Wittgenstein and Psychoanalysis: Image and Imagination in Wittgenstein and Freud


In this text I shall develop a parallel reading of Wittgenstein’s picture theory (Tractatus logico-philosophicus) and Freud’s theory of the vesicle (Beyond the pleasure principle), trying to show the substantial analogies and the identical target of their trajectories, namely the idea that images are things and that signs are the matter out of which our “mind” (Wittgenstein) or “psychic life” (Freud) emerges.

As is well known, after an initial phase of broad agreement with his work, Wittgenstein became more and more critical towards Freud.  Here, I am interested especially in the criticism that he levelled against the hybrid nature of Freud’s major psychoanalytic concepts, which, according to Wittgenstein, swing between the bodily and the mental, the physical and the psychic, the phenomenological and the hermeneutic.  Wittgenstein’s criticism of Freud goes hand in hand with his rejection of  the methodological ‘monism’ of the Tractatus Logico–Philosophicus.  What is more, we must remember that Wittgenstein limits his interest to Freud’s Interpretation of Dreams, and it is within  this context and with this reference in mind  that  he  openly rejects the claim of a possible monistic solution to the problem of the image and, more generally, to that of the mind.  All the causal explanations that Freud presents as final should be rejected with reference to these problems.  Yet, did Freud really think he could propose the ultimate solution to the problem of the mind and its imaginative faculty?

I would like to draw attention to the crucial (albeit rarely cited and considered) conclusion of Freud’s famous essay Beyond the Pleasure Principle.  Here Freud hypothesises that hermeneutic endeavours and psychoanalytic jargon might be one day overtaken by progress in biology and eventually discounted; yet he also adds (and this is a true gem of critical awareness for a ‘positivist’ scientist):  ‘The deficiencies in our description would probably vanish if we were already in a position to replace the psychological terms by physiological or chemical ones.  It is true that they too are only part of a figurative language; but it is one with which we have long been familiar and which is perhaps a simpler one as well’.  What is this figurative language?  What does Freud mean?  Maybe we can find an answer in a famous and often cited letter from Freud to Einstein.  After a concise account of psychoanalytic theories, Freud observes that his words might suggest to his distinguished correspondent that psychoanalysis is some kind of ‘mythology’.  Yet, he adds, is it not the case that every natural and even physical science leads to some kind of mythology?  Is this not what happens today in physics?  That the languages of natural science and even physics lead to mythology is a very firm belief; from it one could conclude that all sorts of theories are ultimately mythologies.  This, interestingly, is precisely the criticism that Wittgenstein directs at Freud.  Psychoanalysis, he says, is a very important theory, full of encouraging ideas, rich in vital implications, yet it runs the risk of becoming a mythology. And I think one might even agree, if it is true that all knowledge is ultimately a mythology. What is important is to be aware of this, because there is no worse mythology than the one which supports knowledge without narrative, thus without mythologies.  Knowledge—any kind of knowledge—is not an indistinguishable equivalent of so-called ‘reality’.  Rather, it is always the translation of the relevant problematic situation into a known language.  To translate the unknown into something known: this is the job of theory, as much as it is the job of a map to translate the territory for the purpose of orientation.  It is a real mess when one is deluded into thinking that the map actually is the territory.

Wittgenstein  accuses Freud of not having an adequate notion of the mind and of having dealt with the problem of the mental image in an unconvincing way.  We should proceed otherwise, but how?  Here, there emerges what I would like to define as the ‘ethical’ issue, an adjective that does not, in my view, have any connection with moral problems.  Let me briefly explain this by referring to the very well-known letter that Wittgenstein wrote to Ludwig von Ficker: the topic of the Tractatus, Wittgenstein says, is ‘ethical’.  Following these famous words , we will now briefly consider the Tractatus in order to understand what it means to deal with the problem of the mental image in an ‘ethical’ way.  For the sake of clarity, I will declare from the outset that I believe Freud did something very similar; Freud too considered the mental image ethically, at least once during his theoretical adventure.

Let us begin with proposition 2.1 of the Tractatus, a proposition that might sound surprising at first, but that is closely linked to what follows:

‘We picture [Bilder] facts to ourselves’.

The same issue underpins the beginning of Freud’s 1919–20 essay Beyond the Pleasure Principle: how do so-called mental images emerge?  What are mental images and what is their purpose?

Wittgenstein deals with this issue with extraordinary radicalism and theoretical courage.  This manifests itself in proposition 2.141 – perhaps even beyond Wittgenstein’s conscious intentions, especially as regards the consequences that such a stance necessarily implies.  2.141 is a very brief proposition that, if correctly understood, de facto reverses the entire history of metaphysics and its so-called substantial ‘Platonism’.  The proposition states: ‘A picture is a fact’.  This assertion wipes out any form of Platonic ‘strategy of the soul’ (as I call it); that is to say, any spiritualistic or psychologistic interpretation of the mind, and even the very notion of a mind.  Wittgenstein’s remark calls into question at its very root any reference to the mental, psychic image, to the meaningful proposition, that is to say any reference to the ‘internal’ of the human being and to her ‘intentionality’, any notion that might oppose the physical to the psychic dimension, the spiritual to the bodily, the psychological to the physiological: the image is a fact and nothing else.  In Wittgenstein’s own words, it is a combination of objects into a state of affairs.  There is nothing special or supernatural in this.

However, to assert that the image is a fact does not involve that one assumes the trivial beliefs of naïve positivism or of materialistic reductionism.  This was Heidegger’s misunderstanding, when, in a private conversation, he called Wittgenstein ‘a crass positivist’: clearly, he had not read the Tractatus with proper attention, if he had read it at all.  To assert that the picture is a fact wipes out at a stroke all ‘easy’ solutions; that is to say, the spiritualistic solutions that in truth do not solve anything at all.  Instead, the assertion highlights the actual problem: how can a fact—which, according to Wittgenstein is a picture—represent another fact?  If one does not have a satisfactory answer to this problem, then the unfortunate thing that Wittgenstein talks about in the above-mentioned letter to von Ficker will happen: people talk nonsense and do not even know what they are saying.

How can a fact represent another fact, put itself in the place of another fact, be the sign of the other?  On close inspection, this is precisely the central issue of the entirety of Western philosophy.  It begins with the enigmatic assertion that we read in Parmenides’ poem: ‘The same thing is for thinking and for being’.  If thinking and being were not the same thing, we would be lost.  Our ‘picturing facts to ourselves’ would be sheer fantasy, devoid of any truth.  Instead, seemingly, we understand facts and we express them in meaningful propositions that can be true or false, whether they conform to ‘being’ or not.  But how are we able to do it?  We have to hypothesise some sort of uniformity between the two facts: the fact of the picture and the fact that is the state of affairs signified by the picture. Here we could mention Wittgenstein’s example of the street accident.  The judge who is investigating the accident asks the witness to describe it.  The witness then takes two matches, places them in a certain way on the table: ‘Here, it went this way…’

Matches are things that represent other things, taken from a precise context or state of affairs.  In proposition 2.16 Wittgenstein says: ‘If a fact is to be a picture, it must have something in common with what it depicts’.  Something in common: ti koinon, Plato said; in other words, there must be a tauton, a ‘sameness’, as Parmenides would have put it.  Considering another of Wittgenstein’s examples we could ask ourselves how the word ‘ship’ or a drawing of the ship can represent a ship: how can states of affairs, such as the sounds made by voices and the marks made with ink represent other states of affairs that are materially completely different?  This is the problem of the ‘logical picture’ in the Tractatus. With this expression Wittgenstein does not mean to refer to a certain kind of image, for instance images that would be ‘logical’ as opposed to ‘non-logical’ ones: this reasoning expresses a ‘logicism’ that does not understand what it is saying.  When Wittgenstein speaks of ‘logical pictures’, clearly he refers to any image, insofar as any image, to be regarded as such, has the power of representing something other, which is a ‘semantic’ faculty of its own.  This is the ‘logical’ capacity of pictures.

Hence, in the Wittgensteinian understanding of ‘picture’, thus, broadly speaking, in the notion of ‘image’, we should include gestures, words, writing, drawing, music, dress codes and so on.  For instance, the score is the image of the symphony, but the CD is also an image of the same symphony and the score is the image of the CD: they are interchangeable states of affairs.  Only when understood in this sense, that is to say in their representational faculty, are pictures and thus images ‘logical’.  In short, ‘logical’ is anything that has the faculty of standing for something else, of functioning as a sign: semiotics, Charles Sanders Peirce said, is the foundation of logics and, ultimately, identifies with it.  The problem is to understand this faculty or function that belongs to images.

In brief, to solve the problem, Wittgenstein resorts to his famous notion of ‘use’, a solution which will be developed by the so-called ‘second’ Wittgenstein.  A state of affairs can function as a sign; a fact can signify another fact when we understand its symbolic value, that is to say, when we put it in practice.  In order to recognise the symbol in the sign, says Wittgenstein, we have to take into account its use; a use which is endowed with meaning (see, for instance, Philosophical Investigations, I, 45).  Wittgenstein therefore appeals to praxis, to habit, to response, to meaningful action; that is to say, to the ethos.  To use one of Peirce’s examples (Peirce who, before Wittgenstein, reached the same conclusion), the meaning of a word is what we are ready to do when the word is uttered in front of us.  Hence, if one tells me that a piece of nougat is very hard and I carelessly bite into it, clearly I am not recognising the symbolic value of the word ‘hard’; this value is in fact entirely defined by the habits of action and response that I am willing to put in practice.  This is the meaning of that word in its entirety.

The ethical solution shows that the problem of representation and of the image cannot find a way out, that is to say a reasonable solution from within.  In other words, the relation between language and the world, between thinking and being, as Parmenides put it, cannot be made the object of a further language or of an ‘objective’ external vision.  Language, says Wittgenstein, is the edge of the world: all that is worldly is immediately also language and vice versa, to the point that ‘I am my language’ means ‘I am my world’ (see 5.6 ff.).  I cannot transcend my language as I cannot transcend my world.  I cannot judge my language, because I have to be in a language to do so; and I cannot judge the world, because I would have to judge the world from elsewhere, from the ‘outside’.  What I can do, and I do it all the time, is to be in the world through language and to be in language as a ‘fact’, that is to say as a part of the world that there is.  And it is in this sense that the image is a fact and that it is a practical and not a theoretical fact.  In other words, the very theory itself is ultimately a praxis, a way of organising signs based on their different contexts of use.  It can be shown then, as I have previously claimed, that every theory is a myth, a narrative, an ‘account’ that teaches us how to ‘behave’ in the world.  World and language are ‘transcendentals’: they cannot be made objects of definitions or final accounts, because every definition stems from them, from the fact of their mutual symbolic coming about, that is to say in the context of the experience of significativity (of language and the world).  The world is everything that is the case says Wittgenstein; but everything that is the case is significative to us; it has a symbolic value, in that we can and we must translate it into a habit of response, that is to say into an ‘ethics’, into a way of being as having-to-be, into a manner of existing which is a constant having-something-to-do and having-to-do-with-it.

If with these few pointers I have succeeded in suggesting a way of reading the theme of the picture/image in the Tractatus, I must now turn to Freud’s essay Beyond the Pleasure Principle: a controversial paper, sometimes criticised even by psychoanalysts for going obviously and declaredly beyond a scientifically grounded discourse.  Freud himself acknowledges that: ‘What follows is speculation, often far-fetched speculation’, a speculation that, as is known, ends with a reference to Plato’s famous myth of the androgyne.  Beyond the Pleasure Principle is as brilliant as it is adventurous and erratic: I am not at all sure to what extent it can find a proper application in clinical practice.  In it, Freud advances the hypothesis of the death drive, but this is not the point I wish to consider here.  Instead, I would like to go back to the question that opens the paper, when Freud asks what is the law that regulates psychic events.  Freud, we should remember, begins from a frank self-criticism: it is not true that all psychic events obey the pleasure principle, as psychoanalysis, drawing on ancient philosophical theories, had believed thus far.  It must be recognised that the pleasure principle cannot account for and explain all psychic events, such as dreams that evoke a traumatic event and, more generally, the very mechanism of the anxiety on which psychoanalytical transference is grounded.  Hence the need to extend the hypothesis on the functioning of the mind, and the attempt to outline a new description of the psychic apparatus, a description that is grounded, as we know, in the idea of the vesicle (elsewhere Freud uses the example of the magic pad with a similar intention).

The basic idea is that the psychic is essentially a threshold, as exemplified by the vesicle: a threshold characterised by the perception-consciousness system (PC system).  This threshold is the place where the events of the world that surround the vesicle write themselves and through which they pass.  To this passage in Freud’s theoretical construction we could apply—effortlessly and in its entirety—the two propositions of Wittgenstein’s with which we started.  To begin with, ‘we picture facts to ourselves’; in other words, we are images incarnated in the world, we are our own inscribed world or one of its well-defined inscriptions.  For instance, we are connected to a vibratile and troubled oscillation of events that leave traces on us.  Hence, as carrier of these traces, every image is a fact. In this way, Freud, on his part, pursues a brave attempt to derive the psychic image not from a presupposed theory of the mind but from those facts that, in physiological terms, refer, as Freud puts it, to the cerebral cortex.  The vesicle is the cerebral cortex when it is invested with the events of the surrounding world, events that thusly translate into psychic facts and mnestic traces. The vesicle protects itself from the excessive forces of the world with a coating that marks the distinction between the exterior and the interior, for instance in the human being. The vesicle makes a shield of the coating, opened only by small loopholes through which it can reduce the forces of the world to small samples which are subsequently interconnected according to spatial and temporal criteria.  In this way, the vesicle keeps the intensity of the world at a distance and at bay, filtering and ordering it.

However, the vesicle, being the place of conscious perceptions (PC system), must also be able to retain the traces written on it by the events of the world.  Precisely like the soul in Aristotle’s De Anima, the vesicle must keep itself ‘clean’ and available for new ‘entries’.  It must be both threshold and medium.  Hence, it is flexible in allowing perceptions to pass through itself so that they can settle in the unconscious in the form of mnestic images.  The vesicle resembles a wax tablet, a magic pad, that is wiped clean every time it perceives, ready to perceive again.  It has to ‘respond’ to the world. Its function is characterised by use: what to do with the world and in front of the world that is announced through the eyes, the ears, etc.

Mnestic traces cannot stay long in the conscious, so they settle in the inner part of the shield.  It is here that a problem emerges: although the vesicle can repair and shield itself from the incident and invading forces of the world, it cannot do the same for the mnestic traces that, after traversing it as perceptions, have then settled, so to speak, behind the shield: on this side, the vesicle is naked, without shelter.  It is the effect of these traces, it is their presence making itself known from within that is responsible for all the phenomena of neurosis and psychosis.  The human being, this conscious vesicle, is a continuously tense being, not only tense towards the world, but also and especially towards those internal drives from which there is no shield or defence.  In fact, there is a defence and it amounts to dealing with these drives with the same criteria adopted for dealing with the forces of the external world; it is a matter of projecting them towards the external, to confuse them with the phenomena of the external world, to sample them and link them in a spatial-temporal continuum.  Thus, the human being imagines a world of favourable or hostile phenomena, populated by friendly or hostile creatures, on whom to unload aggression and fear through superstition, magic, hallucination, that is to say, through pathological forms.

The concept of projection thus becomes the theoretical core that propels all psychoanalysis: the possibility of a clinical approach is grounded in it.  The clinical work is essentially driven by the purpose of eventually ‘seeing the world aright’, Wittgenstein would say; in other words, to see the world without superstitions or ‘mental cramps’.  Thus, for both Wittgenstein and Freud, theory is always praxis, a practice, an ‘ethical’ response to the problems of life.  To see the world aright does not mean to construct a theory that claims to be an ‘objective’ and ‘absolute’ image of the world or to formulate an ultimate theory of the human being.  Instead, seeing the world aright concerns the acquisition of a habit that can appropriately curb aggression and anxiety.  This acquisition is a repeated exercise that can be learned by putting the technique of psychoanalytic transference to good use.  It is somewhat similar to the analysis of forms and the linguistic games that Wittgenstein practised in order to free himself from his own ghosts and ‘metaphysical’ mental cramps.

It is difficult to say to what extent Freud was aware of these ‘ethical’ outcomes. Michel Foucault, though, understood them perfectly, and it is with a quotation of his that I want to conclude.  Towards the end of possibly his most important book, The Order of Things, in the fifth paragraph of the final chapter, Foucault observes that psychoanalysis should be understood neither as a general theory of the human being nor as an anthropology. On the contrary, it is essentially a ‘practice’ that functions by facing the drama of being human, the drama of being tied to the ghosts of language and to the suffering connected to desire.  Any psychoanalytic knowledge is thus characterised by the ‘strangulation produced by the relation between two individuals, one of whom is listening to the other’s language, thus freeing his desire from the object it has lost (making him understand he has lost it), liberating him from the ever-repeated proximity of death (making him understand that one day he will die).  This is why nothing is more alien to psychoanalysis than something resembling a general theory of man or an anthropology’.  In this sense, I would say that psychoanalysis is, essentially, an ‘ethics’.

Translated from the Italian by Alvise Sforza Tarabochia


Accursio, A. (2005) Freud e Wittgenstein: ermeneutica del possibile e paradigma del mentale, in M. Mancia, ed., Wittgenstein & Freud (Torino: Bollati Boringhieri).


Fabbrichesi, R. (2000) I corpi del significato. Lingua, scrittura e conoscenza in Leibniz e Wittgenstein (Milano: Jaca Book).


Foucault, M. (2002[1966]) The Order of Things (London: Routledge).


Freud, S. (1920) Beyond the Pleasure PrincipleSE,18.


Peirce, C.S. (1984) Writings. A Chronological Edition, 2 (1867–1871) (Bloomington: Indiana University Press).


Sini, C. (1989) I segni dell’anima. Saggio sull’immagine (Roma-Bari: Laterza).


Wittgenstein, L.:

–        (2001[1921]) Tractatus Logico-philosophicus (London: Routledge & Kegan Paul).

–        (1966) Lectures and Conversations on Aesthetics, Psychology and Religious Belief (Berkeley: University of California Press).

–        (1969) Briefe an Ludwig von Ficker (Saltzburg: Müller Verlag).

–        (1980) Remarks on the Philosophy of Psychology (Oxford: Blackwell).

–        (1985) Geheime TagebücherSaber 5


Carlo Sini (Bologna 1933) lectured in Theoretical Philosophy at the University of Milan. He is a member of the Accademia dei Lincei in Rome and the Institut International de Philosophie in Paris. He has published nearly sixty books and a selection of his Works is due to be published by Jaca Book (Milan). His texts have been translated into English, French, German, Catalan, Dutch, Arabic, Turkish and Persian. His most recent works include Il gioco del silenzio (“The Game of Silence”, Milan: Mondadori, 2006; Milan: Mimesis 2013); L’uomo, la macchina, l’automa. Lavoro e conoscenza tra futuro prossimo e passato remoto (“Man, Machine, Automaton. Work and Knowledge between the near Future and the Remote Past”, Turin: Bollati Boringhieri, 2009); Del viver bene. Filosofia ed economia (“On Good Living. Philosophy and Economics”, Milan: Jaca Book, 2011); Ethics of Writing (New York: Suny Press, 2010); Il sapere dei segni. Filosofia e semiotica (“The Knowledge of Signs. Philosophy and Semiotics”, Milan: Jaca Book, 2012); Dante. Il suono dell’invisibile (“Dante. The Sound of the Invisible”, Milan: Et Al., 2013); Incontri. Vie dell’errore, vie della verità (“Encounters. Paths of Error, Paths of Truth”, Milan: Mimesis, 2013); Enzo Paci (Milan: Feltrinelli, 2015); Scrivere il silenzio. Wittgenstein e il problema del linguaggio (“Wittgenstein and the Problem of Language“, Milan: EGEA, 1994; Bologna: Castelvecchi, 2013).

Publication Date:

November 19, 2014

Share This Article


European Journal of Psychoanalysis