What Shows Through
Lacan’s three registers are laid down one over another. You learn what each one was doing by watching it fail.
A man with schizophrenia once said that the films he cannot watch are the ones that break the fourth wall. When a character turns from the scene and looks into the lens—addresses you, the watcher, directly—something in him comes loose. He put it more exactly than a theorist could: in schizophrenia, there is no fourth wall.
The fourth wall is a convention, which is to say it is a piece of shared agreement so reliable that we forget it is there. It is the silent clause that says: this is a representation, it is framed, it is not happening to you, and the people inside it do not know you are watching. You do not decide to believe this before each film; you do not believe it at all, in the sense of holding it as a proposition. You simply live inside it, the way you live inside grammar. It is a layer. And like most of the layers we are made of, you only find out what it was doing once it stops doing it. When the frame fails—when the screen turns and speaks to you—the failure does not feel like a clever flourish. It feels like the floor of a category giving way.
Selfhood is built in layers like this one, each laid down over the one before, each covering it. Lacan named three, and the most useful thing about his scheme is not the names but the order, and the fact that the order can break. You read each layer off its failures. You learn what the cover was for by seeing what comes through when it tears.
The first layer, the lowest, Lacan called the Real, and the word is a trap, because it does not mean reality. Reality—the world of tables and weather and other people—is already organized, already cut into objects and meanings, and that organizing is the work of the layers above. The Real is what is left when you subtract all of that. The formless, the undifferentiated, the not-yet-cut. For the infant at the start it is a continuum with no edges: no inside and no outside, no me and no not-me, no objects because there is nothing yet to hold an object apart from its background. The Real is not a place you can go. It is what the other layers exist to cover, and it makes itself known mainly as the thing that resists them—the snag, the leftover, what will not go into words.
Over the Real comes the Imaginary, and here Lacan has one of the great images in the literature: the mirror stage. Somewhere between six and eighteen months, before it can reliably control its own limbs, the infant catches its reflection—or the gestalt of another child, or its own form in the mother’s arms—and recognizes itself in it. The recognition is jubilant and it is a lie. The body the infant actually lives is uncoordinated, in pieces, a flux of sensation that does not yet add up to a thing. The body in the mirror is whole, bounded, unified, out there and ahead of what the child can do. The infant takes that image as itself, and in doing so it acquires, for the first time, an inside and an outside, a surface, an edge—a me set off against a not-me. The Imaginary is the register of the image and of likeness, of the bounded body and the ego built on it, of resemblance and rivalry and the love between like and like. It gives the formless a shape by lending it a likeness. It is also, from the start, a kind of estrangement, because the self you have is a picture you will never quite climb inside.
Over the Imaginary comes the Symbolic: language, and everything language drags in with it. Where the image works by resemblance—this looks like that—the signifier works by difference: a word means nothing in itself and means only by not being the other words. To enter the Symbolic is to be threaded into a system that was running long before you arrived and will run long after, the order of names and laws and kinship and the unconscious that Lacan said is structured like a language. Lacan put the entry under the sign of the father—the Name-of-the-Father, the “no” that breaks the closed circuit between infant and mother and installs the law, the third term that turns a dyad into a structure. And the Symbolic exacts a price for what it gives. The word, as Hegel’s readers liked to say, is the murder of the thing: to name something is to trade the thing itself for a counter that can stand in its absence, which is why entering language is also a loss, a first castration, the opening of the gap that we will spend the rest of our lives calling desire.
A caution, because it matters and because the scheme invites the error. Lacan did not finally believe these were three floors of a building, poured in order, the Real first and the Symbolic last. He resists pure chronology, and he ends not with a stack but with a knot—the three registers Borromean, looped so that no one of them is simply beneath the others and cutting any one frees all three. The layering is a way of reading, a heuristic that earns its keep, not the last word. But it earns its keep especially well at one task, which is making sense of what happens when a register fails to cover the one below.
Here is the grid, and here is where the reading turns from exposition into something with edges.
Each register is supposed to overwrite the one beneath it—the Imaginary to cover the Real, the Symbolic to cover the Imaginary—and you can sort the great structural disturbances of the mind by which overwriting failed.
Psychosis is the failure of the Symbolic to overwrite the Imaginary. This is Lacan’s own, not a gloss: he called the mechanism foreclosure. The paternal signifier, the one that would have buttoned the whole symbolic order into place, is not merely repressed but never inscribed at all—and what is foreclosed from the Symbolic, he said, returns in the Real. The marks of it are exactly what you would predict from a missing cover. Meaning will not stay fixed, because the anchoring points that tack words to their senses are loose, and so significance slides and floods. The image is no longer held at the distance the Symbolic keeps it at, and so it can address the subject directly—which is the man at the start of this essay, for whom the screen that turns and speaks is not a trick but a breach. And in place of the neurotic’s endless doubt there is certainty, delusional certainty, the dreadful conviction that admits no gap, because doubt requires the symbolic distance between a subject and its world, and that is the distance that has collapsed. (The capacity to dwell in not-knowing, the negative capability I have written about elsewhere as the cognitive face of maturity, is precisely what the foreclosed structure cannot reach. There is no gap for it to inhabit.)
Now take the same logic one register down, and you arrive at a claim that is mine and not Lacan’s, and I want to mark it clearly as a borrowed engine running on new fuel. If psychosis is the Symbolic failing to overwrite the Imaginary, then autism might be the Imaginary failing to overwrite the Real. The image fails to take. What you would expect from such a failure is a relation to surfaces that cannot be imbued with likeness—surfaces that stay literally themselves, that will not become signs of anything, that resist the as-if. And you would expect trouble at exactly the boundary the mirror stage installs: the inside and the outside, the me and the not-me, that the Imaginary is supposed to set and that here does not reliably set.
Lacan did not write this, but a whole body of clinical work circles the same phenomena from another school, and it is worth borrowing because it dignifies the intuition rather than decorating it. Esther Bick wrote about the skin as the first thing that holds a self together, and about “second-skin” defenses thrown up against the terror of spilling out or falling to pieces when that holding fails. Frances Tustin wrote about autistic surfaces and shapes—sensation organized by hardness and edge and rhythm rather than by likeness and meaning, a world of contiguity, of things touching, before it is a world of things resembling. And Thomas Ogden gathered this into what he called the autistic-contiguous position, the most primitive of the ways we make experience mean anything: a mode of pure sensory surface, of skin and edge and rhythm, beneath and prior to the others. None of them is talking about Lacan’s registers. All of them are mapping the same seam—the place where raw sensation either does or does not get taken up into a bounded, meaningful surface.
And this is the moment to say the thing that keeps the whole reading honest. It is a structural claim, not a causal one. To say that in autism the Imaginary fails to overwrite the Real is to describe a position, a logical relation among the registers—not to explain how anyone got there, and emphatically not to lay it at anyone’s feet. The history of psychoanalysis contains a catastrophe on exactly this point: Bettelheim took a structural description of autistic withdrawal and ran it backward into a cause, the “refrigerator mother,” and the wreckage that did to real families is one of the field’s lasting shames. The error was not the structure. The error was causalizing the structure. What protects this reading from repeating it is the small phrase that Lacan used about the mirror stage and that I want to put to work here: a logical moment. The mirror stage is not an event in a particular nursery on a particular afternoon; it is a logical juncture in the constitution of any subject. So too here. The autistic moment and the psychotic moment are logical positions in the architecture of subjectivity—junctures every one of us passes through and at which, for any of us, the overwriting can come up short. They are not the afflictions of an unlucky few held at arm’s length. They are constitutive risks of becoming a someone at all. We are all, at the foundations, built over a Real that never fully took the cover, and an Imaginary the Symbolic never entirely subdued.
If pathology lives where the overwriting fails, then there ought to be a place where one register first gets stitched to the next—and there is a candidate, though pinning it requires stealing a tool from one school and using it on another’s patient.
Lacan had a name for the stitch: the point de capiton, the quilting point, the upholstery button where the loose stuffing of the signifying chain gets tacked down so that meaning, for a moment, holds still. He meant it for the seam between signifier and signified, up in the Symbolic. But the seam I want is lower—the one between the Imaginary and the Real, between the first bounded image and the formlessness it covers—and the object that sits on that seam is not Lacan’s at all. It is Winnicott’s transitional object: the blanket, the worn rabbit, the first thing that is neither simply me nor simply not-me, that the child both finds and makes and is never asked to say which. Winnicott placed it in an intermediate area, a potential space, between the subjective and the objective. But read it through the registers and it is doing something more specific. It is the quilting point between the Real and the Imaginary—the first button that tacks the formless to a form, the place where not-me becomes thinkable. And if that is right, it reorders the developmental story in a way I find hard to shake: the transitional object would have to come before the mirror stage, because the mirror stage needs a not-me to set the self against, and the transitional object is where the not-me is first made. The bridge has to be built before anyone can cross it.
I should be honest that this is a graft Lacan would not have made and orthodox Winnicottians might not accept; the two schools keep their objects in separate rooms. But it is the graft that makes the seam visible, and it is also the precise place where this essay shakes hands with the one on Klein. Ogden’s three positions—the autistic-contiguous, then the paranoid-schizoid, then the depressive—are stacked along the same axis I have been drawing. The autistic-contiguous is this essay’s territory, the stratum of surface and skin and the seam between sensation and image. The paranoid-schizoid and depressive are the other essay’s, the oscillation between splitting the world and holding it whole. Two frameworks, one architecture: what Klein and Ogden read as positions one moves among, Lacan reads as registers laid over one another. The engine has two halves, and they mesh at the floor.
There is a further yield, and it cashes out a one-line note I have carried for years: that schizophrenia, autism, borderline personality, narcissism, obsessional neurosis each imply a particular epistemological frame. Once you see the structural positions, this stops being cryptic. A position in the registers is also a position toward knowledge—a characteristic way of holding, or failing to hold, the relation between a knower and what there is to know.
The psychotic structure knows by certainty. With the symbolic gap collapsed there is no distance between the subject and its truth, and so no doubt, only the flooding conviction of the delusion, which is not a belief one could argue with but a thing one inhabits. The neurotic structures know by the question. The obsessional is the master of doubt, circling a question he keeps from ever closing—am I alive or dead, do I have the right to exist; the hysteric lives a different question, the one about desire and the body, a horizon that recedes the instant the lack gets named. (That recession is itself a clue to certain refusals of the Other, certain ways of holding a position with the body against the word that would pin it—but that is another essay.) The autistic relation to knowledge is the relation of literal surface: the sign that is only itself, knowledge as the thing rather than the metaphor, a world declining the as-if. The borderline knows by a knowledge that will not stay still, the loved object and the hated one refusing to be the same object—which is the imaginary running unbuttoned, and which Klein, from the other side of the architecture, called splitting; the same operation seen through two lenses. And the narcissistic relation knows through the image of the self, every truth routed through the specular ego and what flatters or threatens it. None of these is a list of symptoms. Each is a stance toward the real, a way the registers arrange the knower. Diagnosis, read this way, stops being a catalogue of defects and becomes a typology of how a subject can be put together—and therefore of how any of us is.
Which returns us to the cinema, because art is the place where we play with the seams on purpose.
To watch film violence and enjoy it, you have to be a split subject—and this is a precise thing, not a loose one. You have to be the one safe in the seat, held at the symbolic distance the frame provides, and at the same time the one the violence reaches, the body that flinches and is implicated. The pleasure lives in being both at once, in the gap held open between them. Art works by loosening the overwriting deliberately, by letting a lower register show through under controlled conditions: horror lets the Real seep up around the edges of the frame; the uncanny is the Imaginary slipping a notch out from under the Symbolic, the familiar going subtly wrong; the fourth-wall break reaches through the frame to touch you and then—this is the whole trick—lets the frame close again, so that the breach is a thrill rather than a wound. The frame has to hold for the violation of it to be bearable. The split has to be available for the play to be play.
This is why the same device that one viewer experiences as a flourish another experiences as a trigger. I have known someone who could not take film violence as a split subject—for whom the doubling would not happen, so that what was on the screen was either fully real and unendurable or else flatly nothing, with no safe gap between. Where the seam can be played with, the breach is art. Where it cannot, the breach is just a breach, the floor of a category giving way. The difference is not in the film. It is in whether the layers hold well enough to be teased.
If the registers are real strata, then in principle you could probe them one at a time, and here the framework leans toward the work I actually do. The Rorschach is, in effect, an instrument for the Imaginary: it hands you a formless blot and reads what image you lend it, what likeness you project onto the shapeless—which is the Imaginary’s whole operation run on demand. The note I keep meaning to develop is that there should be a corresponding instrument for the Symbolic: automatic writing, or streams of random text, as the signifier’s Rorschach—what is this song about—where what you project is not an image but a meaning, and where, with enough iterations across enough algorithms, you could build a corpus that reliably draws the projection out and then peg it to other measures. The image-generators I have written about elsewhere belong to the same family of probes, latent space squeezed until the structure of perception bleeds out of it. The grid would not only interpret. It would, in the right hands, measure.
But that is the forward edge, and I will leave it at the edge. The point of the grid is quieter and older than any instrument. We are made of layers we did not lay down, each covering the one beneath, and we go through our days never seeing them—until a film turns and speaks, or a surface refuses its likeness, or the world floods up through a cover that did not quite take, and for a moment the construction shows. The diagnoses are names for those moments, and the moments are in everyone. What the schizophrenic sees when the fourth wall falls is not a private derangement. It is the architecture of all of us, briefly visible through a seam. The wall was always a convention. The astonishing thing was never that it sometimes fails. The astonishing thing is that, for most of us, most of the time, it holds.