What does Everett Immortality look like in the long term?
The general idea of EI is that there is always some small chance you will survive in any given situation, so there will be some multiverse timelines whose present is the same as your present, but in which you keep on living indefinitely. However, some forms of survival are a lot more likely than others; eg, it’s a lot more likely that my cryonically-preserved brain will be scanned and turned into an AI than that a copy of my brain will spontaneously appear out of nothingness. Thus, it makes sense to plan around the most likely sorts of scenarios, and not to bother doing much planning for the least likely ones.
But thinking /very/ long term, to the heat death of the universe… every form of negentropy is going to end up exhausted, with no more energy gradients that life and intelligence could use to survive from; meaning that however extended a life might be, there will be some point at which all of a person’s futures eventually fade away...
… or maybe not. Thermodynamic miracles—events violating ordinary statistics—will, on the long term, happen every so often… so might it be possible for some form of life in that era to rely on them as the last available source of negentropy? Which forms of TMs occur most often, that could most reliably be ‘fed’ from? How often do they occur, compared to the potential stability of patterns of matter-energy at this time-scale?
You’re assuming some sort of pattern theory of identity when you consider uploads a potential form of survival. If you go all-out pattern theory of identity and assume we’re in a big world, is there a reason why the subjectively subsequent moments of awareness need to actually take place at increasing time points on the universe’s timeline? A state of matter that corresponds to your pattern’s subjective t + 1 might have occurred at the universe’s t − 10000 at some distant light cone. If your mind stays at any finite size, it’ll eventually just end up going over the same states again, so you could just get an unbound subjective experience timeline inside a fixed timeslice of a spatially infinite, temporally finite universe.
If the ‘afterlife’ is infinite, then it will have infinitely more integral measure than the normal life.
Infinite as in “if you succeeded to make it into situation X, you are guaranteed to live forever” or merely potentially infinite, as in “for every situation X where you are alive, in some Everett branch will survive it” (in other words, you never run out of quantum immortality)? In the latter version, the integral of the ‘afterlife’ may still be smaller than the integral of ‘normal life’.
During a person’s ‘normal’ life the number of Everett branches containing that person approaches infinity. The way mortality currently works is that there’s a certain probability that you will die during each year, let’s say it’s 0.01 when you’re 20. That percent of Everett branches gets “eliminated” each year. This probability of dying increases each year, until it approaches 1 when you’re close to the age of 120. Let’s ignore life-extending technologies. In Copenhagen interpretation the probability that you’re alive after the age of 120 is effectively zero. In MWI there are few branches that survive beyond this, some of these for very long, potentially forever. So I agree with you, that the intergral of branches during a person’s normal life is probably greater than that of the smaller number of branches that survive almost forever. This is true even if the number of branches or the length of them is infinite, didn’t Cantor prove that there are different sized infinities?
Is this what you were after? I’m a bit confused. Tell me if I made any mistakes.
I don’t have bandwidth for a podcast just now; so ‘infinite’ in what direction? If the number of MWI timelines can be divided infinitely, then that seems like it would suffice, even if the universe is finite in many other ways.
Did he give any reasoning for that belief? Eg, does assuming non-infinitesimal worldlines improve the predictions of the interference of double-slit style experiments?
Again from what I recall: scientists have not found any evidence of infinities, math incompleteness problems go away without infinities, and computer physics models work even though computers have finite memories.
So let’s say you’re a soldier in battle in 2000 BCE. Someone just slashed your stomach open with a sword, you’re in horrible pain, your internal organs are spilling out, but you’re still conscious and aware of what’s happening. How are quantum immortality and the power of science going to work out for you now?
EDIT: I thought quantum immortality was thought as a thing that applies to everyone everywhere. Are we discussing some sort of more constrained version here that doesn’t apply to “your chest just got smashed by an engine block but you’re still conscious for a little while” but does apply to cryonics, uploading etc. information theoretic undeath shenanigans?
The answer is the most likely miracle, but I am not sure what exactly that would be. All necessary miracles are so improbably that I don’t trust my ability to evaluate their relative probabilities.
It could be something like: By random movement of atoms, your organs jump inside and your wounds heal (and your body overcomes the infection). All wittnesses stop fighting and start worshiping you as a god. You don’t understand the situation, but successfully use your new situation to stop the war or escape from the war. You collect smart people around you, supported by your followers’ donations, and together you invent science relatively slowly. It still takes a hundred years or more, that you miraculously survive with sufficient brain function. At the end your team develops a recursively self-improving AI (not necessarily a Friendly one, only one that wants to keep you alive).
Despite all the miracles, this seems like the least miraculous path from “cut with a sword” to “immortality”. (Assuming that the damage really happened, because otherwise the most likely path starts with “you wake up from the nightmate”.)
This is curiously detailed for something where basically the only requirement is that you stay aware of every moment, constant horrible pain and debilitating injuries aren’t any sort of problem unless they keep you from staying conscious, and there’s basically no lookahead beyond whatever the duration between consecutive states of subjective consciousness is, definitely something less than a second.
Sure, someone in the multiverse is going to get the happy shiny human-friendly thermodynamic miracle starting up for them, but it seems like there’d be countless quite a bit less improbable quivering masses of horrible injuries and pain who Just. Can’t. Die.
I mean, think of the lookahead. Sure, the miracle scenario has you having a lot bigger measure of existence after the miracle has taken place, but there doesn’t seem to be a point going directly forward from the lethal injury state where it’s more likely to go down the path of the miracle starting to happen than to just stay improbably aware in your current rapidly decaying state. You’d probably end up with some incredibly measure-sparse weird Boltzmann-brain-like states in the end, but isn’t it possible that at every step along the way there are a lot more pseudo-Boltzmann-brain futures than there are body-repairing thermodynamic miracle futures?
Quantum immortality is based on MWI, which is designed explicitly to match the standard “shut up and calculate” approach to QM, which means that it cannot have any measurable effects outside the standard framework, where “Everett branches” are known as “possible outcomes”. If you expect different consequences for your personal experience in the two pictures, you probably do not understand MWI.
So I can kill myself without worrying about some nasty existential horror shit, if needs be? Because that’s really all I wanted to know and LW seems like the only place that would take a query like this seriously
Does not follow. MWI is orthogonal to “some nasty existential horror shit”, it doesn’t provide evidence either for or against your worries.
I have no idea what do you worry about, but according to our current understanding in this life there is no detectable difference between a Copenhagen world and an Everett world. As to the afterlife, all bets are off—contemporary physics can’t help you there.
Trying to understand quantum physics on the basis of web comics doesn’t strike me as a useful. The lesson you should draw from that comic is that standing near a nuclear bomb when it explodes is a bad idea.
What does Everett Immortality look like in the long term?
The general idea of EI is that there is always some small chance you will survive in any given situation, so there will be some multiverse timelines whose present is the same as your present, but in which you keep on living indefinitely. However, some forms of survival are a lot more likely than others; eg, it’s a lot more likely that my cryonically-preserved brain will be scanned and turned into an AI than that a copy of my brain will spontaneously appear out of nothingness. Thus, it makes sense to plan around the most likely sorts of scenarios, and not to bother doing much planning for the least likely ones.
But thinking /very/ long term, to the heat death of the universe… every form of negentropy is going to end up exhausted, with no more energy gradients that life and intelligence could use to survive from; meaning that however extended a life might be, there will be some point at which all of a person’s futures eventually fade away...
… or maybe not. Thermodynamic miracles—events violating ordinary statistics—will, on the long term, happen every so often… so might it be possible for some form of life in that era to rely on them as the last available source of negentropy? Which forms of TMs occur most often, that could most reliably be ‘fed’ from? How often do they occur, compared to the potential stability of patterns of matter-energy at this time-scale?
You’re assuming some sort of pattern theory of identity when you consider uploads a potential form of survival. If you go all-out pattern theory of identity and assume we’re in a big world, is there a reason why the subjectively subsequent moments of awareness need to actually take place at increasing time points on the universe’s timeline? A state of matter that corresponds to your pattern’s subjective t + 1 might have occurred at the universe’s t − 10000 at some distant light cone. If your mind stays at any finite size, it’ll eventually just end up going over the same states again, so you could just get an unbound subjective experience timeline inside a fixed timeslice of a spatially infinite, temporally finite universe.
.
Infinite as in “if you succeeded to make it into situation X, you are guaranteed to live forever” or merely potentially infinite, as in “for every situation X where you are alive, in some Everett branch will survive it” (in other words, you never run out of quantum immortality)? In the latter version, the integral of the ‘afterlife’ may still be smaller than the integral of ‘normal life’.
Good point.
During a person’s ‘normal’ life the number of Everett branches containing that person approaches infinity. The way mortality currently works is that there’s a certain probability that you will die during each year, let’s say it’s 0.01 when you’re 20. That percent of Everett branches gets “eliminated” each year. This probability of dying increases each year, until it approaches 1 when you’re close to the age of 120. Let’s ignore life-extending technologies. In Copenhagen interpretation the probability that you’re alive after the age of 120 is effectively zero. In MWI there are few branches that survive beyond this, some of these for very long, potentially forever. So I agree with you, that the intergral of branches during a person’s normal life is probably greater than that of the smaller number of branches that survive almost forever. This is true even if the number of branches or the length of them is infinite, didn’t Cantor prove that there are different sized infinities?
Is this what you were after? I’m a bit confused. Tell me if I made any mistakes.
As Max Tegmark mentioned on this Rationally Speaking podcast quantum immortality might only work if the universe is infinite.
I don’t have bandwidth for a podcast just now; so ‘infinite’ in what direction? If the number of MWI timelines can be divided infinitely, then that seems like it would suffice, even if the universe is finite in many other ways.
As I recall, he doesn’t believe the universe is infinite in any direction.
Did he give any reasoning for that belief? Eg, does assuming non-infinitesimal worldlines improve the predictions of the interference of double-slit style experiments?
Certainly not the latter.
If there were any perceptible grain to them, we’d be about a picosecond from the abrupt end of the universe-as-we-know-it.
Again from what I recall: scientists have not found any evidence of infinities, math incompleteness problems go away without infinities, and computer physics models work even though computers have finite memories.
Quantum immortality is a poor atheist’s immortal soul.
That’s the opposite of comforting.
How so? Don’t people find it comforting believing that there are universes where they survive against impossible odds?
Mere survival doesn’t sound all that great. Surviving in a way that is comforting is a very small target in the general space of survival.
Beats dying if you believe that some day you will be saved BY THE POWER OF SCIENCE!
So let’s say you’re a soldier in battle in 2000 BCE. Someone just slashed your stomach open with a sword, you’re in horrible pain, your internal organs are spilling out, but you’re still conscious and aware of what’s happening. How are quantum immortality and the power of science going to work out for you now?
EDIT: I thought quantum immortality was thought as a thing that applies to everyone everywhere. Are we discussing some sort of more constrained version here that doesn’t apply to “your chest just got smashed by an engine block but you’re still conscious for a little while” but does apply to cryonics, uploading etc. information theoretic undeath shenanigans?
The answer is the most likely miracle, but I am not sure what exactly that would be. All necessary miracles are so improbably that I don’t trust my ability to evaluate their relative probabilities.
It could be something like: By random movement of atoms, your organs jump inside and your wounds heal (and your body overcomes the infection). All wittnesses stop fighting and start worshiping you as a god. You don’t understand the situation, but successfully use your new situation to stop the war or escape from the war. You collect smart people around you, supported by your followers’ donations, and together you invent science relatively slowly. It still takes a hundred years or more, that you miraculously survive with sufficient brain function. At the end your team develops a recursively self-improving AI (not necessarily a Friendly one, only one that wants to keep you alive).
Despite all the miracles, this seems like the least miraculous path from “cut with a sword” to “immortality”. (Assuming that the damage really happened, because otherwise the most likely path starts with “you wake up from the nightmate”.)
This is curiously detailed for something where basically the only requirement is that you stay aware of every moment, constant horrible pain and debilitating injuries aren’t any sort of problem unless they keep you from staying conscious, and there’s basically no lookahead beyond whatever the duration between consecutive states of subjective consciousness is, definitely something less than a second.
Sure, someone in the multiverse is going to get the happy shiny human-friendly thermodynamic miracle starting up for them, but it seems like there’d be countless quite a bit less improbable quivering masses of horrible injuries and pain who Just. Can’t. Die.
I mean, think of the lookahead. Sure, the miracle scenario has you having a lot bigger measure of existence after the miracle has taken place, but there doesn’t seem to be a point going directly forward from the lethal injury state where it’s more likely to go down the path of the miracle starting to happen than to just stay improbably aware in your current rapidly decaying state. You’d probably end up with some incredibly measure-sparse weird Boltzmann-brain-like states in the end, but isn’t it possible that at every step along the way there are a lot more pseudo-Boltzmann-brain futures than there are body-repairing thermodynamic miracle futures?
.
What claim?
.
I find it counterproductive to assign probability or truth value to untestables.
.
If your decisions depend on untestables, you need a better decision theory.
.
Quantum immortality is based on MWI, which is designed explicitly to match the standard “shut up and calculate” approach to QM, which means that it cannot have any measurable effects outside the standard framework, where “Everett branches” are known as “possible outcomes”. If you expect different consequences for your personal experience in the two pictures, you probably do not understand MWI.
.
Blanking your comments before retracting them? To hide changing your mind after learning stuff?
No, now that I got a clear picture of this issue I will delete this account among other things. Sorry for bothering you.
I don’t think removing the content from your comments is a good way to react to changing your mind, if that is your reason.
What might these consequences be?
.
I don’t think that’s how MWI works.
.
Does not follow. MWI is orthogonal to “some nasty existential horror shit”, it doesn’t provide evidence either for or against your worries.
.
I have no idea what do you worry about, but according to our current understanding in this life there is no detectable difference between a Copenhagen world and an Everett world. As to the afterlife, all bets are off—contemporary physics can’t help you there.
.
Trying to understand quantum physics on the basis of web comics doesn’t strike me as a useful. The lesson you should draw from that comic is that standing near a nuclear bomb when it explodes is a bad idea.
Whatever happens to you after you die.