There are (at least) two ways to implement a (Star Trek style) transporter: A space-time wormhole takes you “directly” from here to there, or We scan you, send the info, make a new copy at the other end, and destroy the original.
I don't see how anything besides the self-consistent model can genuinely be called time travel. Anything else is really travel to a universe that happens to be in the same configuration as our past universe (plus perhaps the annihilation of our universe and the creation of this time delayed copy).
I mean if you are supposedly traveling back to t=0 then people at t=1 including yourself better have the memories and experiences caused by your travel to t=0 or that's an inaccurate way to describe your journey.
When younger I was unduly impressed by the fact that demanding self-consistency while allowing later events to cause earlier ones could replicate the 'paradoxes' of quantum measurement (EPR like effects) with classical particles.
I think that much of the valid concern with star trek style transporters comes from people with an epiphenominal theory of mind. Since these people don't accept that mimicking IO behavior shows that the new being has genuine experiences they are going to worry that the copied being actually lacks any experiences. To this extent I do care how the super watch is implemented.
I think all the above concerns with identity are misplaced worries and supposing a coherent compelling physical theory of experience had been offered that implied the copies were just as capable of experience as I was I would not care how it was implemented...modulo worries about aging with your time machine implementation.
Lovely idea. One point is that if you are destroyed here, you will be destroyed an infinite number of times over in our infinite Universe. And those alone are the individuals who are perfectly identical to you; but it is also true that you (individuals identical to you in their previous history) will be not-destroyed another infinite number of times. But in none of these will any of the infinite number of you s think "Well, here I am. That went well" If there is no sensation of having gone anywhere, you haven't gone anywhere. There's no reference through memory to having been anywhere else so it doesn't count as a transporter option. Furthermore, they weren't you while they were coexisting with you; why should any of them become you the moment you cease to exist? But the fundamental problem is that what is thought when you say "this body is my body, these thoughts are my thoughts" can't be incorporated into an objective representation of the world. It is not a "fact." It is the essence of subjectivity. Only one of the infinite number of identical yous is the you that you are; but there's no way of understanding that as a fact--as a part of an objective description of the world. Your subjectivity, your sense of personal uniqueness is both everything and nothing. Objectively, it doesn't exist. Subjectively, it is the only sense in which you can be said to exist at all.
The second model would allow to undo all suffering that there ever has been - imagine if we could take back the past 500 million years of distress and agony! It would indeed be worth the loss of life. I find it intuitively implausible, however.
There's a third model of time-travel used sometimes in sci-fi: The self-consistent model, in which the original present is a result of future time travel to the past. Intruiging and brilliant storylines can be woven around this, often with some kind of tragedy element of incomplete knowledge and sudden revelation for the protagonists. Sometimes stories start out with this promise, providing evidence for the self-consistent model - only to suddenly switch to one of the other ones. Deja vu with Denzel Washington did this. This makes me very angry! ಠ_ಠ
This discussion reminds me of an analogous argument I made many years ago with regard to time travel.
There are at least two basic models for time travel stories. In one, multiple time lines exist "simultaneously." The time traveler goes back in time, and by doing so creates a new time line that branches off the one where he originated at the point when he arrives and starts changing things. But the old time line still exists and, in principle, he could jump across to it.
In the other model, there is a single timeline. By going back in time the time traveler erases everything subsequent to his time of arrival, creating a new, and now unique, version of everything thereafter.
One can, no doubt, make a variety of arguments as to why one model is or is not more consistent or plausible than the other. But in my view, if time travel existed, practically everyone would believe in the first version--because it is a much pleasanter version to believe in. You, by jumping back a century, haven't killed off everyone you ever knew--you have merely shifted yourself into a different line, leaving all of them unaffected back in the old one.
I know that most people who think humans can be "downloaded" consider themselves ardent materialists, but from a philosophical perspective the belief that death is not the end if you can do an information download into another body is not that different from the belief that you can escape death by reensoulment in another body.
I think the majority of philosophers would agree. If you want to see a debate on this issue between a transhumanist and a philosopher, check this out:
Observably, people care less about hypotheticals than actuals, and less about your actuals than their own. This makes no ethical sense, and possibly not even logical sense, but there it is.
"[...] to guarantee that my awareness will continue in the duplicated body, rather than coming to an end, and an unrelated awareness arising."
Can you be more specific about what you mean by that differerence (between "my awareness" and "an unrelated awareness")? Specifically, can you point out what natural properties in the physical universe you think constitute the difference? It clearly can't be shared memories, continuitiy of personality, structural identity etc., because they are all preserved. What other properties do you think there are to constitute this alleged difference, and what reason do we have to believe that they exist in our physical universe?
This sounds like a variant on quantum immortality, which implies (in a way that may be more horrific than comforting) that nobody really dies. The problem with it is that not all infinities are equally sized. Most universes will not have me spontaneously appear at the destination -- most will have me not appear anywhere. If we consider only the ones where I do spontaneously appear somewhere, most of those will have me appear somewhere other than my destination. Most of the cases where I do spontaneously appear in some location would place me in completely random universes devoid of any life, let alone humans, let alone the ones I am familiar with. Thus my expectation would not be to arrive at my destination. With the actual teleporter #2 I have the assurance that the vast majority of universes where I experience my next moment are the same one I remember having been in -- exactly as many of them as is the case with transporter #1, or no transporter at all.
One might easily prefer Option #2 to Option #3 using reasoning based on Timeless Decision Theory, the short short version of which is:
Choose as though controlling the logical output of the abstract computation you implement, including the output of all other instantiations and simulations of that computation.
This stuff always makes me dizzy. Here's a variant inspired by recent reading which just adds to my disorientation:-) It's called Transporter Option #3. It's addressed to those of you who are OK with Transporter Option #2.
First, I need you to make a Big (but arguably well motivated) Assumption.
The Big (but arguably well motivated) Assumption:Assume you've been convinced that our universe is infinite in spatial extent, and that consequently any physically possible configuration of particles and energy within any specified volume of space is exactly replicated an infinite number of times over, within our infinite universe.
(Why I Claim this Assumption is well motivated: As best I can tell so far from my reading of Brian Greene's excellent "The Hidden Reality," the validity of the Big Assumption is actually implied by our best current cosmological theory applied to our best current cosmological data. I invite those who have finished the book and are knowledgeable concerning recent cosmological research, to comment on this assertion.)
Now, the question: Are you OK with
Transporter Option #3:We just destroy you.
Sales pitch for this Transporter Option #3: We've not only made Option #3 a little cheaper than Option #2, we've made it greener, too: Specifically, we've done away with all the entropy-inducing bother of determining your desired destination, scanning you, beaming your info, reconstituting you at the other end, etc, etc. Not only that, but Option #3 otherwise achieves the same result as Option #2: After all, wherever you want to go, you're already there, a billion times over (see Big Assumption, above). With bells on, if you like (also without bells on, and - truth be told - whether you like it or not).
Consider again the 'three-fold' hinge on my third ontological axis 'carving reality at the joints':
3rd axis
Structural (STR) level – The configuration of thingsFunctional (FUN) level – What things doRepresentational (REP) level – What things symbolize
The assumptions and scenarios Hanson paints only extend to the second level (FUN- the functional level). Many high-IQers seem to believe that describing something in purely functional or predictive terms (i.e. Bayesian) is enough to provide an 'explanation' of something. Not so! The fact that #1 and #2 are functionally equivalent doesn't mean the two scenarios are equivalent. This is because REP - the representational level, wasn't taken into account. Therefore we should be very wary about thinking that #2 is going to preserve our identities.
Now I have no problem with making multiple em copies and then later having those copies be merged back into one copy. Scenario #1, the time travel scenario) is OK.
But scenerio #2 I would not risk, based on my ontology, which clearly shows that descriptions on the functional level alone do not fully capture what is going on.
I would only use a #2 super-watch to ensure one branch of my conscious experience survives a fatal situation. I would pay more for one which erased the original sooner after creating the copy, though; in a reversal of Robin's setup. If it created an exact duplicate of me in the same place and erased the original, all within a few Planck intervals, I might not even mind using it much.
I don't see how anything besides the self-consistent model can genuinely be called time travel. Anything else is really travel to a universe that happens to be in the same configuration as our past universe (plus perhaps the annihilation of our universe and the creation of this time delayed copy).
I mean if you are supposedly traveling back to t=0 then people at t=1 including yourself better have the memories and experiences caused by your travel to t=0 or that's an inaccurate way to describe your journey.
When younger I was unduly impressed by the fact that demanding self-consistency while allowing later events to cause earlier ones could replicate the 'paradoxes' of quantum measurement (EPR like effects) with classical particles.
I think that much of the valid concern with star trek style transporters comes from people with an epiphenominal theory of mind. Since these people don't accept that mimicking IO behavior shows that the new being has genuine experiences they are going to worry that the copied being actually lacks any experiences. To this extent I do care how the super watch is implemented.
I think all the above concerns with identity are misplaced worries and supposing a coherent compelling physical theory of experience had been offered that implied the copies were just as capable of experience as I was I would not care how it was implemented...modulo worries about aging with your time machine implementation.
Lovely idea. One point is that if you are destroyed here, you will be destroyed an infinite number of times over in our infinite Universe. And those alone are the individuals who are perfectly identical to you; but it is also true that you (individuals identical to you in their previous history) will be not-destroyed another infinite number of times. But in none of these will any of the infinite number of you s think "Well, here I am. That went well" If there is no sensation of having gone anywhere, you haven't gone anywhere. There's no reference through memory to having been anywhere else so it doesn't count as a transporter option. Furthermore, they weren't you while they were coexisting with you; why should any of them become you the moment you cease to exist? But the fundamental problem is that what is thought when you say "this body is my body, these thoughts are my thoughts" can't be incorporated into an objective representation of the world. It is not a "fact." It is the essence of subjectivity. Only one of the infinite number of identical yous is the you that you are; but there's no way of understanding that as a fact--as a part of an objective description of the world. Your subjectivity, your sense of personal uniqueness is both everything and nothing. Objectively, it doesn't exist. Subjectively, it is the only sense in which you can be said to exist at all.
The second model would allow to undo all suffering that there ever has been - imagine if we could take back the past 500 million years of distress and agony! It would indeed be worth the loss of life. I find it intuitively implausible, however.
There's a third model of time-travel used sometimes in sci-fi: The self-consistent model, in which the original present is a result of future time travel to the past. Intruiging and brilliant storylines can be woven around this, often with some kind of tragedy element of incomplete knowledge and sudden revelation for the protagonists. Sometimes stories start out with this promise, providing evidence for the self-consistent model - only to suddenly switch to one of the other ones. Deja vu with Denzel Washington did this. This makes me very angry! ಠ_ಠ
This discussion reminds me of an analogous argument I made many years ago with regard to time travel.
There are at least two basic models for time travel stories. In one, multiple time lines exist "simultaneously." The time traveler goes back in time, and by doing so creates a new time line that branches off the one where he originated at the point when he arrives and starts changing things. But the old time line still exists and, in principle, he could jump across to it.
In the other model, there is a single timeline. By going back in time the time traveler erases everything subsequent to his time of arrival, creating a new, and now unique, version of everything thereafter.
One can, no doubt, make a variety of arguments as to why one model is or is not more consistent or plausible than the other. But in my view, if time travel existed, practically everyone would believe in the first version--because it is a much pleasanter version to believe in. You, by jumping back a century, haven't killed off everyone you ever knew--you have merely shifted yourself into a different line, leaving all of them unaffected back in the old one.
I know that most people who think humans can be "downloaded" consider themselves ardent materialists, but from a philosophical perspective the belief that death is not the end if you can do an information download into another body is not that different from the belief that you can escape death by reensoulment in another body.
I think the majority of philosophers would agree. If you want to see a debate on this issue between a transhumanist and a philosopher, check this out:
http://bloggingheads.tv/dia...
It seems like you have to be some kind of dualist to believe such a thing.
This part of your comment confused me.
Are you sure about this? What of those among us who find the mathematical universe hypothesis plausible?
Observably, people care less about hypotheticals than actuals, and less about your actuals than their own. This makes no ethical sense, and possibly not even logical sense, but there it is.
"[...] to guarantee that my awareness will continue in the duplicated body, rather than coming to an end, and an unrelated awareness arising."
Can you be more specific about what you mean by that differerence (between "my awareness" and "an unrelated awareness")? Specifically, can you point out what natural properties in the physical universe you think constitute the difference? It clearly can't be shared memories, continuitiy of personality, structural identity etc., because they are all preserved. What other properties do you think there are to constitute this alleged difference, and what reason do we have to believe that they exist in our physical universe?
This sounds like a variant on quantum immortality, which implies (in a way that may be more horrific than comforting) that nobody really dies. The problem with it is that not all infinities are equally sized. Most universes will not have me spontaneously appear at the destination -- most will have me not appear anywhere. If we consider only the ones where I do spontaneously appear somewhere, most of those will have me appear somewhere other than my destination. Most of the cases where I do spontaneously appear in some location would place me in completely random universes devoid of any life, let alone humans, let alone the ones I am familiar with. Thus my expectation would not be to arrive at my destination. With the actual teleporter #2 I have the assurance that the vast majority of universes where I experience my next moment are the same one I remember having been in -- exactly as many of them as is the case with transporter #1, or no transporter at all.
One might easily prefer Option #2 to Option #3 using reasoning based on Timeless Decision Theory, the short short version of which is:
Choose as though controlling the logical output of the abstract computation you implement, including the output of all other instantiations and simulations of that computation.
-- Eliezer Yudkowsky (quoted here)
This stuff always makes me dizzy. Here's a variant inspired by recent reading which just adds to my disorientation:-) It's called Transporter Option #3. It's addressed to those of you who are OK with Transporter Option #2.
First, I need you to make a Big (but arguably well motivated) Assumption.
The Big (but arguably well motivated) Assumption:Assume you've been convinced that our universe is infinite in spatial extent, and that consequently any physically possible configuration of particles and energy within any specified volume of space is exactly replicated an infinite number of times over, within our infinite universe.
(Why I Claim this Assumption is well motivated: As best I can tell so far from my reading of Brian Greene's excellent "The Hidden Reality," the validity of the Big Assumption is actually implied by our best current cosmological theory applied to our best current cosmological data. I invite those who have finished the book and are knowledgeable concerning recent cosmological research, to comment on this assertion.)
Now, the question: Are you OK with
Transporter Option #3:We just destroy you.
Sales pitch for this Transporter Option #3: We've not only made Option #3 a little cheaper than Option #2, we've made it greener, too: Specifically, we've done away with all the entropy-inducing bother of determining your desired destination, scanning you, beaming your info, reconstituting you at the other end, etc, etc. Not only that, but Option #3 otherwise achieves the same result as Option #2: After all, wherever you want to go, you're already there, a billion times over (see Big Assumption, above). With bells on, if you like (also without bells on, and - truth be told - whether you like it or not).
Well, are you OK with it? Why or why not?
Consider again the 'three-fold' hinge on my third ontological axis 'carving reality at the joints':
3rd axis
Structural (STR) level – The configuration of thingsFunctional (FUN) level – What things doRepresentational (REP) level – What things symbolize
The assumptions and scenarios Hanson paints only extend to the second level (FUN- the functional level). Many high-IQers seem to believe that describing something in purely functional or predictive terms (i.e. Bayesian) is enough to provide an 'explanation' of something. Not so! The fact that #1 and #2 are functionally equivalent doesn't mean the two scenarios are equivalent. This is because REP - the representational level, wasn't taken into account. Therefore we should be very wary about thinking that #2 is going to preserve our identities.
Now I have no problem with making multiple em copies and then later having those copies be merged back into one copy. Scenario #1, the time travel scenario) is OK.
But scenerio #2 I would not risk, based on my ontology, which clearly shows that descriptions on the functional level alone do not fully capture what is going on.
I'm open to Nozick's experience machine, as well as wireheading, so I'm probably atypical.
Surprised Hanson didn't link to To Be.
I would only use a #2 super-watch to ensure one branch of my conscious experience survives a fatal situation. I would pay more for one which erased the original sooner after creating the copy, though; in a reversal of Robin's setup. If it created an exact duplicate of me in the same place and erased the original, all within a few Planck intervals, I might not even mind using it much.
To the anti-type-#2 people: would you use it to escape a life-threatening situation where there is no other option for survival?