The Non-representational Character of Unconscious Phantasy in Kleinian Psychoanalysis

Phantasy comes from light, φῶς, its etymological root. Those things that appear in light, φαίνεσθαι, that, in other words, are projected in and by it, figured and disfigured by it, split in two ways. First, there are those things that are comprehensible, legible, and ordered in a consciousness (those things that appear according to a logos), which end up in the méconnaissances of phenomenology. What does not submit to discourse and understanding, however, what precedes, lacks, or exceeds its logic, merely appears as phantasy.

This doesn’t clarify much, except perhaps a certain obscurity, which is why phantasy is such a difficult concept—that is, if one can grasp it in a concept at all. Nonetheless, for psychoanalysis, phantasy is as important as it is hard to define. The idea of phantasy was in fact so contentious that, in 1940’s London, an entirely psychoanalytic school of thought had to be created in part just to account for its consequences.[1] And although it is no longer considered to be a particularly controversial issue, there is still no consensus on the exact role of phantasy in psychoanalytic theory. Thus here, far from saying anything definitive, I will merely attempt to trace one particular lineage of the ‘concept’ of phantasy through its development in (Sigmund) Freud and its interpretation and modification (becoming “unconscious phantasy”) by Melanie Klein and Susan Isaacs.

For Freud, the concept of phantasy is linked first to wish fulfillment. Elizabeth Bott Spillius clarifies that Freud “speaks of phantasy as a wish-fulfilling activity that can arise when an instinctual wish is frustrated” (Spillius 361).[2] As such, she continues, “[p]hantasies derive ultimately from unconscious impulses, the basic instincts of sex and aggression” (ibid.). Thus, on Freud’s view, phantasies arise from the transformation of drives. However, in Freud, this origin (of phantasy) in somatic processes is displaced, so that phantasy only comes on the scene after an unconscious wish is filtered through secondary, rationalizing processes (Spillius 362). In terms of his topological model of consciousness, in order for phantasies to occur, wishes in the system Ucs must be distorted, which then, similarly to dreams, appear to the subject as something other than what they are unconsciously. According to Spillius, these transformed wishes find expression as phantasy in the system Pcs and system Cs (ibid.). In the former, phantasies remain ostensibly unconscious, but are “formed according to the everyday logic of the secondary process” (ibid.). In the latter, phantasies appear to consciousness as daydreams. Occasionally, according to this schema, some phantasies might never get out of the system Ucs, and as a result “may become indistinguishable from memories and may also find their way into dreams, symptoms, symptomatic acts, further preconscious and conscious phantasies, and other drive derivatives” (ibid.). In any case, what one finds in this Freudian conception of phantasy is a model of representation, or at least of translation, wherein phantasy acts as a distorted Saussurean signifier for an unconscious wish seeking fulfillment vicariously. The Kleinian phantasy, by contrast, seems to be less prone to mediation.

There is, however, another set of Freudian phantasies, which he describes in 1916 in a lecture entitled “The paths to the formation of symptoms”. These are the so-called “primal phantasies”: namely, the primal scene (i.e. the witnessing of one’s parents copulating), seduction by an adult, and the threat of castration. Freud considers these three phantasies to be phylogenetically inherited. And while the idea that phantasy is phylogenetically transmitted is more or less dated, today rejected for being too Lamarckian (Spillius 363), what the notion of the primal phantasies nonetheless importantly emphasizes is the possibility that phantasies might in fact play a primary role in relation to the unconscious, and that they might be immediately felt rather than purely representative and derivative. It is in this regard that, also in his 1916 lecture, Freud makes the speculative discovery of what he calls “psychical reality”:

The [primal] phantasies possess psychical reality as contrasted with material reality, and we gradually learn to understand that in the world of the neuroses it is psychical reality which is the decisive kind (Freud, quoted in The New Dictionary of Kleinian Thought 6-7).

With the notion of psychical reality, Freud opens up a space for a phantasy that no longer operates at secondary, representational levels. Instead, he emphasizes that the experience of the phantasy itself possesses an efficacy that supervenes on what is considered to be physical reality. In other words, since psychical experiences can have real effects, Freud treats these experiences, however ‘delusional’ or ‘hallucinatory’, as real. Yet, despite this discovery, as the authors of The New Dictionary of Kleinian Thought (NDKT) contend, Freud nonetheless throughout his career maintained that phantasy is ultimately “wish-fulfilling and unrealistic” (NDKT 7). This implicit dismissal of phantasy as unrealistic marks perhaps the most crucial distinction between what would become the ‘Freudian’ and Kleinian Schools.

The novelty of the Kleinian theory of phantasy lies in its recognition of the radicality of Freud’s concept of psychical reality. In the Kleinian view—most famously outlined by Susan Isaacs in her 1948 paper “The Nature and Function of Phantasy”—phantasy becomes essential as soon as psychical reality is given its due. Isaacs writes “when external reality is called ‘objective’ reality, there is an assumption which denies to psychical reality its own objectivity as a mental fact. Some analysts tend to contrast ‘phantasy’ with ‘reality’ in such a way as to undervalue the dynamic importance of phantasy” (269). For Klein and Isaacs, in fact, phantasy is so central that they consider its psychical reality to be a primary feature of unconscious life, rather than a distortive representation of it. In other words, rather than characterize it as a more or less rational translation of unconscious wishes, Klein and Isaacs understand phantasy as an experience sui generis that originates in the unconscious without any need of wish fulfillment. Instead of translating the unconscious into phantasy, in Kleinian thought, phantasy is unconscious. Therefore, unconscious phantasy becomes the primary motivator for conscious action, as well as dreams and symptom formation: “the primary content of all mental processes are unconscious phantasies. Such phantasies are the basis of all unconscious and conscious thought processes” (Isaacs 272). If this is the case, then it constitutes an extreme deviation from the traditional Freudian notion of phantasy.[3] In other words, if phantasy is no longer a secondary representation of the processes in the unconscious, but in fact takes place in or as these processes, immediately, then this fundamentally changes its structure. Phantasy would no longer signify; it would no longer be a signifier of an unconscious signified, it would instead be non-representational.

Klein and Isaacs turned to infantile life as the paradigm for understanding unconscious phantasy in this way. In their view, the infantile experience of phantasy serves as a prototype and prefiguration of the (phantasies of the) adult unconscious.[4] This seems to be in part because the principal characteristic of infantile life is at first purely affective:

Now in the infant, experience and mental process must be primarily, perhaps at first entirely, affective and sensorial. Then full perceptual experience gradually develops as relationships of time and space and external reference become possible, owing to the development of the sensory organs and of movement. Much later, conceptual experience dawns, with the rise of memory images as material of experience, and as relationships other than those given in the here and now of what is perceived can be focused in experience (Isaacs 274).

So inasmuch as the infant has (unconscious) phantasies,[5] it experiences them first as affective or as sensorial—which is to say that the first instantiations of psychical reality are intimately linked with somatic processes. Such a claim can be further justified by turning to the most concise definition that Isaacs gives of phantasy, which, interestingly enough, comes from her interpretation of a passage by Freud (from a lecture on “The Dissection of the Psychical Personality”), in which he discusses the unconscious ‘id’ as “a cauldron full of seething excitations”, as “being open at its end to somatic influences, and… there taking up into itself instinctual needs which find their psychical expression in it” (Freud, SE 22: 73). Isaacs writes: “I believe that this ‘mental [psychical] expression’ is unconscious phantasy. Phantasy is the mental corollary, the psychic representative of instinct. And there is no impulse, no instinctual urge, which is not experienced as (unconscious) phantasy” (Isaacs 277). This leaves no doubt of the fundamental relation between somatic drives/instincts and the experience of phantasy in the Kleinian model. This raises the question, however, about exactly what Isaacs might mean by a “mental corollary” or a “psychic representative” of the drives/instincts. Inasmuch as I am arguing for the non-representational nature of phantasy in Klein, I would contend that there is a qualitative distinction to be made between a “corollary” or “representative” and an order of representation where phantasy as a collection of signs would designate something ‘deeper’ below it, waiting to be better understood. Rather, phantasy is what is to be understood. To make a banal analogy, a representative in a republican government is ostensibly supposed to work in the interests of her or his constituents; but once the representative is elected, the link between their actions and whom they represent is severed. There is only the representative, the simulacrum of representation, who, as experience shows all too often, can equally fail to represent anyone. Phantasy would thus be a non-representational representative of somatic processes: its corollary, not its likeness. Phantasy and somatic processes would be the obverse and reverse of each other: two sides of the same coin, in an incomprehensible economy.

And yet, Isaacs insists that the psychic life of phantasy possesses meaning: “[t]he word ‘phantasy’ serves to remind us always of this distinctive character of meaning in the mental life” (Isaacs 272). Certainly there is meaning in phantasy; as Isaacs observes, meaning would be what distinguishes phantasy from being merely “a neutral term such as ‘process’” (ibid.). But this meaning is not “a” meaning. Rather than conforming to linguistic, narrativistic, or even metaphorical frameworks, the unmediated meaning of Kleinian unconscious phantasy pushes the limits of what meaning itself, understood within the purview of traditional hermeneutics, can even mean. Meaning here is an affective obverse of instinct. Keeping in mind the distinction made at the very outset of this short paper, phantasy operates as the psychical mere appearance—not representation, which would imply a secondary (phenomeno)logical organization—of physical processes: the psychically real aspect of a somatically induced ‘meaning’. In other words, the meaning of phantasy is felt, or sensed; it is not understood. Thus phantasy is not the solely ideal interpretation of a physical excitation; rather it is already both psychical and somatic, each coenesthetically imbricated in one another from the start—psychosomatic.

What follows are some illustrative examples. Isaacs writes of an infant’s introjective phantasy about its mother’s breast:

It is not correct to put it that when the infant feels he is incorporating his mother’s breast or some other good thing, he is dramatizing or picturing what is ‘really’ an abstract process or introjecting. At the first level of experience, introjection is felt to be incorporation…[Only] At a somewhat later level, it is imagined to be, incorporation. (Isaacs 275).

The mechanism of introjection is a function of phantasy through which the infant’s sensations appear to it as concretely felt ‘internal objects’. In other words, the infant phantasises that its interroceptive pleasures and pains are caused by real objects lodged in its abdomen. Klein explores internal objects and the notion of introjection throughout her career, often as privileged instances of early phantasy-life.[6] But to return to another more general illustration of phantasy by Isaacs:

Phantasy expresses the specific content of the urge (or the feeling, e.g., hate, anxiety, fear, love, or sorrow) which is dominating to the child’s mind at the moment, e.g., when he feels desires towards his mother, he experiences these as ‘I want to suck the nipple, to stroke her face, to eat her up, to keep her inside me, to bite the breast, to tear her to bits, to drown and burn her, to throw her out of me’ and so on and so forth (Isaacs 277).[7]

These hypotheticals fairly straightforwardly show the lack of mediation between the sensation, the somatic experience of a good feeling, or of hate, fear, etc., and its phantasmatic correlate. The psychical reality of desire toward the mother is not, for Kleinians, a coded message, but a jumble of potential actions that express the infant’s reaction to a bodily enervation. As Isaacs remarks, “[l]ater on…phantasy resides in plastic images—visual, auditory, kinaesthetic, touch, taste, smell, etc. And, since the child understands many words at the end of his first year, his phantasy life by that time certainly makes use of verbal images as well” (284). But only after phantasy is already directing the psychic life of the infant do images intervene to mediate it and specify it, to begin to dissociate the psychical from the somatic, where at first the two are indissociable.

Accordingly, by the time the infant has weathered its first year, its phantasies become so complexly structured by the formation of images and symbols (what Klein called the epistemophilic impulse), that its phantasies are even (unconsciously) projected into its play. Whence Klein’s emphasis on analyzing the play of children. Isaacs offers the following example, among others:

[C]onsider the following play in a girl of 16 months, one who talks very little. She has a favourite game, which she plays with father and mother. She picks small imaginary bits off a brown embossed leather screen in the dining-room, carries those bits carefully across the room in her finger and thumb, and puts them into the mouth of father and mother alternatively. Could the child tell us any more plainly in words that she wants to feed her farther and mother as they had fed her? But we can link two further facts. Of all possible objects in the room, various in colour and shape, she chooses this one, the brown screen with small raised lumps on it—to feed her parents with. Is it not likely that this is connected with the fact that the child was found several times during the previous few months (between 12 and 18 months) to have defecated in her cot, smeared herself and put the faeces into her mouth? Is she not, in giving imaginary bits of the brown screen to her parents, trying to overcome the anxieties which her smearing and eating of faeces—and the reproaches of her parents—had aroused in her? Trying to prove that faeces were good and that her parents could eat them too? (Isaacs 312).

At this the point in the child’s development, phantasy has reached such a level that it is seemingly no longer the corollary of direct bodily experience. But even here, the phantasy underlying this young girl’s game is far from a representation. Instead, it is the game that represents the phantasy—a phantasy that presumably compensates for an affective disequilibrium caused by the reproaches she received from her parents for eating shit. So the phantasy is, for all its elaboration, still psychosomatic—still fundamentally entwined with somatic impulses, which for the infant/unconscious are unnamable and unknowable as anything in particular, and that thus merely appear.

Only much later can phantasy be made sense of. But at that point it is no longer phantasy.





Freud, Sigmund. (1900). The Standard Edition of the Complete Psychological Works. Volume 22. James Strachey (Trans.).


Isaacs, Susan. (1992). “The Nature and Function of Phantasy”, in The Freud-Klein Controversies, 1941-5. Pearl King and Riccardo Steiner (Eds.).


Klein, Melanie. (1935). “A Contribution to the Psychogenesis of Manic Depressive States”.


Spillius, Elizabeth Bott, et al. (2011). The New Dictionary of Kleinian Thought. New York, NY: Routledge Press.


Spillius, Elizabeth Bott (2001). “Freud and Klein on the Concept of Phantasy”. The International Journal of Psychoanalysis (82). Pp. 361-373.




[1] The debates over phantasy at this time, especially between Anna Freud and Melanie Klein, are recorded in The Freud-Klein Controversies, 1941-5 (Eds. Pearl King and Riccardo Steiner). The necessity for a Kleinian school of psychoanalysis emerged from these debates, or rather from their lack of resolution.

[2] It is important to note, however, that Freud rarely explicitly addresses phantasy, and that therefore his take on phantasy is open to interpretation.

[3] However, it should be noted that the above characterization of Freud’s views on phantasy is superficial. In fact, many of Klein’s and Isaacs’s own arguments are drawn from their nuanced interpretations of Freud’s lectures and texts.

[4] Isaacs describes “the particular level of experience with which we are concerned—the infant’s early life, or the unconscious mind of the adult” (275).

[5] At this point in the infant’s development, to call phantasies unconscious would be pleonastic.

[6] Condisder Klein’s 1935 essay “A Contribution to the Psychogenesis of Manic-depressive States”, in which she explores introjection and projection as they relate to phantasy and to her notion of the paranoid-schizoid position and the depressive position.

[7] With regard to this example, Isaacs is compelled to give the caveat: “Needless to say, I am not suggesting that these phantasies are experienced in words” (Isaacs 277).