One of the rare adorable passages in Freud’s work lies at the beginning of Beyond the Pleasure Principle, in which he describes a toddler who likes to throw his toys under the bed while shouting “o-o-o-o!” Freud spends much of the first half of this book trying to demonstrate the bizarre thesis that humans are driven to repeat themselves simply for the sake of repeating, and small children are of course fertile ground for this thesis, with their mad demands to hear the same story exactly the same way, to eat exactly the same meal. These requirements are often excruciating because they’re incomprehensible: an adult can scarcely read the same book even twice with pleasure, but a child will require him to read it dozens of times without changing a single word. If we adults didn’t make sure to systematically misrecognize the obvious insanity of the children around us in order to render them bearable, these desires would give us some pause—what are the children hoping for in the eightieth agonizing repetition of Hop on Pop? How can they stand it? Children are the strangest of strangers.

Freud is aware of this will to word-for-word repetition in children, but it’s not the example he begins with. Instead he begins with a game. Games open up a different perspective on repetition than stories: no two games of chess are exactly alike. If we can play the same game again and again, it’s because the rules don’t change, though what happens in the domain produced by those rules is subject to infinite variation. A game exists insofar as the explicit decision is made not to suspend its rules, to pretend that they are binding. Let’s pretend this piece can only move diagonally, let’s pretend aces are high. One can play the same game repeatedly because, from within the game, nothing can happen to the rules of the game: they’re not at stake in the play. How different from, say, a conversation: in a conversation the rules governing what can be said can be, and very often are, the stakes of the conversation. A conversation can transform its conditions of possibility. In this sense, the spats of old married couples are often only games: the freedom for variation in subject and tone is guaranteed by the decision not to transform or end the relationship, not to fight in a way that could change things. In this sense, too, the psychoanalytic situation is a game, so long as the transference is managed: in Adam Phillips’ words, “Psychoanalysis is about what two people say to each other if they agree not to have sex.”

The game with which Freud begins his investigation of repetition is a very simple one, played by a boy of one and a half. The boy in question was not precocious, but Freud notes that “tributes were paid to his being a good boy”—that is, a compliant one—and in particular he didn’t cry as many children will when his mother left for a bit. The only trouble this boy gave his caretakers was a game he liked to play in which he would throw his toys across the room or under the bed while shouting ‘o-o-o-o!’ [Strachey transliteration], which made cleaning up after him a bit of a pain. His mother and Freud were in agreement that o-o-o-o was his way of saying “Fort!”, “Gone!” He appeared to enjoy this game tremendously, and indeed it constituted his main recreation at the period in which Freud was observing him: “The only use he made of his toys was to play the ‘gone’ game with them.” The only use he made of his possessions was to throw them away.

Freud found a clue to this enigmatic little practice when he caught the child one day playing a slightly more complex version of the game: he had a reel attached to a bit of thread, and he would throw it over the edge of his crib so he couldn’t see it, shouting ‘o-o-o-o!’ Then he’d yank it back into sight with a joyful “Da!” [There!] At this point the meaning of the game became clear to Freud: it was related to his “great cultural achievement,” his capacity to let go of his mother without crying. The game did not merely repeat itself, it repeated a real event: the mother’s disappearance. This interpretation was confirmed by a still more complex version of the game:

One day the child’s mother had been away for several hours and on her return was met with the words “Baby o-o-o-o!” which was at first incomprehensible. It soon turned out, however, that during this long period of solitude the child had found a method of making himself disappear. He had discovered his reflection in a full-length mirror which did not quite reach to the ground, so that by crouching down he could make his mirror-image ‘gone’.

The game, then, is “about” the mother’s leaving. But this interpretation leaves a number of questions wide open. Most prominently: if real separation is painful, why is it repeated with pleasure? One obvious answer—that it’s not separation but reconciliation which brings pleasure, that “Fort!” is just a way of staging “Da!”—contradicts the observed fact that the “Fort” game is often played without a “Da.” Merely throwing something away and then announcing its absence is a satisfying game all by itself. What transformation allows a trauma to be the subject of a game?

Freud entertains for a moment the idea that the game may be, shall we say, “therapeutic”: a way of dealing with or “mastering” the separation. What in reality is an experience of passivity and impotence becomes in play a choice: the child can make it happen as often as he likes, and in this repetition it can be studied, familiarized and comprehended. Freud rejects this interpretation because it fails to take account of what he calls the “economic motive,” the yield of pleasure required to initiate action: in this reading the game results from a free-standing educative instinct, operating teleologically toward adulthood without further need for explanation. And yet the game does yield pleasure, again and again; whatever uses it may have, it is played for its own sake.

Still, the notion of mastery is suggestive: it is indeed true that the transformation from real, traumatic separation to pretend, joyful separation corresponds to a transition from passivity to activity. More specifically: in reality, the child is abandoned, while in the game he casts away. What’s repeated is not the child’s experience of separation, the absolutely passive experience of a child who cries and no one comes, but the act of separation itself—the mother’s act. Clearly a reversal of roles is required. But what reversal? Here we must be cautious.

In the “Fort” game, is the toy that gets thrown away a child? In this case, the boy playing the game is a mother, in the way that a little girl playing with dolls may play the mother: he’s acting out the central relationship in his life from the other side. The yield of pleasure in this case would arise from identification with a position of power: how glorious it must be to be the mother, the adult who can leave whenever she wants and return just as capriciously! In this case the game is a precise repetition of reality, but from the side of the winner—a side the child may hope to occupy someday.

Alternatively, the toy which gets thrown may itself be the mother. The child may play in order to “revenge himself on his mother for going away from him... ‘All right then, go away! I don’t need you. I’m sending you away myself.’” In this case the pleasure is not identification but vengeance—the fantasy is of making the mother know what it means to be abandoned. The yield of pleasure is sadistic, and the motive is vengeance.

So which is it? The psychoanalytic response to this kind of dilemma is of course to say: “Both!” That response finds its confirmation in the last and strangest version of Fort-Da, the mirror game. Here the child is both the toy thrown away and the one doing the throwing: not the mother who makes the child disappear, or the child who makes the mother disappear, but the child in the act of disappearing. The mirror game combines the pleasure of identification with power (“I too can make this baby disappear”) and that of vengeance (“Leave me, will you? I’ll leave you!”) to form a threat well known to children everywhere: “You’ll be sorry you were mean to me when I die!”

This strategem is easy to patronize, but it corresponds to the difficulties of the child’s position perfectly. The child is the adored focus of the parents’ affection, their lives revolve around him and his needs, and yet this position gives him no power at all: he is entirely dependent, to a degree that’s hard for adults to imagine and harder yet to recall. This dependence, combined with the unreliability of his caretakers, is what gives childhood its terror. Why not be afraid of the dark, when if no one sees you you’ll starve to death? The parents’ love can itself become a source of anxiety in such circumstances; witness all those fairy tales in which children get eaten—an answer of sorts to a question every child must ask: what do the parents want, after all, with this object that I am?

What can the child trade for power? He has nothing but himself. He is already the stakes in his parents’ game; how can he become a player? By developing the capacity to take himself away. Leave me one more time, mom, and “baby’s gone!” Mirrors aren’t the only way to play this game, of course, or even the most common: most children come to power first of all as “picky eaters.” The child needs to eat in order not to starve, and depends on her parents for food; but the parents need the child not to starve, and they can’t make the child eat. It’s a game of chicken which the child will occasionally win, a tiny source of power. Dearly bought, to be sure, at the cost of a terrific amount of self-control: to starve the parents of her affection, the child must starve herself. To avoid being abandoned, the Fort-Da boy must learn to throw himself away. Perhaps, indeed, this is where we should look for the beginning of self-control: the child can frustrate the desires of his parents only by frustrating his own.

Initially, of course, this game of chicken really is only a game: the child will eat eventually. I remember how frustrated I was, as a small child, at the futility of my angry resolutions never to talk to my parents again in protest at their injustice: I could barely hold out for half an hour. Humiliating as it was to come crawling back, they still held most of the cards. Even if they needed me as much as I needed them, I needed them a lot sooner.

But the pretend quality of the child’s disappearance isn’t simply the result of a lack of self-control. The classic elaboration of the disappearance-strategem, Tom Sawyer eavesdropping on his own funeral, is at once perfect wish-fulfillment and a very good joke. Tom gets to watch his family mourn him and say what a lovely boy he was, how they were never kind enough. He gets to be present at the site of his absence. But alas! He’s the only one. The trouble with disappearance for the rest of us is that once you’ve done it, you’re not there. The narcissistic gratification to be had from making one’s absence felt can be had only in play. The child would love to see how much his parents missed him after he ran away; he would love this precisely because he loves them and needs them too much to run away. Fort-Da boy disappears from the mirror, but not from the room: the world he likes to glimpse through the looking-glass, the world where he’s really gone, is a place he can never get to, and even if he could he wouldn’t dare.

The specter of real death lingers around the edges of the child’s world: Freud tells us that when Fort-Da boy’s father went off to war he would throw his toys on the ground, shouting “Go to the front!” He adds in a footnote that the boy’s mother had another child soon, disrupting their relationship, and that she died when he was only five. The developmental psychologists tell us that it takes a while for babies to develop “object permanence,” the capacity to recognize that objects persist even when they can’t be perceived; a very young baby, watching a ball go behind a screen, will believe that it has blinked out of existence. When it reappears on the other side she’ll think it was created anew. The Fort-Da boy’s experiments with the transition from visibility to invisibility and back again are in part investigations into permanence: what happens to the object, what becomes of the world, when I’m not there? Disappearance is not identical to destruction, but neither is the difference absolute: to risk invisibility is to play games at the edge of oblivion.

Psychoanalytic narratives of childhood have an unfortunate tendency to end in independence, as though the child starts out needing the mother for everything and ends up sufficient unto himself. It’s a comforting story to tell: if we can’t idealize away the humiliations of childhood, as most people do, at least we can leave them behind. But as impressive as our accomplishments in feeding and bathing ourselves really are, adult independence has its limits. We can’t stop falling, more often than we might like to admit, haplessly and abjectly in love. And so we find ourselves once again desperately in need of someone not much more reliable than our mothers, and needing retains all its terror. The here-gone game isn’t just for children: in all its many meanings, it’s a game lovers play with each other endlessly. Lovers leave each other constantly, sometimes many times in the course of an afternoon, and for many reasons: to take vengeance for this need, to discover how deeply they’re needed, to remind themselves of what freedom was like, to discover how deeply they love. Love in this sense is the strangest kind of war: I can deny you only by starving myself, but that doesn’t mean I won’t do it! I can still do it; watch me go away. These are dangerous games to play, and yet they’re necessary, even beautiful. Without a strategy of separation, the pleasure of togetherness would be unbearable; it’s scary to need someone so much. And then, the wonder of the lover’s “There!” borrows something from her ever-approaching “Gone!”

---