What kind of evidence would you accept as sufficient to be persuaded that it works?
Probably something like this scenario (I just made up):
Bob signs up for cryonics. Then Bob dies of something. So Bob gets frozen some time later. Then at some point in the future, Bob is brought back to life right as rain.
Basically, the process working ever would be evidence that the process might ever work. Until then, consider me in the ‘control group’.
I think it was Mike Li who analogized this to refusing to get on an airplane until after it has arrived in France. The whole point of cryonics is as an ambulance ride to the future; once you’re in the future, you don’t need cryonics any more. I severely, severely doubt that anyone will ever again be frozen after the time a cryonics revival is possible.
Isn’t there some gut, intuitive level on which you can see that your objection obviously makes no sense, because conditioning on the proposition that cryonics with present-day vitrification technology does in fact work as an ambulance ride to the future, we still would not expect to see a revival in the present time?
I think it was Mike Li who analogized this to refusing to get on an airplane until after it has arrived in France.
I take it more to be like refusing to get on an airplane until any one has arrived anywhere, ever.
For all I know, cryonics makes it harder to revive people. Not that I think it’s likely that’s the case, but it certainly doesn’t seem worth my time and money.
It’s like being the guy who checks the Wright brothers’ calculations, finds them correct, and still refuses to leap onboard their untried prototype to escape a tiger, but instead prefers to stand and be eaten.
Look, conventional death makes it maximally hard to revive a person. Their information has dissipated. You would essentially need a time machine. Cryonics is a guaranteed improvement over that—at least you have something to work with.
It’s like being the guy who checks the Wright brothers’ calculations, finds them correct,
Perhaps more like the Wright brothers were planning to figure out how to land the plane after they throw it off a cliff. And your example throws out the benefits of not signing up for cryonics, which are a major factor for me.
Note that if Wright brothers didn’t believe that there was a considerable chance of the plain not crashing, it would be a bad investment to build the plain in the first place. The question is about the cost: does the current state of knowledge support the positive outcome sufficiently to think of designing a plane? To design a plane? To build a plane? To perform an experiment, risking its destruction? To test-pilot a plane, risking one’s life?
The same goes for cryonics, here you risk something like 100 bucks a year.
Sure, I could risk 100 bucks a year on cryonics, and 100 bucks a year on a million other live-forever schemes, if I had 100 million dollars to waste every year.
Or I could invest my money intelligently and use it to help raise my children.
So they haven’t figured out the landing gear. So you might break your neck, might break your arm—but the tiger is sprinting towards you! It certainly will eat you!
Sure—if Einstein signed up for cryonics, I might even follow suit. But a lot of really smart people are signing up for ‘heaven’, and I’m not listening to them, either.
Missing the point I think. Einstein wasn’t stating this as any sort of appeal to authority. He was expressing his confidence in his mathematical proofs.
Mathematical proofs are an appeal to authority. Their standards rest entirely on the ability of experts on Mathematics to understand them. If we had a canonical mechanical proof-checker or something, it might be a different story.
But they were Einstein’s proofs. He was confident in the math that he understood. If he were trying to convince someone else, then yes he’d be using himself as an authority.
That was an allusion to this question, which you still haven’t answered. If, in principle, you could indeed decide that successful revival is possible, based only on theoretical knowledge, before any successful revival was ever performed, then you should be able to explain what kind of indirect evidence it would take to persuade you that successful revival is sufficiently likely for you to decide to sign up for cryonics.
I severely, severely doubt that anyone will ever again be frozen after the time a cryonics revival is possible.
This is too unintuitive an assumption to use in a basic refutation. I doubt it’s even true, if revival is performed by non-AGI means, simply because of improved preservation technology, which may well become possible at some point.
Agreed. Suppose we simply learn how to revive someone who’s frozen first (unlikely, I know). Then, we would selectively freeze/unfreeze people based on the further limitations of medicine at the time (can treat gunshot wounds / can’t treat lukemia)
Yes, that’s one use case. I’m really not competent to estimate with any certainty how biologically feasible is that, and I assume it’s not very feasible. If I remember correctly, the brains of currently preserved, even after vitrification, get cracked during the freezing, so they won’t work even if unfrozen, detoxicated, etc. I don’t know whether it’s possible to find a solution to this problem with anything from the repertoire of current technology.
But the decision concerns the current situation. What do you answer on thesequestions?
Aside: it looks a lot more feasible to me if you don’t try to repair the original biology, but rather try to extract information from it for re-instantiation. Then for example brain cracks become a problem in image-alignment rather than in nanosurgery.
This argument forced me to change my mind a little: indeed, to do the neurosurgery, you need an image anyway, possibly of the same order of resolution or even greater than required for scanning, so emulation may be easier than repair, and realignment of the image should be relatively easy once you have a scan. Still, I don’t see emulation working for a long time still, I’d give it expected 60 to 150 years, and it’s hard to say how the process will look at that point, on the progress of what kinds of technologies the feasibility of this process will depend.
That was “Whole Brain Emulation Roadmap” from the Future of Humanity Institute. (You shouldn’t post bare links.) I’m sure aware of it. It’s a feasibility study, and I’m sure it’s feasible, so no great revelations there.
The problem is that this study is roughly analogous to estimating that the progress in steam engine technology will allow very fast and efficient trains eventually, which means that there will be fast trains based on some technology, if they are still needed, at least that good.
Which was basically my sentiment: you list all these technologies, but they, specifically, may be of little relevance. This isn’t an argument for emulation to be infeasible. I also reserve an option for revival not being the best thing you can do with a dead body, but this is an argument this thread is too small to contain.
What kind of evidence would it take to convince you that cryonics has a small, but considerable chance of working in the future, prior to there being any successful revivals?
I don’t like to deal in probabilities, but I’d reckon a successful revival of a dolphin would count. Short of that? Probably nothing, if by ‘considerable’ you mean ‘worth spending my money on’. Things other than evidence might convince me though—like my wife wanting to sign up for cryonics for whatever fool reason.
Does it have to be a dolphin, or would successful revival of a mouse count?
Try not to look up if that’s been done before you answer. If you do know, try to imagine whether you’d count it as evidence, if you didn’t already know.
I don’t like to deal in probabilities, but I’d reckon a successful revival of a dolphin would count.
No, that’s out.
Short of that? Probably nothing, if by ‘considerable’ you mean ‘worth spending my money on’.
Yes, I do mean that.
This means, that no matter what you observe, you always estimate the probability of cryonics working as very low, right up to the point where it does succeed (if that ever happens). Which is equivalent to a priori estimating the probability of it working eventually very low also.
Do you believe that progress will never be made, that it will never be possible to revive a very slowly changing frozen body? In 100 years? In 10000 years? Never ever?
You are not following the argument. You said that you accept the possibility of knowing the outcome from purely theoretical (indirect) argument (that is, not the kind of data where you are presented with successful revival of anything), as in the Einstein’s anecdote. I ask what kind of indirect data/argument that would be, that is enough to convince you to sign up. That you may do that for signaling reasons is irrelevant to the question.
Probably something like this scenario (I just made up):
Basically, the process working ever would be evidence that the process might ever work. Until then, consider me in the ‘control group’.
I think it was Mike Li who analogized this to refusing to get on an airplane until after it has arrived in France. The whole point of cryonics is as an ambulance ride to the future; once you’re in the future, you don’t need cryonics any more. I severely, severely doubt that anyone will ever again be frozen after the time a cryonics revival is possible.
Isn’t there some gut, intuitive level on which you can see that your objection obviously makes no sense, because conditioning on the proposition that cryonics with present-day vitrification technology does in fact work as an ambulance ride to the future, we still would not expect to see a revival in the present time?
I take it more to be like refusing to get on an airplane until any one has arrived anywhere, ever.
For all I know, cryonics makes it harder to revive people. Not that I think it’s likely that’s the case, but it certainly doesn’t seem worth my time and money.
It’s like being the guy who checks the Wright brothers’ calculations, finds them correct, and still refuses to leap onboard their untried prototype to escape a tiger, but instead prefers to stand and be eaten.
Look, conventional death makes it maximally hard to revive a person. Their information has dissipated. You would essentially need a time machine. Cryonics is a guaranteed improvement over that—at least you have something to work with.
Perhaps more like the Wright brothers were planning to figure out how to land the plane after they throw it off a cliff. And your example throws out the benefits of not signing up for cryonics, which are a major factor for me.
Note that if Wright brothers didn’t believe that there was a considerable chance of the plain not crashing, it would be a bad investment to build the plain in the first place. The question is about the cost: does the current state of knowledge support the positive outcome sufficiently to think of designing a plane? To design a plane? To build a plane? To perform an experiment, risking its destruction? To test-pilot a plane, risking one’s life?
The same goes for cryonics, here you risk something like 100 bucks a year.
Sure, I could risk 100 bucks a year on cryonics, and 100 bucks a year on a million other live-forever schemes, if I had 100 million dollars to waste every year.
Or I could invest my money intelligently and use it to help raise my children.
So they haven’t figured out the landing gear. So you might break your neck, might break your arm—but the tiger is sprinting towards you! It certainly will eat you!
I’ll take my odds against a tiger rather than a cliff any day. How confident are you that you won’t live forever?
Do you believe in the “Then I would feel sorry for the good Lord. The theory is correct.” situations?
Sure—if Einstein signed up for cryonics, I might even follow suit. But a lot of really smart people are signing up for ‘heaven’, and I’m not listening to them, either.
Missing the point I think. Einstein wasn’t stating this as any sort of appeal to authority. He was expressing his confidence in his mathematical proofs.
Mathematical proofs are an appeal to authority. Their standards rest entirely on the ability of experts on Mathematics to understand them. If we had a canonical mechanical proof-checker or something, it might be a different story.
But they were Einstein’s proofs. He was confident in the math that he understood. If he were trying to convince someone else, then yes he’d be using himself as an authority.
So, you concede that it’s possible to know the outcome in advance without empirical observation of success.
Now, what makes Einstein a special person for this purpose? Can it be you that decides?
Sure, it could be me that decides. That’s why I’ve decided. What’s your point?
That was an allusion to this question, which you still haven’t answered. If, in principle, you could indeed decide that successful revival is possible, based only on theoretical knowledge, before any successful revival was ever performed, then you should be able to explain what kind of indirect evidence it would take to persuade you that successful revival is sufficiently likely for you to decide to sign up for cryonics.
This is too unintuitive an assumption to use in a basic refutation. I doubt it’s even true, if revival is performed by non-AGI means, simply because of improved preservation technology, which may well become possible at some point.
Agreed. Suppose we simply learn how to revive someone who’s frozen first (unlikely, I know). Then, we would selectively freeze/unfreeze people based on the further limitations of medicine at the time (can treat gunshot wounds / can’t treat lukemia)
Yes, that’s one use case. I’m really not competent to estimate with any certainty how biologically feasible is that, and I assume it’s not very feasible. If I remember correctly, the brains of currently preserved, even after vitrification, get cracked during the freezing, so they won’t work even if unfrozen, detoxicated, etc. I don’t know whether it’s possible to find a solution to this problem with anything from the repertoire of current technology.
But the decision concerns the current situation. What do you answer on these questions?
Aside: it looks a lot more feasible to me if you don’t try to repair the original biology, but rather try to extract information from it for re-instantiation. Then for example brain cracks become a problem in image-alignment rather than in nanosurgery.
This argument forced me to change my mind a little: indeed, to do the neurosurgery, you need an image anyway, possibly of the same order of resolution or even greater than required for scanning, so emulation may be easier than repair, and realignment of the image should be relatively easy once you have a scan. Still, I don’t see emulation working for a long time still, I’d give it expected 60 to 150 years, and it’s hard to say how the process will look at that point, on the progress of what kinds of technologies the feasibility of this process will depend.
http://www.fhi.ox.ac.uk/Reports/2008-3.pdf
That was “Whole Brain Emulation Roadmap” from the Future of Humanity Institute. (You shouldn’t post bare links.) I’m sure aware of it. It’s a feasibility study, and I’m sure it’s feasible, so no great revelations there.
The problem is that this study is roughly analogous to estimating that the progress in steam engine technology will allow very fast and efficient trains eventually, which means that there will be fast trains based on some technology, if they are still needed, at least that good.
Which was basically my sentiment: you list all these technologies, but they, specifically, may be of little relevance. This isn’t an argument for emulation to be infeasible. I also reserve an option for revival not being the best thing you can do with a dead body, but this is an argument this thread is too small to contain.
What kind of evidence would it take to convince you that cryonics has a small, but considerable chance of working in the future, prior to there being any successful revivals?
I don’t like to deal in probabilities, but I’d reckon a successful revival of a dolphin would count. Short of that? Probably nothing, if by ‘considerable’ you mean ‘worth spending my money on’. Things other than evidence might convince me though—like my wife wanting to sign up for cryonics for whatever fool reason.
Does it have to be a dolphin, or would successful revival of a mouse count?
Try not to look up if that’s been done before you answer. If you do know, try to imagine whether you’d count it as evidence, if you didn’t already know.
No, that’s out.
Yes, I do mean that.
This means, that no matter what you observe, you always estimate the probability of cryonics working as very low, right up to the point where it does succeed (if that ever happens). Which is equivalent to a priori estimating the probability of it working eventually very low also.
Do you believe that progress will never be made, that it will never be possible to revive a very slowly changing frozen body? In 100 years? In 10000 years? Never ever?
You are not following the argument. You said that you accept the possibility of knowing the outcome from purely theoretical (indirect) argument (that is, not the kind of data where you are presented with successful revival of anything), as in the Einstein’s anecdote. I ask what kind of indirect data/argument that would be, that is enough to convince you to sign up. That you may do that for signaling reasons is irrelevant to the question.