Formation via nucleation of boltzmann brains
post by Zeruel017 · 2022-06-01T18:05:34.450Z · LW · GW · No commentsThis is a question post.
Contents
Answers 1 JBlack None No comments
hi beautiful people, in this forum I feel very stupid and it's good, it means that I can learn more!
I would have a doubt as I expressed in my previous question about boltzmann brain formation via nucleation (https://www.lesswrong.com/posts/LGMSLXkpKAofebjfi/a-terrifying-variant-of-boltzmann-s-brains-problem [LW · GW]), I read this concept in this Wikipedia page (https://en.m.wikipedia.org/wiki/Boltzmann_brain) but it is not clear to me how it actually happens? for example: does it form instantaneously or slowly? (obviously this is not the only thing I doubt about this way of formation)
Answers
In any practical sense, it does not actually happen. It's a prediction of a model derived from lots of assumptions, many of them pushed far beyond any experimental support, and some pushed beyond any possible experimental support.
The simplest idea is that the known laws of physics are time symmetric at a suitably microscopic level. A person in space can die, dessicate, and over billions of years erode and evaporate away to basically become a uniform addition to the interstellar medium. So in principle, the reverse can happen too.
So a few atoms at a time of interstellar medium can by chance be travelling in the right direction and speed to eventually form a dead dried husk of a person, which again by chance accumulates water molecules and other compounds that (by chance) formed far away, and hydrate and form complex structures until you end up with a living person - or just a living brain in a bottle with fake memories and sensations - which will almost certainly die again very quickly. The smaller and simpler the object formed, the higher probability it is by an enormous margin.
The phrase "by chance" is doing an enormous amount of work there. Calling the chances merely microscopically small is doing them a gross disservice. Even their logarithms have a more digits than is convenient to write. We have no idea if the laws of physics by which the universe actually runs even support such things, and there is no way to ever know since the prediction is so low in probability that pretty much everything else that is possible to observe is immensely more likely.
Notably, one of the incredibly more likely possibilities is that someone has the mistaken impression that they are likely to be a Boltzmann brain.
↑ comment by Zeruel017 · 2022-06-02T12:14:50.925Z · LW(p) · GW(p)
so are you telling me that the current laws of physics don't necessarily support boltzmann's brain hypothesis or any other object formed in this way? And if you have patience, could you also briefly list the unproven assumptions to make Boltzmann's "things" a realistic scenario?
note: from what you know, by chance do you know if all matter will become iron and therefore the hypothesis of boltzmann's brains can be denied? (source from the concept of iron matter: (https://en.m.wikipedia.org/wiki/Future_of_an_expanding_universe)
Replies from: JBlack↑ comment by JBlack · 2022-06-03T09:15:57.239Z · LW(p) · GW(p)
Current models of physics do not rule out Boltzmann brains, but don't really support them either. Even the slightest variations in our expectation of the long-run future of the universe make them either essentially impossible, or possible but vanishingly rare. In no scenario are they anything like common.
The main ingredient in any Boltzmann brain scenario is staggeringly large quantities of time in which nothing else happens except random thermal fluctuations. None of this tiny little trillion year nonsense, we're talking about a number of years that is so large that it would take you many trillions of years to even read the number. Obviously we do not have, and can not have, any support for that actually happening.
On the timescales being discussed here, whether or not all matter becomes iron is irrelevant. Even if an iron atom has a 10^-1000 chance of encountering a ridiculously unlikely random gamma ray photon that spallates it into lighter elements, over these timescales that happens so many times that you can't hope to keep track of them all even if you had some entropy-proof enormous computer. It even happens to trillions of atoms in close proximity simultaneously so many times that you can't hope to keep track of how many times it happened. I cannot state strongly enough how mindbogglingly long these periods of time are. Immortal monkeys eventually typing at random the complete works of Shakespeare have nothing on these lengths of time; they packed up their typewriters with the job well done long before the first Boltzmann bacterium. Then unimaginably many Boltzmann bacteria would have to form and decay before the first instance of a Boltzmann bacterium that was one atom larger, and so on until you get enough atoms to form an actual brain.
Predicting the future is hard. Being confident in predicting the future is foolhardy. Confidence in the existence of Boltzmann brains requires an immensely precise prediction about how reality must proceed over unimaginable timescales in the future, far beyond any conditions that we have observed or even could possibly hope to observe, and is empty intellectual arrogance of the greatest order.
Replies from: Zeruel017↑ comment by Zeruel017 · 2022-06-03T11:11:14.023Z · LW(p) · GW(p)
Thanks for the long answer, the only thing I did not understand is the fact of the irrelevance of the iron matter, I thought that maybe if the matter decays with iron tunneling, a boltzmann brain or other things can no longer form
Replies from: JBlack↑ comment by JBlack · 2022-06-04T02:58:43.089Z · LW(p) · GW(p)
Over long enough timescales any interaction will reverse, and happen again, and so on. It's just a question of what fraction of time is occupied in each state. Or so the theory goes, at least.
Replies from: Zeruel017↑ comment by Zeruel017 · 2022-06-04T12:37:01.449Z · LW(p) · GW(p)
therefore the matter which will perhaps decay into iron is not proof against the Boltzmann Brains hypothesis
Replies from: JBlack↑ comment by JBlack · 2022-06-05T01:45:14.041Z · LW(p) · GW(p)
Yes, there is no proof that they cannot possibly happen. It is a thing that theoretically could happen in a sufficiently boring universe, and there is no way to know whether our universe is sufficiently boring (but my prior odds are very much against it).
Replies from: Zeruel017, Zeruel017, Zeruel017No comments
Comments sorted by top scores.