Boltzmann Brains

Sean Carroll has posted a few times on Boltzmann Brains, and he has a new entry today Richard Feynman on Boltzmann Brains. The original motivation for considering such things is the fact that the Universe seems to have started with a very low entropy - very low compared to all possible universes. This is fortunate since it allowed the development of all the things that make us possible - galaxies, solar systems, life and evolution. So why, we may ask, was the initial state of our universe so low in entropy. Nobody knows, but Boltzmann, who started the whole business, had an idea: maybe it was a random statistical fluctuation.

The so-called Boltzmann Brain Paradox is intended to show that the fluctuation hypothesis is not a profitable assumption. From Wikipedia:

This leads to the Boltzmann brain concept: If our current level of organization, having many self-aware entities, is a result of a random fluctuation, it is much less likely than a level of organization which is only just able to create a single self-aware entity. For every universe with the level of organization we see, there should be an enormous number of lone Boltzmann brains floating around in unorganized environments. This refutes the observer argument above: the organization I see is vastly more than what is required to explain my consciousness, and therefore it is highly unlikely that I am the result of a stochastic fluctuation.

The Boltzmann brains paradox is that it is more likely that a brain randomly forms out of the chaos with false memories of its life than that the universe around us would have billions of self-aware brains.

So far as I know, Feynman didn't actually consider Boltzmann Brains, but he makes a somewhat similar argument against the fluctuation theory based on familiar observations of physics. I don't think anybody thinks that Boltzmann Brains actually exist in our part of the multiverse, but they do serve as a useful reductio ad absurdum.

Lumo weighs in with a commentary, which, not unexpectedly, seems to utterly miss the point. It's a deep oddity that Lubos, who masters the intricasies of string theory with apparent ease, can't seem to follow a three step scientific, philosophical, or political argument.

Dennis Overbye, writing in the New York Times about a year ago, has an excellent article on why the subject has come up again in Cosmology.

A different, but perhaps related, form of antigravity, glibly dubbed dark energy, seems to be running the universe now, and that is the culprit responsible for the Boltzmann brains.

The expansion of the universe seems to be accelerating, making galaxies fly away from one another faster and faster. If the leading dark-energy suspect, a universal repulsion Einstein called the cosmological constant, is true, this runaway process will last forever, and distant galaxies will eventually be moving apart so quickly that they cannot communicate with one another. Being in such a space would be like being surrounded by a black hole.

Rather than simply going to black like “The Sopranos” conclusion, however, the cosmic horizon would glow, emitting a feeble spray of elementary particles and radiation, with a temperature of a fraction of a billionth of a degree, courtesy of quantum uncertainty. That radiation bath will be subject to random fluctuations just like Boltzmann’s eternal universe, however, and every once in a very long, long time, one of those fluctuations would be big enough to recreate the Big Bang. In the fullness of time this process could lead to the endless series of recurring universes. Our present universe could be part of that chain.

In such a recurrent setup, however, Dr. Susskind of Stanford, Lisa Dyson, now of the University of California, Berkeley, and Matthew Kleban, now at New York University, pointed out in 2002 that Boltzmann’s idea might work too well, filling the megaverse with more Boltzmann brains than universes or real people.

Blame that troublemaker Susskind!