Quantum Immortality without MWI?

[Physicuser]

Hello.

There is the idea that Many Worlds interpretations could imply subjective immortality (known as Quantum Immortality).

However, could this work even if MWI is not true? If the universe is infinite in space or time, or there are an infinite number of universes like ours (several mainstream theories imply it), is supposed that all possible happens, so there are infinite versions of you out there, and some of them will scape death miraculously.

If you are just a succession of mental states/observer-moments, it doesn't matter if these are produced by your original "you", a person like you in another galaxy or a brain that popped out in the middle of space, so even if there is no causal connection between them, there is a subjective sensation of continuity, so you won't experience die. Even more, if your continuity of identity is based on your memories, it doesn't even need a real succession, so can appears a brain with memories of being a 1000 years old you (this raise the question on the nature of consciousness and identity).

Is this possible with other interpretations of QM?

 

[EPR]

Google 'Bolzmann brain' by Austrian physicist Ludwig Boltzmann.

String theorist Brian Greene:

You mention in your book that we might be “Boltzmann brains” – fleeting aggregations of particles in space that happen to create the sense of being us. Does it keep you awake at night?
I don’t worry in the sense of it giving me some kind of existential angst. I am confident that I am not a Boltzmann brain. However, we want our theories to similarly concur that we are not Boltzmann brains, but so far it has proved surprisingly difficult for them to do so. So, I see Boltzmann brains as a mathematical problem that we need to solve, as opposed to an existential affront. I believe it is a problem that we will one day overcome.

https://www.theguardian.com/science...cal-physicist-interview-until-the-end-of-time

 

[Physicuser]

Yes, I know the concept. Boltzmann Brains are usually dismissed saying that if you were one, it would be more probable to don't observe a coherent world. Mallah makes a similar objection to QI: if it was true and you could be live indefinitely, since there are more "survival" observer moments, it would be more probable to you being experiencing one of these than one of the few "normal".

A doubt I have is if the important thing here is the number of observers or observer-moments, I mean, if there are 100 versions of you, 99 drinking water and 1 drinking coke, since the 99 observer are same observer-moment, what are the chances? 99-1 or 1-1?

 

[akvadrako]

However, could this work even if MWI is not true?

There are two ways that SI could work with MWI and it depends how one experiences life; what it means to have a consistent identity over time. If your sense of self requires a connected physical component that evolves over time, then it seems that SI would only work with MWI and only if you apply a branching model, where matter really splits along higher dimensions when measurements happen.

On the other hand, we might experience a consistent identity via self-locating uncertainty, which could be like you said, just a succession of mental states. Then SI could work in any multiverse, including a diverging model of MWI, where the worlds are physically separate to begin with. Or a cosmological multiverse or even just an infinitely large single spacetime.

Is this possible with other interpretations of QM?

That depends on (1) are there other multiverses (Level I, II or IV) and (2) is self-locating uncertainty a real phenomenon. Assuming the answer to either is no, QI might be the only form of SI.

For Bohmian mechanics, consistent histories and objective collapse theories, I would say there is no possibility of QI, since there is just one approximately classical world.

For Copenhagen, the statistical ensemble and QBism, I don't think the question can even be defined. Those interpretations require an observer who sits outside the system being measured to record the outcome. So if the observer ceases to exist there will not be any collapse.

 

[entropy1]

My two cents:

If you, a creature, are traversing a path of experience, this could be traversing a worldline as in MWI. But in this context, you are experiencing a single worldline, so that would not require MWI, but just an unambiguous history. The experience you have in case of MWI would be the same experience without MWI.

However, the probability of getting beyond 200 years old are (necessarily) equally improbable.

And that exposes a weakness of MWI, namely that you are certain to get beyond 200 years old. However, at the price of dying astronomically more often.

 

[PeterDonis]

a weakness of MWI, namely that you are certain to get beyond 200 years old

No, the MWI does not say this. The MWI does not say people don't age, or don't die of old age. The MWI only says that, if a particular event which might or might not kill you depends on quantum uncertainty, there will be a branch where you live and a branch where you die. But it is highly unlikely that all potentially fatal events that happen to people as they age depend on quantum uncertainty. If there is no quantum uncertainty about an event, and it is fatal, there is no branching according to the MWI; you just die.

 

[entropy1]

@PeterDonis, but given the branching that is part of the MWI, in any particular branch the probability that you reach age 200 is small, wouldn't you agree?

BTW, doesn't the impact of any particle into a macro object constitute a measurement?

EDIT: I see I misunderstood you; you mean that it is not certain that you reach age 200 in some branch, right?

 

[PeterDonis]

@PeterDonis, but given the branching that is part of the MWI, in any particular branch the probability that you reach age 200 is small, wouldn't you agree?

The very concept of "probability in any particular branch" doesn't make sense in the MWI. MWI is deterministic. We've had this discussion before in other threads. If you're going to use the MWI at all, you need to use it properly.

BTW, doesn't the impact of any particle into a macro object constitute a measurement?

Pretty much, but the question is not whether a "measurement" is happening, but whether it's a measurement, subject to quantum uncertainty (meaning, there will be branching in the MWI), whose result makes a difference to whether a person lives or dies. My point is that once a human body reaches a certain age, there may be no measurements at all that do that--that no matter what measurement gets made, the result still leads to the person dying. And if there is such an age, and it's less than 200 years, then no human would ever reach 200 years of age even if the MWI is true.

EDIT: I see I misunderstood you; you mean that it is not certain that you reach age 200 in some branch, right?

Again, if you're going to use the MWI, you need to use it correctly. The question is whether there is any branch, for a given human, in which the human reaches age 200. You are simply assuming without argument that there must be at least one such branch. I am simply pointing out that you can't just assume that; you need to actually make an argument for why there should be at least one such branch. And your argument would need to address the issue I raise earlier in this post.

 

[PeterDonis]

If you're going to use the MWI at all, you need to use it properly.

Perhaps I should expand on this a little more. Many people seem to assume that the MWI means anything at all that they can imagine will occur in some branch. You seem to be assuming that, since you can imagine that a human could live to be 200 years old, there must be some branch in the MWI in which that occurs. But that assumption is not valid. The MWI does not guarantee that any outcome you can imagine will have a nonzero amplitude in the wave function (which is what is required for there to be some branch in the MWI in which it occurs). If you want to say that there will be a branch in which some outcome occurs, you have to actually make an argument for why there should be a nonzero amplitude for that outcome in the wave function. You can't just assume it.

 

[Jarvis323]

Perhaps I should expand on this a little more. Many people seem to assume that the MWI means anything at all that they can imagine will occur in some branch. You seem to be assuming that, since you can imagine that a human could live to be 200 years old, there must be some branch in the MWI in which that occurs. But that assumption is not valid. The MWI does not guarantee that any outcome you can imagine will have a nonzero amplitude in the wave function (which is what is required for there to be some branch in the MWI in which it occurs). If you want to say that there will be a branch in which some outcome occurs, you have to actually make an argument for why there should be a nonzero amplitude for that outcome in the wave function. You can't just assume it.

Is there much we can say about how divergent branches can get? Based on the nature of chaos, the vast number of branches and time, it would seem like there would be an unfathomably large number of branches, and that there would likely be all manner of branches, spanning the spectrum of nearly identical to very different to the one you occupy.

Popularizers like Sean Carroll have really emphasized this idea, making it seem like anything imaginable would happen. A better explanation might be that anything that can happen will, but that's trivially true.

If a person is playing roulette, would there be different outcomes of the same spin? Would one version of you win more that day? Would they win only somewhat more? Would one of them get the exact same outcome over and over the entire night? Would there be a universe where every ball thrown into every roulette table had the exact same outcome every time; a world where there was a bizarre apparent roulette law of physics?

If roulette is not a sufficiently chaotic process, then what about one that were? One where very very slight differences to the initial conditions change the outcome?

 

[PeterDonis]

Is there much we can say about how divergent branches can get?

We can't really say anything without knowing the wave function. Any claim about some particular outcomes being possible in the MWI is a claim about the wave function. And in most cases the person making the claim has no actual argument for why the wave function should have a nonzero amplitude for the particular outcome being claimed; they're just waving their hands and assuming it.

Based on the nature of chaos

Chaos has nothing to do with it. We are talking about the MWI, in which the wave function evolves by linear, unitary evolution and that's it. There is no chaos in linear, unitary evolution.

Popularizers like Sean Carroll have really emphasized this idea, making it seem like anything imaginable would happen.

Yes, and that's just another reason not to try to learn science from pop science sources. Even actual scientists can't resist the temptation to make claims they would never get away with in a textbook or peer-reviewed paper.

If a person is playing roulette, would there be different outcomes of the same spin?

Depends on whether quantum uncertainty is involved in how the spin comes out.

If roulette is not a sufficiently chaotic process

Chaos is irrelevant; we are talking about linear, unitary dynamics in which there is no chaos. See above.

 

[Jarvis323]

We can't really say anything without knowing the wave function. Any claim about some particular outcomes being possible in the MWI is a claim about the wave function. And in most cases the person making the claim has no actual argument for why the wave function should have a nonzero amplitude for the particular outcome being claimed; they're just waving their hands and assuming it.
Chaos has nothing to do with it. We are talking about the MWI, in which the wave function evolves by linear, unitary evolution and that's it. There is no chaos in linear, unitary evolution.
Yes, and that's just another reason not to try to learn science from pop science sources. Even actual scientists can't resist the temptation to make claims they would never get away with in a textbook or peer-reviewed paper.
Depends on whether quantum uncertainty is involved in how the spin comes out.
Chaos is irrelevant; we are talking about linear, unitary dynamics in which there is no chaos. See above.

But in our world, we've observed chaotic systems and divergence, and statistical laws. Does many worlds interpretation invalidate all statistical laws and theorems about dynamical systems. So that we are only a little lucky to be a typical world where we see chaos happen at all?

 

[PeterDonis]

in our world, we've observed chaotic systems and divergence, and statistical laws

In the classical approximation, we observe these things. But that's just an approximation. The MWI does not invalidate such approximations; but it also cannot make use of them when analyzing whether branching occurs. To analyze whether branching occurs in the MWI, you have to use the linear, unitary quantum dynamics.

Does many worlds interpretation invalidate all statistical laws and theorems about dynamical systems.

No, of course not. See above.

So that we are only a little lucky to be a typical world where we see chaos happen at all?

Not at all. The MWI does not claim that there are different classical approximations in different branches. That would be inconsistent with the basic math of QM, since the classical approximation comes from the basic math of QM.

 

[Jarvis323]

In the classical approximation, we observe these things. But that's just an approximation. The MWI does not invalidate such approximations; but it also cannot make use of them when analyzing whether branching occurs. To analyze whether branching occurs in the MWI, you have to use the linear, unitary quantum dynamics.
No, of course not. See above.
Not at all. The MWI does not claim that there are different classical approximations in different branches. That would be inconsistent with the basic math of QM, since the classical approximation comes from the basic math of QM.

If there aren't different approximations in different branches, then the point I made about the nature of chaos is not irrelevant. It means that some small differences at one time will observationally lead to big differences over time. It means if macroscopic change is possible due to quantum uncertainty, then we can expect to see diverging macroscopic outcomes, depending on what those small changes are. It means that small changes accumulate into bigger differences in the classical approximation, and it means that we can reason about the questions that I was asking with real valid arguments and evidence. Sure technically maybe that would just be an approximation, as is practically all of physics.

 

[Physicuser]

Popularizers like Sean Carroll have really emphasized this idea, making it seem like anything imaginable would happen

It seems that Carroll don't believe in it, he said this on Reddit:

Briefly: I don't think it makes much sense. As your present self evolves into many selves in the future due to branching of the wave function, those future selves are all different people. Even if one of them lives forever (not obvious, since it's not true that "everything happens," only things that quantum evolution leads to), the others won't. They'll be just as dead as they would in a classical universe.

 

[PeterDonis]

If there aren't different approximations in different branches, then the point I made about the nature of chaos is not irrelevant.

It is to the particular point under discussion here, which is whether the MWI, if true, means that everyone will live to be at least 200 years old in some branch. All of your (correct) statements about how chaos will be observed in the classical approximation in different branches say nothing about that point: sure, there might be wildly divergent macroscopic histories in different branches because of tiny initial differences in some quantum measurement outcome, but it could also still be the case that no human lives to be 200 years old in any of those wildly divergent macroscopic histories, because none of them include a human living to be 200 years old in the resultant classical, chaotic dynamics. And in fact, based on our knowledge of how human bodies work in the classical approximation, chaos and all, that is what we would expect to be true of all MWI branches. "Chaos" is not a magic word that makes anything you can imagine possible, any more than "MWI" is.

 

[PeterDonis]

there aren't different approximations in different branches

In fact, this statement all by itself summarizes the argument I have been making to @entropy1, better than I did in earlier posts. Yes, there will be multiple branches in the MWI, but, for the very reason given in the quote above, all of those branches will have the same classical dynamics. And that means all of the branches will have human bodies decaying in a way that makes it impossible for any of them to live to be 200 years old.

(Note that I am ignoring the possibility of humans doing something to alter how our bodies work that either makes them age slower or undoes the effects of aging. It is certainly possible that, if we figure out how to do that, humans might live to be 200 years old or more; but that possibility has nothing whatever to do with the MWI. It depends on human ingenuity and motivation. Physics is not the discipline to go to if you want to make predictions about those things.)

 

[Jarvis323]

Ok, I guess there was a misunderstanding that I was arguing in favor of people definitely reaching 200 in some branch. My comments were a little off topic I guess.

 

[Physicuser]

Wait... I was seeing this like if there were a lot of people like you somewhere, you will die and maybe some of them survive, but this doesn't affect you...

However, if the universe is infinite in space, you are "rebirthing" continually, and if conciousness has nothing special, and a person exactly like you is you, you are "reliving" your live, like the "eternal retourn", or like if it was reincarnations... Then if Quantum Immortality is possible, in one of these lives you will survive, not a copy of you, not subjectively, but this you...

dang, I hope this to be flawed, some objections?

 

[PeterDonis]

if the universe is infinite in space, you are "rebirthing" continually

No. You are misunderstanding what a spatially infinite universe actually implies and what the actual argument is for there being an infinite number of copies of "you" in a spatially infinite universe.

The actual argument goes like this: a human body is made of a finite number of particles and therefore has a finite number of degrees of freedom, so there are a finite number of possible "states" of a human body (which includes the brain, and we are assuming physicalism so all of your conscious experience is instantiated in your body and brain). A human has a finite lifespan, and if we assume that human experiences are finitely divisible in time, there are then a finite number of possible sequences of human experiences--each such sequence being a discrete sequence of a finite number of "states of a human body and brain", and therefore a finite number of human experiences. Therefore, in a spatially infinite universe, each one of the finite number of possible sequences of human experiences will be instantiated infinitely often (since any finite sequence of events will).

Each such possible sequence of human experience defines a "person"--so "you" and "me" and every other possible human person is identified with one such sequence. So in a spatially infinite universe, each possible human person is instantiated infinitely often. But each such instantiation is the same exact finite sequence of human experiences. That is what defines "you" and "me" and every other person, at least as far as this argument is concerned. And that definition includes your finite lifespan. (Notice that if we remove the requirement that the sequence of human experiences defining a person must be finite, the argument no longer works--since if we allow infinite sequences of experiences, we now have an infinite number of possible human beings and it is no longer necessarily the case that any of them must be instantiated more than once even in a spatially infinite universe.) In other words, each one of the infinite number of "copies" of you in a spatially infinite universe lives exactly the same life, with exactly the same experiences, from start to finish. So having an infinite number of copies of "you" doesn't make you immortal; it just means there are an infinite number of copies of "you" each living exactly the same finite lifespan.

Note, btw, that the hypothesis of the universe being spatially infinite has nothing whatever to do with quantum mechanics or the MWI or "quantum immortality" (more on the latter below). It's a classical hypothesis.

if Quantum Immortality is possible

The only kind of "quantum immortality" that is possible (which is only possible if the MWI is true) does not mean "living forever", for the reasons already discussed in my subthread with @entropy1. In other words, the term "quantum immortality" is really a misnomer; it doesn't mean what it seems to mean.

 

[Physicuser]

Interesting, I have to think about it, thanks... If you or somebody else wants to make more remarks on this, please, do it.

Note, btw, that the hypothesis of the universe being spatially infinite has nothing whatever to do with quantum mechanics

Yes, I know it, I am talking about infinites because it seems to be the only way this could maybe work without MWI, since these events are almost impossible... But I'm not even sure, I mean, a ball tunneling a table with Copenhagen is totally improbable, but with infinit time it will happens, however, the probabilities of the ball tunneling continually throw the table up and down are infinite small, I guess, so I don't know.

 

[Demystifier]

No, the MWI does not say this. The MWI does not say people don't age, or don't die of old age. The MWI only says that, if a particular event which might or might not kill you depends on quantum uncertainty, there will be a branch where you live and a branch where you die. But it is highly unlikely that all potentially fatal events that happen to people as they age depend on quantum uncertainty. If there is no quantum uncertainty about an event, and it is fatal, there is no branching according to the MWI; you just die.

Yes, but! Just as there is a very very small quantum probability that a macroscopic body will tunnel through a macroscopic wall, one can argue that there is a very very small probability that a quantum fluctuation will reverse or significantly slow down the thermodynamic time arrow at a macroscopic level, in which case one can avoid death. No matter how small this probability is, it will happen in some branches according to MWI.

 

[PeterDonis]

there is a very very small probability that a quantum fluctuation will reverse or significantly slow down the thermodynamic time arrow at a macroscopic level, in which case one can avoid death

This assumes that such a fluctuation wouldn't also alter the structure of a human body to the point where it could not support life. I see no reason to believe this; in fact, I would expect the contrary, since the workings of the human body depend in many, many ways on the thermodynamic arrow of time working the way we normally see it work. Again, I don't think the MWI gives permission to wave one's hands and say that any possibility one can imagine must have a nonzero amplitude in the wave function.

 

[Jarvis323]

The actual argument goes like this: a human body is made of a finite number of particles and therefore has a finite number of degrees of freedom, so there are a finite number of possible "states" of a human body (which includes the brain, and we are assuming physicalism so all of your conscious experience is instantiated in your body and brain). A human has a finite lifespan, and if we assume that human experiences are finitely divisible in time, there are then a finite number of possible sequences of human experiences--each such sequence being a discrete sequence of a finite number of "states of a human body and brain", and therefore a finite number of human experiences. Therefore, in a spatially infinite universe, each one of the finite number of possible sequences of human experiences will be instantiated infinitely often (since any finite sequence of events will).

Each such possible sequence of human experience defines a "person"--so "you" and "me" and every other possible human person is identified with one such sequence. So in a spatially infinite universe, each possible human person is instantiated infinitely often. But each such instantiation is the same exact finite sequence of human experiences. That is what defines "you" and "me" and every other person, at least as far as this argument is concerned. And that definition includes your finite lifespan. (Notice that if we remove the requirement that the sequence of human experiences defining a person must be finite, the argument no longer works--since if we allow infinite sequences of experiences, we now have an infinite number of possible human beings and it is no longer necessarily the case that any of them must be instantiated more than once even in a spatially infinite universe.) In other words, each one of the infinite number of "copies" of you in a spatially infinite universe lives exactly the same life, with exactly the same experiences, from start to finish. So having an infinite number of copies of "you" doesn't make you immortal; it just means there are an infinite number of copies of "you" each living exactly the same finite lifespan.

There must be something more to the argument. I don't see any apparent reason, even in a spatially infinite, discrete universe with a finite alphabet, that a person needs to be part of the redundancy at all.

In fact, a spatially infinite universe means it is possible that no full configuration of the universe ever repeats. If the universe were discrete, and spatially finite, and describable with a finite alphabet, then we could say that at least one configuration must repeat eventually. That repeating configuration could just be a self loop on one state where nothing much happens at all (like a void going into a void).

An argument that a discrete person must be repeated, would depend on there being infinitely many people (or more than the possible combinations of symbols making them up). But even then, it doesn't need to be every person that would be repeated.

Wouldn't there have to be some argument, probably about chaos, or mixing, and homogeneity?

 

[PeterDonis]

An argument that a discrete person must be repeated, would depend on there being infinitely many people

No, you have it backwards. If there are infinitely many possible people, then no person will necessarily be repeated in a spatially infinite universe. The argument for there being multiple copies of any person in a spatially infinite universe depends on there being only a finite number of possible people.

Tegmark's paper on the Multiverse Hierarchy is here:

https://arxiv.org/pdf/0905.1283.pdf

Note that in this subthread we are only talking about Level I. Tegmark describes it in terms of Hubble volumes and predictions of inflation models, but I believe the underlying argument works out to be basically the same as the one I gave.

 

[Jarvis323]

No, you have it backwards. If there are infinitely many possible people, then no person will necessarily be repeated in a spatially infinite universe. The argument for there being multiple copies of any person in a spatially infinite universe depends on there being only a finite number of possible people.

Tegmark's paper on the Multiverse Hierarchy is here:

https://arxiv.org/pdf/0905.1283.pdf

Note that in this subthread we are only talking about Level I. Tegmark describes it in terms of Hubble volumes and predictions of inflation models, but I believe the underlying argument works out to be basically the same as the one I gave.

I said infinitely many people, not infinitely many possible people. I still think the argument you gave as presented says nothing about whether any people should be repeated at all, in any type of universe. I'll check out the paper though.

 

[PeterDonis]

I said infinitely many people, not infinitely many possible people.

Ok, but then there are infinitely many people in a spatially infinite universe. So your condition is satisfied.

 

[Jarvis323]

Ok, but then there are infinitely many people in a spatially infinite universe. So your condition is satisfied.

Like I said, even supposing there would be infinitely many people in an infinite and discrete universe with a finite alphabet, which there is no valid argument for, there is still no argument for why every person would need to be repeated.

 

[PeterDonis]

a spatially infinite universe means it is possible that no full configuration of the universe ever repeats.

The argument is not that a spatially infinite universe repeats the same person infinitely many times in time. The argument is that a spatially infinite universe has infinitely many copies of the same person in different parts of space.

 

[Jarvis323]

The argument is not that a spatially infinite universe repeats the same person infinitely many times in time. The argument is that a spatially infinite universe has infinitely many copies of the same person in different parts of space.

But there is no valid argument as of yet presented. There must be something missing, some evidence to warrant the speculation (it certaintly is not provable), like homogeneity and mixing or something like that. Otherwise, an infinite void with no people is perfectly fine.

Like maybe if we assume that space and the stuff in it here is like it is there, and mixing ensures that all the things that might pop up here will eventually pop up there too?

 

[PeterDonis]

there is no valid argument as of yet presented

I understand that this is your opinion. I was simply clarifying a point that you might have missed based on the statement of yours that I quoted.

maybe if we assume that space and the stuff in it here is like it is there

The argument assumes that our best current cosmological model of our universe is correct, if that helps.

 

[Jarvis323]

Ahh, it looks like this is the argument.

This quantum mechanism generates initial conditions that are for all practical purposes random, producing density fluctuations described by what mathematicians call an ergodic random field. Ergodic means that if you imagine generating an ensemble of universes, each with its own random initial conditions, then the probability distribution of outcomes in a given volume is identical to the distribution that you get by sampling different volumes in a single universe. In other words, it means that everything that could in principle have happened here did in fact happen somewhere else.

https://arxiv.org/pdf/0905.1283.pdf

 

[PeterDonis]

it looks like this is the argument

That's only the first part of the argument. Look at the continuation of that section on p. 4, left column, and footnote 5 on that page. That is where he makes the argument that the total number of possible Hubble volumes of the same size as ours is finite--which implies that, in a spatially infinite universe, infinitely many copies of each possible Hubble volume will exist.

 

[PeterDonis]

this is the same approach I was using to reason about MWI

The MWI is Level III in Tegmark's categorization. If you look at the section of the paper that discusses that level, you will see a number of similarities with Level I, yes.

 

[akvadrako]

(Note that I am ignoring the possibility of humans doing something to alter how our bodies work that either makes them age slower or undoes the effects of aging. It is certainly possible that, if we figure out how to do that, humans might live to be 200 years old or more; but that possibility has nothing whatever to do with the MWI. It depends on human ingenuity and motivation. Physics is not the discipline to go to if you want to make predictions about those things.)

That is the way QI likely works though – more so than the idea that you'll experience a normal life up until the point where your body is about to decay and die, then suddenly you'll get lucky and regenerate a bit.

It doesn't take that much ingenuity either. What we know causes branching, assuming QM is how the universe works, is running a quantum random number generator. People are regularly using that output to write documents and seeing if they provide interesting instructions. As long as it's physically possible, we can assume somewhere in the multiverse people know how to do it. Things like life extending drugs, cryogenics, brain upload machines or the seed to an AI that will build such things.

QI might not provide complete immortality; the heat death of the universe seems unavoidable. But if it's a real phenomenon it seems like it allows greatly expended lifetimes.