Theodore Redpath remembers that Wittgenstein once said of Schopenhauer: ‘Well, he was a philosopher’, and when asked what he meant by a ‘philosopher’, he replied: ‘A teacher of manners’ (Redpath 1990, 41). As such he clearly influenced Wittgenstein’s early thinking about ethics and the meaning of life. His 1916 notebook (NB 71-91) and the final pages of the Tractatus contain a number of echoes of Schopenhauer. Like him he describes aesthetic contemplation using Spinoza’s expression ‘sub specie aeterni[tatis]’; he repeats Schopenhauer’s criticism of the categorical imperative: that every imperative calls forth the question ‘And what if I do not do it?’ (TLP 6.422); he also agrees with Schopenhauer (and Kant) that the good action should not be motivated by its consequences (ibid.); like Schopenhauer he thinks that science cannot answer questions of value, like him he places ‘the solution of the riddle of life’ outside space and time (TLP 6.4312), and like him he thinks that ‘what is higher’ cannot ultimately be expressed in words (TLP 6.432, 6.522).
In those early thoughts Schopenhauer’s influence on Wittgenstein is obvious1 — but of very limited philosophical interest. The fact is that the young Wittgenstein’s musings and gnomic aphorisms on the meaning of life and the mystical, although extremely significant from a biographical point of view, are of little philosophical value and only tenuously related to his work on logic and language.2 As a soldier at the Eastern Front, enduring physical hardship, a loss of freedom and privacy, insufferable company, and a constant threat of death, he was trying to take stock and groping for a world view that would allow him to come to terms with his situation. Unsurprisingly the result was eclectic, extremely sketchy and in many points half-baked.3 (In a diary entry from 1930, Wittgenstein judges that beside good and authentic things the Tractatus also contains some ‘Kitsch’ (D 30). One may suspect that some of the purple passages from the end of the book belonged to what he had in mind.)
Responding to his situation in 1916, Wittgenstein was trying to formulate a conception of the good and happy life that was doubly independent of the world of facts: it would not change the facts and it would also not be affected by adverse circumstances. It would be happy come what may. The ideal Wittgenstein developed at the front was that of stoic equanimity and content. Apart from a firm religious faith, which Wittgenstein felt unable to obtain, stoicism appeared to be the only ethical view that would provide an answer to his miserable situation. Stoic ethics emphasises the distinction between what depends on us and what does not, and counsels us not to count on the latter at all. That way we will never suffer from having our desires thwarted. In the same spirit Wittgenstein notes:
I cannot bend the happenings of the world to my will: I am completely powerless.
I can only make myself independent of the world — and so in a certain sense master it — by renouncing any influence on happenings. [NB 73: 11. 6. 16; cf. TLP 6.373]
The stoic attains a serene happiness by ceasing to worry about what might happen to him. He is no longer concerned about future events; in particular not about his own death: he ‘lives not in time but in the present’, and ‘for life in the present there is no death’ (NB 75: 8.7. 16; cf. TLP 6.4311). — This last thought is reminiscent of Marcus Aurelius’ remark that we possess only the present moment, and therefore the length of our life is entirely irrelevant (Meditations, Book 2 §14). Wittgenstein’s acquaintance with stoic ethics may well have been derived from Schopenhauer’s account of it in §16 of The World as Will and Representation, vol. 1, and as already mentioned, he threw in ideas he remembered from Schopenhauer’s own system wherever they seemed to fit, together with what had impressed him in Otto Weininger,4 Tolstoy, Dostoevsky, Karl Kraus, and others. His allusions to Schopenhauer are all fairly vague and of a kind one can make from memory even years after one has read him. There is no evidence that he reread him in 1916.5
Of much greater philosophical interest than Wittgenstein’s early remarks on ethics are his thoughts on idealism. According to Elizabeth Anscombe, Wittgenstein had read Schopenhauer as a boy of sixteen ‘and had been greatly impressed by Schopenhauer’s theory of the “world as idea” (though not of the “world as will”); Schopenhauer then struck him as fundamentally right, if only a few adjustments and clarifications were made’ (Anscombe 1959, 11f.)6. Wittgenstein’s enchantment with idealism continued for quite a while: it is very strong in the Tractatus period, and there are still traces of it in his writings of the early 30s.7
Idealism identifies the world with our experiences of the world, our mental representation (Vorstellung) of it. This is the first sentence of Schopenhauer’s main work: ‘The world is my representation’ (WWR1 §1, p.3), repeated by Wittgenstein in his notebook on 17. Oct 1916 (NB 85). In the Tractatus he uses the word ‘life’ to denote a person’s experience of the world, which is then identified with the world: ‘The world and life are one’ (TLP 5.621). It follows from this identification that without an experiencing subject there can be no world. Schopenhauer: ‘[the subject] is accordingly the supporter of the world, the universal condition of all that appears, of all objects, and it is always presupposed; for whatever exists, exists only for the subject’ (WWR1 §2, p.5). The young Wittgenstein agrees: ‘the subject is … a presupposition of [the world’s] existence’ (NB 79: 2.8.16). But conceiving of the subject as the ‘supporter of the world’ means: distinguishing it from the world: placing it in opposition to and outside the world. Schopenhauer makes this point epistemically by introducing the subject as in principle unknowable: ‘that which knows all things and is known by none’ (WWR1 §2, p.5).8 It is comparable to the eye which cannot see itself (WWR2, ch. 41, p.491; [575f.]). Wittgenstein repeats this simile in the Tractatus (TLP 5.633) and combines it with the spatial image of the subject being outside the world of experience (as the eye is outside the visual field) (TLP 5.632). If I wrote a book called The World as I found it, the subject could not be mentioned in it (TLP 5.631). Although we know a priori that we have a subject — it is a logical truth that without a subject there could not be any experience — that subject cannot be encountered in experience. It is not an object (NB 80: 7.8.16) that could stand in a possessive relation to its experiences. Therefore the subject must be identified with its experiences: I am my world (TLP 5.63). Alternatively, the subject could be regarded as an outer boundary of the world (TLP 5.632). Wittgenstein’s positive characterization of the ‘metaphysical subject’ vacillates between these two metaphors for what is strictly speaking unsayable. What matters, however, is the negative point that the subject does not belong to the world (TLP 5.632). Why not? First, it appears that the real subject of experience cannot be one’s physical body or a part of it (AL 23). The argument (not explicit in the Tractatus, but spelt out in later writings) is this. The fact that my perceptions and thoughts are lodged in this particular body is merely contingent. Tomorrow I could find myself looking at the world from a different body while still being myself. Hence my self is not identical with my body. To put the same point in a slightly different way, if ‘I’ meant ‘this body that has such-&-such characteristics’, I might be mistaken in claiming that, say, I am in pain. Perhaps my body no longer has any of those characteristics. But, surely, I cannot be wrong in thinking that I am in pain. So by ‘I’ I must mean something that is quite independent of any of my bodily features (cf. BB 66f.). Then, having distinguished the self from one’s body, one is inclined to look for it in one’s mind or consciousness. But, as Hume was the first to discover, introspection, close attention to one’s feelings and perceptions fails to reveal anything one might call the self. As Wittgenstein puts it:
The experience of feeling pain is not that a person ‘I’ has something.
I distinguish an intensity, a location, etc. in the pain, but not an owner. [PR 94: §65]
So, nowhere in the world, neither in the physical nor in the mental realm, do I encounter a subject (TLP 5.633). In fact, it is not just an empirical datum that the subject is not to be found in the world, it could not be otherwise. For that all experience must depend on a subject is an a priori truth. But an a priori truth can (as such) never be learnt from experience. For ‘no part of our experience is at the same time a priori. Whatever we see could be other than it is’ (TLP 5.634). In order to see that my experience was had by a subject it would have to be possible likewise to see that it was not had by a subject; yet that is impossible.
So far, the young Wittgenstein’s idealism and his conception of the metaphysical, or ‘representing’ (vorstellende) subject (TLP 5.631) are clearly Schopenhauerian, complemented only by some independent support to Schopenhauer’s idea that the subject must be distinguished from my empirical person. But now the stories are continued differently: Schopenhauer gives his idealism a transcendental twist, which is not adopted by the author of the Tractatus, who instead takes a further step which Schopenhauer was very anxious to avoid: the step from idealism to solipsism.
Transcendental idealism is based on the distinction between things as they are in themselves and as they appear to us. In itself the world exists independently of us, but the way it appears to us is determined by our cognitive faculties. Notably that all objects are experienced in space and time and according to causal laws is not due to their independent nature, but imposed on the world by the perceiving subject. This doctrine, originating with Kant and adopted by Schopenhauer, is absent from the Tractatus. Wittgenstein’s most explicit solipsist pronouncements contradict the independence that transcendental idealism accords to the world in itself. For instance: ‘My world is the first and only one’ (NB 82: 2.9.16); ‘at death the world does not alter, but comes to an end’ (TLP 6.431). And although it has occasionally been suggested that logic, in the Tractatus, has the transcendental status that space, time and causality have in Kant and Schopenhauer, that is not so. In the relation between logic and the world it is not logic, but the world that wears the trousers. Logic is not said to shape the world, but merely to pervade it (TLP 5.61a), which suggests that the world has a certain extension independently of logic. Again, Wittgenstein does not say ‘the limits of logic are (i.e. determine) the limits of the world’, but he puts it the other way round: ‘the limits of the world are also its [i.e. logic’s] limits’. Thirdly, the limits of language are not said to be the limits of my world, but to mean or indicate it (bedeuten) (TLP 5.6; 5.62c). Logic and language reflect the form of the world, they do not produce it (cf. TLP 5.511 & 6.13, where logic is said to mirror the world). Anyway, Wittgenstein nowhere entertains the idea that logic might be subjective in a transcendental sense: imposed upon the world by us.
The very word ‘solipsism’ in the Tractatus already suggests that this part of Wittgenstein’s early philosophy was not derived from Schopenhauer, who never uses this term, but speaks of ‘theoretical egoism’ instead.9 Schopenhauer writes:
theoretical egoism … regards as phantoms all phenomena outside its own individual, just as practical egoism does in a practical respect … Theoretical egoism, of course, can never be refuted by proofs, yet in philosophy it has never been positively used otherwise than as a sceptical sophism, i.e. for show [zumSchein]. As a serious conviction, on the other hand, it could be found only in a madhouse; as such it would then need not so much a refutation as a cure. Therefore we do not go into it any further, but regard it as the last stronghold of scepticism, which is always polemical. [WWR1 §19; 104; transl. slightly altered; ]
Wittgenstein’s attitude towards solipsism is strikingly different. The earliest evidence of his occupation with it is a diary entry of 8th December 1914, where Wittgenstein expresses some sympathy for Nietzsche’s hostility towards the Christian religion:
Certainly, Christianity is the only sure way to happiness. But what if someone spurned that happiness?! Might it not be better to perish, unhappy, in a hopeless struggle against the outer world? But such a life is meaningless. But why not lead a meaningless life? Is it without dignity? How can it be reconciled with the strictly solipsistic point of view? [MS 102,41]10 Here ‘the strictly solipsistic point of view’ — far from being dismissed as madness — is treated as a standard by which to gauge an ethical position. This is almost certainly a reference to Weininger’s moral solipsism, which Wittgenstein had encountered about that time.11 In Weininger’s ethics, developed from Kantian ideas, all heteronomy is rejected: I am responsible neither to God nor to anybody else, only to my own self as the source of the moral law.
Truth, purity, faithfulness, sincerity towards oneself: that is the only conceivable ethics. … Duty exists only towards oneself. [Weininger 1903, 206 & 208]12 As a moral agent ‘man is alone in the universe in eternal, immense loneliness’ (Weininger 1903, 210).13 Ethics requires that one become aware of one’s utter loneliness. Hence, a ‘refutation of solipsism would be altogether incompatible with ethics’ (1904, 147):14 Shrinking back from solipsism is being incapable of giving independent value to life, incapable of a rich loneliness, needing to hide in the crowd, to vanish in a great number, to go under. It is cowardly. [Weininger 1904, 148]15 This, to be sure, is not the metaphysical solipsism propounded in the Tractatus. However, being persuaded by Weininger of a solipsistic outlook in ethics,16 Wittgenstein was probably more readily prepared to embrace a form of metaphysical solipsism too. Unlike Schopenhauer, he would certainly not dismiss it as madness.
In fact, Schopenhauer’s distaste for solipsism was shared by many readers of the Tractatus who refused to believe that Wittgenstein really held such an absurd doctrine. They read him as saying merely that solipsism is in some ways tempting, but ultimately mistaken (Black 1964, 309; Pears 1987, 188); so that in this respect Wittgenstein did not disagree with Schopenhauer after all (Magee 1983, 313; Janaway 1989, 324ff.; Weiner 1992). But that is not a plausible interpretation. For one thing, it is flatly contradicted by TLP 5.62, where solipsism is implicitly called ‘a truth’, and explicitly described as something that is ‘entirely correct’ and ‘shows itself’. Moreover, Wittgenstein declares that ‘at death the world does not alter, but comes to an end’ (TLP 6.431), whereas a non-solipsistic idealist would say that it continues to exist in other people’s minds.17 For another thing, the fact that by Tractatus standards any formulation of the doctrine must indeed be dismissed as ‘nonsense’ does not, for Wittgenstein, preclude the doctrine from being true — as according to its author the whole of the Tractatus is in the same situation: its sentences are nonsense (TLP 6.54), yet the truth of the thoughts they express is ‘unassailable and definitive’ (TLP, Preface: h). That is because the author of the Tractatus advocated an extremely restrictive criterion of sense according to which only contingent, empirical, propositions are meaningful, not the necessary truths of logic or philosophy.18
Thirdly: Wittgenstein’s writings and lectures between 1929 and 1935 provide abundant evidence of how compelling solipsism had seemed to him and how difficult he found it to free himself from its grip. Still in December 1929 he regarded solipsism as an unsayable truth:
That statement, that only the present experience is real, seems to contain the last consequence of solipsism. And in one sense it is indeed so; only it [the statement] can say as little as solipsism. — For what belongs to the essence of the world just cannot be said. [MS 108, 2]
Only what we can imagine to be different, language can say. [MS 108-1]19
In the following years Wittgenstein became more critical, and soon regarded both idealism and solipsism as philosophical errors (e.g. AL 29; BT 508). But they were by no means trivial errors. The tenacity with which he repeatedly discussed them over many years shows clearly that overcoming his solipsistic inclinations was for him a formidable philosophical challenge. Anticipating the famous image of the Philosophical Investigations that solving a philosophical problem is like showing a fly the way out of a fly-bottle (PI §309), he writes about 1935:
The solipsist flutters and flutters in the fly-bottle, strikes against the walls, flutters further. How can he be brought to rest? [NfL p.258]
Let us now consider how Wittgenstein got from idealism to solipsism. Why did he think that ‘what the solipsist means is entirely correct’ (ganz richtig)? According to TLP 5.62, ‘the key’ to the issue of solipsism is the remark 5.6:20 ‘The limits of my language mean the limits of my world.’21 By ‘my language’ Wittgenstein means the type of language that is at all possible for me to understand (McGuinness 2002, 137f.; Schroeder 2006, 94). What kind of language does that exclude? For one thing, the change from ‘language’ to ‘logic’ in 5.61 suggests that a (would-be) language that was not in accordance with logic would be disqualified: ‘what we cannot think we cannot say either’ (TLP 5.61d; cf. NB 84: 15.10.16). For another thing, for language to be meaningful, Wittgenstein thought, its names must be correlated with objects (TLP 3.203); hence, if an object is entirely inaccessible to me I cannot give it a name, and so cannot talk about it. So a language containing names of objects beyond my ken would be incomprehensible to me: it would not be ‘my language’, and those objects would not be part of my world.
Now, as explained above, one implication of idealism that Wittgenstein was particularly impressed by is the elusiveness of the subject: nowhere in the world do I encounter my own subject (TLP 5.631) — let alone anybody else’s. But if the subject is not part of the world, a proposition like ‘I see a tree’ would appear to transgress the bounds of sense. For contrary to the requirements of the Tractatus semantics, the word ‘I’ does not stand for anything in the world. Thus it should not appear in a logically more appropriate notation (cf. TLP 4.002). Analysis must transform the proposition into one that no longer contains the objectionable word ‘I’. All that can be meant by ‘I see a tree’ is: ‘A tree is seen’ (cf. ML 309). But with the removal of the term that apparently denoted a subject we have removed the mark that distinguishes between propositions reporting different people’s experiences, for instance, ‘I see a tree’ and ‘You see a tree’, or ‘Smith sees a tree’. So ‘what the solipsist means is quite correct’: we cannot make sense of the idea that there might be different people having experiences. There is only this experience — and here I focus on whatever may be in my field of vision and hearing at the time. As Wittgenstein was to put it later in some lecture notes: ‘what I now see, this room, plays a unique role, it is the visual world!’ (NfL 258; cf. PI §398).
Here one might object that even if the self dissolves, in Humean fashion, into a bundle of experiences, it does not seem to follow that there could not be a multiplicity of such bundles. Experience may only allow me to say ‘A tree is seen’, but could I not easily imagine that at the same time somewhere else a house is seen, a rose is smelt and church bells are heard, each such experiential complex forming another person’s world? This objection is not discussed in the Tractatus, but Wittgenstein provides a reply to it in Philosophical Remarks:
The two hypotheses, that others have pain, and that they don’t and merely behave as I do when I have, must have identical senses if every possible experience confirming the one confirms the other as well. In other words, if a decision between them on the basis of experience is inconceivable. [PR 94f.: §65]
The difference between the view that others have experiences like me and the view that they behave exactly as if they had such experiences without having them is that the former hypothesis goes beyond the latter in assuming the existence of certain entities: others’ experiences on top of their behaviour. But such entities are necessarily inaccessible to me. I could not possibly encounter them — I cannot experience what is not experienced by me. Hence I cannot name them or talk about them either. So the first hypothesis in as much as it goes beyond the second makes no sense to me.
Shall we then call it an unnecessary hypothesis that anyone else has personal experiences? — But is it an hypothesis at all? For how can I even make the hypothesis if it transcends all possible experience? How could such a hypothesis be backed by meaning? [BB 48]
This position, implicit in the Tractatus and spelled out in Wittgenstein’s later writings, can be called Semi-Behaviourism. Behaviourism tries to reduce all psychological states to dispositions to show certain patterns of behaviour. Thus having a headache, for example, boils down to nothing more than a tendency to moan, complain, rub one’s forehead, avoid vigorous movements, take an aspirin etc. etc. Wittgenstein’s semi-behaviourism maintains this only about other people’s psychological states. And isn’t that exactly the way we experience things? In my case, to be sure, pain is not just a tendency to behave in certain ways: it is really felt (and very unpleasant); as regards others, I do not actually experience any pain: what I mean when I say they are in pain is that they are inclined to behave in a certain way. ‘Here it can be seen that solipsism, when its implications are followed out strictly, coincides with pure realism’ (TLP 5.64).22 It is an accurate account of the world as it actually presents itself in a person’s experience: only some pains are really felt (viz. one’s own), all others occur only as observed behaviour.
Roughly speaking, Wittgenstein’s solipsism results from the combination of idealism and an empiricist-verificationist semantics. Indeed, idealism itself — the reduction of the world to the experienced world — can be seen as an extreme form of empiricism. As Wittgenstein remarked in 1930:
What is true about idealism is that the sense of a proposition derives entirely from its verification. [MS 110, 240; BT 500]23 Consequently, when later Wittgenstein reverts to these issues he tends to discuss the two doctrines together, one just being a more radical manifestation of the tendency that also motivates the other.
From the early 1930s on Wittgenstein becomes concerned that these two related doctrines are in conflict with our grammar: Contrary to idealism, there are significant grammatical differences between the uses of the words ‘(mental) representation’ (Vorstellung) and ‘thing’ (BT 501). However, realism is not entirely correct either if it holds that representations are only subjective pictures (Bilder / Abbilder) of things, for this is based on a false analogy between the mental representation of a thing and the picture of a thing: as only the latter, but not the former, can be compared with the thing (BT 502). — It is perhaps no coincidence that at this point Wittgenstein takes realism to be indirect realism, assuming that a realist would hold a representative theory of perception; just as Schopenhauer does when he complains that realism entails an absurd duplication of worlds (WWR2 §1, pp.9f. [16f.]).
Solipsism is criticised, at about the same time, for giving the word ‘I’ an unduly exalted position, which in fact it does not have in our grammar (BT 508). But for quite a while Wittgenstein continues to insist that there is something right about his earlier view: ‘the solipsist is right in treating “I have toothache” as being on a different level from “He has toothache”’ (AL 23; cf. ML 103). Wittgenstein no longer believes in a metaphysical subject, but he is still attracted by the asymmetric solipsist phenomenology: the observation that there is felt pain (one’s own) and mere pain behaviour (others’). He wrestles with this intuition, trying to break its spell, in the ‘Notes for Lectures’, written about 1935, where it is expressed again and again:
[The solipsist] will say that behind the sentence ‘I see …’ when he says it and it’s true, there stands something which does not stand behind “he sees” or “I see” when the other man says it. [NfL 228]
“But I am in a favoured position. I am the centre of the world.” [NfL 257]24 To pursue the details of Wittgenstein’s increasingly critical thoughts about solipsism and idealism would require a book-length study.25 Suffice it to say that this discussion leads directly to the very centre of Wittgenstein’s mature philosophy of mind, the so-called private language argument. The reason why solipsism cannot be stated in meaningful language is not only that it is an attempt to describe what cannot be otherwise, it is also that my solipsistic talk about what is really seen or felt would appear to be incomprehensible — nonsense — to anybody else. Thus out of Wittgenstein’s earlier fascination with idealism and solipsism grows the idea of a necessarily private sensation language, critically discussed in §§243ff of the Philosophical Investigations.26 It is noteworthy that the argument that Wittgenstein employed originally to take the step from Schopenhauerian idealism to solipsism is re-employed in the Investigations with an entirely different thrust. As explained above, his idea had been that I cannot make sense of others’ really feeling pain because as far as actual experiences are concerned, only the ones I have are part of my world. The thought reappears in his later work:
If one has to imagine someone else’s pain on the model of one’s own, this is none too easy a thing to do: for I have to imagine pain which I do not feel on the model of the pain which I do feel. [PI §302]
But now it is used as a reductio ad absurdum of the solipsist’s starting point, namely the idea that psychological concepts are grounded in introspection. ‘If one had to imagine someone else’s pain on the model of one’s own’ — one wouldn’t be able to: one could not even be sceptical about others’ feelings: the very idea would be nonsense! The fact that we do shows that such concepts cannot be grounded in introspection, but must essentially involve the idea of expressive behaviour.27
Schopenhauer’s influence on Wittgenstein’s mature philosophy is, however, not entirely indirect and negative. Wittgenstein’s discussion of voluntary action (PI §§611-32) contains Schopenhauer’s insight that voluntary movements are not caused by mental acts of will. And in all likelihood that is not a coincidence. Schopenhauer writes:
Every true act of [a subject’s] will is also at once and inevitably a movement of his body; he cannot actually will the act without at the same time being aware that it appears as the movement of the body. The act of will and the action of the body are not two different states objectively known, connected by the bond of causality; they do not stand in the relation of cause and effect, but are one and the same thing, though given in two entirely different ways, first quite directly, and then in perception for the understanding. [WWR1 §18; p.100 ]
This passage is echoed by the young Wittgenstein when on 4th November 1916, thinking about the nature of the will, he notes:
This is clear: it is impossible to will without already performing the act of the will.
The act of the will is not the cause of the action but is the action itself.
One cannot will without acting.
The fact that I will an action consists in my performing the action, not in my doing something else which causes the action.
[NB 87f.: 4.11.16]
However, Wittgenstein’s pre-Tractatus notebooks were really just that: notebooks, where he jotted down ideas as they occurred to him. One can see how in many points his thinking was still very tentative and unclear, not only regarding ethics and the mystical. Occasionally, a thought written down on one day is contradicted by the opposite view expressed a few days later. Thus, the Schopenhauerian rejection of a causal link between act of will and bodily movement is in conflict with some other ideas Wittgenstein had about the will. As already remarked, he was, at the time — probably as a result of his personal circumstances as a soldier in World War I — profoundly impressed by the idea that the world was beyond his control: ‘The world is independent of my will’ (NB 73: 5.7.16).28 Although it occurred to him that, in response, one might distinguish between some aspects of the world that are under my control and others that are not (I can move my arm, but I cannot move the moon); he dismissed that distinction as ‘intolerable’, for he was strongly inclined to see the world as a whole, without any distinctions, in neat opposition to the subject whose world it is (NB 88: 4.11.16).29 But if all the world, including my body, is regarded as independent of my will, then acts of my will must be seen as distinct from movements of my body — contrary to Schopenhauer’s identification of the two. Acts of the will can at best be the causes of movements of the body, and the young Wittgenstein firmly believed that:
There is no compulsion making one thing happen because another has happened. The only necessity that exists is logical necessity. [TLP 6.37]
Of course, this line of reasoning is flawed. Even if I cannot necessitate any events, it does not follow that events are altogether independent of my will. Still, it was this stoical vision of a person’s utter powerlessness that prevailed with Wittgenstein at the time and made him adopt in the Tractatus the empiricist account of voluntary movements as caused by mental acts of will — in spite of the fact that he had already seen the force of Schopenhauer’s repudiation of that picture. Again, in his 1916 notebooks Wittgenstein introduced the distinction between willing and mere wishing, which he may well have found in Schopenhauer (WWR1 §55, p.300 [376f.], WWR2 ch.20, p.248 ; FW ch.2 p.15 ): wishing is indeed a mental event that may precede an action, but it fails to account for voluntariness:
[the willed movement] is not accompanied just by a wish! But by will.
The wish precedes the event, the will accompanies it.
Suppose that a process were to accompany my wish. Should I have willed the process? [NB 88: 4.11.16]
In the Tractatus, however, Wittgenstein slid back into a careless conflation of wishing and willing:
Even if all that we wish for were to happen, still this would only be a favour granted by fate, so to speak: for there is no logical connexion between the will and the world … [TLP 6.374]
It was only in the 1930s, when Wittgenstein had realised that the Tractatus contained ‘grave errors’ (PI, Preface) and was going over a lot of the terrain again, that he came back to those earlier thoughts about willing, which would not fit into the framework of his first book. In Eine philosophische Betrachtung (1936) he reverts to his earlier distinction:
Here it is dangerous to confuse willing and wishing. — For when I raise my arm, it is not that I first wish that my arm will rise, and then indeed it does. [EPB 235]30 The point is repeated in Philosophical Investigations (PI §616) where Wittgenstein argues, more generally, that voluntariness cannot be accounted for by any kind of mental event (of which one might then ask the embarrassing question whether it was voluntary or not). As for Schopenhauer’s positive suggestion simply to identify the willing with the bodily action, it is now presented in inverted commas, indicating that Wittgenstein finds it attractive, but not entirely correct:
“Willing, if it is not to be a sort of wishing, must be the action itself. It cannot be allowed to stop anywhere short of the action.” [PI §615]
His immediate critical comment is that if one were to identify willing with acting one would have to identify it with all sorts of acting: not just bodily movements, but also imagining or trying to do something. That contradicts the passage from Schopenhauer quoted above where willing is identified with bodily action, but elsewhere Schopenhauer gives a more inclusive account of willing that agrees fairly well with Wittgenstein’s reminder:
For not only willing and deciding in the narrowest sense, but also all striving, wishing, shunning, hoping, fearing, loving, hating, in short all that directly constitutes our own weal and woe, desire and inclination, is obviously only affection of the will, is a stirring, a modification, of willing and not-willing, is just that which, when it operates outwards, exhibits itself as an act of will proper. [WWR2, ch. 19; p.202 ; cf. FW ch. 1: 10 ]
However, Wittgenstein’s comment is only conditional (‘If it is the action…’); he does not really identify willing and doing. In fact, he already rejected the identification two sections before, reminding us that ‘“Willing” [“Wollen”] is not the name of an action’ (PI §613). ‘Ich wollte aufstehen’ (‘I wanted to get up’) does not report an action; certainly not the action of getting up. Against this, it might be said that it is not Wollen, in the ordinary sense of the word, that Schopenhauer means to identify with acting, but the act of will (Willensakt). This expression — as the English ‘willing’ — is rarely used outside philosophy, and so cannot easily be claimed to mean something else. However, Wittgenstein’s objection is not based on ordinary usage. The point is rather that an act of will, in philosophical terminology, is a mental occurrence, yet unprejudiced inspection shows that very often there is no such thing: It is simply not true that whenever we act voluntarily we experience a certain mental occurrence that accounts for the action’s voluntariness. Schopenhauer suggests that seen from the inside the action appears to the agent as an act of willing, but that would still require the existence of a distinctive mental event of willing, which in fact is not there. And even if we could find some mental experience of willing, it would be hard to see how it, a mental occurrence, could be the same as a bodily action.
Perhaps, on a charitable interpretation, by an ‘act of will’ Schopenhauer should not be taken to mean a mental occurrence. After all, his crucial insight is that beside the action itself there is no second event involved; and by insisting on an act of will as a mental event he would appear to be committed to what he is trying to rule out. Perhaps his use of the term ‘act of will’ is best seen as just a — slightly misleading — hangover from the traditional account of voluntary action, which he rejected. On this reading, by ostensibly identifying a bodily action with a mental event Schopenhauer might be taken to make two points: first, that the assumed mental event does not really exist as such, separately from the action; and secondly, that the voluntariness of the action is something one is directly aware of (as one is directly aware of a mental occurrence). When you raise your arm you are not in doubt that it is your own doing. Both these points Wittgenstein could agree with.
Attempts have been made to find affinities in yet other respects between Schopenhauer and the later Wittgenstein, notably by S. Morris Engel and Bryan Magee. The latter’s unsupported claim that ‘the Kantian-Schopenhauerian framework … remains much the same’ all through Wittgenstein’s philosophical writings need not detain us.31 Perhaps slightly less fanciful, or at least less vague, are the suggestions that Wittgenstein was inspired by Schopenhauer’s views on language and philosophy, and that he borrowed from him key terms such as ‘family resemblance’ (Familienähnlichkeit) and ‘form of life’ (Lebensform).32
The latter point is another of Magee’s unsupported claims. I suspect what he had in mind was a passage in §54 of volume I of The World as Will and Representation, where Schopenhauer writes that ‘the form of all life is the present’. However, the German does not contain the compound noun ‘Lebensform’, but ‘Form des Lebens oder der Realität’ . Indeed, I don’t know of any occurrence of the word ‘Lebensform’ in Schopenhauer. More importantly, what Schopenhauer means by ‘Form des Lebens’ — that one always lives in the present moment — has nothing whatsoever to do with Wittgenstein’s concept of a ‘Lebensform’, which means roughly the way we live and interact with other people.33
The word ‘Familienähnlichkeit’ does occur in Schopenhauer (e.g. WWR I §28b; ) — as in hundreds of other 19th and 20th century German writers. It is not a technical term, but a commonly used German compound noun. Wittgenstein employed it to explain the way a concept can be held together without an underlying definition specifying necessary and sufficient conditions. This philosophical idea was to a certain extent anticipated by William James’s (1902) remarks on the concept of religion (which Wittgenstein had read in 1912). It also occurs (definitely unknown to Wittgenstein) in Robert Musil’s novel Der Mann ohne Eigenschaften (1952, 1173f.). But there is not a trace of this philosophical idea in Schopenhauer, whose uses of the word ‘Familienähnlichkeit’ have nothing to do with language.
A more interesting point of contact is that Schopenhauer rejects the idea that understanding words is a process of translating them into mental images (WWR I §9). In that he is very sensible and fully in agreement with the later Wittgenstein’s views. Did Wittgenstein derive his anti-imagist stance from Schopenhauer? We cannot tell for certain, but it seems rather unlikely. There is no clear evidence that Wittgenstein reread Schopenhauer while working on the Tractatus,34 and it is even less likely that he should have reread him after 1920. The later Wittgenstein’s occasional remarks about Schopenhauer are markedly unenthusiastic. About 1939 he writes:
One could call Schopenhauer an altogethercrude mind. I.e., he does have refinement, but at a certain level this suddenly comes to an end & he is as crude as the crudest. Where real depth starts, his finishes.
One might say of Schopenhauer: he never takes stock of himself. [CV 41]
And in 1948 Wittgenstein expressly rejects his friend Drury’s suggestion that his ‘fundamental ideas’ might have been inspired by Schopenhauer:
No; I think I see quite clearly what Schopenhauer got out of his philosophy — but when I read Schopenhauer I seem to see to the bottom very easily. He is not deep in the sense that Kant and Berkeley are deep. [Drury 158]
Some of the ideas of a ‘shallow’ philosopher may well be useful and memorable — as Wittgenstein would always remember and occasionally refer to some of Schopenhauer’s tenets or observations35 — but one doesn’t feel the need to go back to them. Apparently, the mature Wittgenstein regarded Schopenhauer as a thinker whom he had exhausted: who would not repay further study. If in those later years he had had another look into The World as Will and Representation, it had obviously not made a good impression on him and he would hardly have read on for very long.
Assuming, then, that Wittgenstein’s only serious reading of Schopenhauer was before the completion of his early philosophy, it seems rather unlikely that the passage in which Schopenhauer remarks that words are not always accompanied by mental images had made a great impression on Wittgenstein. For there is no reflection of it in the Tractatus or the early Notebooks, which, on the contrary, lean towards the idea that semantic representation is indeed mirrored by some form of mental representation.
It should also be noted that Schopenhauer’s observation is very limited. He correctly observes that words are not always accompanied by mental images, but he doesn’t seem to realize the philosophical significance of this negative observation. Locke had invoked images (‘ideas’) to explain how arbitrary signs are invested with meaning. If this is wrong, the question arises how else words become meaningful. Yet Schopenhauer, unaware of the problem, has nothing to suggest. All he says is:
The meaning of the speech is immediately grasped, accurately and clearly apprehended, without as a rule any conceptions of fancy being mixed up with it. [WWR1 §9; p.39 ]
Finally, it has been suggested that Wittgenstein’s conception of philosophy owes something to Schopenhauer (Engel 1980, 250-2). The tertium comparationis is Kant’s project of critical philosophy, his rejection of speculative metaphysics, which Schopenhauer to some extent endorsed and inherited, and which Wittgenstein might be said to have developed further (cf. Glock 1999, 427-35). There is no evidence of a Kantian influence on Wittgenstein’s subversive account of philosophical doctrine as the result of conceptual confusion.37 More likely is the surmise that the Tractatus project of ‘drawing a limit to thought’ and meaningful discourse (TLP, Preface) is not only analogous to Kant’s attempt to determine the limits of knowledge, but actually to some extent inspired by it, at least indirectly, through Schopenhauer’s account of those Kantian ideas. However, the similarity between the two projects should not be exaggerated. Whereas Wittgenstein rejected all metaphysics as nonsense, Kant’s aim was, on the contrary, to rehabilitate metaphysics. He rejected hopelessly speculative metaphysics only in order for it not to be mixed up with respectable, transcendental metaphysics: metaphysics of a kind that ‘will be able to be presented as science’, as Kant promised in the title of one of his books.38
Anyway, the influence in this case would ultimately be Kant’s, not Schopenhauer’s, whose attitude to Kant’s rejection of speculative metaphysics was notoriously ambivalent. Although Schopenhauer adopted Kant’s framework of transcendental idealism, he did not in the end hold on to Kant’s definition of the noumenal side as entirely beyond the reach of human understanding, but advanced the explanation that the thing in itself could, at least approximately, be understood as will. Overall, his conception of philosophy was markedly different from that of either Kant or Wittgenstein. Schopenhauer’s philosophy tries to answer the questions to which traditionally religion would have given answers. It is a secular attempt to solve ‘the riddle of the world’ (WWR1 428 ): to decipher the meaning of the world (which, as in a religious world view, is taken to lie beyond the world of experience) in response to the experience of suffering and death (which are treated as inductive evidence for the sought explanation) (WWR2 ch.17, 161 [187f.]). Philosophy as Weltanschauung, as a secular substitute for religion, was never Wittgenstein’s project. It is true that at a time of personal crisis he attempted in his philosophical diary to clarify his own ethics and world view, and a few laconic remarks from those passages were inserted in the Tractatus. But (as noted in section 1 above) their link with the main body of the book is tenuous and they are certainly in conflict with (even the young) Wittgenstein’s conception of philosophy as ‘not a body of doctrine, but an activity’ of linguistic clarification (TLP 4.112). This conception of philosophy would be taken more seriously by the later Wittgenstein (cf. WVC 183), and the Philosophical Investigations contains not even sporadic aphorisms about the meaning of life. It is not that he had lost interest in the big questions of morality and man’s place in the world — far from it. What is more, he no longer maintained the unduly narrow criterion of sense propounded in the Tractatus, according to which only the propositions of natural science are meaningful (TLP 6.53). One can, according to the later Wittgenstein, meaningfully discuss questions of moral value and religious, or quasi-religious, attitude. It is only that he didn’t regard them as a concern of his academic philosophy. His aim in philosophy was merely ‘to shew the fly the way out of the fly-bottle’ of conceptual confusion (PI §309).
Schopenhauer’s influence on Wittgenstein was considerable, but for the most part indirect and negative. I argued that the most obvious traces of Schopenhauerian ideas in Wittgenstein’s writings are the least interesting ones. As a young man, in times of crisis, trying to formulate his ethics and attitude towards life, he remembered and adopted various thoughts from Schopenhauer, some of which he tacked on to his logical-philosophical treatise; but they have very little to do with his philosophical achievements. His real debt to Schopenhauer lies elsewhere. For one thing, the young Wittgenstein was persuaded by Schopenhauer’s idealism (minus its transcendental side), and that proved extremely fruitful for his own thinking all through his life. Early on the empiricism, that recommended idealism to him in the first place, was pushed further and (perhaps with some additional inspiration from Weininger) led him to solipsism. Later, when he had become more critical of these ideas, they spurred him into a thorough investigation of the grammar of our psychological concepts; which is justly regarded as one of the main achievements of his mature masterpiece, the Philosophical Investigations. For another thing, he learnt from Schopenhauer that a voluntary action is not a bodily movement caused by a mental act of willing; and this thought was not just copied and pasted, but developed further and supported independently as part of a sustained discussion of the issue in his later writings.39
Bibliography Anscombe, G.E.M. (1959): An Introduction to Wittgenstein’s Tractatus, London: Hutchinson.
Berkeley, George (1710): A Treatise concerning the Principles of Human Knowledge, London: Dent, 1910.
— (1713): Three Dialogues between Hylas and Philonous, ed.: J. Dancy, Oxford: OUP 1998.
Black, Max (1964): A Companion to Wittgenstein’s ‘Tractatus’, Cambridge University Press.
Candlish, Stewart (2001): ‘The Will’, in H.-J. Glock (ed.), Wittgenstein: A Critical Reader, Oxford: Blackwell; 156-73.
Drury, M. O’C. (1981): ‘Some Notes on Conversations with Wittgenstein. Conversations with Wittgenstein’, in: R. Rhees (ed.), Recollections of Wittgenstein, Oxford: OUP; 76-171.
Engel, S. Morris (1980): ‘Schopenhauer’s Impact on Wittgenstein’, in: M. Fox (ed.), Schopenhauer: His Philosophical Achievement, Brighton: The Harvester Press; 236-54.
Gabriel, G. (1995): ‘Solipsismus’, in: J. Ritter, K. Gründer, G. Gabriel (eds), Historisches Wörterbuch der Philosophie, Bd. 9, Basel: Schwabe.
Glock, Hans-Johann (1999): ‘Schopenhauer and Wittgenstein: Representation as Language and Will’, in: C. Janaway (ed.), The Cambridge Companion to Schopenhauer, Cambridge: CUP; 422-58.
— (2000): ‘Forms of Life: Back to Basics’, in: K. Neumer (ed.), Das Verstehen des Anderen, Frankfurt/Main: Peter Lang.
Hacker, P.M.S. (1986): Insight and Illusion: Themes in the Philosophy of Wittgenstein, rev. ed., Oxford: Clarendon Press (1st ed. 1972); repr. Bristol: Thoemmes, 1997.
James, William (1902): The Varieties of Religious Experience. A Study in Human Nature, London: Longmans.
Janaway, Christopher (1989): Self and World in Schopenhauer’s Philosophy, Oxford: Clarendon Press.
Kant, Immanuel (1783): Prolegomena zu einer jeden künftigen Metaphysik, die als Wissenschaft wird auftreten können, Riga: Johann Friedrich Hartknoch.
Magee, Bryan (1983): The Philosophy of Schopenhauer, Oxford: Clarendon Press.
McGuinness, B. (1988): Wittgenstein, A Life: Vol.1 Young Ludwig 1889-1921, London: Duckworth.
— (2002): Approaches to Wittgenstein. Collected papers, London: Routledge.
Musil, Robert (1952): Der Mann ohne Eigenschaften, Reinbek bei Hamburg: Rowohlt, 1978.
Pears, David (1987): The False Prison. A Study of the Development of Wittgenstein’s Philosophy. Vol.1, OUP.
Redpath, Theodore (1990): Ludwig Wittgenstein: A Student’s Memoir, London: Duckworth.
Schroeder, Severin (2006): Wittgenstein: The Way Out of the Fly-Bottle, Cambridge: Polity.
Weiner, David Avraham (1992): Genius and Talent: Schopenhauer’s Influence on Wittgenstein’s Early Philosophy, Rutherford: Fairleigh Dickinson University Press.
Weininger, Otto (1903): Geschlecht und Charakter. Eine prinzipielle Untersuchung, repr. Munich: Matthes & Seitz, 1997.
— (1904): Über die letzten Dinge, repr. Munich: Matthes & Seitz, 1997.
ALWittgenstein’s Lectures, Cambridge, 1932-1935, ed.: A. Ambrose, Oxford: Blackwell, 1979.
BBThe Blue and Brown Books, Oxford: Blackwell, 1958.
BTThe Big Typescript, = TS 213, published as WA vol. 11.
CVCulture and Value, rev. ed., ed.: G.H. von Wright; tr.: P. Winch, Oxford: Blackwell, 1998.
D Denkbewegungen: Tagebücher 1930-32 1936-37 (MS 183), ed.: Ilse Somavilla, Innsbruck: Haymon Verlag, 1997. [Quoted by MS page numbers.]
EPBEine philosophische Betrachtung, in Werkausgabe Band 5, Frankfurt/Main: Suhrkamp 1984; 117-282.
FBWittgenstein. Familienbriefe, eds: B. McGuinness, M. C. Ascher, O. Pfersmann, Vienna: Hölder-Pichler-Tempsky, 1996.
ML ‘Wittgenstein’s Lectures in 1930-33’, notes by G.E. Moore, reprinted in: Philosophical Occasions 1912-1951, eds: J. Klagge & A. Nordmann, Indianapolis: Hackett, 1993; 46-114.
MS Unpublished manuscript, numbered in accordance with von Wright’s catalogue, reprinted in: Philosophical Occasions 1912-1951, eds: J. Klagge & A. Nordmann, Indianapolis: Hackett, 1993; 480-510.
NBNotebooks 1914-1916, eds: G.H. von Wright & G.E.M Anscombe; tr.: G.E.M. Anscombe, revised edition, Oxford: Blackwell, 1979.
NfLNotes for Lectures on “Private Experience” and “Sense Data” (1934-36), in Philosophical Occasions 1912-1951, eds: J. Klagge & A. Nordmann, Indianapolis: Hackett, 1993; 202-88.