EDWARD HARCOURT
Of the extraordinary roll call of Viennese cultural celebrities who were Wittgenstein’s rough contemporaries, some were certainly far closer to Wittgenstein than Freud was. But though there is no evidence that Freud and Wittgenstein ever met, there were a number of indirect personal connections between them. And though psychoanalysis was not a major theme of Wittgenstein’s work, it was a theme that Wittgenstein could not leave alone, and we find him going over much the same small set of questions about it for approximately the last 20 years of his life. After describing briefly some of the personal connections that relate Wittgenstein to psychoanalysis, I turn to psychoanalysis as it features in Wittgenstein’s writing (which in practice means Freud). The subject can be divided into three main themes: the unconscious; dreams, jokes, and the nature of psychoanalytic explanation; and the relation between psychoanalysis and Wittgenstein’s method in philosophy (on this latter theme see also Chapter 13, PHILOSOPHY AND PHILOSOPHICAL METHOD). All three themes have had, and continue to have, a considerable afterlife. Wittgenstein’s continuing influence on philosophical thinking about the unconscious and about psychoanalytic explanation will be evident in my discussion of these themes in Wittgenstein’s work itself. To capture some further echoes of Wittgenstein’s writing about psychoanalysis, I also add two further sections, one on Wittgensteinian philosophy that has been self‐consciously “psychoanalytical” in its method, the other on Wittgensteinian influences on post‐Freudian psychoanalytic practice.
Though Freud continued to publish up to his death in 1939, the only works by him to which Wittgenstein refers in writing were written before World War I: The Interpretation of Dreams (1900), The Psychopathology of Everyday Life (1901), and Jokes and their Relation to the Unconscious (1905). (Wittgenstein also refers to Breuer, whose Studies in Hysteria, co‐authored with Freud, appeared in 1895.) It is possible that Wittgenstein first read these works only around 1930, when he began to discuss psychoanalysis in his writing (MS 109, p.174), though the fact that these discussions are so short on detail and the early publication dates of the works themselves might be taken to suggest that Wittgenstein was relying on memories of his reading that go much further back. Wittgenstein would surely have been acquainted with psychoanalysis – whether through reading or not – well before his first written mention of it. The Breuer and Freud volume was in his family’s library and it is reported that the Wittgenstein siblings took special pleasure in trading jokes with one another, under the influence of Freud’s 1905 book (Prokop, 2003, p.104). Wittgenstein’s sister Margarete took her adolescent son Thomas for analysis with Freud on account of his stammer (Prokop, 2003, pp.202, 222) which, since the boy was born in 1906, may be presumed to have occurred some time between 1916 and the early 1920s; “suppose you were analysed when you had a stammer” introduces some 1938 remarks of Wittgenstein’s on the criteria for the truth of psychoanalytic interpretations (LC 25).
But even if Wittgenstein was acquainted with psychoanalysis well before 1930, indirect personal connections brought it closer to the forefront of his mind from the mid‐1920s onwards. Wittgenstein’s meetings with the Vienna Circle began in 1928; according to a junior member of the Circle, Heinrich Neider, “numerous members of the Vienna Circle were in analysis” (Bouveresse, [1991] 1995, p.7); certainly Schlick was (Money‐Kyrle, 1979, p.266). More importantly, Frank Ramsey traveled to Austria in 1924 to have analysis with Freud’s follower Theodor Reik (Forrester, 2004, p.11). Ramsey visited Wittgenstein in Lower Austria to ask him questions about the Tractatus and had an argument with him about Freud in England in 1925 (Forrester, 2004, p.17; see also Paul, 2012), before the two saw a great deal of each other after Wittgenstein’s return to Cambridge in 1929. A third connection runs via Wittgenstein’s sister Margarete, with whom he was in contact more or less throughout his life. She was already thinking about Freud’s views of dream interpretation in 1918, complaining – as Wittgenstein himself did later – about Freud’s determination to find sexual meaning in dreams (“pity he’s so […] one‐track,” Prokop, 2003, p.100). She also read Freud’s Civilization and Its Discontents when it came out in 1930, finding it “dreadfully bad.” As long as Freud stuck to “the bodily and the psychical” he got it “about 90% right,” but “when he gets philosophical and deals with guilt, happiness and such, he comes out with unfortunate rubbish” (Prokop, 2003, p.202). She nonetheless entered an analysis with Freud in 1937, and in 1938 was instrumental in securing permission for him to leave Austria for England, receiving from Freud an inscribed copy of The Future of an Illusion on the day of his departure (Nedo and Ranchetti, 1983, p.301) and a letter from him from England (Subotincic, 2000, p.60). Margarete Wittgenstein also left an unpublished manuscript, apparently from the early 1940s, a “psychoanalytical investigation” of the success of the Nazis (Prokop, 2003, p.245, n.415). Even if Wittgenstein’s interest in psychoanalysis was less long‐standing than his sister’s, it seems unlikely that his interest in it was not in part stimulated by hers (though perhaps also vice versa).
There is an odd atmosphere to Wittgenstein’s remarks about psychoanalysis. On the one hand he is on record as making some very enthusiastic remarks about it. He was “greatly impressed when he first read Freud” (WAM 39); on first reading The Interpretation of Dreams, he thought: “Here at last is a psychologist who has something to say” (RW 136). Nor was it just a matter of a good first impression: to Norman Malcolm in the 1940s he praised Freud’s “extraordinary scientific achievement”; to M. O’C. Drury in 1948 he said “no one today can do psychoanalysis the way [Freud] did” (RR 154); and to Rush Rhees in the mid‐1940s he described himself as “a disciple of Freud,” “a follower of Freud” (LC 41), and as “one influenced by Freud” (PCR 12). On the other hand he can also be very negative about Freud: his “whole way of thinking wants combatting”; it requires “a very strong and keen and persistent criticism to see through the mythology” (LC 50); Freud offered “fanciful pseudo‐explanations” (CV 62); “unless you think very clearly psychoanalysis is a dangerous and a foul practice” (WAM 39); Freud’s followers had made “an abominable mess” (M 107).
One possibility, then, is that there is no settled judgment to be extracted from Wittgenstein’s remarks about Freud, that is, nothing to be got by taking the balance of the positive and the negative: the only reality is his veering between the two. Another possibility is that, as Rhees says, Wittgenstein’s remarks on psychoanalysis are a sustained attempt “to separate what is valuable in Freud” from what should be rejected (LC 41; cf. DB 16–17). The first possibility derives some support from the fact that Wittgenstein returns again and again to the same narrow set of psychoanalytic ideas: unless his interest in Freud had something of the character of an obsession, how could he have sustained himself for so long on such slender fare? But even if there is something to this, it is not a reason for us not to try to take the balance of the positive and the negative, and see what’s left. The questions in relation to Wittgenstein’s treatment of Freud that I shall address are therefore the following: where (in Wittgenstein’s view) does the line fall between the objectionable and the unobjectionable in Freud? Are Wittgenstein’s objections to Freud well founded? And, supposing they are, is the philosophically tidied‐up version of psychoanalysis – which has been built upon Wittgenstein’s remarks by more than one commentator – both coherent and more than mere common sense?
Sometimes Wittgenstein mentions the unconscious more or less en passant, to illustrate a more complex philosophical point; sometimes as a topic of interest in its own right. His most extensive remarks about it, which are of both kinds, occur in “The Blue Book” (BB 23, 57; cf. AWL 39 ff.). When “psychoanalysis talks of unconscious thoughts, acts of volition, etc.,” no philosophical mistake is made, since all that has happened is that a “new notation” – that is, a new use of the words “thought” and so on – has been introduced. Indeed this usage couldn’t involve a mistake, since notations on their own do not say anything, and can “at any time be retranslated into ordinary language” (BB 23). The notation is of philosophical interest, however, because it is so easy to think – falsely, thanks to the notation’s “calling up new pictures and analogies” (BB 23) – that its use reports a new discovery. If we fall into that trap, we will “be misled into thinking that a stupendous discovery has been made” (like “the psychoanalysts [who] […] were misled by their own way of expression into thinking that […] they had, in a sense, discovered conscious thoughts which were unconscious” (BB 57; AWL 40; Freud, [1933] 1953–66a, p.159), but “the contribution [of psychoanalysis] to science lies precisely in having extended research to the mental field”); or else be “tempted to deny the possibility” of unconscious thoughts (BB 57, like those “revolted” by “the idea of there being” such things, ibid.), though these objectors to psychoanalysis were – if only they knew it – objecting only to a form of description.
Had the psychoanalysts then discovered nothing? The “unconscious toothache” example in the “Blue Book” suggests this view, since the phenomena dressed up as unconscious toothache by the new notation – a toothache in a particular tooth that comes and goes, perhaps – are boringly familiar (cf. PG 48, 106, 181; PI §149). But, according to the “Blue Book” account, the psychoanalysts did discover something: “new psychological reactions” (BB 57; M 102). Moreover Wittgenstein does not seem always to have regarded the language of unconscious states of mind as a terminological innovation (and so a fortiori not as an innovation misunderstood as a discovery). In a typescript from 1946–1947, he says that “we” – not just “the psychoanalysts” – would (“perhaps”) say that a man who “suddenly climbs on a chair and then gets down again” without being able to say why, though “he reports having noticed this and that from the chair, and that it seems as if he climbed up in order to observe this,” had “acted with unconscious intention” (RPP I §225; cf. LC 22–3; PPF 282). And in a 1931 manuscript, Wittgenstein speaks, without a trace of a scare‐quote, about “unconscious contempt,” and simply goes on to explain what that useful expression means (MS 155, pp.30v–31r).
On dreams, Wittgenstein agreed with Freud – though of course also with a great many others (Freud, [1900] 1953–66d, pp.1–5) – that
dreams […] seem to have something puzzling and in a special way interesting about them — so that we want an interpretation of them.
(LC 45; cf. PCR 12; CV 75, 79; LW II §§195–6)
However, he thought Freud was wrong to claim that all dreams are wish‐fulfillments (LC 42; CV 50; RR 154), or that they all have sexual meaning (LC 23–4, 47–8). He also said that “in Freudian analysis the dream is as it were dismantled. It loses its original sense completely,” because it substitutes an interpretation for “the dream story [which] […] has its own charm, like a painting that attracts & inspires us” (CV 78–9).
One might think from this last passage that Wittgenstein thought – in the spirit perhaps of “we must do away with all explanation, and description alone must take its place” (PI §109) – that dreams should simply be contemplated, and that any interpretation of them spoils them. But that suggestion is belied by the fact that Wittgenstein frequently interpreted his own dreams (DB passim), so the objection seems to be only to the straitjacket of wish‐fulfillment or sexual interpretations. The guiding thought behind Wittgenstein’s objections to the thematic monotony of Freudian dream‐interpretation seems really to be “I will show you differences” (Monk, 1990, p.537).
Beyond this point, however, dreams are also an occasion – jokes are another – for Wittgenstein to charge Freud with confusion about the interrelation between, on the one hand, the criteria for the truth of psychoanalytic interpretations and, on the other, the standing of psychoanalysis as a would‐be scientific enterprise. This objection extends beyond dreams to distinctively psychoanalytic claims to knowledge more generally. The objection may be introduced by Moore’s summary of it from Wittgenstein’s 1932 lectures:
[Psychoanalysis] does not enable you to discover the cause but only the reason of, e.g., laughter. […] [P]sychoanalysis is successful only if the patient agrees to the explanation offered by the analyst, and […] since this is so, what is being agreed to isn’t a hypothesis.
(M 108)
That is, Freud claims that the criterion for the correctness of a psychoanalytic interpretation of a joke or dream is the patient’s assent (cf. AWL 39–40). But assent, either the patient’s or anybody else’s, could not possibly be the criterion for the truth of a causal hypothesis, so it cannot be the case – as Freud also claims – that the interpretation is a hypothesis about the dream’s or the laughter’s unconscious causes. The link with the objections summarized in the previous section should be obvious: It’s “the hypothetical part of his theory, the subconscious, […] which is not satisfactory” (AWL 39; Cioffi, 1998, p.206). (Freud, incidentally, warns against the use of the word “subconscious” – as opposed to “unconscious” – “which has become so popular in the more recent literature of the psychoneuroses” ([1900] 1953–66d, p.615), a caution that evidently fell on deaf ears as far as Wittgenstein or at least his note‐takers were concerned.)
Now there are many phenomena, dreams and laughter apparently included, for which in Wittgenstein’s view the kind of explanation that depends on assent is entirely proper. One is having something on the tip of one’s tongue (the speaker’s saying “‘that’s it!’ […] certifies the word as having been found” (CV 68; cf. LC 18)); another is overruling another’s claims about their feelings (MS 110, p.230). A third such type of explanation seems to be the wide class of what Wittgenstein calls “aesthetic explanations” which are not “causal” but do “what aesthetics does: puts two factors together”; and certainly Wittgenstein sees some of Freud’s explanations as of this kind, e.g., of jokes (AWL 39), and Freud’s connection between the fetal position and sleep (AWL 39; cf. LW II 86). Indeed it is even in order, in explanations where the patient’s assent is the criterion for correctness, to say that the explanation gives the patient’s unconscious state of mind (e.g., the unconscious reason for the joke), as long as we do not make the mistake of thinking that in so saying, we are saying something “as to what was happening at the moment when he laughed” (M 108). If that was all there was to it, the objection would be very mild: not to Freud’s explanations themselves, nor to the word “unconscious,” but only to Freud’s self‐understanding (a self‐understanding that would be proper to a “psychologist,” since “in psychology” we are “interested in causal connections” (AWL 38), though Freud himself – had he only been able to see it – wasn’t one).
However, this is not quite all there is to it. For a start, Freud’s scientistic self‐misunderstanding has consequences: he doesn’t stick to the limits of interpretation that the assent criterion imposes but corrects the patient if their explanation doesn’t accord with his “hypothesis.” This is a mistake he as it were would not have dared to make had he realized what he was doing. Secondly, the explanations Freud offers are, in Wittgenstein’s view, unreliable in a way that saying what was on the tip of one’s tongue is not: apropos Freud’s explanation of a patient’s “beautiful dream” Wittgenstein says “this ugly explanation makes you say you really had these thoughts, whereas in any ordinary sense you really didn’t” (LC 20). People are “charmed” by the kind of interpretation Freud is prepared to recognize as correct, so they assent to it, but this complicating “charm” is no part of the operation of the assent criterion per se. Whereas in saying what’s on the tip of one’s tongue one is free to answer without undue influence from elsewhere, in Freud assent is contaminated by the “charm” of his various “mythologies” which attract us overwhelmingly to certain kinds of explanation. As to what this “charm” is, Wittgenstein makes various suggestions: the charm of any explanation of the form “this is only that,” the charm of the ugly (LC 23), the charm of “origins” (LC 43) or the “secret cellar” (LC 25), the “new mythology.” It’s not clear which Wittgenstein loathes in Freud the more: the particular mythology that he is (in Wittgenstein’s view) merely campaigning for, or the fact that he dresses up this campaign as a kind of science.
I would now like to offer two sets of remarks by way of commentary. First of all, Wittgenstein’s assimilation of Freudian dream‐ or joke‐interpretations to “aesthetic investigations” needs to be treated with great caution. In such investigations, various things are “laid alongside” the initial object of investigation, and whether anything is accomplished by so doing depends on somebody’s assent. But whose? Wittgenstein’s idea is that if dream‐interpretations were not contaminated by Freudian suggestion, it’s the patient’s assent that would properly be decisive. But although you can (Wittgenstein says) “make a person see what Brahms was driving at by showing him lots of different pieces by Brahms, or by comparing him with a contemporary author” (M 106), it is one’s interlocutor’s assent, not Brahms’s, that’s relevant to the success of this exercise in “plac[ing] things side by side” (AWL 40). Moreover if this method fails to make your interlocutor “see what you see,” this is not proof that you didn’t, after all, see anything, but simply “an end of the discussion” (M 106; cf. LC 20–1). So if dream‐interpretation were exactly like “aesthetic investigation,” the (in Wittgenstein’s view false) assumption that psychoanalytic interpretations are “hypotheses” would not be needed to show that Freud was not making any mistake in ignoring his patient’s dissent and sticking to his own interpretation: it would just be a further case of one person not seeing what another sees. This wouldn’t matter if all Wittgenstein’s objections to Freud come to are that he thinks Freud regularly chooses objects of comparison that Wittgenstein would not (and that Wittgenstein finds distasteful, etc.). But if the objections are – as they seem to be – also in part methodological, the comparison of psychoanalytic interpretation with “aesthetic investigation” (which puzzled Moore (M 105)) is unhelpful.
Secondly, how plausible is it that patient assent is the criterion of correctness for explanations that refer to unconscious thoughts and feelings? Wittgenstein’s discussion of unconscious motivation in the “Lectures on Aesthetics” is relevant here:
Suppose Taylor and I are walking along the river and Taylor stretches out his hand and pushes me in the river. When I ask why he did this he says: “I was pointing out something to you,” whereas the psycho‐analyst says that Taylor subconsciously hated me. […] When would we say that Taylor’s explanation was correct? When he had never shown any unfriendly feelings, when a church‐steeple and I were in his field of vision, and Taylor was known to be truthful. But, under the same circumstances, the psycho‐analyst’s explanation may also be correct. […] The explanations could in a sense be contradictory and yet both be correct.
(LC 22–3)
This is striking: The unconscious explanation “may […] be correct,” and yet there is no mention of its correctness depending on Taylor’s assent. What is relevant to its correctness is a further fact, that “the person pushed in had a similarity with the father of the other person” (LC 22–3; cf. RPP I §225). The explanatory model Wittgenstein has in mind here seems closer to “aesthetic investigation” than to the patient’s‐assent‐as‐criterion model: even if it would be troubling to concede that, absent one’s interlocutor’s assent, reason‐giving is simply at an end (because, for instance, further reasons might lie in general observations about the emotions to which human beings are subject), certainly the reasons in the Taylor case “are in the nature of further descriptions” (M 106). One might try to relieve the tension between Wittgenstein’s treatment of “unconscious motive” in the Taylor case and his treatment of dream‐interpretations by suggesting that further facts about the patient are relevant in the former case but not in the latter. This might be true of claims about what was on the tip of one’s tongue, but it is surely not so for dreams: a psychoanalyst (or indeed anyone else) may be familiar with details of the dreamer’s preoccupations, and with more general facts about what people in the patient’s predicament think or feel, which suggest interpretations of the dream that the dreamer may well not acknowledge.
Indeed Wittgenstein’s own remarks about the way in which patient assent may be contaminated – Wittgenstein dwells on the “charm” of certain kinds of explanation, but he might just as well have mentioned the prestige of the analyst, or the thrill of being of one mind with the analyst that assent can create (Caper, 1999, p.115) – gives us further reason to doubt whether patient assent can, on its own, be the criterion of correctness of the interpretation of a dream or joke. For it makes no sense to speak of contamination unless one can specify – as one surely can – what it would be for assent to be uncontaminated. But if it can be either contaminated or uncontaminated then, though assent could still be a criterion, it won’t be the criterion. Does Wittgenstein need to say that it is in order to push through his two chief objections to Freud? As to the first – that Freud contaminates patient assent by suggesting interpretations that are appealing but bad – clearly not. As to the second – that dream‐ or joke‐interpretations, or ascriptions of unconscious motives, are not causal hypotheses – things are less clear. The patient’s‐assent criterion excludes the explanation‐as‐hypothesis view. But that criterion is not needed in order to exclude it, for it’s also excluded by the idea that interpretation is a matter of “further description,” if this is just another species of “aesthetic explanation” where failure to get the other to “see what you see” is just “an end” of the discussion (M 106), in the sense that there’s no fact of the matter rather than that one has reached an epistemic impasse. If on the other hand exchanges of reasons in the psychological case are not to be thought of in this “no fact” way, why couldn’t psychological explanations proceed by “further description” precisely because adding further descriptions is a way, if not of establishing a particular causal hypothesis, at least of ruling out inadmissible ones? The argument from the claim that various psychological explanations proceed by “further description” to the conclusion that they are not about causal hypotheses seems to need some extra premises.
There is no doubt that in a certain phase of his philosophical development, Wittgenstein took the comparison between psychoanalysis and what he saw as the correct method in philosophy very seriously. This attitude on Wittgenstein’s part may also have lasted long enough to explain some well‐known methodological remarks in Philosophical Investigations. The chief question to which the comparison gives rise is whether it really casts light on Wittgenstein’s method, or whether it is based to a greater or lesser extent on a misunderstanding of Freud’s psychoanalysis.
Wittgenstein’s explicit comparisons between Freud’s method and his own occur between the time he began what became the so‐called Big Typescript (around 1932) and the writing down of TS 220, in 1937 or 1938 (Baker, 1997/2004; 2003; Majetschak, 2008), and most saliently in the Big Typescript itself and in “Dictation for Schlick” and “Our Method,” dictated to Waismann (VW 1–83; 277–311). The starting point for the comparison is the idea that philosophical problems stem from false “analogies” or “pictures” suggested to us by language – an idea that, as far as I know, has no echo in Freud. But once that point is granted, the comparison with psychoanalysis can take off. Philosophical puzzlement is a cause of mental disquiet or unease (M 114) – also presumably the condition of anyone who refers himself for psychoanalysis. Philosophy, like psychoanalysis, cures this unease by bringing what is unconscious to consciousness (“A simile at work in the unconscious is made harmless by being articulated” (VW 69; cf. MS 109, p.174; BT 409; PG 381–2)); and, though this comparison is more rarely drawn than the previous one, philosophy like psychoanalysis has to deal with resistances to giving up such analogies (“What has to be overcome is not a difficulty of the intellect but of the will,” BT 407). Furthermore the success of this kind of philosophical cure depends – as, in Wittgenstein’s view, the success of psychoanalytic treatment or interpretation (LC 25, 44, 52) – on the “patient’s” assent to the “diagnosis” offered (BT 410; Baker, 1997/2004, p.159). Once assent is given, the effect is not only to remove the destructive power of the analogy but also to bring about “a new way of seeing things” (Baker, 1997/2004, p.158).
Some of these comparisons with Freud – that psychoanalysis proceeds by making the unconscious conscious, and that the difficulties in doing so are due to resistances on the patient’s part – are comparisons Freud himself would have had no difficulty in recognizing (Majetschak, 2008, p.52). Others depend on Wittgenstein’s own (counter‐Freudian) understanding of what Freud was really doing. But even if all that is accepted, there is at least one point at which the comparison between their methods appears to falter. Baker has put it very clearly: while psychoanalysis addresses conflicts of “conative or affective states,” philosophy seems to address only conflicts “in ways of seeing things,” in people’s “prejudices or dogmas [that] clash with each other and creat[e] fogs of confusion,” or between different things that – thanks to these dogmas – we feel driven to say; in psychoanalysis we are concerned with “patterns of behaviour (e.g. manifestations of an Oedipus complex),” whereas in philosophy – seemingly – our concern is “with patterns in the uses of our words” (Baker, 1997/2004, pp.153, 159). The question is what we should make of this.
One response is exemplified by Alice Ambrose, Morris Lazerowitz, and their followers (Lazerowitz, 1985; Ambrose, 1966; 1972; Lazerowitz and Ambrose, 1984; Kennick, 1970). According to them, (non‐Wittgensteinian) philosophers mistakenly believe their words are “being used to express a theory” when in fact they simply “herald a redefinition” (Lazerowitz, 1985, p.209). This is very close to Wittgenstein’s own view of philosophical disagreements over solipsism, or over the existence of an unconscious mind, in the “Blue Book.” (Ambrose of course was one of the students who transcribed the “Blue Book.”) Part of the Wittgensteinian philosopher’s task is therefore to “expose the verbal content behind the ontological façade” (Lazerowitz, 1985, p.209). However, Ambrose and Lazerowitz’s conception of the proper Wittgensteinian method in philosophy has a further articulation. What makes it difficult for philosophers to acknowledge their theories as “mere linguistic contrivance[s]” (Lazerowitz, 1985, p.211) is that they are held in place by “unconscious ideas” of a kind that are the stuff of Freudian dream‐ or joke‐interpretations, and that it is psychologically costly for the philosopher to renounce (1985, pp.236–7). Thus in Lazerowitz’s view philosophers have been kept from seeing the claim that “one cannot think of what does not exist” as a mere “redefinition” because this would rob it of its psychological function of “fend[ing] off the invasion of anxiety,” for the unconscious meaning of these words is that “there is something [sc., the penis] […] not possessed by some and whose loss is feared by others” and “whose non‐existence is too painful to be thought of,” this latter point supplying the double meaning for “one cannot think” of it (1985, p.238).
Lazerowitz, Ambrose, and others have taken to heart Wittgenstein’s idea that the method of philosophy is the method of psychoanalysis but, noting the apparent limits to the analogy claimed by Baker, have filled in the gap – the third layer of the “three‐layer structure of a philosophical theory” (Ambrose, 1972, p.25) – on Wittgenstein’s behalf, suggesting moreover that Wittgenstein only failed to do so himself because he couldn’t bear to (1972, p.25). But their “completion” of Wittgenstein’s method is unsatisfactory, and not just because the particular unconscious explanations they suggest would have attracted Wittgenstein’s scorn. For one thing, solipsists and others usually offer arguments for their views and one will surely be the more inclined to look for an explanation in terms of unconscious defenses the weaker the argument offered, and in particular, the less satisfactorily the solipsist is able by his own lights to reply to objections. But Ambrose et al.’s diagnoses apply to philosophical positions per se, independently of the arguments anyone offers for them.
Secondly, philosophical problems are usually addressed in a way that bypasses the personal life of the “sufferer”: we don’t ask fellow philosophers about their dreams or their past or their current special others with a view to helping them to get clear on a philosophical issue, and it is not clear that doing so would help; but this sort of probing is essential in psychoanalysis. This is not to say that philosophical problem‐solving isn’t personalized (Baker, 1997/2004, p.210): different people may well need different approaches to help them get unstuck philosophically. But to say that is not to say that any particular person’s way out of it need involve the examination of their personal lives. Of course it might be said that even if this investigation of the personal is usually missing from philosophy, it shouldn’t be, and Ambrose et al. remedy that defect. But in fact it’s precisely this kind of shaping of the investigation to the particular patient that is missing from their version of therapeutic philosophy: philosophical solipsism (for example) is at once a “symptom,” but a symptom whose meaning we can ascertain without knowing anything in particular about the person whose symptom it is supposed to be (Brearley, 1984, p.183).
Eugen Fischer’s view, which also takes the analogy between philosophy and psychological therapy very seriously, makes for an instructive contrast (2004; 2012). Fischer is impressed, like Baker, by the fact that the conflicts philosophy deals with aren’t emotional ones – Wittgensteinian “drives to misunderstand” (Fischer, 2004, p.107) are precisely “diseases of the intellect” (CV 50); philosophical disquietude is sufficiently explained by the thought that mere platitudes get “distorted” “through inadvertent misinterpretation or mindless inference, in line with ideas [one] unreflectively rejects” (Fischer, 2004, p.112). So, though Wittgenstein’s philosophy is indeed therapeutic in some sense, the correct object of comparison is not psychoanalysis but cognitive behavioral therapy (CBT):
the problems Wittgenstein tries to cope with in sections 138–97 of the Investigations are in some pertinent ways similar [to those described by CBT and] […] Wittgenstein attacks [them] much in line with the ‘cognitive approach’, which proceeds by identifying and […] breaking the relevant cognitive habits.
(Fischer, 2004, p.91)
Were the analogy with CBT not available, then – supposing philosophical tangles are indeed cognitive and not emotional – the response to Ambrose and Lazerowitz might still be that a philosophical theory can be like “a kind of neurotic symptom” (Lazerowitz, 1985, pp.236–7) without actually being one. But once it is available, a necessarily incomplete analogy with psychoanalysis seems pointless, because it breaks down at just the point at which it might begin to be illuminating. Thus one would be forced to conclude that Wittgenstein made as much of the comparison with psychoanalysis as he did only because it was (unsurprisingly) the only form of psychological therapy he had ever encountered.
But are philosophical problems, as Wittgenstein saw them, merely cognitive tangles? The fact that Ambrose and Lazerowitz’s psychoanalytical alternative is unsatisfactory doesn’t show that they are, nor does the intellectualist view do justice to Wittgenstein’s assertion that philosophical problems are “deep disquietudes” (PI §111), still less – though such remarks are rare in Wittgenstein – to the idea that in struggling with philosophical problems we need to overcome resistances of the will (BT 407). Wittgenstein’s struggle with philosophical problems seemed to engulf his whole being, so it is not credible that he should have thought of them merely as intellectual problems, however troubling they too can be. The high color of the language in which he speaks of them and their solution surely betrays the presence of powerful emotions (Lazerowitz and Ambrose 1984, p.13). What’s needed, then, is an account that captures the fact that for Wittgenstein philosophical problems were existential not merely intellectual, but without burdening his words with Freudian “conjectures” (Lazerowitz, 1985, p.211) about their unconscious meaning. I suggest that the more than merely intellectual significance that philosophical problems had for Wittgenstein stems not, indeed, from the fact that he saw them as symptoms of neurosis but from the fact that he saw ridding himself of them as a way of transforming himself into a “decent” person (which he constantly thought he wasn’t (Gesamtbriefwechsel 25 June 1919; cf. ibid., 16 January 1918, 24 February 1925), of redeeming himself from the fallen state that philosophical puzzlement betokens. (I develop this suggestion further in Harcourt, 2012; cf. Forrester, 2004, p.17.)
Wittgenstein has had a significant influence on philosophical commentary on psychoanalysis, and some influence on the development of psychoanalysis itself.
As to the former, one of the abiding criticisms of Freud has been that his work is not properly scientific, whether in the form of the claim that it is bad science, or that its claims are unfalsifiable and so not science at all. (The criterion of scientificity is Popper’s, though Popper (1963, pp.37–8) was ready to forgive Freud a good deal: “Much of [Freud’s work] is of considerable importance, […] describ[ing] some facts, but in the manner of myths” and containing “most interesting psychological suggestions, but not in a testable form.”) So the appeal of Wittgenstein’s view to psychoanalysis is obvious: if Freud misunderstood the nature of his own enterprise, the criticism falls only on what Freud thought he was doing, not – at least as far as this line of criticism goes – on what he did. Wittgenstein’s reconstruction has been developed, by applying his general approach to a wider range of Freudian concepts and writings than those Wittgenstein himself dealt with, to great effect by e.g., Alisdair MacIntyre (1958/2004) and Ilham Dilman (1983; 1984; 1988).
As to the latter, some Wittgensteinian versions of psychoanalysis (Dilman, MacIntyre) sound very like British “independent” psychoanalysis, and some British “independent” psychoanalysis – in its antitheoretical stance, its prizing of ordinariness of utterance, and its hostility to sexual reductionism or to reductionism about art and religion – sounds like Wittgenstein (Winnicott, 1985; Rycroft, 1991; Lomas, 1973). However, one must be cautious in claiming Wittgensteinian influence on “independent” psychoanalysis. Psychoanalysis had established itself in (and perhaps begun to be transformed by) Cambridge before Wittgenstein returned there in the late 1920s: analyst Susan Isaacs established the Malting House Lane school there in 1924; Isaacs, Ramsey, James Strachey, and others read papers to the Cambridge “1925 Group,” a condition of membership of which was having been psychoanalyzed (Forrester, 2004, p.3); and many of the same people were also members of the Heretics, whose Secretary was John (A.J.T.D.) Wisdom and to whom Wittgenstein read his “A Lecture on Ethics” (PO 36–44) in 1929. So later British psychoanalysts and psychotherapists were able to discover a community of thought between Wittgenstein and their own immediate psychoanalytic ancestors (e.g., Michael Brearley who, like Dilman, studied philosophy with Wisdom; or John Heaton, Trinity undergraduate in the late 1940s: see Heaton, 2000; 2010, p.x). The history of these complex lineages is still to be written.
In any case, let’s suppose that the Wittgensteinian rereadings succeed in showing that all the objectionable bits of Freud, his “bawdy” (LC 24) included, are “philosophical froth” (Dilman, 1983, p.3). They nonetheless give rise to an awkward question: is what’s left over once the froth is gone more than mere common sense? It’s often said by defenders of Freud that he didn’t discover the unconscious, but only a theory of the unconscious. That is meant to block criticisms of Freud based on his sometimes exaggerated claims to originality, and the defense is worth making, since unconscious explanations of behavior by way of thoughts, emotions, intentions, and the like are a commonplace at least in imaginative literature way before Freud (and continue to be so now). But if what Freud added to such explanations is just a series of philosophical mistakes then, when – as Rhees puts it (LC 41) – we “separate what is valuable in Freud” from what should be rejected, we are left only with something we had plenty of already. That substantive challenge requires an answer, and unless it is answered, an interpretative challenge follows close on its heels: what on earth did Wittgenstein – after all, the originator of the substantive challenge – think was Freud’s “extraordinary […] achievement”?
The substantive question is a large one and I can only offer some indications here as to how it should be answered. First of all, if the Wittgensteinian reconstruction of psychoanalysis includes the thought that psychoanalysis is simply an exercise in “interpretation” or “giving reasons,” it cannot be right. Psychoanalytic discourse is too various for any one simple theory about it to stand a chance of being correct (Farrell, 1981), not least because in between its (quite possibly bad) philosophical components and the interpretations of which “hermeneuticism” (Cioffi, 1998, p.130) makes so much lie a number of claims, both general and particular, about human beings and their behavior that are open to empirical confirmation or refutation. Many of those Freud made seem to be false (e.g., that “the mental apparatus endeavours to keep the quantity of excitation present in it as low as possible,” Freud, [1920] 1953–66b, p.9), but his successors have made considerable efforts to repair the problem that “hermeneuticism” pretends to magic away.
Secondly – and focusing only on those aspects of the psychoanalytic enterprise that might be called interpretative – the worry that Wittgensteinian reconstruction narrows the gap between psychoanalysis and common sense almost to nothing rests, in part, on the thought that psychoanalytic explanations are all of one kind, and of a kind indeed that would be familiar even if Freud had never existed. Unfortunately Wittgenstein does little explicitly to discourage this thought with his insistent contrast between causal explanations and “giving reasons,” which covers at least two sorts of reason, reasons for action and the “reason why [someone] laughed” (AWL 40). But there is in fact considerable variety to the explanations psychoanalysis offers.
In one very common kind of case, explanation appeals to an unconscious intention. This is so in Wittgenstein’s case of the man who “suddenly climbs on a chair and then gets down again” (RPP I §225): he “noticed this and that from the chair,” and in the circumstances, seeing those things would be a good enough reason for him to climb up, so we say he climbed up intentionally, though because he can’t say why he did so, we say the intention was unconscious. If that were the only type of explanatory appeal to the unconscious to be found in Freud, there would be some substance to the worry that he added little to what we already knew. But this is not the only one, and I shall mention four further types (with no pretense that the list is exhaustive).
In the first type of case, we form an intention to say something, akratically fail to act on it (e.g., out of fear), but the thing “slips out” anyway (Freud, [1901] 2002, pp.87–8). Here it need not be the intention that’s unconscious: the interesting feature of these cases is rather that the phenomenology of utterance has, for the agent, the character of the involuntary. But the action is so well explained by the agent’s intentions – indeed by intentions he may be fully capable of avowing – that we say the “mistake” was an intentional action. A second kind of case of “symptomatic error” is exemplified by the following:
Frau F. said, of her first lesson in a language course […]: ‘It’s really interesting, the tutor is a nice young Englishman. In the very first lesson he indicated to me durch die Bluse [through my blouse]’ – and corrects herself: ‘durch die Blume [lit., ‘through the flower’ but, colloquially, ‘with veiled hints’] – that he would rather give me private tuition’.
(Freud, [1901] 2002, p.78)
Here, unlike the last type of case, there is no unconscious communicative intention on Frau F.’s part: her attraction to the tutor simply betrays itself involuntarily in her utterance, much as emotions betray themselves in facial expressions.
Two further types of explanatory appeal to the unconscious may be illustrated by Freud’s case of a girl’s obsessive bedtime ritual, which involved arranging her bedding in a particular way (Freud, [1916/17] 1953–66c, pp.264–9). The arranging behavior consists of a series of (consciously) intentional actions. But these intentions – as it might be “I am making sure the pillow doesn’t touch the headboard” – fail satisfactorily to explain the action, since the agent’s further reasons for so acting give out immediately, and yet – unlike whistling or running one’s fingers through one’s hair – actions of this kind do not remotely make sense without a further reason. Now unconscious further reasons are – Freud says – ready to hand: the girl’s ritual prevented her parents from going to bed, so it manifested her unconscious desire to prevent her parents from having sexual intercourse, thereby averting an outcome she (unconsciously) very much did not want, namely a sibling who would be her rival. But, as MacIntyre has recently emphasized, this is not – like the “chair” case – a matter of the girl’s unconsciously acting for reasons, but rather of her “acting as if unconsciously guided by reasons.” For here, “the motives that control [her] behaviour” “preclude [her] from acting as a practical reasoner does,” since they preclude her from asking whether her reasons for action are good reasons (MacIntyre, 1958/2004, pp.25–6). (Compare also the way panic – which is of course conscious – both drives action and precludes one from evaluating reasons.) However, the girl could after all have prevented her parents from going to bed in any number of ways, so this last explanation doesn’t reach to the details of the ritual, on which the girl was insistent. To focus on just one detail, the girl would fluff the eiderdown in such a way that it made a hump, then smooth it again. According to Freud, this was in some sense the girl’s undoing of her mother’s (imagined) pregnancy ([1916/17] 1953–66c, p.268). Though various accounts of what’s going on are possible, it’s attractive to say that the girl, in fluffing the eiderdown, is intentionally undoing her mother’s pregnancy. The intention is of course unconscious (because she cannot say what she is doing). What’s more she cannot evaluate her reasons for so doing. But even if the intention were conscious – as when a child strokes an injured parent to “make it better” – the intention alone wouldn’t be a full explanation: the child’s action might have no healing effect and it might know this. To see how undoing the pregnancy is what’s intentionally being done, we must see that the relation between the action and the intention is mediated by the symbolic connection – yet another thing the girl cannot articulate – between the hump in the eiderdown and pregnancy (Gardner, 1993, p.116).
Of course it is a substantive historical question to what extent even explanations of these further kinds were already part of the cultural currency before Freud. Be that as it may, they all differ substantially from the explanatory pattern exemplified by Wittgenstein’s “chair” case, and whose availability we did not need psychoanalysis to grasp. Moreover, all four patterns invoke ideas of meaning, symbol, what’s of value to the patient and what’s of value generally in order to supply a kind of intelligibility to the actions they explain, and so fall within a broad category of explanations that “give reasons.” So even once psychoanalytic explanation is reconstructed along Wittgensteinian lines, there is a large gap between what psychoanalysis has presumptively taught us about the explanation of human behavior and mere common sense.
It’s a further substantive question whether what occupies this gap deserves to be thought of as distinctively psychoanalytical, or rather as simply marking the overlap between psychoanalysis properly so called and non‐psychoanalytical psychodynamic psychotherapies (Gardner, 1993). Be that as it may, I suggest that the gap is sufficient to explain why Wittgenstein said – and meant – that Freud’s achievement was “extraordinary” (though I can still make nothing of his describing Freud’s achievement as “scientific”). Moreover the particularistic character of what Freud, on this reconstruction, adds to common sense fits the fact that what Wittgenstein especially valued in Freud was finding “new psychological reactions,” or “phenomena and connexions not previously known” (BB 57; M 102). There is further evidence that this is where Wittgenstein located the value of Freud’s contribution in the use Wittgenstein makes of Freudian “phenomena and connexions” in his own nonphilosophical writing. In a 1930 diary entry, for example, commenting on his love of cinema, Wittgenstein compares films to dreams and says “Freudian thoughts/methods can be applied to them directly” (DB 28–31). Wittgenstein’s diaries frequently record dreams, some of them with interpretations (DB passim; Gesamtbriefwechsel 14 March 1944). Though the interpretations are not narrowly Freudian in suggesting a fantasized wish‐fulfillment or sexual meaning, they show that Wittgenstein expected his dreams to express matters of personal importance that were on his mind (e.g., racial identity, or his then intended wife Marguerite Respinger). And in a 1948 letter to one of his sisters, and not in the context of any philosophical discussion of psychoanalysis, he offers a psychoanalytical interpretation of the fact that he couldn’t get a Mendelssohn passage out of his mind (Gesamtbriefwechsel 15.3.48 cf. RPP I §262; PPF 268). Passages such as these show how far Wittgenstein internalized some – as he saw it – psychoanalytical ways of thinking.