The Information - James Gleick [171]
At first astrophysicists focused on matter and energy falling in. Later they began to worry about the information. A problem arose when Stephen Hawking, adding quantum effects to the usual calculations of general relativity, argued in 1974 that black holes should, after all, radiate particles—a consequence of quantum fluctuations near the event horizon.♦ Black holes slowly evaporate, in other words. The problem was that Hawking radiation is featureless and dull. It is thermal radiation—heat. But matter falling into the black hole carries information, in its very structure, its organization, its quantum states—in terms of statistical mechanics, its accessible microstates. As long as the missing information stayed out of reach beyond the event horizon, physicists did not have to worry about it. They could say it was inaccessible but not obliterated. “All colours will agree in the dark,” as Francis Bacon said in 1625.
The outbound Hawking radiation carries no information, however. If the black hole evaporates, where does the information go? According to quantum mechanics, information may never be destroyed. The deterministic laws of physics require the states of a physical system at one instant to determine the states at the next instant; in microscopic detail, the laws are reversible, and information must be preserved. Hawking was the first to state firmly—even alarmingly—that this was a problem challenging the very foundations of quantum mechanics. The loss of information would violate unitarity, the principle that probabilities must add up to one. “God not only plays dice, He sometimes throws the dice where they cannot be seen,” Hawking said. In the summer of 1975, he submitted a paper to the Physical Review with a dramatic headline, “The Breakdown of Physics in Gravitational Collapse.” The journal held it for more than a year before publishing it with a milder title.♦
As Hawking expected, other physicists objected vehemently. Among them was John Preskill at the California Institute of Technology, who continued to believe in the principle that information cannot be lost: even when a book goes up in flames, in physicists’ terms, if you could track every photon and every fragment of ash, you should be able to integrate backward and reconstruct the book. “Information loss is highly infectious,”♦ warned Preskill at a Caltech Theory Seminar. “It is very hard to modify quantum theory so as to accommodate a little bit of information loss without it leaking into all processes.” In 1997 he made a much-publicized wager with Hawking that the information must be escaping the black hole somehow. They bet an encyclopedia of the winner’s choice. “Some physicists feel the question of what happens in a black hole is academic or even theological, like counting angels on pinheads,”♦ said Leonard Susskind of Stanford, siding with Preskill. “But it is not so at all: at stake are the future rules of physics.” Over the next few years a cornucopia of solutions was proposed. Hawking himself said at one point: “I think the information probably goes off into another universe. I have not been able to show it yet mathematically.”♦
It was not until 2004 that Hawking, then sixty-two, reversed himself and conceded the bet. He announced that he had found a way to show that quantum gravity is unitary after all and that information is preserved. He applied a formalism of quantum indeterminacy—the “sum over histories” path integrals of Richard Feynman—to the very topology of spacetime and declared, in effect, that black