“Momentism”: Ethics for Boltzmann Brains
4Seth Herd
3Jordan Arel
3Seth Herd
2silentbob
3Viliam
1Jordan Arel
3Cleo Nardo
6Richard_Kennaway
12JBlack
1Jordan Arel
2Mikhail Samin
2avturchin
6Mikhail Samin
2avturchin
New Comment
You're not a Boltzmann brain, because even the simplest thought required uncountable successive states to unfold. Even if there's some simplification, the odds of the time sequence of even the simplest thought unfolding are minuscule raised to the power of the many necessary successive states.
1My understanding is that in the case of heat death, if this state continues to persist for infinite time, then every possible event that can happen will eventually happen due to sheer infinite time and every possible thermal fluctuation that can happen eventually happening; although of course much smaller fluctuations are always more likely than bigger ones.
A brain fluctuating into existence and then existing for long enough to take an instantaneous mental action is absurdly more unlikely than the brain popping into existence and then scattering just as fast; but the infinity, if true, would make infinite such brains inevitable and much more likely than real brains.
Again, this is just my understanding of the argument as an amateur, definitely not a professional in this and I think there are physics and epistemics arguments which might overcome this.
Thanks, I hate it! I guess I hadn't seen a full presentation of the argument or didn't remember it. Now I see why it's troubling enough to want to resolve. Those physics and epistemics arguments seem important but I'm going to resist getting into them just in case I'm real and should be working on solving alignment.
If you are a Boltzmann brain, then you will probably more or less instantaneously disintegrate.
NOPE! 🚨🚨🚨🚨
Dynamics in thermal equilibrium are time-symmetric, so Boltzmann brains disintegrate in the exact same way they integrate, except in reverse.
And they integrate very very very slowly.
And they integrate very very very slowly.
How so? Boltzmann brains, as I understand them, are just random thermal fluctuations that happen to take the form of a thinking brain. Fluctuations on average decay at a rate determined by how frequently energy is exchanged between the degrees of freedom. By time symmetry, when they happen they develop at that rate also. There is no painstaking process of assembly, as there is for ordinary brains.
1Boltzmann brains (in universes that permit them) will be overwhelmingly dominated by the least improbable of all possible methods of formation. Matter is sticky, so that once partly assembled it has a much higher probability of continuing to remain assembled especially at the very low temperatures usually envisaged.
There are far more paths for slow formation of assemblies of matter in such an environment than instant ones. To randomly assemble a trillion correct atoms in exactly the right place at the same time is much less likely than accumulating a trillion atoms in the wrong places which then migrate through any of an immense number of possible products of paths to the right places over different times.
It seems very much more likely that the modal Boltzmann brain has had essentially all of its matter in a pre-functioning state for far longer than mere geological timescales. It will almost certainly not be biological matter due to every atom contributing orders of magnitude less probability of formation. I would guess the most likely form to be some sort of dense reversible computing substrate that will operate very near absolute zero, and for a conscious brain of this type to be immensely more common in the space of thermal configurations than even one cell of familiar biological life.
You’re right about time-symmetry. But the key point is that minimal fluctuations are exponentially more probable than large ones.
The most likely Boltzmann brain is one that exists for just an instant - particles briefly converge into a brain configuration then immediately scatter. A fluctuation that maintains a brain for seconds or minutes is exponentially less likely because it requires sustaining an improbable state longer.
So yes, formation and disintegration are time-symmetric, but the most probable Boltzmann brains still form and disintegrate nearly instantly - not because symmetry breaks, but because brief fluctuations vastly outnumber sustained ones.
The absurdity of your continued influence climbs exponentially
Having a memory of continued existence isn’t absurd if you’re a Boltzmann brain.
What’s absurd though is that you continue to not start seeing white noise; there are more Boltzmann brains that see white noise than those that see something meaningful.
(Oh well, I guess “you’re a Boltzmann brain” loses a lot of Bayes points to “most of your reality-fluid comes from faithful simulations by physics->atoms->neurons->algorithms that make you up, not by random fluctuations”)
For any BB, there is another BB somewhere which looks like as if it is causally affected by the first BB. As a result, there are chains of BBs which look like causal chains of minds.
There is almost no difference between them and the real world.
There might be a brain that looks like a continuation of the previous one, but there won’t be any causal impact, because the probabilities of some of them appearing have no influence on the probabilities of others of them, so the math of causality says there isn’t any arrow of causality present.
Curated and popular this week