Thoughts from a conversation on quantum immortality
[Today, I participated in a conversation about the idea of quantum immortality. I decided to summarise some of the thoughts that came up in this short post. Therefore, it should be viewed as a report on a discussion rather than an attempt at a proper post.]
Assume that many-worlds interpretation is correct and quantum immortality is true. Essentially, in at least one universe you survive no matter what dangerous things you try. Since there is no defined biological “expiry date” on your body, you end up in state where your body just continues avoiding terminal shutdown. Your body is failing but the space of probabilistic events (such as a given organ failing, or a given blood vessel rupturing, or two given molecules interacting, or others, however minor) which lead to terminal shutdown is sufficiently large to last you for a while, with at least one universe where you still happen to be alive.
The above process probably takes a while(?). Until the entire space of events is explored (your body is finite) and you die in all universes. But in this case we don’t have quantum immortality, only “maximally delayed mortality” (MDM).
We can only observe quantum immortality with relation to ourself. Thus, in the universe where you survive, everyone else around you is likely to be dead, since the probability of two individuals surviving to this stage in the same universe is much smaller than the probability of you alone surviving. Therefore, you end up in an incapacitated state of continuous (eventually, lethal) failing of your body, completely alone in a universe where everyone else is dead.
The argument in the previous paragraph does not account for new people being born. So assuming no catastrophic event killing everyone else but you occurred, there may be other people in the universe where you are. But then you will not be in a state to appreciate that towards the end of your MDM.
So, if many-worlds is true, are we all going to end up experiencing a slow gradual fade-out of life just as we experienced a gradual fade-in?
Even a many-worlds literalist would require at least one more assumption: that there is a common time flow for all the universes, and that your self-perception is averaged over all of them using some reasonable measure. Note that this is very hard to make consistent with relativity, as in some of the worlds you are spirited away at near light-speed by evil aliens, and the measure of those worlds becomes relatively large once your “stationary” (in some sense) copies die out.
What matters, presumably, is the relative measure of worlds within the “branch” that you’re “already in”. Or, in other words, once you’re in branch a_1 (rather than b_1) at time t_1, the probability that you end up in branch a_{12} at time t2 is the conditional probability P(a_{12}|a_1) rather than the prior P(a_{12}) at time t_0. So it’s very unlikely to jump into a b-branch future from an a-branch past.
There is a subset of worlds in the branch “a” (or any other branch) where the aliens carry away your body at high speed. And this branch will eventually dominate any other subbranches (where you are still alive) of the same branch.
“Eventually” doesn’t do it. They have to dominate right then, at the moment of branching.
Let’s say that at 9:00 am you split into two branches, one where you stay on Earth, and one where you’re kidnapped by aliens traveling at high speed. At 9:05 am in the Earth branch, you fall off a cliff, and so most Earthly copies of you die, except for a few who get saved by some weird miracle. Suppose that at 9:30 am one of miraculously-saved copies attempts suicide. Again, that kills off most of the remaining Earth branches, but the few that remain have now experienced two miracles. Et cetera.
You seem to be saying that, eventually, after a sufficiently large number of miracles, the remaining branches have even smaller measure than the”alien” branch that diverged at 9:00 am, so therefore suicidal Earth copies at 10:00 am are likely to experience a sudden relativity-violating teleportation to an alien ship. And I’m saying no, that doesn’t follow, because of the survivor branches diverging at 10:00 am, most involve being saved by some relativity-obeying Earthly miracle, not teleportation to an alien ship. The overall greater measure of the 9:00 am alien branches is irrelevant to the 10:00 am Earth copies; what matters to the 10:00 am Earth copies is the relative measure of the branches splitting off on Earth at 10:00 am. Branches that split off before that don’t count.
Maybe I misunderstand the setup. I thought that in this model the state of one’s life at 10am is determined by averaging over all the remaining copies in all branches, including those on the alien ship from 9am. As you keep trying to suicide, you fade out of your existence on Earth and into the ship. Did I get the setup it wrong?
Very much so, as far as I understand. Indeed, if it worked the way you suggest, presumably we’d be fading in and out of places all the time as the wavefunction evolves, rather than having a single coherent conscious experience.
And how would we tell?
I’d appreciate if you point me to the setup you are working with.
I can’t math, but as I understand it, branches merge only when they happen to become identical. So, unless your earthbound suicide attempts end up putting you in a situation physically indistiguishable from your MWI-cousin on an alien ship, you won’t merge.
Interesting.
I was willing to go with the (completely unreasonable) assumption that “many-worlds interpretation is correct and quantum immortality is true” and see where it leads.
John begins to notice that he seems to be immortal. He has had a number of lucky escapes, and begins to push his luck more and more. Each time, he survives by some bizzare coincidence, and decided to publicise his talents. He calls all the major television networks, and soon the amazing surviving man manages to avoid the most deadly events.
Scientists across the world are stunned by this turn of events. Across the world, skeptics say that John is either incredibly lucky, or a hack. They predict that either one he’ll die in one of these stunts, or he’ll be disproven. John begins to do more dangerous, more public events, and continues to survive improbably each time. He sets up stunts that he simply shouldn’t be able to deal with, and walks out fine.
Of course, in many millions upon billions of other universes, the skeptics are proved right, and John dies, or is horribly injured, or just chickens out at the last minute. But in this one universe, he does the impossible.
There are versions of us which might live in those worlds and become 150+. However, most versions die earlier, which means that the large mass of weight of every person is in some everyday-state.
The fact that you’re not in palliative quantum hell right now is strong empirical evidence that if quantum immortality is true this interpretation is. (Whew!)
Only under SIA. Under SSA, most observer-moments with memories and experiences like yours haven’t just experienced 50 years of excruciating near-death.
These two statements are not equivalent. Quantum immortality refers to the hypothetical (supposedly inevitable) subjective experience of ever-continuing life, the latter sentence seems to describe the continuing existence (in atleast one branch) of some person you could currently consider to be a future version of your current self.
I believe in the latter, but not in the former. There’s nothing mystical and immortal about my self-perception of continuity, or memory, that it should persist in the branch that “I” follow; even if we limit ourselves to the branches where there’s some continuing qualia-producing processes, why should the sense of self be the process that remains functioning last?
In my opinion, if the many-worlds interpretation of quantum mechanics is correct, we will certainly experience inmortality in some way. But what I imagine is very different from what you are picturing. Branching of universes doesn’t have to happen in dichotomic life/death situations, it would be happening in every moment, thousands of millions of times per second, as all the fundamental processes happening everywhere are quantum.
At this moment, some quantum process in my neurones has decided whether I was typing this sentence now or a milisecond later. Two universes have been created, one in which I typed it at 18h24m05,000s and another one in which I did it at 18h24m05,001s. None of the copies of myself will ever communicate between each other so all of them feel they are still living in the only possible universe.
Of course this take us to endless possible universes, and in an infinity of them I will live forever. You are saying our body is finite but that doesn’t have anything to do with it. For example, to not get old, it just takes to stay in the branches in which the irreversible biological processes that make us decay, which are quantum at their microscopic nature, don’t happen. There is a ridiculously small but non-zero chance that that happens, so the branching will occur, and it could keep occuring forever with millions of variations within it. In some universe I will be 1000 years old and have a Coke and in another one I’ll be having a Pepsi.
Am I my future copies? I wouldn’t say so, but all of them are me. The copy of me whose quantum working brain decided not to write this comment ten minutes ago is not me anymore, but we both are the “me” that existed ten minutes ago. The copies of you people in that other universe won’t have to go through this boring comment.
MWI is the memetic uniform, worn by many on this list. Years ago, some like cyphergoth and Yudkowsky behaved quite nasty when I wanted to object against it.