A Conference Incident, Paris, October 1975
The symposium was called Structures of Knowledge After Structure, which Derrida thought was a fine title and Baudrillard thought was redundant, and Lyotard thought both of them were missing the point, which was that there was no point, which was itself a kind of point, and so on.
It was held in a conference room on the fourth floor of the Ăcole Normale SupĂ©rieure â a room that smelled of old carpet and ambition â and it was going more or less normally (three audience members asleep, one furiously annotating, one who had wandered in from a tax conference and was too embarrassed to leave) until the refrigerator incident.
The refrigerator in question stood in the adjoining kitchenette, a squat white Fagor with a dented handle. It had been placed there for the philosophers' coffee breaks. It contained: two wheels of brie, a bottle of Sancerre, someone's yoghurt labelled NE PAS TOUCHER, and a small internal light.
The light was the problem.
It was raised not by any of the thinkers on the program but by the facilities manager, a compact Breton named Gérard, who had come to fix the coffee urn and who made the mistake of mentioning, in passing, while threading a new gasket, that he had always wondered: when you close the refrigerator door, does the light go off?
He meant it rhetorically. He was thinking about something else entirely. He did not expect what happened next.
Foucault, who had been leaning against the doorframe smoking, straightened immediately.
"The light," he said, "exists only insofar as it is observed by a subject produced by the very discourse that grants legitimacy to the category 'refrigerator.' To ask whether it persists in the dark is to ask whether power persists when unwitnessed â which is precisely the wrong question, because power is the act of witnessing."
"That is beautiful," said Lyotard, refilling his cup, "and completely untestable."
"Beauty is not obligated to be testable."
"I didn't say it was. I said it was untestable. Those are different statements."
Derrida, seated at the small table with his coffee, had been quiet. He now looked up.
"The door," he said carefully, "is not simply a door. It is a threshold â the arche of inside and outside â and the light on one side of that threshold cannot be simply present or absent, because presence itself is always diffĂ©rance, always deferred." He paused. "I think the light is â in some meaningful sense â both."
Baudrillard entered from the hallway where he had been on the telephone. He had caught the last sentence.
"The light is neither," he said, pouring himself wine despite the hour. "What we call the light is already a simulation of light. The original has been lost. We live in the age of the refrigerator's sign, not its content. There may not even be brie in there. There is only the idea of brie, which we consume in place of the brie, which is â " he opened his hands, " â fine. Better, even."
Gérard looked at the refrigerator. He looked at Baudrillard. He opened the refrigerator door. There was brie.
"There is brie," said Gérard.
"Naturally," said Baudrillard, without missing a beat. "But notice how its presence confirms your expectation rather than surprises it. You have consumed the hyperreal brie already. The physical brie is merely its alibi."
Deleuze arrived late, as always, with Guattari. They had been walking because Guattari had wanted to stop and look at some pigeons. Now they settled onto the windowsill together in a way that made it hard to tell where one ended and the other began.
"The light," said Deleuze, without being caught up, as if he'd heard everything, "is a line of flight. The refrigerator is a territory. The door is a fold. When it closes, the light does not go off â it goes underground. It deterritorialises."
"It becomes rhizomatic," Guattari agreed, eating someone's yoghurt.
"Yes. The light escapes the striated space of the visible and occupies the smooth space of the potential."
"Is that different from being off?" asked Gérard.
Deleuze and Guattari looked at each other.
"Yes," they said, in unison. "Qualitatively."
The conversation had now attracted the rest of the conference attendees, including the tax man, who had followed because everyone else was moving. He was a small, sensible person named Claude Renard â no relation â who worked for the Direction GĂ©nĂ©rale des ImpĂŽts and had spent seventeen years determining the actual value of actual things for purposes of actual taxation.
He listened for four minutes.
Then he said: "Why don't we just put a camera in there?"
"A small camera," he elaborated, feeling the silence, "would record whether the light is on or off when the door is closed. We would then have an answer."
The silence continued but changed in quality.
Derrida was the first to speak. "The camera," he said slowly, "is itself a text."
"I know what a camera is," said Renard.
"The recording is not the event."
"No, but it's evidence of the event."
"Evidence," said Foucault, "is a category produced byâ"
"By the refrigerator?" said Renard.
Renard looked around the room. He was not hostile. He was genuinely curious â the look of a man who has spent his life trying to determine whether a declared asset is real and has developed, through this practice, a refined nose for the difference between complexity and evasion.
"Forgive me," he said, "but it seems like some of you don't want to know the answer. Is that right?"
Another silence â but this one was different again. Productive, almost.
Lyotard laughed first. It was a real laugh. "He's not wrong," he said.
"The answer would be boring," said Baudrillard, but he said it quietly, without conviction.
"It would foreclose the question," said Derrida, more carefully, "and the question is where we live."
GĂ©rard, who had been standing by the kitchenette all this time, waiting to return to the coffee urn, said: "My father was a fisherman. He lived inside a question his whole life â where are the fish? But he still needed to throw the net."
Everyone looked at Gérard.
Gérard looked slightly alarmed to have said something.
Deleuze reached for his notebook.
They did not put a camera in the refrigerator. This was the consensus, reached implicitly, without vote. Instead they stood around the open refrigerator door for a while, looking at the light, which was unambiguously on. Then Foucault closed the door. They stood around the closed refrigerator. The light was â somewhere. Doing something. In a state that was either on or off or indeterminate or socially constructed or simulated or rhizomatic.
Nobody opened the door to check.
This was, Renard thought on the train back to his office, the most expensive, educated, and elaborate way he had ever watched a group of adults refuse to open a door.
He also thought â and this was the part that stayed with him â that the light was probably off. But he wasn't completely sure anymore. And he'd been on the train for forty minutes before he noticed he was no longer sure, and couldn't say exactly when or why.
He looked out the window at the passing suburb.
"Merde," he said quietly, to no one.
The question of the refrigerator light was settled empirically in 1981, by a General Electric engineer who simply used a thermocouple. The light goes off. The philosophers were not notified. Gérard fixed the coffee urn. The brie was eaten.