Critique of the quantum suicide experiment

[Demystifier]

At another thread, it has been pointed out that Tegmark said the following:

• a quantum suicide is the only experiment he can think of that allows us to distinguish between MWI and Copenhagen.
• it will only convince the person in front of the gun, not the person behind it.
• the probability that the person in front of the gun will hear a click (indicating that the gun isn't going to fire) is 1.

Here I want to criticize it. More precisely, I want to show that the suicidal guy cannot be objective about his own chances for survival. Therefore, since his own conclusions cannot be objective, he cannot objectively conclude that his experiment confirms the many-world interpretation.

To understand this, for the moment forget about quantum mechanics. Instead, assume that we have a CLASSICAL gun that kills with the probability 50% (classical Russian roulette). We give that gun to a suicide person who wants to kill himself, but we do not tell him what the probability of killing himself is. We ask him to try to kill himself as many times as he wishes and ask him to tell us his conclusions. No matter how many times he tries to kill himself, he will either not exist as a person which can make any conclusion (because he will be dead), or will find out that he survives whenever he tries to kill himself. Thus, he will either conclude nothing, or will highly overestimate the probability of surviving. There is no chance for him to conclude by this experiment that the gun kills in 50% cases. If he will conclude anything, it will be that the probability of surviving is larger than 50%.

But we know that this conclusion is wrong. So we must conclude that conclusions about probabilities for surviving found by someone who attempts suicides and survives - are unreliable. So we cannot trust even our own conclusions about our survival probabilities in our own attempted suicides. In this case, we are biased observers who cannot make objective conclusions about probabilities for survival. Subsequently, we cannot make objective conclusions about anything which otherwise could be concluded from those probabilities.

So far I was talking about classical probabilities, but quantum probabilities are not much different. If I try suicide many times and I survive each time, I cannot objectively conclude anything about my probabilities for survival. Subsequently, I cannot objectively conclude from that experiment that the many-world interpretation is correct. If I am objective, I know that I am a biased observer, so I know that a naive conclusion that the many-world interpretation seems to be right - is in fact an invalid conclusion.

To conclude, objective suicide persons cannot use the quantum suicide experiment to convince themselves about the validity of the many-world interpretation. That's because they are biased observers, so cannot be objective about estimating probabilities.

In other words, the first Tegmark's statement above is incorrect.

 

[Fyzix]

I already debunked this in the other thread.

If some hypothetical model of MWI could be made to work, it would have Born Rule and thus the exact same probabilities as any single universe model (bohm, copenhagen, other hidden variable etc.).
So if you try to commit a quantum suicide and you survive, this would just confirm Born Rule, not whether MWI is more likely or less likely, if that had been the case MWI would be required to not obey quantum probabilities.

If surviving the quantum suicide is 1% and dying is 99% you surviving this would only confirm that you were extremley luckily, this would also happen 1 out of 100 times in a single universe model.
The probabilities doesn't change, so you could do this to infinity and it would only confirm QM and Born Rule, not MWI at all.

Jacques Mallah has also written a good paper on the whole illusion that you PERSONALLY survive a quantum suicide:
http://arxiv.org/ftp/arxiv/papers/0902/0902.0187.pdf
(this is a must read for anyone who believes in this fallacy, as it is potentially life threatening delusion amongst people who believe in MWI)

I also seem to remember that Don Page retracted his belief in the validity of QS because he realized what I described above.

 

[xts]

Frankly: I got a bit bored by Mallah's paper - he could present the same ideas on two pages...
I can't agree with his argumentation. He redefines 'personality' in a way far from common meaning and from our introspective sense, by introducing strange concepts like 'total amount of consciousness' (recombining all Everett's worlds back together), while he misses major Tegmark's point:
- in MWI there always exists some worlds with the experimenter alive;
- in single-world interpretations, consciousness of the observer often totally ceases.

(this is a must read for anyone who believes in this fallacy, as it is potentially life threatening delusion amongst people who believe in MWI)

I don't think this threat is so serious. Vast majority of people believe in some kind of afterlife immorality (usually in worlds better than ours) and only few of them is led to suicide by those religious delusions.

If surviving the quantum suicide is 1% and dying is 99% you surviving this would only confirm that you were extremley luckily, this would also happen 1 out of 100 times in a single universe model.

You are missing the same Tegmark's point as Mallah does: MWI implies that you are always extremely lucky in death situations - your consciousness always survive. While on single world your consciousness may just stop to exist.

 

[Demystifier]

Since nobody commented my argument, let me add a clarification:

I DO think that many-world interpretation implies quantum immortality.
Yet, I argue that a rational experimentalist conducting the quantum suicide experiment cannot conclude that the result of experiment is an evidence for the many-world interpretation.

In fact, if you are immortal (for whatever reason you are so, which may have nothing to do with many worlds), you cannot prove it to yourself by a Russian-roulette experiment, provided that you are rational and understand the theory of probability and sampling.

 

[xts]

In simpler words, if you are immortal, you cannot prove it to yourself by an experiment

Yes, and no. You can't prove it with absolute certainity. But you can do it with any marginal error you like. Just perform the suicidal attempt with this marginally low chance to survive.

The paradox is opposite: you cannot prove to yourself that you are mortal!

 

[Fyzix]

Frankly: I got a bit bored by Mallah's paper - he could present the same ideas on two pages...

Well, if you didn't get it in that many pages I doubt 2 would do the trick...

I can't agree with his argumentation. He redefines 'personality' in a way far from common meaning and from our introspective sense, by introducing strange concepts like 'total amount of consciousness' (recombining all Everett's worlds back together), while he misses major Tegmark's point:
- in MWI there always exists some worlds with the experimenter alive;
- in single-world interpretations, consciousness of the observer often totally ceases.

Ok let us break it down in simple terms if MWI had been true, you would be like a amoeba always "splitting" into new versions of yourself, all of which are the "original" in a sense.
Now imagine that they did some sort of experiment where 9 of the future splittings of the "original" amoeba will die and 1 will live.
9 actually ****ing dies and 1 survives.
It is really that simple. The consciousness of these amoebas do not magically transfer over to the surviving amoeba, so 9 will infact DIE just like in a single universe model and 1 will survive, get it?

If I had a flash light and conducted an experiment where in 9 out of the 10 worlds the flashlight would go out of battery and thus stop shining the photons will not magically jump into this single universe where the flash light still works.
it is THAT simple.

A single last example: if MWI had been true this would be true for everything.
If in another universe right now a brick forms right over your head and falls on top of your head, YOU in this Universe does not experience that pain, because it doesn't happen to YOU, it happened to your twin.
Just like death in a parallel universe!
People need to stop thinking that consciousness is a soul that could just jump from universe to universe...

I don't think this threat is so serious. Vast majority of people believe in some kind of afterlife immorality (usually in worlds better than ours) and only few of them is led to suicide by those religious delusions.

Because in most religions suicide is also what they call a "sin", a guaranteed into whatever made up hell they believe in.
If you believe that MWI is true and that it implies quantum immortality for you personally, there is no reason for you not to play quantum lottery and continue killing yourself until you end up in a branch where you are a billionaire...
So yes QS/QI belief (which is a fallacy) is dangerous...

You are missing the same Tegmark's point as Mallah does: MWI implies that you are always extremely lucky in death situations - your consciousness always survive. While on single world your consciousness may just stop to exist.

A consciousness survives, not YOUR consciousness survives.
Consciousness is nothing but interactions in a ****ing brain, it's not a soul that magically jumps to the next universe.

Luckily MWI is most likely wrong and people are finally starting to realize this so this falsehood will most likely die out before it claims too many lives.

 

[xts]

People need to stop thinking that consciousness is a soul that could just jump from universe to universe...[...]A consciousness survives, not YOUR consciousness survives.

In MWI consciousness is not 'jumping' but 'splitting' or 'forking'. No consciousness ever dies. It sometimes forks, and sometimes does not fork. Even in an example with heavy wound as an outcome of russian roulette, in some worlds the medical help comes quickly enough to stop bleeding. Of course - heavy wound leading to spending the rest of your (still immortal) life on a wheelchair - is definitely not a desired (but possible) outcome of russian roulette experiment in MW.

If you believe that MWI is true and that it implies quantum immortality for you personally, there is no reason for you not to play quantum lottery and continue killing yourself until you end up in a branch where you are a billionaire...
So yes QS/QI belief (which is a fallacy) is dangerous...

Fortunately, religious and metaphysical beliefs are not so dangerous. Except of few fanatics of various religions, majority uses common sense rather than follow logical implications of their beliefs.

ADDED:
By end of 19th century it was quite popular to bet the whole property on horse racings and shot himself in case the bet failed. Do you think all of them were MWI precursors?

 

[Fyzix]

In MWI consciousness is not 'jumping' but 'splitting' or 'forking'. No consciousness ever dies. It sometimes forks, and sometimes does not fork. Even in an example with heavy wound as an outcome of russian roulette, in some worlds the medical help comes quickly enough to stop bleeding. Of course - heavy wound leading to spending the rest of your (still immortal) life on a wheelchair - is definitely not a desired (but possible) outcome of russian roulette experiment in MW.

I don't even understand how you don't understand it...
That puzzles me more than this whole QS puzzle.
There is nothing "magical" about life, it's just processes, they can stop and then you are dead, consciousness does not jump over to the other branches.

If I conduct a quantum blind-myself experiment, sure there will be branches where successors of me will against all odds not go blind, but there is still a huge chance that I will become blind on one eye, I won't jump over to the branch where I'm not blind...

Another comment by Tegmark on this whole thing seems to suggest even he isn't 100% sure you are guaranteed immortality.

In response to questions about "subjective immortality", Max Tegmark made some brief comments: He acknowledged the argument that "everyone will be immortal" should follow if a survivor outcome is possible for all life-threatening events. The flaw in that argument, he suggests, is that dying is rarely a binary event; it is a progressive process. The quantum suicide thought experiment attempts to isolate all possible outcomes for the duration of the thought experiment. That isolation delays decoherence in such a way that the subjective experience of the superposition is illustrated. It is only within the confines of such an abstract quantum scenario that an observer finds they defy all odds.[6]

ADDED:
By end of 19th century it was quite popular to bet the whole property on horse racings and shot himself in case the bet failed. Do you think all of them were MWI precursors?

Lol, no :P
Very very few actually believe in MWI at all

By the way, you should send a email to jacques mallah (email on top of the paper I linked you to earlier in the thread) and have him participate in this debate and explain his paper in more detail.

 

[xts]

There is nothing "magical" about life, it's just processes, they can stop and then you are dead,

That is an external (objective) point of view. From subjective, conscious, introspective point, you are never dead.

consciousness does not jump over to the other branches.

I think you miss Everett's idea. I'll repeat: 'fork', not 'jump'. At every event the world splits into two (more), and all copies of me are the same ME at the very moment of split, then they start to continue their own life, inheriting my personality and my history. In case of suicidal attempt the split does not occur: my personality continues only in one branch: where the gun misfired.
That's a huge difference to quantum-blind experiment: in this case your personality splits and one of yous may say and think: "****, I lost my eye!"

Dying slowly is only minor problem to MWI(consciousness) QS interpretation. You just should use the killing machine acting quickly enough to be unnoticed by mind. I believe the shockwave caused by magnum .44 bullet penetrating your brain is sufficiently quick.
Or e.g. - put yourself into the narcosis, and then apply either wake up cure or poison.

 

[Fyzix]

You still don't understand it!
We are talking about POST-FORKING, after the fork/branching takes place you can't jump from branch 1 to branch 15 because you died in branch 1 and survived in branch 15

I give up... Please contact Jacques Mallah and bring him into the discussion, he will clarify way better than i can.

 

[xts]

I doubt if J.Mallah can explain it, as I understand your argument, but I don't see it valid.

From external observer view, before the experiment we have one world with my conscious presence and after the experiment we have two worlds, one with me alive, and one with him mourning over my dead body.

From my perspective ('my' = successor of my current consciousness) before we have one world, but after we also have one world with me happy and alive.

No "post-forking-jumps" are needed for such explanation.

You helped me to realize, that QS experiment is convincing only for atheistic MWI believers. Religious experimenters, in many cases, would find themselves in heaven (or rather hell - punished for suicide).

 

[Fyzix]

You make no sense.

And that bs about "Only atheistic" wtf?
No, sorry its the other way around, you need to postulate a magical branch jumping consciousness to make Quantum Immortality valid.

Just contact Mallah and see what he has to say, give him a chance to defend his paper.
I'm sure he is more than glad to defend it and answer any critique you have of it.
It would be helpful for other people who later on read this thread too.

 

[Demystifier]

Yes, and no. You can't prove it with absolute certainity. But you can do it with any marginal error you like. Just perform the suicidal attempt with this marginally low chance to survive.

No. As I explained in the first post, in that way, if you will conclude anything, you will conclude that your chances to be immortal are very high, even if in reality these chances are very low or zero. Therefore, the conclusions obtained in this way are not reliable. Therefore, if you are rational and understand this, you cannot conclude that in the first place.Consider also the following game. (It's a variant of the game above, but nobody dies.) You flip a coin. If it is head, the result is not shown to you and the game stops. If it is tail, the result is shown to you, after which you flip the coin again with the same rule. It is repeated, until eventually it is head in which case the result is not shown to you and the game stops. You play the whole game only once. But others can independently play the same game as well. They all tell me their final conclusions about the coin. How many of them will tell me that the coin is probably biased? How many of them will tell me that the coin is probably not biased? How many of them will not be able to tell? Whom should I believe?

Hint: The problem with that game is that the rules are such that the OBSERVER is biased, rather than the coin.

 

[xts]

@Demystifier:
I think you think in terms of 'objective reality', where MWI must be thought of in somewhat solipsistic way.

Conclusion are not reliable for others. Even if you meet a guy, who already survived 1000 russian roulette attempts, you'll probably (as a honest experimentalist, but cruel man) ask him to repeat it 10 times more. Then you'll probably (1023/1024) go to his funeral being convinced, that the guy was mortal. But the guy is still alive in this small branch, pretty sure that if he survived 1010 attempts, he must be immortal. If he is not quite convinced yet - he'll make another 1000 attempts. For him - he is immortal.

Stanisław Lem wrote a story (one of "Cyberiad" series) about 'probability amplifier' - bringing even very unlikely events to reality (dragons still occurred more often than fairies, as just dragons are a bit more probable). In Single World we have no way to build such "probability amplifier". In Many Worlds the suicide experiment is the only available: it filters out all those worlds, where you are not conscious, so you may purposely construct very unprobable scenario (but the only you may survive) and find yourself at its end, encountering dragons and other phenomena you may explain only as miracles. Thus: miracles proves MWI (unless someone prefers religious explanation). But in MWI it is you, who may create miracles - you don't have to rely od God's good will to organise them.

The MWI of suicide experiment is as convincing, as explanation of happy life on Earth with the background of billions dead planets. It explains why something very unlikely to happen really happened.

 

[Demystifier]

@Demystifier:
I think you think in terms of 'objective reality', where MWI must be thought of in somewhat solipsistic way.

Not really. I think in terms of probability theory and biased observers, independent on quantum mechanics, objective reality, many worlds and solipsism. Please see my coin game in the previous post.

 

[xts]

I fully agree with you in the following:
1. your interpretation of the coin game - coin is not biased;
2. full equivalence of 'quantum' and classical (revolver gun) suicide.

But the point of QS experiment is not to show that your revolver is biased, nor that probabilities of the repetable experiments are not statistically verifiable. Its point is that you may design an experiment, whose result is a priori very unprobable, which may be extended and prolonged, but still giving subjectively more and more unprobable results.

Mere existence of your consciousness in the world where something very unprobable happened, leads either to religious explanation (miracle - God's intervention), or to acceptance of Many Worlds. The difference between your coin experiment and MW-suicide is that even if you take all people on Earth the experiment will stop after about 1000 turns. In MW-suicide - you may continue it as long as you like, and you always get tails and tails, even as you are able to examine a coin and make test flips (which gives about 1/2 count of heads).

In my view MWI (and suicide experiment) just exploit the paradox of infinity - every possibility mst be realized. In single world there is nobody who may say: I flipped coin 1,000,000 times and got all tails, while in MW you may become such person in one of 2^1,000,000 parallel worlds.

 

[Demystifier]

Mere existence of your consciousness in the world where something very unprobable happened, leads either to religious explanation (miracle - God's intervention), or to acceptance of Many Worlds.

If we accept that it is rational for the suicide guy to conclude from this experiment that MWI is probably true, then we have a paradox. The paradox is the fact that his neighbor witnessing all these unsuccessfull attempts of suicide will NOT interpret the same statistical data as evidence that MWI is probably true. How would you resolve this paradox (if not in my way by arguing that the suicide guy also should NOT conclude that MWI is probably true)?

 

[xts]

Scenario: your colleague already survived 1000 shots, and you are not still convinced. So you ask him to make 10 more. Now we go to that 1/1024 world, in which you witness the guy is still alive. In this world you also saw a miracle. So you'll probably are convinced. If not - ask him to make another 20 atempts (would 10⁶ narrowing of the margin convince you?). And again - take the world, where he survives. In this world you are now fully convinced that the miracle happened.
The difference between your, and your colleague views is that your consciousness in vast majority of cases see his funeral, while he never sees that.
So - in those rare small branches of many worlds, where you witness your colleague successfully playing russian roulette - you should also got convinced.

But after the second additional test - your consciousness may get convinced with only 2^-30 probability, while his - with 100%.

 

[Demystifier]

So - in those rare small branches of many worlds, where you witness your colleague successfully playing russian roulette - you should also got convinced.

No, that's totally wrong. Indeed, as indicated by the second item in the first post on this thread, the inventor of quantum suicide explicitly said the opposite. Even if the many-world interpretation is correct, it cannot offer an explanation for the colleague which would explain why the suicide guy always survives. It is still a very unprobable event for him. Hence, the colleague cannot conclude from such a result of this experiment that MWI is probably right (even if in fact it is). That's because the colleague, unlike the suicide guy, has not an a priori reason to believe that he will always be in the branch in which the suicide guy survives.

 

[xts]

He may convince a collegue, if he allows him to plan the experiment.
We may bet in less bloody way:
You don't believe in God's intervention, so if 1,000,000 coin tosses show all tails, then you both believe we are in a so unprobable place, that only analogue of anthropic principle may explain it. MWI is such analogue.
If something like this happens (which is rather unlikely: 2^-1,000,000) - you are both convinced.
Now change the coin tossing to russian roulette. You (witness) are getting convinced again with probability 2^-1,000,000. While your colleague gets convinced with 100% probability.

 

[Demystifier]

No, he can't. Or at least, not if he understands probability theory in the same way I understand it. And I am pretty confident that my understanding of probability theory is correct. But I guess you are equally confident that your understanding of probability theory is correct, so it seems that we can't make any further progress in this discussion (unless someone else involves it too).

In the meanwhile, you can test your understanding of probability theory by thinking about the raven paradox:
http://en.wikipedia.org/wiki/Raven_paradox
It's not directly related to the problem we discuss here, but it also has to do with experimental data which do or do not support given scientific theory. It is a good exercise, and a failure to understand it could be a sign that one should work more to improve his understanding of that stuff.

 

[xts]

I think it is not a difference of our views of probability, but rather how they relate to Many Worlds. You take probability as externally objective. While I try to advocate Everett's subjective view (I don't share it by any means, btw, I just take Everett's postulates as an assumption for tricky logical game...). In MW terms you see probability as a ratio of numbers of all worlds in which some event happens to all worlds, while Everett suggests that it is restricted to those worlds, where your consciousness is present.

Wiki improperly describes raven paradox: this (4) statement is logically equivalent to (1) all ravens are black is false (maybe that's Hempel's fault). It is not equivalent - it is only implied by: (1) => (4). The same (3) (my raven is black) is not a proof - it is just a single sandgrain in an inductive justification of the claim (1) that 'all ravens are black'. Similar sandstone is added by showing non-black (green) non-ravens (apples). The existence of green apples makes sense of speaking about colours. If there wasn't any objects like green apples, the word 'black' would be meaningless. Thus, existence of green apple makes the sentence 'all ravens are black' meaningful.

ADDED satisfactory resolution should also explain why there naively appears to be a paradox
Because:
1. our use of language puts stress on nouns, rather than adjectives;
2. ravens form the pretty small part of the world known to us;
2a. black is pretty popular colour (and non-black colours are even more common)
thus we take the sentence 'all ravens are black' as expressing some property of ravens (they are black), rather than about non-black items (they can't be ravens), and we take an example of green apple as unrelated to ravens we were speaking about.

 

[xts]

You flip a coin. If it is head, the result is not shown to you and the game stops. If it is tail, the result is shown to you, after which you flip the coin again with the same rule. It is repeated, until eventually it is head in which case the result is not shown to you and the game stops. You play the whole game only once. But others can independently play the same game as well. They all tell me their final conclusions about the coin. How many of them will tell me that the coin is probably biased? How many of them will tell me that the coin is probably not biased? How many of them will not be able to tell?

That's a great question!
I'd passed it to my friend - teaching Psychology at provincial college. She's eager to make an experiment like this (and - of course - with fake series rather than fair coin) with her students, but we must wait over a month for results (students have vacation till end of September)

 

[Ken G]

I DO think that many-world interpretation implies quantum immortality.

I agree with your argument about the biased nature of the observer, so I wanted to focus on this statement instead. I'm not sure what you mean here by "quantum immortality", but it seems to me that too much can easily be read into this concept. First of all, we must distinguish a case where there could be a consciousness, somewhere in the many worlds, that are older than any age you can mention, from the concept of "immortality." Immortality generally means you could point to a given individual or consciousness, and say that that individual or consciousness will never die. But that usually comes with an expectation that we are also specifying the world-- that individual will never die in this world. Of course it doesn't mean that here, so we have to ask, what else doesn't it mean?

The first thing I would say it doesn't mean is that a given individual, say you, will live forever. To say otherwise would be tantamount to claiming that you personally have multiple existences, in all these other worlds. But you don't perceive those other existences, so what is it about them that make them you? Their genes? That doesn't seem to be what a person is. Their memories or personalities? Those will all vary. It may even be hard at some point to find more similarity between you in all those other worlds, and your next door neighbor in this world. So either your consciousness and identity are singular to this world, or else all consciousnesses are just as much "the same person" as yours in the other worlds. (In other words, if my consciousness is constantly branching, and only one continues to be me, then who's to say the one that survives the suicide is still going to be me.)

Now we can certainly argue that a consciousness is just a consciousness, and that identity is a kind of illusion based on a self-image we create in any given instant using whatever memories are in our head, rather than something that follows our consciousness around making it "our" consciousness. But just like your argument about the false "test" of the MWI, if we take this approach to what consciousness and identity are, we have lost any contact with MWI or quantum mechanics-- a perfectly classical universe, or a Copenhagen universe, could contain infinite consciousnesses in infinite space with no contradiction. If so, they would also always have consciousnesses that survive the various calamities of each of their worlds, and so if consciousness survives, then so does something that is conscious, and all that I am is something that is conscious.

Now you might argue that the QS position focuses on a consciousness that shares the memories that I have now, but that won't hold up, because there's nothing in quantum mechanics that says I have to remember my youth. What if my constant surviving of all these near-death experiences has so degraded my mind that I no longer remember anything, am I still immortal? How much do my mental faculties have to diminish before I need to start looking to some other "branch" to find the immortal me?

And finally, we must certainly all hope that the QS promise of immortality is bogus, because if it is really true that the consciousness I call my own will indeed survive all these near-death experiences, and become some spectacularly unlikely and highly selected old consciousness, it is almost certain that this eternity I have to look forward to would be a vegetable-like existence at best, and a living hell at worst. The body deteriorates, the mind gets fuzzy, and most people over 100 are in considerable discomfort a lot of the time. Those are just the people in our world, what about that rare 200 year-old that we don't have any of? What about the 500 year old, what kind of life is that going to be? Imagine a life so useless and helpless and infirm and sickly that you want more than anything to die, and cannot, because of that darn QS. A bleak promise, that!

And my final criticism of the QS viewpoint is that since it is fundamentally a philosophical viewpoint that we agree cannot be tested, it should be held up to the standard of a philosophical stance. A general rubric for a philosophical stance is that it should advise us in how to live. How does QS tell us we should live? It is a philosophy that tells us we can put a gun to our heads and pull the trigger, and we won't die. What would that feel like, pretty exhilarating yes? Great, we have a philosophy that tells us we can all get our kicks with firearms at our temples-- does that really sound like a reasonable philosophical stance?

 

[xts]

(In other words, if my consciousness is constantly branching, and only one continues to be me, then who's to say the one that survives the suicide is still going to be me.)
First: both are 'you'. Second: in case of QS your consciousness see no branch.
QS is based on a kind of tautology: it says that you are alive in all worlds in which your consciousness is present. (As only those worlds are available to you) This tautology is combined with a consequence of MWI (or 'infinite universe'), that every arbitrary small, but non-zero probability is realized. Thus, there always exist some worlds with you alive and conscious.

What if my constant surviving of all these near-death experiences has so degraded my mind that I no longer remember anything, am I still immortal?
Pure QS experiment says about killing machine, which either kills you quickly and painlessly, or leaves you totally unharmed. If you consider other calamities - of course, there will be lots of worlds populated with barely alive yous, vegetating on wheelchairs or at intensive therapy hospitals.

[...]A bleak promise, that!
You are totally right. Gulliver traveling to Luggnagg already noticed that promise of immortality is not a nicest perspective. But if some consequence of some assumption is sad or nightmarish, that is not an argument neither pro nor cons validity of that assumption.

A general rubric for a philosophical stance is that it should advise us in how to live.
That's a veeeery narrow view what philosophy is for. Majority of philosophy, especially metaphysics, gives no such advices.

It is a philosophy that tells us we can put a gun to our heads and pull the trigger, and we won't die.[...] does that really sound like a reasonable philosophical stance?
For me (atheist) it is not reasonable. But I admit it is not less reasonable that those of many religions, promising that after death your consciousness would continue eternal life in some kind of heaven/hell, or will be reincarnated.

 

[Ken G]

First: both are 'you'.

In that case, how are you defining "me"? Apparently it is anything that branches from the current me. But I don't perceive any of those branches, only the one that is writing this. So just what is my relationship to those other branches, and should I care if any of them die in the next minute? Let's say you say that I should not care if any of them die as it won't affect me-- well then they are not me. Or you could say I should care about all of them because they are all having "my experiences", I just don't remember them. Then what separates them from all the other consciousnesses around me right now whose memories I also don't share? Genes?

Second: in case of QS your consciousness see no branch.

Certainly I agree, but that is the whole point. If I see no branch, then what is your basis for associating anything in that other branch with me? Their memories will soon be different from mine, their thoughts and personality will soon be different, their bodies might not even look like me eventually. All that remains is genes, which hardly seems like a good basis for what is "me."

That's a veeeery narrow view what philosophy is for. Majority of philosophy, especially metaphysics, gives no such advices.

In this case, it certainly does-- that's the problem with it.

 

[xts]

In that case, how are you defining "me"?

That is a paradox of MWI - in my opinion one of the most important its flaws.
MWI, in order to get rid of problem of our experience not facing superpositions of states, introduces forking of consciousness - something equally weird to our experience.

But MWI 'me' may be defined as any copy of current 'me'. The copies separates as they start to differ: something happens (and got remembered) by one of them, but something else to other one.
'Me' is 'me now', 'mine history', but not 'mine alternative history started/forked some time ago'

 

[Ken G]

That is a paradox of MWI - in my opinion one of the most important its flaws.

It's not really a paradox of MWI, MWI needs take no stance on what "me" means, or many different stances are consistent with MWI. But it is certainly a problem with QS, because QS requires a concept of "me" to survive the suicide. What I mean is, MWI destroys the very class of definitions of "me" that QS relies on, so the two seem incompatible with each other.

MWI, in order to get rid of problem of our experience not facing superpositions of states, introduces forking of consciousness - something equally weird to our experience.

Yes, I agree-- solving one problem with another.

But MWI 'me' may be defined as any copy of current 'me'. The copies separates as they start to differ: something happens (and got remembered) by one of them, but something else to other one.
'Me' is 'me now', 'mine history', but not 'mine alternative history started/forked some time ago'

The problem there is that we are not talking about copies of me-- we are talking about me's with a bullet in their temple, and me's without a bullet in their temple. Hardly copies! The point is, it doesn't matter how quickly or slowly the changes begin to diverge-- when divergence happens, we have to be able to say "which one is me" if we will claim that I will survive the suicide.

 

[yoda jedi]

what about consciousness in artificial systems, i mean for discursive insight...

 

[Ken G]

It seems it wouldn't matter if the consciousness was artificial or not, because we aren't really even talking about the attributes of other consciousnesses-- as long as we ourselves are not trying the suicide, we can conclude nothing about consciousnesses. If people really believed the QS contention, then we nonbelievers would just see a bunch of QS believers dying right and left. The only way anything at all strange happens (other than the strangeness of people killing themselves for their beliefs, often a kind of cult behavior rather than an example of a responsible philosophy) is if we ourselves try to commit suicide. So there is not much anyone can really say, unless it is "I tried to kill myself 100 times and failed them all"-- and of course the chances of anyone like that being in our universe is miniscule because we'll get all the ones who succeed. So the bottom line is, it can't be from anyone else, it has to be from you-- you the person in this discussion. This is purely a form of self-knowledge we are talking about, no one can do this for you. That's what suspends the rules of science-- it is a form of science that allows you to draw only from your own experience to test it (bringing in Demystifier's questioning of your objectivity if you try the test). But even that lack of objectivity is somewhat moot, because the simple fact is, you have not tried to commit suicide, and no amount of discussion can replace the knowledge you can only get from actually trying it. That's the problem-- it is a terrible philosophy, it forces you to do something incredibly stupid if you want to test it.

 

[Delta Kilo]

This whole topic is a red herring. You don't need MWI to show that. Imagine you have an army of volunteers. At your command they go into individual sound-proofed rooms and start shooting themselves. Those lucky few who survive may well consider themselves invincible. This is no different from multiple copies of the same person in MWI. So what?

 

[Demystifier]

This whole topic is a red herring. You don't need MWI to show that. Imagine you have an army of volunteers. At your command they go into individual sound-proofed rooms and start shooting themselves. Those lucky few who survive may well consider themselves invincible. This is no different from multiple copies of the same person in MWI. So what?

Excellent point! But note that I have also presented two analogies in classical probability without many worlds. And your example also reinforces my central conclusion: Even if MWI is right, the suicide experiment cannot be used to prove it.

 

[Ken G]

This whole topic is a red herring. You don't need MWI to show that. Imagine you have an army of volunteers. At your command they go into individual sound-proofed rooms and start shooting themselves. Those lucky few who survive may well consider themselves invincible. This is no different from multiple copies of the same person in MWI. So what?

I think the key difference is that, you cannot use an army of volunteers to test MWI, we all agree there. You can only test it yourself! And for that, you need no army at all, you just need one gun, and a few bullets. If you're still here after that, MWI is clearly right, and so is the QS interpretation of MWI that there is such a thing as "me" that forks along with the experimental outcomes. I don't recommend you try this, because I don't think you will make it. That's because I think MWI and QS cannot both be right-- they are having a little fight over the meaning of "me." MWI must reject the entire concept of "me", because I'm forking all the time, yet only perceive one of me, so the whole "me" concept is a kind of illusion in MWI (a weird kind of illusion that passes every test, which is not normally what we call an illusion). But QS invokes the same "me" concept that MWI just destructed, to argue that "I" will survive the suicide. So banking on not ending your existence is like betting that two different horses will win the same race.

 

[Delta Kilo]

If you're still here after that, MWI is clearly right, and so is the QS interpretation of MWI that there is such a thing as "me" that forks along with the experimental outcomes.

I don't see how is it going to prove anything at all except that I am one lucky son-of-a-gun.

 

[Ken G]

In QS, you will eventually be 1000 years old. I think that will tell you something.