after, Mach writes: “The primary fact is not the ego, but the elements (sensations)… The elements constitute the I… The ego must be given up.” ( Ernst Mach, The Analysis of Sensations, pp. 23-24; originally published 1897, Engl. transl. C.M. Williams, Dover Publications, New York 1939.)
By 1918 Wittgenstein must have been aware of these ideas. He was certainly familiar with Schopenhauer by then, he had also studied parts of Kant’s Critique of Pure Reason, knew some of Nietzsche’s writings, and was acquainted with Mach’s Analysis of Sensations from which the quoted sentences are taken. But anti-Cartesianism is only one component in Wittgenstein’s thinking about the mind and not the most original at that. What makes his position unusual is rather the view that underlies his anti-Cartesian stand. We may call it, for short, Wittgenstein’s anti-objectivism. In his Notebooks he writes, for instance, in accordance with his anti-Cartesian position that “the thinking subject is surely mere illusion,” but he adds to this what I call the anti-objectivist claim that “the I is not an object.” (NB, p. 80) I take this to mean that the I cannot be a constituent of the world at all, that the word “I” can neither be a name for a simple object nor a description of a complex. ( My interpretation of the statement “The I is not an object” assumes that Wittgenstein is not using the word “object” to refer exclusively to the simple objects of the Tractatus. My assumption is, rather, that his terminology corresponds here to that of other parts of the Notebooks where he speaks of complex as well as of simple objects. (NB, pp. 50, 59, 60 etc.) Understood in this sense, the assertion can be taken to mean that the I is not a constituent of the world at all and that a full description of the world, i.e., a description which lists all simples and all complexes, would not mention the I. That view is, indeed, made explicit both in the Notebooks and in the Tractatus. In both places Wittgenstein insists that a complete description of the world, “a book called The World as I found it”, would not mention the I, the subject. (NB, p. 50 and TTLP)
A superficial reading of these remarks might convince one that Wittgenstein commits himself here to a simple-minded eliminative reductionism according to which all there is, in any possible sense of the word, is the physical world describable by natural science and which concludes that within this world there occurs no subject. One might, in other words, conceive of Wittgenstein’s position in the Tractatus as a radical objectivism of a behaviorist, materialist, or (more generally speaking) physicalist kind. But this would be to misunderstand his views altogether. While he took the physicalist picture of empirical reality for granted, he also thought that physicalism was philosophically insufficient because it could not provide an understanding of human subjectivity. The observation that the I is not an object means for him that objectivism proves unworkable with respect to the self. Wittgenstein holds that the I cannot be conceived as any kind of object: it can be neither a mental, Cartesian substance, nor a material thing; but the subject is thereby not abolished. This is made explicit in his further observation that “I objectively confronts every object. But not the I.” And, if this was not sufficiently clear as an expression of anti-objectivism, he adds: “So there really is a way in which we can and must acknowledge the I in a non-psychological sense in philosophy.” (NB, p. 80)
The sentence and its counterpart in the Tractatus (TLP, 5.641) have proved difficult for the translators. Wittgenstein writes in German that “vom Ich die Rede sein kann und muss”. Since “Rede” commonly means “speech,” Anscombe translates the sentence in the Notebooks as saying that “there can and must be mention of the I;” Pears and McGuinness render it in the Tractatus as “philosophy can talk about the self.” Both translations make Wittgenstein’s view self-contradictory. For if we can meaningfully speak only of items in the world, as he asserts over and over again, then a self we can talk about must be part of the world and then it we can also talk about it objectively and scientifically. The translators have failed to notice that the German “Rede” derives from Latin “ratio” and that the original meaning is preserved in the phrase “davon kann die Rede sein” which says roughly “such and such is reasonable.”
To acknowledge the I here in a non-psychological sense means to consider it as something which is not subject to scientific theorizing and which is not objectively given as part of the world. The I has not been reduced to nothing, but it also does not have the status of an object in the world, of a something. We are reminded here of Wittgenstein’s analogous later point that the sensation of pain is “not a something, but not a nothing either.” (PI, 304) In parallel we might say here that according to the Tractatus the I is not a something, but not a nothing either. To this we must add another idea. For the anti-objectivism of the Tractatus is linked to a view which we may label “referentialism” for short. The referentialist view assumes that every meaningful expression which is not a sentence or a logical symbol has meaning by either naming a simple object or describing a complex. Since Wittgenstein has concluded, in accordance with his anti-objectivism, that the I is neither a simple object nor a complex, it follows that any word we might want to use to speak about the subject must, according to the assumptions of referentialism, be meaningless. In the Notebooks and in the Tractatus the peculiar status of the self is explained by the observation that “The I makes its appearance in philosophy through the world’s being my world.” (NB, p. 80 and TLP, 5.641) While there is no such thing as a worldly subject or self for Wittgenstein, there still remains for him the phenomenon of subjectivity. That is made visible in the fact that a complete description of the world will not (and need not) mention the I, but the world so described is still called “The World as I found it”. In other words: the objective world has to be conceived as a world given to a subjectivity and it is in this that the subject makes its appearance.” (For an earlier discussion of these ideas see my essay “Subjectivity in the Tractatus”, Synthese, vol. 56, 1983.)
On the Tractarian account there can, in any case, be no objective account, no science of the subject at all, since science deals only with objects in the world and their relations. Psychology as a science of the soul is therefore impossible. But the meaning of non-objectivism is for Wittgenstein not exhausted in this negative observation, as we have seen. Non-objectivism means for him, rather, that the I is given in a non-objective way and not as an object in the world. As Wittgenstein writes: “The philosophical self is not the human being, not the human body, or the human soul, with which psychology deals, but rather the metaphysical subject, the limit of the world – not a part of it.” (TLP, 5.641) And it is useful to remind ourselves here that “the limits of my language signify the limits of my world.” (5.6) Hence, that which is conceived as the limit of the world must also be conceived as being at the limit of language. While we can give an exhaustive objective, scientific description of the world, according to Wittgenstein, that description cannot touch on the (transcendental) fact that the world is after all my world. This fundamental feature of subjectivity cannot be accounted for by postulating an objectively available subject (or objectively available subjects) within the world. The mental is not a sphere within the world nor is it an object outside the world; “the metaphysical subject” is, rather, the non-objective condition of the possibility of the objective world. On this account the self is neither a Humean or Machian logical construct, nor a Kantian subject that is somehow both an empirical and a transcendental consciousness, nor a causal construct as Nietzsche and Freud would have it. All these philosophers remain caught within the framework of objectivism.
For Kant the empirical self is an object within empirical reality and thus has its own causal efficacy; Nietzsche and Freud see themselves as psychologists, as investigators of an objectively constituted ego; even Hume and Mach who speak of the self as a fictional object treat it thereby as an object. Wittgenstein’s seeks to set himself off from all of them. But that attempt carries a heavy price. The relation of the Wittgenstein’s philosophical self to the everyday self of which we commonly speak remains unspecified. Unlike the Cartesian self, the philosophical self is unindividuated and Wittgenstein describes it accordingly also in his Notebooks as a “world soul.” Wittgenstein may have taken this notion from one of a number of different sources. A plausible one is William James, The Principles of Psychology, Henry Holt, New York 1890, vol. 1, p. 346. When we conceive it in this way, it becomes impossible to speak about a plurality of subjects. Wittgenstein’s view thus appears to force him into a transcendental solipsism for which “there is really only one world soul, which I for preference call my soul.” (NB, p. 49)” (The road to this conclusion and its difficulties are spelled out in “Subjectivity in the Tractatus”, loc. cit.).
It took Wittgenstein great effort to overcome this transcendental solipsism, which initially seemed to him as both problematic and unavoidable. His later reflections on the mind can be read as one long argument designed to show that one can hold on to the insights of anti-Cartesianism and anti-objectivism without falling into such a solipsism. However, in order to reachthat conclusion, he first had to change three assumptions about language and meaning.
(1) In the Tractatus Wittgenstein had assumed that language constitutes a formal unity. While he had allowed that “man possesses the ability to construct languages” (4.002), he had also argued that insofar as these languages are intertranslatable, “they are all constructed according to a common logical pattern… They are all in a certain sense one.” (4.014) By the early 1930s Wittgenstein came to think that language is made up of a number of distinct language-games. In illustration of this point he wrote at the time: “The picture we have of the language of the grown-ups is that of a nebulous mass of language, his mother tongue, surrounded by discrete and more or less clear-cut language-games, the technical languages.” (BrB, p. 81) And still later he wrote more generally: “We see that what we call ‘sentence’ and ‘language’ has not the formal unity that I imagined, but is the family of structures more or less related to one another.” (PI, 108)
(2) This new openness went hand in hand with the rejection of the idea that our sentences are meant to mirror the logical structure of the world. We cannot justify the choice of a system of notation by reference to the structure of the world. Wittgenstein is ready to accept now the legitimacy of the solipsist’s preferred way of speaking. ”There is, as we have said, no objection to adopting a symbolism in which a certain person always or temporarily holds an exceptional place… What, however, is wrong, is to think that I can justify this notation… There is nothing wrong in suggesting that the others should give me an exceptional place in their notation; but the justification which I wish to give for it: that this body is now the seat of that which really lives – is senseless. “(BB, p. 66)
This principle of tolerance is made possible, because Wittgenstein holds no longer that language serves a single function, that of depicting reality. In the Tractatus he had written: “A proposition is a model of reality as we imagine it.” (TLP, 4.01) And: “The totality of propositions is language.” (TLP, 4.001) But now he argued that language may serve very different needs. “Thus we sometimes wish for a notation which stresses a difference more strongly, makes it more obvious, than ordinary language does, or one which in a particular case uses more closely similar forms of expression that ordinary language. Our mental cramp is loosened when we are shown notations which fulfill these needs. These needs can be of the greatest variety.” (BB, p. 59) The crucial point for him was that language is primarily a “system of communication” rather than one of representation. (BrB, p. 81)
(3) And this recognitions of different functions of language went together with the rejection of Wittgenstein’s earlier referentialism. He called referentialism now “one of the great sources of philosophical bewilderment.” (BB, p. 1) The mistake, he said, was “that we are looking for the use of a sign, but we look for it as though it were an object co-existing with the sign.” (p. 5) This had the consequence that, whenever we look for such an object and cannot find it, we may start looking for it in an “aethereal” sphere. We appeal to the gaseous and aethereal in philosophy “when we perceive that a substantive is not used as what in general we should call the name of an object, and when therefore we can’t help saying to ourselves that it is the name of an aethereal object.” (p. 47) The assumptions of an invisible Platonic realm, of an inaccessible transcendent reality, of a transcendental perspective, and of Cartesian mentalism are all dismissed as a subterfuge “when we are embarrassed about the grammar of certain words, and when all we know is that they are not used as names for material objects.” (Ibid.)
These changes in Wittgenstein’s thinking about language and meaning were based on a variety of grounds. The Blue Book shows, however, that they are intimately linked to his concerns with the nature of the mind. His remarks about the aetheral as a subterfuge in that text are characteristically followed by the assertion: “This is a hint as to how the problem of the two materials, mind and matter, is going to dissolve.” (Ibid.) The Blue Book gives us, in fact, reasons to think that the shifts in Wittgenstein’s thinking about language are intimately connected with his worries about the nature of the mind.
In considering how that work tackles this issue we must pay attention to the fact that Wittgenstein dictated it to his students at Cambridge over a period of two terms and that the text therefore into two natural parts of roughly equal size. The first deals with various problems surrounding the concept of meaning while the second deals almost exclusively with questions in the philosophy of mind, and specifically with the problem of solipsism.
The general point of the discussions in the first half of the book is summed up at the end of that section when Wittgenstein says that the investigation of how we use words like “thinking”, “meaning”, “wishing” etc. “rids us of the temptation to look for a peculiar act of thinking, independent of the act of expressing our thoughts, and stowed away in some peculiar medium.” (p. 43) The conclusion, he also says, is that “the experience of thinking may be just the experience of saying” (Ibid.) One might calls this an anti-mentalistic conclusion and this reading is confirmed by the details of Wittgenstein’s exposition. He writes, for instance:“We are tempted to think that the action of language consists of two parts; an inorganic part, the handling of signs, and an organic part, which we may call understanding these signs, meaning them, interpreting them, thinking. These latter activities seem to take place in a queer medium, the mind; and the mechanism of the mind, the nature of which, it seems, we don’t quite understand, can bring about effects which no material mechanism could.” (p. 3) Wittgenstein rejects this view by insisting that what we might call “the life of the sign” is really its use. “And when we are worried about the nature of thinking, the puzzlement which we wrongly interpret to be one about the nature of a medium is a puzzlement caused by the mystifying use of our language.” (p. 6) When we think of “the meaning” of a sign as something necessarily located in the medium of the “the mind” we are making the same logical mistake twice over, i.e., the mistake of “looking for a ‘thing corresponding to a substantive.'” (p. 5)
Wittgenstein seems to be saying that there is no such thing as the mind and no such thing as the meaning of a sign, that there are only uses of language. Insofar as using language is a behavior we might classify this view as a form of behaviorism. While this appears to be the conclusion of part one of the Blue Book, it is not, however, the conclusion of the book as a whole. Part two of the Blue Book begins, rather, with the warning that once we start to consider the nature of personal experience “all we have said about signs and the various objects we mentioned in our examples may have to go into the melting-pot.” (p. 44) There follows a lengthy metaphor according to which handling philosophical problems is like arranging the books in a library.
As we proceed we may have to move books again and again to new shelves. The point is that “no philosophical problem can be solved until all philosophical problems are solved; which means that as long as they aren’t all solved every new difficulty renders all previous results questionable.” (p. 44) Wittgenstein draws our attention next to the different kinds of propositions in our language. “There are propositions of which we may say that they describe facts in the material world (external world)… There are on the other hand propositions describing personal experiences, as when the subject in a psychological experiment describes his sense-experiences.” (pp. 46-47) This acknowledgement removes, in one stroke, his earlier claim in the Tractatus that all meaningful propositions will have to be propositions of natural science describing facts of external reality. His principle of tolerance allows him now to acknowledge the existence of meaningful descriptions of personal experience.
Wittgenstein goes on to say that the existence of these two types of proposition has encoraged a number of different interpretations. “At first sight it may appear… that here we have two kinds of worlds, worlds built of different materials; a mental world and a physical world.” (p. 47) Another interpretation holds that “the mental phenomena, sense experience, volition, etc., emerge when a type of animal body of a certain complexity has been evolved.” (p. 47) A third view maintains that “personal experience, far from being the product of physical processes, seems to be the very basis of all that we say with any sense about such processes.” (p. 48) Finally, there is the view that “the whole world, mental and physical, is made of one material only.” (Ibid.) What Wittgenstein describes here are the positions of Cartesian dualism, materialist emergentism, idealism, and neutral monism.
It is, however, clear that he wants to adopt none of these – and, hence, presumably also not the materialist behaviorism that seems to be suggested in the first part of the Blue Book. He insists, rather, that the common-sense man, with whom he identifies here, “is as far from realism as from idealism”. (p. 48) The remark reminds us of his earlier assertion that metaphysics “leads the philosopher into complete darkness” (p. 18) and is in accord to the disparaging later comment in the Philosophical Investigations that disputes between idealists, solipsists and realists are merely disagreements about forms of expression and not about “facts recognized by every reasonable human being.” (PI, 402) Or, as he puts it in the Blue Book: “We are up against trouble caused by our way of expression.” (p. 48)
The difference between propositions describing the external world and propositions describing sense-experience is then not to be explained metaphysically. Anti-objectivism remains an integral component of Wittgenstein’s reflections on the mind in the Blue Book. But in the light of his modified views about language and meaning it now takes on a new form. Where the distinction between the objective and the non-objective had previously been thought to coincide with the distinction between that which is sayable and that which is not sayable, it is now taken to coincide with a distinction between two different ways of speaking that serve different human needs. This insight is now put to the service of rethinking the whole area of the philosophy of mind. Wittgenstein is now ready to admit that the word “I” can serve a meaningful function in language. He assumes it to have, in fact, two distinct functions. There is, as he puts it, its “use as object” and its “use as subject.” (p. 66) His examples make clear that he means to say that we use “I” (or “my”) to refer to an object when we use it to speak of a human body and its bodily characteristics. Hence, we have “My arm is broken”, “I have grown six inches”, “I have a bump on my forehead”, and “The wind blows my hair about.” The word “I” is used as subject, on the other hand, when we speak of mental states, mental processes, and sensations such as seeing, hearing, trying, thinking, feeling pain. Wittgenstein’s examples are: “I see so-and-so”, “I hear so-and-so”, “I try to lift my arm”, “I think it will rain”, “I have toothache.” (pp. 66-67)
But what are the different functions of the term in these two uses? What different needs are being served by them? In the first kind of sentence, Wittgenstein says, an object is identified and something is said about it. In the second no object is referred to. It follows that “to say, ‘I have pain’ is no more a statement about a particular person than moaning is.” (p. 67) By uttering “I have pain” I am not trying to state a fact; instead I am trying to “attract attention” to myself. By means of a number of thought experiments Wittgenstein seeks to show that in such utterances the word “I” does not serve as a description of a body. It has, in fact, no referential function at all. We fail to recognize this only because we are inclined to take every meaningful noun to be standing for an object. “We are up against one of the great sources of philosophical bewilderment: a substantive makes us look for a thing that corresponds to it.” (p. 1) Referentialism in conjunction with the correct observation that the word “I” does not denote a physical body in certain contexts, is then seen as the source of the Cartesian conception of the self. The last sentences of the Blue Book says accordingly: “The kernel of the proposition that that which has pains or sees or thinks is of a mental nature is only, that the word ‘I’ in ‘I have pains’ does not denote a particular body.” (p. 74) We might say that referentialism in conjunction with the observation that the word “I” does not denote a body drives us into Cartesianism. In order to escape from this metaphysical position we must, first of all, abandon referentialism and that, in turn, will allow us to abandon objectivism. This is, in effect, the message of the crucial statement from the Blue Book with which I initially substantiated his anti-Cartesianism. For Wittgenstein says here that “we feel that in the case in which ‘I’ is used as subject, we don’t use it because we recognize a particular person by his bodily characteristics; and this creates the illusion that we use this word to refer to something bodiless, which, however, has its seat in our body.” (p. 69)
The argumentative strategy of the Blue Book seems clear. Having convinced ourselves in the first half that the mind is not a “queer medium” and that thinking is just an operating with signs, we are made to see in the second half that materialism and behaviorism cannot explain propositions like “I am in pain.” The word “I” cannot be taken to refer to something material and the predicate “being in pain” cannot be taken to refer to some physical state or physical behavior of such a material thing. Given the assumption that either materialist behaviorism must be true or a Cartesian mentalism, the insoluble difficulties of the former seem to drive us into the latter – a conclusion that is in flat conflict with the results of part 1 of the Blue Book. We can resolve the dilemma only when we come to see that the alternative “either materialist behaviorism or Cartesian mentalism” rests on the assumption that objectivism is true: If objectivism is true, then either materialist behaviorism or Cartesian mentalism. Since part 1 of the Blue Book has produced compelling arguments against mentalism and part 2 equally compelling arguments against behaviorism, we must conclude that objectivism is false.
There remain two serious problems with the Blue Book account. The first is that Wittgenstein admits the existence of connections between statements like “My arm is broken” and “I am in pain” but fails to account for these connections. As a result he comes close to a “conceptual dualism” according to which the two utterances belong to two different language games and are, thus, logically and conceptually independent of each other. The shortcomings in the argument are taken up in the Philosophical Investigations where Wittgenstein offers a more detailed and more satisfactory treatment of this difficulty. The second problem left unresolved in the Blue Book concerns the positive content of Wittgenstein’s non-objectivist understanding of the mind. In the Blue Book he says very little about this matter. In the passage from which I have just quoted he answers the question whether on his conception there is then no mind by saying: “The word ‘mind’ has meaning, i.e., it has a use in our language; but saying this doesn’t yet say what kind of use we make of it.” (pp. 69-70) That response might strike one as evasive, since he does not go on to say what meaning the word “mind” has for him. And this evasion, if that is the right word, continues in Wittgenstein’s later writings. Neither the Philosophical Investigations nor the subsequent notes overcome it, though these writings give evidence of his lasting preoccupation with these matters.
In the Philosophical Investigations Wittgenstein continued to affirm that the physical and the mental belong to different language games, but also now acknowledged the need to explain their relations in detail. Accordingly he wrote:”Here we have two different language-games and a complicated relation between them. – If you try to reduce their relations to a simple formula you go wrong. “(PI, p. 180)The crucial insight that takes him beyond the position reached in the Blue Book is expressed in the formula that “an ‘inner process’ stands in need of outward criteria.” (PI, 580) It assures us of a linkage between utterances of the form “His arm is broken” and “He is in pain” as well as between “My arm is broken” and “I am in pain.” Wittgenstein illustrates the point as follows:”This is how I think of it: Believing is a state of mind. It has duration; and that independently of the duration of its expression in a sentence, for example. So it is a kind of disposition of the believing person. This is shown me in the case of someone else by his behavior; and by his words. And under this head, by the expression “I believe…” as well as by the simple assertion. (PI, p. 191-192)
The behavior, Wittgenstein says, serves here as my criterion for ascribing an enduring belief. Without such external criteria we would not be in a position to ascribe inner states. It follows that “only of a human being and what resembles (behaves like) a living human being can one say: it has sensations; it sees; hears; is deaf; is conscious or unconscious.” (PI, 281) For that reason, too, we can imagine pain in the wriggling fly but not in the dead rock, unless we imagine a living being embedded in it. Such considerations might make it appear that Wittgenstein has now, after all, slipped into the materialist behaviorism that he had tried to resist in the Blue Book. He is well aware of the possibility of this misunderstanding and, therefore, repeatedly and vehemently rejects the charge of behaviorism. He writes in five successive remarks: (304.) “But you will surely admit that there is a difference between pain-behavior accompanied by pain and pain-behavior without any pain?” – Admit it? What greater difference could there be?…” ( 305.) “But you surely cannot deny that, for example, in remembering, an inner process takes place?” – What gives the impression that we want to deny anything…”(306) “Why should I deny that there is a mental process?…( 307) “Are you not really a behaviorist in disguise? Aren’t you at bottom really saying that everything except human behavior is a fiction?” – If I speak of a fiction, then it is of a grammatical fiction.”(308) “… And now it looks as if we had denied mental processes. And naturally we don’t want to deny them.”
To understand the tenor of these remarks, it is useful to focus on the aphorism that immediately follows them: “What is your aim in philosophy? – To show the fly the way out of the fly-bottle.” (PI, 309) Fly-bottles, we must know, are devices for catching flies. Attracted by a sweet liquid in the bottle, the fly enters it from an opening at the bottom and when it has stilled its hunger tries to leave by flying upwards towards the light. But the bottle is sealed at the top and so all attempts to escape by that route must fail. Since it never occurs to the fly to retrace its path into the bottle, it will eventually perish inside. The aphorism is thus evidently meant to alert us to two sentences in the preceding remark where Wittgenstein had said of the philosophical problems surrounding mental processes, mental states, and behaviorism that “the first step is the one that altogether escapes notice” and that, as a result, “the decisive movement in the conjuring trick has been made, and it was the very one that we thought quite innocent.” (PI, 308)
The aphorism tells us then that behaviorism is a dead end; it is the fly-bottle from which Wittgenstein is trying to extract us. More generally speaking, he is trying to help us escape from our philosophical problems concerning the nature of the mental, but he is aware that the way we pursue them is driving us into behaviorism. So behaviorism, far from being the position Wittgenstein advocates, is what he wants to liberate us from by and this attempted liberation takes the form of retracing the steps by which we had got into our behaviorism.
But what is the decisive first step in our thinking about the mind that takes us in the end all the way to behaviorism? Wittgenstein’s answer is short and decisive: “We talk of processes and states and leave their nature undecided… But that is just what commits us to a particular way of looking at the matter… For we have a definite concept of what it means to learn to know a process better.” (PI, 308) He goes on to speak of an analogy which falls apart at this point. And the result of the failed analogy is that “we have to deny the yet uncomprehended process in the yet unexplored medium.” Since the analogy was meant to explain the nature of our thoughts, it now looks as if we had denied mental processes.
The misleading analogy seems to be this. In physics we may have an idea of certain processes going on inside elementary particles, but we may also think that we understand neither these processes nor the particles well enough, and so we see ourselves as struggling with yet uncomprehended processes in a yet unexplored medium. Now we turn to mental phenomena and start talking about them, as if they, too, were yet uncomprehended processes in a yet unexplored medium. We think of these processes as going on “in the mind” and then add that the mind is something aethereal and difficult to understand. But, Wittgenstein suggests, the assumed analogy between the problems of physics and our questions about the mind is false and quickly falls apart. He does not tell us immediately how the analogy fails, but he is sure that it is mistaken and that it is only this mistaken analogy that drives us into our usual views aboyut the mind.
In order to understand how talk about physical particles differs from talk about the mind and its states and processes, we must look more closely at how statements about the human body are connected to psychological utterances. Here we must distinguish two cases: the case where we are speaking about a third person (“He is in pain”) and the case where we are speaking in the first person (“I am in pain”). He tells us, accordingly, that “My own relation to my words is wholly different from other people’s.” (PI, p. 192) When I say of someone else that he is in pain, I depend directly on the availability of outer criteria for my assertion. I say that he is in pain because I see his pain-behavior. Wittgenstein writes: “I noticed that he was out of humor.” Is this a report about his behavior or his state of mind?… Both; not side-by-side, however, but about the one via the other. A doctor asks: “How is he feeling?” The nurse says: “He is groaning”. A report on his behavior. (PI, p. 179)
Though outer behavior serves here as a criterion for the ascription of a psychological state, the latter is not just a statement about the behavior. There are a number of reasons for this. One is that the criterial relation connecting behavior and pain is not absolutely tight. It is possible that someone may feel pain and yet not show it and it is equally possible for someone to simulate pain, i.e. to exhibit pain behavior without feeling actual pain. On the other hand, it is also obvious that our practice of ascribing pain to others would not get off the ground, if there were no general and natural relation between pain and pain-behavior.
But there is still another and more powerful reason for denying that an ascription of pain is a statement about a behavior and for that we must look at first-person utterances. When I say that I am in pain, I do not do so on the basis of observing my own behavior. In this case, “words are connected with the primitive, the natural, expressions of the sensation and used in their place… the verbal expression of pain replaces crying and does not describe it.” (PI, 244) For this practice to get going, for children to learn to say “I am in pain” as a replacement for crying demands, of course, a linkage between non-linguistic behavior and the utterance. For children are taught to say “I am in pain” by adults who speak the language and they will teach the child to use the utterance when they see the child’s pain behavior. “A child has hurt himself and he cries; and then adults talk to him and teach him exclamations and, later, sentences. They teach the child new pain-behavior.” (PI, 244)The first-person case makes the difference between a description of a behavior and a pain utterance clear. When I say “I am in pain” I am not describing anything, I am rather expressing pain. My utterance has a different function from a description. That holds true even in the third-person case. When I say of someone “I believe that he is suffering”, I am not describing his behavior, though my ascription is surely based on his behavior; I am rather expressing an attitude towards him, “my attitude towards him is an attitude towards a soul.” (PI, p. 178) The problem with behaviorism is that it has correctly diagnosed the existence of connections between pain and the expression of pain (pain behavior), but it has misinterpreted this fact by arguing that pain utterances are descriptions of behavior. To overcome behaviorism means to “make a radical break with the idea that language always functions in one way, always serves the same purpose: to convey thoughts – which may be about houses, pains, good and evil, or anything else you please.” (PI, 304) And this questionable assumption behaviorism shares, in fact, with Cartesian mentalism, its apparent opposite. For that, too, operates on the assumptions of referentialism and objectivism. It, too assumes, that words always have meaning by standing for something. Both behaviorism and mentalism are thus driven into paradoxes, because both of them are forms of objectivism.
These considerations are at the heart of what is known as Wittgenstein’s private-language argument. For if we construe the grammar of the expression of sensation on the model of “object and designation,” we are left with the idea that there are inner objects designated by our sensation language. In his private-language argument Wittgenstein seeks to show that the inner objects that are here presupposed can play no role in our language. “The thing in the box has no place in the language-game at all; not even as a something; for the box might even be empty.” (PI, 293) The conclusion is that if we follow the objectivist and construe meaning on the model of “object and designation”, then “the object drops out of consideration as irrelevant.” (PI, 293)
The private-language argument can, thus, be read as directed against the objectivism embedded in the Cartesian conception of the mind. The private language that Wittgenstein means is evidently the kind of language a Cartesian subject would speak. In Wittgenstein’s account: “The individual words of this language are to refer to what can only be known to the person speaking… So another person cannot understand the language.” (PI, 243) A private language, as here characterized, is essentially private in the sense that what it talks about is in principle accessible only to the speaking subject. Since the language which the subject of the first two Cartesian Meditations speaks cannot be an external language – for it is accepted that we may be confused and deceived about everything external – it must be an internal language which only the speaking subject can use. But the conception of such a language is incoherent. According to O.K. Bouwsma, Wittgenstein said in conversation: “Now a scrupulously honest Descartes will not say: There goes my horse. A! A bird singing up in the tree, etc. There’s a woman holding an umbrella.’… So Descartes can present nothing. One can say that he might say: ‘Ah!’ or ‘This!’ or ‘Awareness!’ But if he now said anything of this sort, his words would have no meaning. There would be nothing to provide a contrast. ‘I think’ is or would be like: ‘Ah!’.” (O.K. Bouwsman, loc. cit., p. 13-14)
It is easy to misread these anti-Cartesian arguments as supporting behaviorist conclusions. We are likely to succumb to such a misreading, when we fail to notice that Wittgenstein’s argument is really directed against assumptions that the mentalist and the behaviorist share, i.e., the assumptions that the subject must be conceived as an object and that any meaningful noun or pronoun in our language must be a name or description of an object.
The question whether and in what sense one can speak of a self – what plausibility there attaches to the idea that there exist individual selves or that there is only one world soul of which individual selves are aspects or manifestations or that the self is some sort of construct or even just an illusion – was for Wittgenstein connected with profound moral and personal concerns. That is evident already in the Tractatus where he speaks of a link between the way we see ourselves and the problem of the meaning of life. Having argued that there is no such thing as the soul in the world and that the metaphysical subject must be conceived as the limit of the world, not as part of it, he insists that “the solution of the problem of life in space and time lies outside space and time.” (6.4312) Against first impressions he is not hankering here for a solution in some transcendental or transcendent sphere. The point of the remark is rather that the problem of life does not have the kind of solution which problems concerning things in space and time have, i.e., it has no scientific or theoretical solution. Hence, Wittgenstein’s conclusion that “the solution of the problem of life is seen in the vanishing of the problem.” (6.521) The compelling thought is that anyone beset by that problem (even to the extent of advancing a theoretical solution to it) shows thereby that he has not yet solved it. We have solved the problem only when it no longer concerns us and when our attention is turned to the process of living itself. Wittgenstein expresses, thus, an unwavering conviction that our deepest human problems call for practical resolution not for a theoretical answer.
This anti-theoretical attitude forms the ultimate underpinning to both the anti-objectivism of the Tractatus and the anti-referentialism of the later writings. The problem of the nature of the self is not resolved by advancing a theory. The I is not an object and the word “I” is not a name or description of anything. Just like the problem of life, the problem of the nature of the self finds its solution in the disappearance of the problem. It is resolved only when we no longer concern ourselves with the I and have learned to face the world without being bothered over the question of its nature.
Wittgenstein offers us, for that reason, no positive account of the nature of the self. Readers who have searched for such an account, have found no more than a few sparse and unsystematic remarks that might with an effort be taken for such. One is struck by this most forcefully when one compares Wittgenstein’s discussion of the mental with the rich theoretical accounts advanced by Freud and his followers. The difference is not simply that Wittgenstein has a less fully developed picture, but that the aim of his discussion is utterly different and that he is deeply suspicious of any attempt to offer a positive account of the mental. Rush Rhees reports that during the 1940’s Wittgenstein spoke of himself as a “disciple” and “follower” of Freud; but throughout this period he remained hostile to Freud’s theoretical claims. What appealed to him was the psychoanalytic practice not the theory that was meant to justify the justify and support that practice. When he spoke of his own enterprise as therapeutic he was, no doubt, thinking of similarities to the kind of therapy promised in psychoanalysis. But he never bought into Freud’s belief that the self was a causal construct whose characteristics and principles of constructions could be the subject matter for a new kind of science.
It is, in fact, unrewarding to try to characterize his thinking about the mental in positive terms. His key insights are unfailingly expressed in negative terms: “There is no such thing as the self,” “the word ‘I’ is not the name of a person,” and ”there is no such thing as a private language,” etc. His position is best described in the way I have done here, as antagonistic to certain common philosophical view-points. Thus, we can say that he is anti-Cartesian, anti-Russellian, anti-Freudian, anti-objectivist, and anti-behaviorist in his thinking about the mind, without being able to identify anything positive from which these negative conclusions might be thought to derive.
This approach is ultimately grounded in the moral temper of Wittgenstein’s thinking. When he insists in the Tractatus that the subject cannot be a thing in the world we must surely link this to the book’s austere moral outlook which treats the world as “a matter of complete indifference for what is higher” (6.432), which insists that “the sense of the world must lie outside the world” and that “all that happens and is the case is accidental,” (6.41) which inclines, moreover, to the belief that not wanting is the only good,” (NB, p. 77) and sees salvation in a life of knowledge: “How can man be happy at all, since he cannot ward off the misery of this world? Through the life of knowledge.” (NB, p. 81) His later views avoid the solipsistic tone of these early remarks, but he still ties moral insights to the discovery that the I is not an object and that to speak of the subject is not to refer to a thing. In the Philosophical Investigations he links his remarks about the relations between the feeling of pain and its expression to the observation that my attitude towards another human being can be “an attitude towards a soul.” (PI, p. 178) “How am I filled with pity for this man?” he asks (PI, 287) and supplies an answer in a preceding remark: “Look at a stone and imagine it having sensations… And now look at a wriggling fly and at once these difficulties vanish and pain seems able to get a foothold here.” (PI, 284) The thought that conceptions of the self are linked to differing moral ideals is not new. Interpreters have frequently pointed out how the Cartesian conception of the self historically accompanied the emergence of an ethics of individual self-fulfillment. In opposition to this moral view, later thinkers like Schopenhauer and Mach held that an ethics of compassion and unselfishness required the abandonment of the belief in individual minds. In The World as Will and Representation Schopenhauer wrote: “ If that veil of Maya, the principium individuationis, is lifted from the eyes of a man to such an extent that he no longer makes the egoistical distinction between himself and the person of others, but takes as much interest in the suffering of other individuals as in his own, and thus is not only benevolent and charitable in the highest degree, but even ready to sacrifice his individuality whenever several others can be saved thereby, then it follows automatically that such a man, recognizing in all beings his own true and innermost self, must also regard the endless sufferings of all lives as his own, and thus take upon himself the pain of the whole world.” (Arthur Schopenhauer, The World as Will and Representation, vol 1, sect. 68, transl. E.F.J. Payne, Dover, New York 1969, p. 378f.)
There can be little doubt of the affinity between Schopenhauer’s thought at this point and Wittgenstein’s remarks in the Notebooks that “the world and life are one” (p. 77, repeated in TLP, 5.621), that “the spirit of the snake, of the lion is your spirit,” (p. 85) and that “my will is the world-will.” (Ibid.) Mach argued similarly that to abandon the assumption of a substantive ego would have positive moral consequences. In The Analysis of Sensations he wrote: “In this way we shall arrive at a freer and more enlightened view of life, which will preclude the disregard of other egos and the overestimation of our own.” (Loc. cit., p. 25)Wittgenstein held that neither Cartesianism nor behaviorism can give us an adequate account of the subject since both are committed to the assumptions of objectivism. The moral implications of his own view derive from this anti-objectivism, the recognition that we can think neither of ourselves nor of others in fully objective terms. Practical and moral attitudes are, rather, inherent in the way in we understand ourselves and others.
Appealing as this maybe, it presents us with questions Wittgenstein has done little to answer. Does a phenomenology of experience not suggest that there are forms of consciousness in which we objectify other human beings and even ourselves, i.e., in which we look at them or ourselves as objects in a detached, disinterested, objectifying manner? If another person’s behavior can alert us to his suffering and can engender in us a sense of pity, there is also that other possibility: that his suffering fails to ignite our pity or that we see his behavior and refuse to recognize his suffering, that we dehumanize the other and close ourselves off from him. As far as the relation to ourselves is concerned, we may also discover in ourselves the possibility of a detachment from our own feelings and of looking at ourselves, as if we were looking at another person or even at something inanimate. Wittgenstein is not unaware of such possibilities. In passing he notes: “If listened to the words of my mouth, I might say that someone else was speaking out of my mouth.” (PI, p. 192) Do we not at times look ourselves in such a detached fashion, e.g., when we consider past or future or when we think about our own fixed chararacter. (Cf. Jean Paul Sartre’s Being and Nothingness.) Wittgenstein never thinks about the self in temporal terms. It may well turn out to be the case that we cannot clearly distinguish between a self and its state when we focus only on current states of consciousness, as both Hume and Lichtenberg had, in fact, observed long before Wittgenstein, but it may still be true that such a distinction becomes plausible to us when we think about ourselves over time, about changing states of consciousness which are nevertheless all experienced as being mine.
We can put the matter in terms familiar from Wittgenstein’s writings. Consider, then, the following types of utterance: (1) “My arm is broken,” and “I have blue eyes;” (2) “I am in pain,” and “I am seeing the table;” (3) “I am not a genius, I am only a talent,” (something Wittgenstein once said of himself, CV, p. 18) and “Knowing myself as I do, I will now act this way,” (an example he once considered, LWPP II, 8); and finally(4) “I am L.W.” Let us tentatively agree with Wittgenstein that in utterances of type (1) the word “I” refers and that it refers to something that has bodily characteristics as Wittgenstein writes in the Blue Book. Let us also tentatively agree that in type (2) utterances the word “I” does not refer and that such utterances function as expressions of pain or perception. But how are we supposed to treat utterances of type (3)? Here we are certainly not talking of bodily characteristics as in type (1) utterances. Should we assimilate them to utterances of the second type? Should we treat them as expressions of feeling? When Wittgenstein wrote in his diary: “I am not a genius, I am only a talent” he may, indeed, have been giving voice to a feeling of inner turmoil or a sense of despair. But even so the remark seems to be more than an expression of feeling; it seems to be saying something true or false about Wittgenstein’s mind. And when someone says that “Knowing myself as I do, I will now act this way” or “These days I am inclined to say…” (LWPP II, p. 9) he may be giving us simply a cool assessment of his own mental resources. Such utterances appear to be therefore more similar to type (1) than to type (2) utterances. When we utter them we seem to be saying something factual about ourselves; but we seems to be speaking of the mind rather than of the body. How is that possible without assuming that there is, after all, a mind or self to which we can attribute objective characteristics? Do utterances of type (3) commit us to the assumption that in them the word “I” refers to a something that has definite, objective, fixed characteristics? Before deciding on this issue, it may be useful to consider the last type of utterance. Wittgenstein says with a view to such utterances: The word “I” does not mean the same as “L.W.” even if I am L.W. nor does it mean the same as the expression “the person who is now speaking”. But that doesn’t mean: that “L.W.” and “I” mean different things. All it means is that these words are different instruments in the language.(BB, p. 67)
It should be obvious that “I” and “L.W.” cannot mean the same thing, for otherwise the sentences “I am L.W.” and “L.W. is L.W.” would mean the same thing; but the second is a tautology whereas the first can be a very informative (true or false) assertion. It should also be clear that “I” cannot mean the same as “the person who is now speaking,” even though that is widely held to be the correct explication of the meaning of “I”. But when we are sitting in an auditorium listening to a lecture and I whisper into your ear “The person who is now speaking is L.W.” I don’t mean to be saying that I am Ludwig Wittgenstein and you are not likely to be in doubt about that. The phrase “the person who is now speaking” can occasionally mean the same as “I” – though we might then chide the speaker for being prolix -, but it generally does not.
So far it is easy to follow Wittgenstein’s reasoning in the quoted remark; but what about his conclusion that “I” and “L.W.” do not “mean different things”, that they are, rather, “different instruments”? The first part of this claim implies, presumably, that while “L.W.” refers to something, the word “I” does not; it contains what I have called Wittgenstein’s anti-objectivism. But his terse remark that “I” and “L.W.” are different linguistic instruments appears uncomfortably elliptical. What kind of instrument is each one of them? What functions does it serve?
We can clarify the point by considering that on Wittgenstein’s account the sentence “I am L.W.” cannot be taken to be an identity-statement. An identity-statement says that such and such an object is the same as such and such another object. (“The morning star is the evening star.”) But if “I” is not a referring term, then “I am L.W.” cannot be an identity statement. What then are we to say about it? We must look back here at a remark we have quoted early on to the effect that the word “I is not a name but that it is connected with names and that names are explained by means of it. In illustration of this we might observe that when someone reads out a list of names and asks in a crowded room “which of you is L.W.?” the answer “I am L.W.” may serve as an explanation of the meaning of the name the questioner has found on the list. By saying “I am L.W.” Wittgenstein is making it clear which of a number of people has that name. But is that the distinctive function of the utterance “I am L.W.”? The explanation of who L.W. is, may also be given by someone else saying “He is L.W.” (pointing to Wittgenstein) or “The person sitting in the corner is L.W.” and so the utterance “I am L.W.” is not essential for such an explanatory task. But there is another and more distinctive function which the utterance performs. When Wittgenstein says “I am L.W.” he is, we might put it, identifying himself.
The observation suggests an answer to the question how we should understand utterances of the form “I am not a genius, I am only a talent.” Perhaps, we should consider them, too, as identifying statements, i.e. as statements in which the speaker makes certain observable and objectively determinable characteristics his own, in which he “recognizes” himself as a mere talent rather than a genius, as having such and such fixed characteristics. And having taken this step, we may want to go further and say that even in utterances of type (1) we are not simply using the word “I” as a name for a body – as the Blue Book has it -, but here, too, the word “I” indicates that utterance serves the purpose of identification. Finally, we may extend this account even to utterances of the second type involving sentences like “I am in pain”. Wittgenstein’s own account of these types of utterance as verbal expressions replacing natural expressions of sensations fails to tell us why they have the specific syntactic structure they possess. Is it not possible that they actually serve a double function: as expressions of feelings and as statements of identification in which we declare those feelings to be our own?
But this seems to lead us back to the question what it is we are speaking of in such utterances. Is there an I, a something we identify in such statements of identification? Our account commits us to neither a Cartesian nor a behaviorist analysis of the self. We are saying rather that the utterances under consideration have a dual function and that only one of them is those described by Wittgenstein. To say “I am in pain” is certainly also to express pain. In other “I”-utterances we are likewise giving vent to a subjectivity that cannot be conceived as some object in the world. But our utterances have, in addition, a second function. By speaking again and again of an “I” these utterances define a self with a location in space and time, a self that has a fixed character, that has desires, purposes, goals, and hopes. It is this self that is identified in statements of identification. But this self is not a real thing, it is rather a conception and image we construct, in terms of which we make sense of ourselves, of our states, experiences, and thoughts and in terms of which we project a coherent future for ourselves.
We are, thus, forced to modify Wittgenstein’s account of the self by adducing a notion of self-image or self-conception as being inherently appealed to in our “I”-utterances. Formally speaking, such a self-conception or self-image is, of course, an object just like the fictional land of cockaigne; but it is not a real object existing with causal powers and in this sense Wittgenstein was surely right when he said that the self is not an object. But Wittgenstein said almost nothing about this notion of self-conception, though it is surely of crucial significance when we think about ourselves historically backward-looking or forward-looking as moral agents.
We can contrast his thought here with that of Michel Foucault. He, too, started his reflections on the self with a denial of Cartesian mentalism, but eventually he came around to seeing that such negative conclusions had to be supplemented with an analysis of “the practices by which individuals were led to focus their attention on themselves, to decipher, recognize, and acknowledge themselves as subjects of desire, bringing into play between themselves and themselves a certain relationship that allows them to discover, in desire, the truth of their being, be it natural or fallen.” (Michel Foucault, The Use of Pleasure, transl. R. Hurley, Vintage Books, New York 1986, p. 5.)
For Foucault “it seemed appropriate to look for the forms and modalities of the relation to the self by which the individual constitutes and recognizes himself qua subject.” (Ibid., p. 6) Holding that “all moral action involves a relationship with the self,” he argued: “The latter is not simply ‘self-awareness’ but self-formation as an ‘ethical subject,’ a process in which the individual delimits that part of himself that will form the object of his moral practice, defines his position relative to the precept he will follow, and decides on a certain mode of being that will serve as his moral goal.” (Ibid., p. 28) In his last books, The Use of Pleasure and The Care of the Self, and in a series of important late interviews Foucault has shown what wealth of philosophical insight can be gained from an exploration of these issues.
But these are not Wittgenstein’s issues. He stays, on the whole, with the negative conclusions that the word “I” either refers to a body or does not refer at all. The moral conclusions he reaches are those that follow from the negative discovery that the self is not an object. Hence, he never gets to the positive ideal of an “aesthetics of existence” that Foucault envisaged. Wittgenstein’s moral attitudes, as he expressed them in his own life and in his remarks to friends, tend rather towards a life of ascetic denial. In this respect he remains close to Schopenhauer’s ethics whereas Foucault’s affirmation of a positive self-constitution links him to Nietzsche. The critique of Wittgenstein’s views suggested in these observations is, in effect, that which Nietzsche directed against what he called Schopenhauer’s nihilism. What is certain is that Wittgenstein saw his task in demolishing objectivist accounts of the self and that he did not address the philosophical issues arising from the process of self-formation. In this we can surely see a limitation in his thinking about the self; it is a limitation, moreover, which restricted the scope of Wittgenstein’s reflections on moral issues.
Only once did Wittgenstein touch on these issues and then it was only in a single, passing remark which did not bear on the moral significance of the notion of self-conception. The reference is to be found in the remark from which we started and which for that reason has a unique importance in Wittgenstein’s work. The remark alerts us to a lacuna in Wittgenstein’s thinking and to the fact that Wittgenstein was vaguely aware of its presence. In saying in section 398 that the house in the imaginary landscape “might be” said to belong to the farmer in the picture, Wittgenstein is, first of all, re-affirming his commitment to the idea that there is no such thing as the self. I may of course, say that I imagine the landscape, the house, and the farmer, but I cannot take this to imply that there is a real self that owns these things. If I want to speak of someone owning the imaginary house or anything else in the imaginary landscape, I can only talk of an imaginary object, the farmer sitting in front of the imagined house. But in this case I am really only adding an item to the contents of my imagination, an item I invest with a symbolic relation of ownership of the others. This imaginary self is not a real object; it is rather a self-conception, the kind of thing that Foucault has in mind when he speaks of the care of the self. Because it is merely imaginary, this self has no causal powers. That is suggested by Wittgenstein’s observation that the imaginary farmer cannot enter his imaginary house. We can imagine him doing such things, but he cannot do them in reality. As such the remark is one more nail in the coffin of objectivism. For Wittgenstein is trying to show us that the belief in a real self results from confusing this self-conception with an objectively real thing.