6 ms·

That paradox implies that it's possible (even theoretically) to "records your molecular composition" and then "re-creates you ... each atom in exactly the same

by Miiko 5y ago

That paradox implies that it's possible (even theoretically) to "records your molecular composition" and then "re-creates you ... each atom in exactly the same relative position". However, my understanding is that such things are not possible because of quantum Uncertainty principle - one cannot simply discard atom's momentum, as that would kill person being recorded, IMHO. Doesn't that solves the paradox, does it?

alisonkisk 5y ago
Sure, "it's impossible" is always a resolution to a paradox. But not a very interesting one.
NoblePublius 5y ago
In Star Trek, this is handled by the Heisenberg Compensator https://memory-alpha.fandom.com/wiki/Heisenberg_compensator https://memory-alpha.fandom.com/wiki/Heisenberg_compensator
superposeur 5y ago
More precisely, the “no-cloning theorem” of quantum mechanics implies that one cannot replicate the exact quantum state of a clump of matter in a second clump of matter (although it is possible to “teleport” the exact quantum state by transmitting its information, provided one is willing to mess up the original). However, although no one knows for sure, it seems likely that the operations of consciousness are purely “classical” in the sense that the exact quantum state is not needed (Roger Penrose’s ideas not withstanding).
- pontus 5y ago
  Turns out that while the no cloning theorem does prohibit cloning, it does not prohibit perfect replication as long as the original is destroyed. This is the premise of quantum teleportation. Using some clever applications of entanglement it is possible to perfectly replicate the quantum state of a particle in a different location, it's just that the original particle is necessarily destroyed in the process. So, in short the no cloning theorem does not prevent quantum teleportation. In this sense, the follow-up in the paradox where you create 10 copies all of which are identical would not work. Quantum mechanics effectively ensures that there's ever only at most one of you.
  - superposeur 5y ago
    Yes, hence my parenthetical phrase: > … (although it is possible to “teleport” the exact quantum state by transmitting its information, provided one is willing to mess up the original).
    - pontus 5y ago
      When I first wrote my comment, this part of your comment was not there. You must have added it after the fact (hence my comment)
      - superposeur 5y ago
        Definitely was there (note the lack of “edited” tag) :) no biggie, not trying to pick a fight
        pontus 5y ago
        My bad, sorry! I could have sworn it wasn't there but you're obviously correct.
- zepto 5y ago
  > it seems likely that the operations of consciousness are purely “classical” in the sense that the exact quantum state is not needed There really is no evidence for this.
  - wizzwizz4 5y ago
    People don't die in MRI machines. That's evidence.
    - zepto 5y ago
      That magnets don’t kill people, but not of much else.
grumbel 5y ago
Not really. The issue is that you aren't the same from one moment to the other in the natural world either. You constantly change, large parts of your body even get rebuild regularly. So demanding 'quantum level perfection' of a copy just doesn't mesh with the ever changing mess that is the real you. As long as the copy is 'good enough', which can be far away from the uncertainty principle, it really shouldn't make a difference to the though experiment.
- filoeleven 5y ago
  Extend the thought experiment. Given: If the technology exists to create “good-enough” copies of a human, the tech to create “good-enough” copies of a habitat should also exist. Experiment: 1) Create X copies of a controlled habitat 2) Create X copies of N humans, where N > 1 3) Place human copies into habitat copies 4) Observe What is the expected percentage that the humans will behave identically? For how long? It’s probably best to start with one human per habitat, but I think “more than one human” is the more interesting question. I can’t find the name of the condition right now, but sometimes people fall into loops, where they will ask the same series of questions and give the same responses each time until they recover. It happened to my grandmother once after a mini-stroke, and there’s a Radiolab episode about it which I also can’t find. So that lends credibility to the hypothesis that if the environments are “good enough,” there will be no divergence. I’m curious to know how long that could possibly last.
  - ben_w 5y ago
    > I can’t find the name of the condition right now, but sometimes people fall into loops, where they will ask the same series of questions and give the same responses each time until they recover Anecdotally, my mother did that sometimes during the mid-to-late stages of her Alzheimer’s (before we had to send her to a care home, but after she was failing the Clock test). My one experience of edible cannabis (in retrospect I don’t think it was just cannabis), my partner at the time said I was repeating the same sentences. Not that I know for certain: myself, I had the subjective feeling of all working memories going direct into long-term memory without touching short-term memory.
ben_w 5y ago
The uncertainty principle is deeper than that. Not only is it impossible to perfectly measure both momentum and position, but they are not both simultaneously well-defined even in perfect conditions — the uncertainty in each is itself a real property. I would also be somewhat surprised if the most detailed scan possible under QM is even necessary for a functionally-perfect copy of consciousness: The perceptron model used in artificial neural nets is a mere toy in comparison to a real neurone, yet very effective at duplicating interesting behaviours.
- TeMPOraL 5y ago
  > but they are not both simultaneously well-defined even in perfect conditions — the uncertainty in each is itself a real property. This also implies trying to "dig deeper" doesn't matter - there's nothing to dig up. From this follows that either 1) "the most detailed scan possible under QM" is at least sufficient to capture the entirety of consciousness, or 2) there's something beyond QM we have no clue about, that makes uncertainty not fundamental, or 3) consciousness is magic and the universe makes no sense. I find 2) unlikely, and 3) is religion (the "separate magisterium from science" kind).

by Miiko 5y ago

alisonkisk 5y ago

Sure, "it's impossible" is always a resolution to a paradox. But not a very interesting one.

NoblePublius 5y ago

In Star Trek, this is handled by the Heisenberg Compensator https://memory-alpha.fandom.com/wiki/Heisenberg_compensator https://memory-alpha.fandom.com/wiki/Heisenberg_compensator

superposeur 5y ago

More precisely, the “no-cloning theorem” of quantum mechanics implies that one cannot replicate the exact quantum state of a clump of matter in a second clump of matter (although it is possible to “teleport” the exact quantum state by transmitting its information, provided one is willing to mess up the original). However, although no one knows for sure, it seems likely that the operations of consciousness are purely “classical” in the sense that the exact quantum state is not needed (Roger Penrose’s ideas not withstanding).

pontus 5y ago
Turns out that while the no cloning theorem does prohibit cloning, it does not prohibit perfect replication as long as the original is destroyed. This is the premise of quantum teleportation. Using some clever applications of entanglement it is possible to perfectly replicate the quantum state of a particle in a different location, it's just that the original particle is necessarily destroyed in the process. So, in short the no cloning theorem does not prevent quantum teleportation. In this sense, the follow-up in the paradox where you create 10 copies all of which are identical would not work. Quantum mechanics effectively ensures that there's ever only at most one of you.
- superposeur 5y ago
  Yes, hence my parenthetical phrase: > … (although it is possible to “teleport” the exact quantum state by transmitting its information, provided one is willing to mess up the original).
  - pontus 5y ago
    When I first wrote my comment, this part of your comment was not there. You must have added it after the fact (hence my comment)
    - superposeur 5y ago
      Definitely was there (note the lack of “edited” tag) :) no biggie, not trying to pick a fight
      - pontus 5y ago
        My bad, sorry! I could have sworn it wasn't there but you're obviously correct.
zepto 5y ago
> it seems likely that the operations of consciousness are purely “classical” in the sense that the exact quantum state is not needed There really is no evidence for this.
- wizzwizz4 5y ago
  People don't die in MRI machines. That's evidence.
  - zepto 5y ago
    That magnets don’t kill people, but not of much else.

grumbel 5y ago

Not really. The issue is that you aren't the same from one moment to the other in the natural world either. You constantly change, large parts of your body even get rebuild regularly. So demanding 'quantum level perfection' of a copy just doesn't mesh with the ever changing mess that is the real you. As long as the copy is 'good enough', which can be far away from the uncertainty principle, it really shouldn't make a difference to the though experiment.

filoeleven 5y ago
Extend the thought experiment. Given: If the technology exists to create “good-enough” copies of a human, the tech to create “good-enough” copies of a habitat should also exist. Experiment: 1) Create X copies of a controlled habitat 2) Create X copies of N humans, where N > 1 3) Place human copies into habitat copies 4) Observe What is the expected percentage that the humans will behave identically? For how long? It’s probably best to start with one human per habitat, but I think “more than one human” is the more interesting question. I can’t find the name of the condition right now, but sometimes people fall into loops, where they will ask the same series of questions and give the same responses each time until they recover. It happened to my grandmother once after a mini-stroke, and there’s a Radiolab episode about it which I also can’t find. So that lends credibility to the hypothesis that if the environments are “good enough,” there will be no divergence. I’m curious to know how long that could possibly last.
- ben_w 5y ago
  > I can’t find the name of the condition right now, but sometimes people fall into loops, where they will ask the same series of questions and give the same responses each time until they recover Anecdotally, my mother did that sometimes during the mid-to-late stages of her Alzheimer’s (before we had to send her to a care home, but after she was failing the Clock test). My one experience of edible cannabis (in retrospect I don’t think it was just cannabis), my partner at the time said I was repeating the same sentences. Not that I know for certain: myself, I had the subjective feeling of all working memories going direct into long-term memory without touching short-term memory.

ben_w 5y ago

The uncertainty principle is deeper than that. Not only is it impossible to perfectly measure both momentum and position, but they are not both simultaneously well-defined even in perfect conditions — the uncertainty in each is itself a real property. I would also be somewhat surprised if the most detailed scan possible under QM is even necessary for a functionally-perfect copy of consciousness: The perceptron model used in artificial neural nets is a mere toy in comparison to a real neurone, yet very effective at duplicating interesting behaviours.

TeMPOraL 5y ago
> but they are not both simultaneously well-defined even in perfect conditions — the uncertainty in each is itself a real property. This also implies trying to "dig deeper" doesn't matter - there's nothing to dig up. From this follows that either 1) "the most detailed scan possible under QM" is at least sufficient to capture the entirety of consciousness, or 2) there's something beyond QM we have no clue about, that makes uncertainty not fundamental, or 3) consciousness is magic and the universe makes no sense. I find 2) unlikely, and 3) is religion (the "separate magisterium from science" kind).