The Brain Preservation Foundation’s mission to validate structural preservation of the brain has been very successful but the link with mind uploading as a means of personal survival raises some important questions. Alexandre Erler makes the case for a distinct cryonics research program based on biological survival.
INTRODUCTION: CRYONICS AND THE BRAIN PRESERVATION FOUNDATION
As someone who is fully supportive of the ultimate goals of the cryonics enterprise, but still views the current state of the practice with some degree of skepticism, I make a point of acquainting myself with the latest evidence regarding the quality of cryonics procedures and their ability to preserve the foundations of a person’s identity through time. Over the past two years or so, I have increasingly seen a recent achievement by 21st-Century Medicine (21CM) cited by some cryonics supporters as demonstrating the scientific validity of those procedures: namely 21CM’s research on aldehyde-stabilized cryopreservation (ASC). This new technique has allowed them to win, in 2016, the Small Mammal Prize, and just this year, the full Technology Prize awarded by the Brain Preservation Foundation (BPF), by demonstrating excellent preservation of the ultrastructure in, respectively, a whole rabbit brain (McIntyre and Fahy, 2015) and a whole pig brain (The Brain Preservation Foundation, 2018). Were I to follow this line of reasoning, I could happily set aside my concerns about the adequacy of today’s cryopreservation procedures, which had now been verified by scientific experts; the proper focus would now need to be on how to responsibly introduce those procedures into a clinical setting, for patients at the end of their lives who might request them.
It turns out, however, that things are not so simple. ASC is no doubt a step forward for the field of brain banking, and as its name indicates, it it is indeed a form of cryopreservation, since it involves vitrification of the brain at -135 °C. Nonetheless, ASC does not count as cryonics, insofar as it uses a fixative solution prior to vitrification and cooling, which could potentially preclude revival of the original biological brain (an essential part of cryonics as traditionally understood). And indeed, biological revival with the help of future technology is not a priority for the BPF’s president, Dr Kenneth Hayworth. Rather, he envisages brain preservation as conducive to life extension via mind uploading: a process that would involve cutting the preserved brain into thin slices, scanning each slice, and feeding the resulting data to an advanced computer that would thereby be able to map out the entire network of neural connections in the person’s original brain, and ultimately to emulate that person’s mind (Hayworth, 2010). This is quite different from cryonics.
Assuming that a technique like ASC is compatible with mind uploading, but not with the revival of the original brain, it should not be treated as a landmark in cryonics research. Admittedly, there is some uncertainty about the truth of that assumption. It seems at least conceivable that the chemical cross-links created by the fixation process could be reversed, and the original brain revived, using future technology. Nonetheless, ASC introduces empirical and philosophical uncertainties (e.g. could we really restore, as opposed to recreating, the original neural structure following the various molecular changes involved?) to a much greater degree than traditional cryonics does.
But why, it might be asked, should one remained fixated on pursuing biological revival via cryonics, if the evidence in favour of good ultrastructure preservation is better for ASC than it is for contemporary cryonics procedures? It is for instance known that, up to now, 21CM’s cryonics protocol involving the use of cryoprotectant M22 has been causing the brain to shrink to almost 50% of its natural size due to osmotic dehydration, hindering our ability to establish the quality of ultrastructure preservation using electron microscopy (The Brain Preservation Foundation, n.d.; De Wolf, 2017). If so, why not join the BPF in focusing simply on the type of brain preservation that seems to yield the best evidence of success, even if this means turning away from cryonics towards mind uploading?
In what follows, I will argue that, given the current state of our scientific and philosophical knowledge, doing so would be irresponsible. The BPF’s commitment to holding brain preservation research to the highest standards of scientific rigour is laudable, and worth emulating. Nonetheless, for those interested in brain preservation with a view to enabling life extension, supporting cryonics-specific research remains the safer bet. We should not simply rely on the BPF’s approach if our goal is to try and save those whom medicine in its current state cannot restore to life and health.
TWO DIFFERENT VIEWS ABOUT PERSONAL IDENTITY AND SURVIVAL
To see why this is so, let us begin by noting the two main philosophical theories of personal identity through time that are relevant when discussing the respective merits of cryonics and mind uploading in this context. The first one, which we can call the “Physical Continuity” (PhyCon) theory, asserts that a person is identical with the physical substratum from which her mind emerges: that is to say, her brain, with its intricate web of neurons and synaptic connections. (For a good exposition of the theory, see e.g. McMahan, 2002.) According to this theory, saving a person from destruction after she has been pronounced dead requires preserving enough of her brain, in a state in which that brain retains at least its potential for viability. What exactly counts as “enough” of the brain is of course a difficult question that would deserve much more discussion. While we can safely say that, all else being equal, it is always preferable to preserve as much of the original brain as we can, the survival of the person arguably does not require perfect preservation. Intuitively, people can survive limited forms of brain damage, such as those caused by strokes. What is more, as cryonicists have pointed out, brain damage that causes significant disability today might no longer be a serious problem (as long as it is limited enough not to undermine personal identity) in a future where cryonic revival has become possible, as the technological means will then likely exist to fully repair that damage, e.g. based on inferences from the state of the person’s brain prior to repair.
The second relevant theory can be referred to as the “Psychological Continuity” (PsyCon) theory. Roughly speaking, it says that you are to identical with the set of psychological features (memories, beliefs, desires, personality traits, etc.) that constitutes your mind. On this view, preserving you after you have been pronounced dead requires ensuring the persistence of enough of those psychological features, in an embodied mind of some sort (but one that need not be embodied in your current biological brain). One variant of PsyCon, endorsed by many supporters of mind uploading including Hayworth, states that preserving a person after legal death requires preserving her connectome, understood as the mapping of neural circuitry encoding one’s memories, skills, and other psychological features – that is to say, the connectome as an informational entity rather than a physical one (Hayworth, 2010), even though the information in question will by necessity be stored in some physical substratum, whether a brain or a computer.
Like virtually all philosophical theories, both the PhyCon and PsyCon theories have their partisans and detractors. PhyCon, for instance, has been said to imply that there is a fundamental difference between a scenario in which a person had her brain suddenly destroyed and replaced by an exact copy of it, perhaps produced via scanning and 3D printing using neurons as basic material; and a scenario in which the person’s brain cells were gradually replaced by new ones over an extended period of time, in the same way as the rest of the human body regularly regenerates itself. While most PhyCon theorists would agree that the second scenario is compatible with the preservation of the person’s identity through time, they will deny that the first is – if the original brain gets destroyed, they will say, so must the person as well, and the new replica brain must belong to a new person not numerically identical with the first one. Some find this difference of treatment between the two scenarios arbitrary (e.g. Parfit, 1984).
Some versions of PsyCon, on the other hand, imply that multiple copies of yourself could all be you. Indeed, suppose that after scanning your brain to obtain a map of your connectome, we then created two identical copies of your mind running on two different computers. Since both copies would demonstrate the same degree of psychological continuity with your previous self, we would have to conclude that both are you – something many find intuitively unacceptable. Other versions of PsyCon strive to avoid that implication by stipulating that you are only identical with an upload of your mind if no more than one copy of it has been created, yet this move leads to other philosophical problems. Hayworth, however, happily endorses the implication that multiple copies of a single individual can co-exist at the same time, and contends that those who object to that implication are simply confused (Hayworth, 2010).
HOW TO MAKE A PRUDENT CHOICE UNDER (PHILOSOPHICAL) UNCERTAINTY
For the record, I personally find PhyCon more plausible than PsyCon (although I also agree that the preservation of one’s psychological features after cryonic revival is highly desirable, regardless of its significance for sheer survival). On that basis, I do not support further animal research targeted exclusively at the development of mind uploading technology. However, my personal opinion on the matter can be set aside for the sake of the present discussion. The important fact is that there are reasonable, honest and intelligent people on both sides of that debate, and that neither side has so far managed to present arguments that would convince all reasonable people on the other side. In such a situation, the intellectually responsible path to take is surely to eschew certainty, and acknowledge that the other side could be right, even if one thinks that this is unlikely and that the arguments favoring one’s own position are very strong.
If that is the case, what is the prudent choice to make for those who wish to promote life extension through brain preservation? I submit that traditional cryonics is the more prudent option to pursue. (This remark could be extended to ASC if one could show that it is in principle compatible with the revival of the original brain, and provided that it is not combined with destructive mind uploading.) This can be demonstrated using a simple argument that considers what the implications are if we assume that PhyCon and, respectively, PsyCon are true.
Suppose first that PhyCon is true. If so, a cryonics procedure carried out properly will save a person’s life, whereas using a technique like ASC that compromises the brain’s potential for viability, followed by destructive scanning and uploading, will kill that person. If PsyCon is true, on the other hand, both methods can ensure survival. Indeed, adequate cryonic preservation of a person’s brain would also preserve the ultrastructure grounding the various psychological features that defined that person. Insofar as traditional cryonics (at least once sufficiently perfected) can secure survival whether PhyCon or PsyCon is true, whereas mind uploading of the kind envisaged by the BPF can only do so if PsyCon is correct, traditional cryonics is the safest bet.
This conclusion is reinforced by the fact that the success of mind uploading at securing personal survival might depend on an additional factor, namely the possibility of creating conscious or sentient machines. If, for whatever reason, computers – which, unlike biological brains, rely on hardware rather than “wetware” – happen to fundamentally lack the capacity for consciouness, regardless of how powerful and sophisticated they might be, then uploads turn out to be no more than computer “zombies” mimicking now deceased people. It’s unclear that someone could “survive” as such an entity. And even if we assume that they could, the value of such survival, devoid of the conscious experiences that make our lives worth living, would be dubious, somewhat like the value of surviving with only a brain stem. This point about machine consciousness equally applies to the idea of a “Moravec transfer”, i.e. a procedure involving gradually uploading a person’s mind to a computer (neuron by neuron if necessary), unlike the BPF’s proposed method (Moravec, 1988). Traditional cryonics, by contrast, can succeed at preserving a person regardless of whether or not computers can be conscious.
Hayworth would presumably deny that any such doubts about the possibility of machine consciousness are legitimate. Indeed, he seems to confidently embrace the so-called computational theory of consciousness, according to which consciousness is fundamentally the product of – highly complex – computation, which we know computers to be capable of at least in principle (e.g. Hayworth, 2015). However, there is currently no general agreement among philosophers of mind or neuroscientists that the computational theory of consciousness is correct, and Hayworth does not demonstrate that it is (although he dogmatically equates the idea that there might be physical properties required for the production of conscious experience which are found in wetware, but not computer hardware, with invoking “magic”).
Furthermore, even if taken for granted, the computational theory of consciousness cannot, absent additional philosophical arguments, show mind uploading to be consistent with personal survival. Assuming that R2-D2 from Star Wars is conscious does not commit us to accepting that a perfect replica of R2-D2 built from fresh parts, after – let us assume – it was destroyed by the Empire’s forces, is numerically identical with the original robot. In response, Hayworth could perhaps abandon PsyCon and instead invoke the claim, famously defended by philosopher Derek Parfit, that personal identity does not actually matter in the way many of us tend to think – rather, psychological continuity is what really matters (Parfit, 1984). (In his reply to an article by neuroscientist Michael Hendricks critical of cryonics, Hayworth actually appears to move in that direction: see Hayworth, 2015.) However, besides the fact that this is again a controversial philosophical view, it notably led Parfit to dissociate psychological continuity from personal survival, and to conclude that the latter was, in itself, also overrated. This position is very much at odds with the life extension project, which the BPF claims to be pursuing.
I cannot but see some irony in the fact that Hayworth, the author of an essay titled “Killed by Bad Philosophy”, should show a degree of overconfidence in his philosophical views that might potentially lead his followers to experience the very same outcome his essay is warning against.
The differences previously highlighted between traditional cryonics and the BPF’s approach are summarised in the table below:
||ASC + mind uploading*
|Physical continuity theory is true
|Psychological continuity theory is true
|Machines can’t be conscious
||Compromised (creates computer “zombie”)
(*It is assumed that the relevant procedures are performed in accordance with the highest standards of quality)
WHAT THE CRYONICS MOVEMENT CAN LEARN FROM THE BPF
None of this is meant to imply that the work of the BPF is without merit. On the contrary, the Foundation’s approach demonstrates a number of virtues that can provide a model for the cryonics movement to follow. These include a commitment to rigorously and impartially evaluating the quality of brain preservation procedures, in accordance with the standards of scientific peer-review. Another example is the BPF’s successful effort at crowdfunding its incentive prizes for brain preservation research, such as the two prizes won by 21CM. For those seeking to promote life extension through brain preservation, who I have argued need to prioritize cryonics-specific research, this suggests two main paths worth pursuing in the future:
(a) Incentive prizes. Such prizes are a powerful tool for stimulating research, particularly in neglected areas of science. Unfortunately (and perhaps unsurprisingly, given Hayworth’s philosophical beliefs), the BPF does not at this point appear inclined to set up any new prize that would include a requirement to preserve the brain’s potential for viability. On the basis of the arguments I have provided so far, I submit that the institution of a prize (possibly crowdfunded) incorporating that requirement would be highly desirable.
How demanding such an incentive prize should be with regards to the winning entry is a matter for further debate. A relatively modest version would require demonstrating adequate ultrastructure preservation in a small mammalian brain, but using a procedure that could in principle be reversed by future technology, making it possible for the original brain to eventually be “re-started”. However, based on a recent talk by Dr Greg Fahy from 21CM, which I attended in May 2017 at the International Longevity and Cryopreservation Summit in Madrid, such a goal may soon be achieved. Indeed, Dr Fahy reported having found a way to largely overcome the abovementioned problem of dehydration and shrinking that has so far prevented a proper assessment of the quality of ultrastructure preservation offered by traditional cryonics protocols (e.g. using M22). Assuming Fahy has now reached that milestone (and I look forward to the publication of his paper on the topic), one could set up a prize with a more ambitious goal: for instance, one could add that besides showing good ultrastructure preservation and retaining the preserved brain’s potential for viability, one should also demonstrate actual viability, say via measurable electrical activity, either at the level of the whole brain or in slices obtained from that brain. Ultimately, the details of such a prize should be worked out by scientists with the relevant expertise (as long as the constraints I have outlined here are respected).
Those who believe that ASC holds greater promise could support a prize rewarding the first team who found a way to reverse the fixative process involved in that procedure, and restore the original neural structure.
(b) Research funds. Such a fund, which could also be crowdfunded, would be managed in a transparent manner by an organization committed to promoting cryonics-specific research. In accordance with standard practice when it comes to funding scientific research, project proposals would be solicited from active researchers in cryobiology (and other relevant fields), and a committee of experts would select the proposals that it deemed most worthy of funding. The organization would then help disseminate the results of the completed projects (e.g. as laid out in peer-reviewed publications).
The scientific experts tasked with evaluating the submissions for either an incentive prize or a research fund should ideally be publicly identified, and sufficiently independent of both the authors of the submissions and of cryonics companies (e.g. they should not be receiving research funding from those companies). Furthermore, while an organization that might implement solution a) or b) could be created de novo, existing institutions might already be able to fulfill that role. One example would be the UK Cryonics and Cryopreservation Research Network (http://cryonics-research.org.uk), led by Dr João Pedro de Magalhães, who has connections with other scientific experts, including Ken Hayworth. Despite the scientific rigour with which it approaches the issue of cryonics, the network is currently underfunded.
While most people may understandably not be able to commit substantial amounts of resources to supporting cryonics research, the success that the BPF has enjoyed so far with its incentive prizes demonstrates that large numbers of even small donations can foster impressive technical breakthroughs and help strengthen the credibility of research projects of the most audacious sort. I believe it is now time to apply a similar approach to the safer bet of cryonics-specific research. Further raising the public profile of such research, and improving its status in the eyes of the mainstream scientific community, can help promote a virtuous cycle leading in turn to more funding and greater professionalization. The sooner we can make this happen, the better.
An earlier version of this article appeared in Cryonics magazine, November-December, 2017.
DE WOLF, A. August 21 2017. Cryonics Without Cerebral Dehydration? Evidence-Based Cryonics [Online]. Available from: https://www.biostasis.com/2017/08/21/cryonics-without-cerebral-dehydration/.
HAYWORTH, K. 2010. Killed by Bad Philosophy: Why Brain Preservation Followed by Mind Uploading Is a Cure for Death. Available: http://www.brainpreservation.org/content-2/killed-bad-philosophy/.
HAYWORTH, K. September 16 2015. Ken Hayworth’s Personal Response to MIT Technology Review Article. Brain Preservation Foundation [Online]. Available from: http://www.brainpreservation.org/ken-hayworths-personal-response-to-mit-technology-review-article/.
MCINTYRE, R. L. & FAHY, G. M. 2015. Aldehyde-Stabilized Cryopreservation. Cryobiology, 71, 448-58.
MCMAHAN, J. 2002. The Ethics of Killing : Problems at the Margins of Life, Oxford, Oxford University Press.
MORAVEC, H. P. 1988. Mind Children : The Future of Robot and Human Intelligence, Cambridge, Mass., Harvard University Press.
PARFIT, D. 1984. Reasons and Persons, Oxford, Clarendon Press.
THE BRAIN PRESERVATION FOUNDATION. n.d. Overview of 21st Century Medicine’s Cryopreservation for Viability Research [Online]. Available: http://www.brainpreservation.org/21cm-cryopreservation-eval-page/.
THE BRAIN PRESERVATION FOUNDATION. 2018. Aldehyde-Stabilized Cryopreservation Wins Final Phase of Brain Preservation Prize [Online]. Available: http://www.prweb.com/releases/prweb15276833.htm.