In a sense, the string situation is even worse than that: there is no agreement on how to approach cosmology in string theory. All these landscape papers are based on ansatze that just guess how the big picture works.
The Boltzmann brain problem arises because the volume of eternally inflating de Sitter space is vastly bigger than the non-inflating regions, big enough for an exponentially unlikely entity to fluctuate into existence. But de Sitter space is quantum-mechanically unstable, and how to do quantum gravity there is probably the outstanding issue in that subject right now.
One possibility that Linde examines is that de Sitter space is unstable enough that it transitions to something else before it can become vast. The best toy model of de Sitter space right now is (in my opinion) dS/CFT duality in Vasiliev gravity theories, which are like a truncated version of string theory, and there’s very recent work on the instability there. So that might end up providing the answer.
My general response to the original post is therefore that you shouldn’t attach too much significance to papers on Boltzmann brains. They are a bit like papers on the Drake equation: they make as many assumptions as they need to make, in order to reach a conclusion, but those assumptions aren’t very well-founded. It’s the hard technical work on quantum gravity in de Sitter space which will decisively clarify the issue.
Does this generally lower your estimate of multiverse theories (vs single universes) or are initial arguments solid (there seems to be significant consensus in that direction), but further details are questionable?
Boltzmann brains and multiverse are logically distinct issues. A single universe with an eternal de Sitter future potentially has this problem. I regard any theory that predicts a Boltzmann brain majority as wrong, but it’s not clear how generic a prediction this is for eternal inflation, or even whether eternal inflation is the right cosmology (e.g. for the test case, Vasiliev gravity).
In a sense, the string situation is even worse than that: there is no agreement on how to approach cosmology in string theory. All these landscape papers are based on ansatze that just guess how the big picture works.
The Boltzmann brain problem arises because the volume of eternally inflating de Sitter space is vastly bigger than the non-inflating regions, big enough for an exponentially unlikely entity to fluctuate into existence. But de Sitter space is quantum-mechanically unstable, and how to do quantum gravity there is probably the outstanding issue in that subject right now.
One possibility that Linde examines is that de Sitter space is unstable enough that it transitions to something else before it can become vast. The best toy model of de Sitter space right now is (in my opinion) dS/CFT duality in Vasiliev gravity theories, which are like a truncated version of string theory, and there’s very recent work on the instability there. So that might end up providing the answer.
My general response to the original post is therefore that you shouldn’t attach too much significance to papers on Boltzmann brains. They are a bit like papers on the Drake equation: they make as many assumptions as they need to make, in order to reach a conclusion, but those assumptions aren’t very well-founded. It’s the hard technical work on quantum gravity in de Sitter space which will decisively clarify the issue.
Does this generally lower your estimate of multiverse theories (vs single universes) or are initial arguments solid (there seems to be significant consensus in that direction), but further details are questionable?
Interesting, I just came across the discussion here http://physics.stackexchange.com/questions/32979/why-arent-we-boltzmann-brains-in-an-infinite-universe where you gave the argument from immediate disintegration. This is what I’ve been thinking also (some theist friends of mine were attempting to “prove” that multiverse theories are crazy based on BB).
Boltzmann brains and multiverse are logically distinct issues. A single universe with an eternal de Sitter future potentially has this problem. I regard any theory that predicts a Boltzmann brain majority as wrong, but it’s not clear how generic a prediction this is for eternal inflation, or even whether eternal inflation is the right cosmology (e.g. for the test case, Vasiliev gravity).