Ludwig's Revenge

Dennis Overbye, writing in the New York Times, has a nice article today on the so-called Boltzmann brain problem in cosmology. It's a new incarnation of a familiar problem in statistical mechanics and the physics of time, the problem that pure statistics seems to imply that entropy should increase in the past as well as the future: it's a priori more probable that the present is a momentary fluctuation rather than part of a long history extending into the past. Overbye, like cosmologists, recognizes that this way madness lies - the question is what to do about it.

It could be the weirdest and most embarrassing prediction in the history of cosmology, if not science.

If true, it would mean that you yourself reading this article are more likely to be some momentary fluctuation in a field of matter and energy out in space than a person with a real past born through billions of years of evolution in an orderly star-spangled cosmos. Your memories and the world you think you see around you are illusions.

This bizarre picture is the outcome of a recent series of calculations that take some of the bedrock theories and discoveries of modern cosmology to the limit. Nobody in the field believes that this is the way things really work, however. And so there in the last couple of years there has been a growing stream of debate and dueling papers, replete with references to such esoteric subjects as reincarnation, multiple universes and even the death of spacetime, as cosmologists try to square the predictions of their cherished theories with their convictions that we and the universe are real. The basic problem is that across the eons of time, the standard theories suggest, the universe can recur over and over again in an endless cycle of big bangs, but it’s hard for nature to make a whole universe. It’s much easier to make fragments of one, like planets, yourself maybe in a spacesuit or even — in the most absurd and troubling example — a naked brain floating in space. Nature tends to do what is easiest, from the standpoint of energy and probability. And so these fragments — in particular the brains — would appear far more frequently than real full-fledged universes, or than us. Or they might be us.

It seems that the current incarnation of this puzzle is somehow connected to the problem of dark energy, though I couldn't quite understand Overbye's discussion of the problem. I think I may need to go read Sean Carroll's discussion of the subject, if I can still find it.