= 27d567bc2d48d9f40e9ccc20ea370b15)
lsparrish comments on On the unpopularity of cryonics: life sucks, but at least then you die - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (465)
The magic bean scenario has a low prior, and you haven't provided any evidence, just a claim, so it's not really comparable. On the other hand it is the sort of thing that would be easy to prove if it worked (assuming you have multiple beans) -- just feed one to someone who is about to die.
Cryonics is not easy to prove if it works because the level of technology we are at is not anywhere near the theoretical maximum where repairing the damage of cryopreservation is concerned. You would have to base your opinion on indirect evidence, such as whether the structure of the brain appears preserved.
As to the brain/metabolic activity question, that's possible to answer by examining samples that are cryopreserved and thawed. Turns out they resume electrical and chemical activity just fine. The fundamental barrier is structural preservation and avoiding/limiting toxicity effects from the cryoprotectants.
Now, you might argue rationally in favor of allocating funding towards cryobiology research instead of your own future arrangements, but I'm inclined to think people with cryonics arrangements will allocate more money towards cryobiology research than disinterested outsiders.
Agreed, but I was arguing specifically against saturn's apparent claim that you should spend money on any process that promises to allow you to live forever, no matter how unlikely (assuming that you actually want to live forever, of course, which we do).
Is there good evidence to believe that the damage of cryopreservation (as well as the damage due to the brain losing oxygenation while it's waiting to get cryopreserved) is actually reversible ? Furthermore, is it the case that preserving the physical structure of the brain is sufficient ? To use a crude analogy, if I were to unplug my computer from the wall socket at this very moment, the contents of this post that I am now typing would be lost forever, and no amount of future technology would bring them back -- despite the fact that the physical structure of my computer hardware will be perfectly preserved, and despite the fact that my computer will resume functioning normally after I plug it back in.
I think that the question of "how should you allocate your money given that you want to live forever" is rather more complex than that. Here are some possible choices:
These are just some possible answers to the question of how you should spend your money; I'm sure there are many others.
Yes there's pretty good evidence that the physical structure of the brain is sufficient for long-term memories. The hard drive of your computer would survive if you unplugged your computer. Also if you put the RAM chips of your computer in LN2 immediately after powering off, the data would be readable later (or so I've heard). If human personalities were stored as dynamic processes, that would be a strong argument against cryonics, however this is unlikely to be the case; if they were, we would expect that hypothermia causes amnesia.
The damage done by the brain losing oxygen is mostly done later, after bloodflow has been restored. (This is known as the ischemic cascade.) It is not likely to be structurally damaging, provided there is prompt cooling. If there is adequate perfusion of cryoprotectants, the only damage is that of cryoprotectant toxicity, chilling injury, and cracking, none of which seem likely to render the structure unreadable or irreparable. All of these forms of damage are desirable to avoid (and each increases your risk), but none of them seem to me to be deal-killers.
I am guessing these problems can be resolved long before true antiaging therapies become possible, because it seems to be an inherently simpler problem. We already have a foot in the door so to speak, with the cryopreservation of tissue samples and even small organs such as the rabbit kidney. Note that unlike antiaging treatments, you can get immediate results, so even if they are comparable in difficulty, we should expect cryobiology to progress faster given the same amount of funding.
I find it hard to figure out whether the singularity is near or far. If it is near, that is an argument against not just cryonics but practically everything else. I tend to regard the singularity as moderately (but not ridiculously) low probability.
I was under the impression that hypothermia is not the same as cryonic freezing, and that even when a person is undergoing hypothermia, he still has brain activity. I am reasonably sure that the person is at least undergoing metabolic activity, because otherwise, in the absence of cryoprotectants, his cells would freeze and burst. I could be wrong about this, though.
As far as tissue damage is concerned, AFAIK our current technology is not advanced enough to even preserve human hearts, livers, and other organs (for the purposes of transplanting them into other patients). That is, we can freeze an organ, and we can thaw it, but in the process it is damaged beyound repair -- and we're talking about relatively simple organs here, not brains. As I understand it, the damage is caused by the following factors:
You say that this damage is not a "deal-killer", but if this were the case, we could at least cryopreserve individual organs today (not to mention whole mammals). Cryonics advocates usually agree with me here, but postulate some form of future technology which will be able to repair this damage. I'm not sure what kind of technology could do that, though. Molecular nanotechnology is the most popular candidate, but I am not convinced that it could, in fact, exist (which is one of the reasons I'm not too enthusiastic about the Singularity, as well).
I'm not sure what you mean by this, though:
What do you mean by "immediate results" ? Cryonics is kind of the opposite of "immediate", by definition.
Yes there is some metabolic activity in hypothermia patients. However, metabolic activity isn't the reason for no ice crystal formation in hypothermia patients. The relatively high temperature (vs. cryogenic temperatures) is responsible for both phenomena. However, because it is much lower than ordinary body temperatures, hypothermia slows metabolism down and reduces (and halts) electrical activity in the brain.
This is correct. However, vitrification has been used to vitrify and revive a rabbit kidney. The brain has a higher degree of vascularization than other organs, which is positive for cryonics purposes. However, nobody is suggesting that the brain can be perfectly preserved as a functioning organ at present. Its size and the conditions under which cryonics is practiced make this unlikely, although it is a reasonable goal for researchers to work towards (and is much easier than full body).
Cracking due to temperature differentials is macroscopic in vitrification, thus doesn't apply to "cells" so much as to tissues. It is also preventable, by stopping cooling at the glass transition temperature rather than continuing down to LN2 temperature.
Entirely eliminated in parts of the tissue that are well-perfused. In ideal cases, my understanding is that the entire brain can be vitrified. The cost of this is the toxicity of the required concentrations of cryoprotectants.
Largely prevented (under ideal conditions) with prompt cooling.
My understanding is that this is not a problem when you get to the glass transition temperature (-135C).
I don't know why your interpretation of the phrase "deal-killer" would exclude the possibility of repairs using future technology. That's explicitly part of the deal when discussing cryonics.
Classic MNT is the most extreme candidate, and is easy to visualize as something likely to work if it does come to exist. However, there is a spectrum of possible candidates for implementing the repairs, ranging from bioengineered microbes to synthetic proteins, or other constructs ("soft machines") that mimic life's mechanisms.
Note that the most compelling arguments against MNT are because it would be too fragile to work in warm, wet conditions. Cryonics puts the body in non-warm, non-wet conditions, which could be considered ideal for forms of nanotech (and/or a spectrum of microtech featuring nanoscale components) that are not suitable for higher temperatures.
There's also no reason (of which I am aware) that you could not convert the cryopreserved body into very small micro-scale blocks or slices, fix them individually using advanced lithographic techniques applied to the exposed surface area, then reassemble them. This sort of thing could require extreme amounts of resources to pull off, which (at least given a non-MNT universe) indicates that a large amount of money should be set aside to generate interest for reanimation.
I haven't yet mentioned uploading, because it's worth addressing the least convenient possible universe the skeptic can imagine first. But this is another possible track which could proceed in the absence of MNT, so if you accept that an upload is a valid continuation of self (which plenty of lesswrongers will be happy to defend), this would also need to be eliminated for cryonics to be rendered implausible.
I was referring to the outcomes of cryobiology experiments. You can experiment with new cryoprotectants and get meaningful results right away. Virtually all of the damage (toxicity) is done on the way to and from the reduced temperature, with virtually none occurring during the long-term storage.
With anti-aging interventions by contrast, you cannot tell whether something is helping until the experimental animal reaches old age. Also, the more effective the anti-aging treatment is, the longer it takes to test. The shorter-lived the animal is the less humanlike its genome, and thus the less likely the treatment will successfully translate to humans. Thus we might take hundreds or thousands of years to reach actuarial escape velocity or comprehensively cure aging.
Sorry, my reply came out differently from what it sounded like in my head. What I meant to say was something like, "hypothermia patients still undergo noticeable metabolic activity, as opposed to cryonic patients, because the temperatures involved are much higher (among other reasons)". So, you and I mostly agree here; still, what I tried to say is that hypothermia and cryopreservation are not entirely analogous.
Can you explain why ? Is this because the higher density of blood vessels allows for higher densities of cryoprotectants to be delivered ? But isn't this advantage offset by the disadvantage of having a more delicate structure that needs to be preserved ? AFAIK, you can lose a chunk of liver or a piece of skin with only a minor loss of function, but losing a piece of brain is a different story.
In any case, is there a reason why we can freeze and revive a rabbit kidney, but not a human one ?
That actually sounds worse, not better...
How prompt are we talking here ? And isn't this requirement at odds with your other comments ?
Fair enough, I think I interpreted your comment too strongly. As for the future technologies, you bring up molecular nanotech, molecular biotech (for lack of a better word; MBT for short), and uploading (which I do accept as a continuation of self, so at least we can agree on this point). You say that the "most compelling arguments against MNT are because it would be too fragile to work in warm, wet conditions"; I actually don't think that this is the most compelling argument against MNT, but that's another topic. I also believe that uploading will require some sort of MNT, or something very close to it.
The problem I have with MNT and MBT is twofold:
This entire talk of "future technologies" sounds a little hand-wavy to me, to be honest. Yes, I understand that all kinds of really awesome future technologies could hypothetically exist, but I would need to see some compelling evidence before I become convinced that they are likely to exist.
Oh, ok, I see what you meant. I have a few objections, though (as you probably knew I would, heh):
Don't get me wrong, I'm not saying that cryonics is a generally bad idea. I'm just saying that, as of today, there are better ways to spend your resources (i.e., money).
For the sake of argument let's assume the worst case regarding current cryopreservation technology. Assume it does too much damage to work regardless of what future tech is possible. That does not rule out advancements in cryopreservation tech in the near term that permit either a) good enough, or b) perfect preservation of the brain (or the full body for that matter) within our lifetimes, without needing to achieve anything approaching MNT or what you term MBT. So this is an independent variable.
Cryobiology has a solid track record of progress, to the point where reversibly preserving embryos and cellular cultures is routine. Tissue slices (which are easier to load and unload cryoprotectants by immersion) can be preserved with 100% cellular survival rates relative to a control, and vitrification, while usually too toxic, can provide good morphological preservation for whole organs. We've yet to postpone aging in any mammal, except via caloric restriction. Even SENS, which aims to avoid most of the work of unraveling metabolism, is very complex and relies on radical gene therapy and quite a bit of guesswork.
Rabbit kidneys are smaller, which affects the rate cryoprotectants can be loaded and unloaded, and the cooling rate which affects toxicity time. Like the brain, kidneys are heavily vascularized, with the exception of the medula at the center which (according to my understanding) is what usually doesn't (but in some cases does) survive cryopreservation.
Losing a piece of the brain does not always translate to loss of function, in fact many areas seem to operate independently of each other. It is true that such injuries (when they are survived) can often result in personality changes, such as Phineas Gage who survived with memories intact, but with dramatic personality changes after an iron bar went through his skull. But there is a fairly strong argument that brain function could be restored after losing chunks, by using stem cells, growth factors, scaffoldings, etc. to grow new analogous chunks where they are missing that make you at least approach the functionality of an average person possessing your DNA. Conceivably, chip-based digital or analog prostheses could also be used for the missing bits.
The parts of the brain most likely to survive are the outer layer (cerebrum). The cerebellum is harder to perfuse, so it is comparatively unlikely to survive. Fortunately, the personality and higher functions seem to be in the cerebrum.
MNT or MBT are not all or nothing. "Future technology" may sound hand-wavy, but it is a compact way of describing a very large set of potential technologies, all of which could independently or in conjunction lead to reanimation of a sufficiently well preserved person. Bear in mind that plain-vanilla biology is already something that operates on a molecular level, and accomplishes very sophisticated results, despite having had to evolve without goal or guidance in an environment where fluctuations of temperature, differences in available nutrients, genetic mutations, etc. are constantly placing limits on what can be done reliably enough to be passed on.
To me it seems rather more burdensome than less to doubt future ingenuity to the degree that is necessary to rule out cryonics from working, particularly in cases where the degree of morphological preservation of the brain is high.
It's more complicated than that, even assuming you are more or less right. If I spend money on cryonics now, that makes it more likely to be more obviously worth someone else's while later. If someone else spends money on it later, that makes it more likely for it to have retroactively been worth my while to spend it now. So this would be a classic case of "rational irrationality" -- much like wasting your individual time voting in a popular election, cooperating in the prisoner's dilemma / stag hunt, or one-boxing in Newcomb's Paradox.
Of course there are third options to be explored. As it happens, currently I'm allocating money towards living expenses and paying down my credit cards rather than cryonics, at least temporarily until my finances improve. Nonetheless, I do contribute significant time towards arguing for the cause, something I feel is more valuable than being signed up. One might also simply send a check to the cryobiology researchers, for example.
In general, I agree, but I'm not sure about the "within our lifetimes" part. I understand that cryobiology has made progress, but I am not convinced that it's moving fast enough. There's a huge difference between preserving tissue slices or even small organs for a relatively short amount of time, and preserving entire organisms (or at least their brains) for centuries. We can't even preserve plant seeds for that long, AFAIK.
I am ashamed to admit that I don't know what SENS is.
Agreed, I never meant to imply that this is always the case. But for each Phineas Gage, there are many more patients who suffer brain damage and never wake up. In addition, I would argue that what we really care about is the preservation of the person's personality (which just happens to be powered by the brain). Yes, this is not an all-or-nothing proposition, and there are degrees of success. Still, if the person who is revived has a totally different personality from the person who went into the cryotank (as I believe was the case with Phineas Gage), I'd count that as a failure.
Sorry, I don't mean to sound too adversarial, but I'll have to press you on this point, as this still sounds hand-wavy to me -- more so than before, in fact. I am not doubting "future ingenuity", but you can't justify cryonics by invoking some sort of an unimaginably ingenious future technology about which we currently know nothing. Violating Occam's Razor makes your argument weaker, not stronger.
Agreed, but there's a huge difference between existing bacteria, and even viruses (which I fully agree are amazing), versus the kind of precise molecular machinery that would be needed to reconstruct a brain. It would have to be small enough to fit into the intercellular matrix, for one thing. I don't want to get too far off-topic, but basically I'm not convinced that MBT is any better than MNT in terms of feasibility.
Ok, I'm going to press you again. If you are convinced that cryonics is the way to go, and this is the best possible prospect for gaining eternal (or, at least, sufficiently long) life, then wouldn't it be logical to put every available dollar toward your own cryopreservation ?
Well, yes, but you could also send a check to Alzheimer's researchers, or cancer researchers, or physicists and other pure science researchers, or even to SIAI.
Hopefully it's equally obvious that I disagree with you that your proposition is "obvious" :-) In addition, it all depends on what you mean by "later". How much later are we talking ? A hundred years later ? A thousand ? A million ? The further into the future you look, the more nebulous it becomes; and thus your expected utility gets lower and lower as the probability decreases. Which was kind of my original point -- it's the low expected utility of cryonics that gives me pause, not its absolute payoff.
Yes we can.
The fact that you keep mentioning timescales suggests that you haven't internalized the fact that we are talking about a temperature at which most chemical reactions are effectively stopped. The major concern is damage on the way down, and whether it is reversible or not.
Here you go.
I already responded to the argument about personality changes due to missing chunks of brain matter. To the extent that you have the same personality as your identical twin, this should be fixable. The relevant concern is memories formed during your lifetime. Furthermore, if the brain's missing chunks can be cloned back into existence, the functionality problem vanishes and takes with it any related mortality. With the scanned upload scenario this is even less of a concern.
The trouble with your claim to parsimony is that you're basically doubting the existence of any and all relevant repair strategies that we don't yet know about. Sure it's possible that we live in a universe where there aren't any repair strategies we can't yet imagine, but again that seems more burdensome to me, not less.
The brain is capable of functioning and growing in the first place. Why wouldn't it be something capable of being repaired?
I've already addressed the concern about the intracellular matrix to some extent: slice it small enough, and you can operate on the surface with much bulkier machines. I imagine this would work best if you keep it super-cold (indicating MNT), but there is the possibility that we develop high-temperature vitrification methods analogous to fixation in amber. Alternately, find reversible ways to disrupt the cells and move them apart to make room.
Not if cryopreservation's chances can be improved more by putting money towards other things.
Huh? I was thinking more like 10 to 50 years. We're talking ordinary business/marketing/infrastructure network effects, plus relatively near term (as I must insist) incremental scientific advances.
"Effectively stopped" is not the same as "stopped"; and of course there are other effects that add up over time, such as mechanical damage an even cosmic rays. But I think my biggest mistake was in vastly overestimating the time scale that you're talking about. I assumed that you were thinking in terms of centuries, but you say:
Does this mean that, should you be cryopreserved today, you expect yourself to be successfully revived after 10 to 50 years ? IMO, that's a very strong claim. I want to address it, as well as the rest of your points, but first I want to make sure we're on the same page.
Er, thanks, but that still doesn't help me figure out which of the possible expansions of the acronym you're referencing.
It might be helpful to ask each other questions about different, unrelated future technologies to see how intuitions differ, and what the patterns in thinking are.