Wednesday, 19 August 2015

Frontiers 5 - Downloading Minds

In early July, the Guardian reported that "Scientists have linked together the brains of three monkeys, allowing the animals to join forces and control an avatar arm, in research that raises the prospect of direct brain-to-brain interfaces in humans." As with all such technology, it is 'early days' - a Borg-like collaborative ability to do something mechanical is interesting but not yet useful. However, if minds can be connected to other minds, we are moving into an area once deemed pseudo-science or magic - telepathy - and it suggests that something deemed intrinsically absurd - uploading minds to another platform than the biological human body (or into another biological form) may not be so absurd in the far future. If brain-to-brain, why not brain-to-alternative-substrate.

The possibility raises all the questions about 'what it like to be a bat' in a new form. What will it be like to be a human being whose sensory inputs are radically changed either in a soft form, by inheriting the subtly different senses of another human being, or in a hard form by being given new senses in a new form of material embodiment that might need very different processing tools to cope with the data. Will the first people to explore new worlds in inner space go mad or not? And will we have issues of neural privacy and hacking that make our issues with internet privacy seem relatively easy to resolve by comparison?

The practical short term benefit of current brain research is relatively simple and holds very few existential terrors and only tactical opportunities. If scientists can get the human mind to be able to 'move' prosthetic limbs and mechanical additions to the human body, then this would be a stunning move forward for the quality of life of the severely disabled, a liberating use of technology for a significant minority that could materially reduce human misery and frustration. On the other hand, "The scientists said that in the future, the concept might be extended to produce neurally connected “swarms” of rats with collective intelligence" [Guardian] which I suggest we need like we need a hole in the head.

There is another aspect of the case. Scientists have been dismissive of telepathy and perhaps a lot of telepathic discourse is wish fulfilment but the folk beliefs of many people attest to practical experience of small-scale examples of mental connection over long distances, especially between family members and sexually bonded people. This is before we even get into the even more contested territory of psi, clairvoyance and so forth. I have witnessed directly at least two examples of 'death telepathy' - my mother waking in the night to report her father coming towards her and calling out her name and the phone call coming to say that he had died without warning that night (though not entirely unexpectedly) and my own experience of waking with a sharp intolerable knife like feeling of something slicing through my brain and getting a phone call from my father an hour or so later that my brother had been shot on military exercises - in the head - though the timing of the shot and my waking was out of synch (though not perhaps his actual death). These could be coincidence or chance but reports of this phenomenon are widespread, unprovable (though both 'wakings' were witnessed by third parties) and tend to run in families as if there is some genetic aspect, some lost sense, which is to be found in some gene lines and not others.

The science of the 'super brain' (the ridiculous hyperbole of the media-friendly scientist) tells us nothing yet about this folk phenomena which remains scientifically very elusive. Other similar phenomena I have experienced, and which I accept as 'normal' no matter how talked away by positivists, all have one thing in common ... they are triggered by extreme external events or frustrations, usually 'in the zone' (that is distracted from reasoning). There is no means of willing or controlling or redoing what has happened. It is a form of altered state (which we discussed in the last Posting in this series).  At the end of the day, the monkeys and the rats involved in the scientific experiments were connected by material arrays making a physical connection between the brain matter and the man-made external tools. There is no evidence here of thought leaping across space and perhaps time through the ether (though if radio can, why not mental transmissions?) but only of the ability to connect matter in its formal atomic sense.

For this reason, we have to separate the random, probably real but uninvestigable business of mental transmission without any obvious material connection from this business of wires and cables - so the question becomes not whether we can enter into someone else's mind through telepathy but whether our minds can manipulate other minds, merge with other minds or be uploaded into other bodies. This is the frontier that we are looking at here. As for the monkeys, "Although their brains were not directly wired together, [they] intuitively started to synchronise their brain activity, allowing them to move the arm collaboratively to a reach for a virtual ball on the screen" which is suggestive at least.

As for Whole Brain Emulation [WBE] which is the business of taking a brain and so a mind and enabling it to run on a non-biological platform or substrate, the consensus among speculative scientists is that the technology is theoretically perfectly possible at a mechanical level even if it does not exist yet and may not do so for some considerable time. It is another question whether it is a practical proposition. Another again whether the uploaded mind will actually be conscious. So, although not immediate, the philosophical, ethical and social issues raised by WBE are going to have to be faced at some stage if our society continues to develop technologically as it is doing.

The obviously troubling questions to those who are candidates for uploading are those of identity ... is the uploaded person the same person? If there is an upload and the original remains in place, do we have two separate persons or not? If the upload is a mass upload - ten thousand uploads of the same individual - what are the implications? ... and ... another set of questions along different lines ... what changes will happen to identity and personality with massively different sensory inputs or increased or different processing power or less or more mobility or a virtual environment into which the mind is lodged? And what of death if the uploaded person is apparently immortal (or near to it) yet reliant, just as we are, on energy sources being maintained and on mechanics and systems outside the uploaded self? What becomes of the person when the biochemical basis for feeling is eliminated? Are neuroses transferred with the mind?  What happens to a mind that no longer feels hunger or suddenly has to cope with the different desires and hungers of another human body? And what of chimera or hybrid bodies that have mechanical and biological elements or merge the biologies of different species? And so on and so on ad infinitum.

But this is all high-end speculation which we frequently warn about. Speculative science is a form of mental activity somewhere between science and philosophy but it is not necessarily either good science or good philosophy. It can, too quickly, become a genre within science fiction. The 'frontier' is not (any more than in our space-related postings) way out there with the mental equivalents of star travel but in the more near-at-hand whose applications are likely to be prosthetic and related to the techno-enhancement of our own species - a subject we will deal with in a later Frontiers. Given that predicted dates for full human brain simulation through super-computation have passed already, it is probably true that Kurzweil's 2029 predicted date will pass in the same way. It is true that "a massively parallel electronic counterpart of a human biological brain in theory might be able to think thousands to millions of times faster than our naturally evolved systems" [1] but thinking is not all there is to being - a common mistake of the enthusiastic nerd.

Something that thinks at those speeds is not a human brain but something different, far faster than a human brain. It would be a brain-like thing, that is all. In any case, the speed of the connections (where electronics are far faster than biology) is not at issue but the number of possible connections under conditions where the number of connections in an average human brain would require an enormously large supercomputer. Something that large is self-evidently not relating to the environment in the same way as a brain-sized brain in an ambulatory android. Reproducing human brains (before we even get to the framework for uploading) may not require a revolution in processing speed but it would require a nanotechnological revolution in the hardware that does the processing and then a choice of embodiment that at least approximates not only the sensory capacities of the human but their mobility in the sensory world that humans inhabit (rather than a simulated silicon sensory world). It would be wrong to be dismissive of progress to date - not only is nanotechnology in itself a leading edge technology of fearsome potential but there are time-sharing options that supercomputers can use and which are already getting close to approximating mammal brains albeit at slower speeds. Something brain-like with an intellectual capacity exceeding that of our species will undoubtedly appear within the next three decades or so. But any uploaded human minds entering this system will not be human but post-human. We are speaking here of species-replacement and not enhancement.

A key question is one of data capture when it comes to the matter of moving from creating a brain-like thing that emulates the human brain, but is not like any actually existing human brain, to emulating an actually existing human brain (the upload of a personality). If you look deeply, the extrapolations from (say) Moore's Law rather evade the issue of computational complexity - the enthusiasts for mind uploading simply have no idea what the actual computational requirements, likely to be very huge indeed, would be for successful uploading and, of course, there is the other side of the equation, how one abstracts data from the really existing brain so that it might be uploaded then or later without killing it (creating a copy). Since 2005, there has been a Blue Brain Project at the École Polytechnique Fédérale de Lausanne which aims to reverse engineer a mammalian brain to create an artificial electronic brain but the researchers have no illusions about what they are doing. One of the key researchers said in 2004 that "in the brain, every molecule is a powerful computer and we would need to simulate the structure and function of trillions upon trillions of molecules as well as all the rules that govern how they interact. You would literally need computers that are trillions of times bigger and faster than anything existing today."  It is not likely that we have seen a supercomputer, even hidden away in the defence establishments of the great powers, that is "trillions of times bigger and faster than anything existing" in 2004 appear since then. In a moment of enthusiasm later, this researcher predicted a detailed functional artificial brain by 2019 so maybe ... since the funding arrived, that same researcher has been noticeably rather silent.

There is a simulation model that this writer imperfectly understands but based on reverse-engineering the blueprint of the brain's data system but this has the huge assumption embedded that there are no quantum mechanical processes involved. The jury is out on that one. A full brain map is technically feasible (we understand) in terms of data storage of the brain's system fixed in time but the sheer complexity of the functioning biology of the brain may make the final tally of data far too big to handle. The logic of the situation is, once again, not an uploaded brain of an actual person with memories and all but rather a brain-like thing that mimics a person. Will it be conscious? Hard to say when we only surmise that any other human being is conscious. No current technology appears to be sufficiently robust to reliably capture the actual molecular structure of the brain, bringing us back to the Blue Brain Project's 2004 concerns. Naturally, that does not mean that a future technology or the improvement of current technology might not result in the level of data capture required but the obstacles seem formidable.  And it should be remembered that nearly all (though not all) the research work being undertaken is funded not to upload minds but to understand the brain better for medical purposes - specifically, "various psychiatric disorders caused by malfunctioning neurons, such as autism, and ...  how pharmacological agents affect network behavior."  Some richer than average enthusiasts anxious about death are keen to fund research designed to upload a mind and reboot it in virtual space but offering, for example, $106,000 as a prize (as in one case) is unlikely to speed matters up a great deal.

The science may also be irrelevant, as we reviewed in the last Frontiers posting, if consciousness (such as it is) is not quite as quantifiably physicalist or functionalist or is based on quantum events as many philosophers suspect and argue. The critics are persuasive but there is no need to move into a dualist position to follow them. There is a revised materialist model - somewhat along the lines of Arthur C. Clarke's famous 'magic is just undiscovered science' - in which the hard materialists are right that mind is an emergent property of matter but that they are wrong that it is to be understood in terms of the quantifiable matter of classical physics or in terms of the possibility that it can be understood in real time by a sufficiently intelligent system which can then reproduce it as an 'upload'. We have seen that the simulation model cannot work if there are quantum effects but no model can work if the amount of computation of all states of being within the conscious brain is far greater than anything that is material that is not that brain itself.

One can thus perceive easily enough of a brain being created but not of a brain that can mimic perfectly another brain if only because the mimicking brain will always be sufficiently marginally different to be different from the uploaded person. At best, the person who dies does not simply transfer from one state to another but ceases to exist and a simulacrum emerges that believes itself to be the person that has just died. To all intents and purposes, the second brain, in believing themselves to be the first person, is the first person in a form of self-delusion but the differences between the state of the brain in the recently deceased body and in the new embodiment or virtual state will spin the second person rapidly into a new status altogether unless every possible memory, unconscious behaviour pattern, biological trigger and so forth are also transferred - that is, not only the brain must be transferred but the body including its gut bacteria and neuroses, indeed its somatic memory as well as its actual memory. Yes, we may able to create computational duplicates but not more than this. The fact that the uploaded entity believes itself to be subjectively the person from whom it was uploaded may be useful but it is not true. This is the fallacy of the non-self, fashionable in the dislocated modern world, but self-evidently not the case where a mind inhabits a brain which inhabits a body which inhabits a world with a continuous history of direct material experience. Transfers of minds merely transfer parts of the whole. This is the crux of the debate over whether an uploaded mind has been moved or copied.

Whatever the feasibility (and until brain emulation is judged more feasible most of the theoretical ethical, socio-economic, political, legal and philosophical thinking is pleasurable but rather idle), the actual frontier at this time is that of practical neuroscience. This also means a frontier of psychological manipulation, surveillance and intrusiveness, cognitive enhancement and medical improvement. The neuroscience directed at these ends cannot be isolated from research into artificial intelligence because it might be regarded as a race against time whether the next advanced 'consciousness' will be an enhanced virtualised or embodied human or an embodied or non-embodied artificial intelligence. The bookies would probably put money on an AI becoming at least far more intellectually enhanced than a human (and with its own views on the virtualisation or mental enhancement of its creator) well before a mind is ever uploaded.

The dialectic between the human fear of death that drives the more radical models of mind uploading and the existential risk of emergent AGI is scarcely discussed and yet this is the dialectic that matters - the human species wants to survive as individuals while something is emerging that may want to survive itself and may see the species or individuals as threat. Both sets of entity are, bluntly, being explored by capitalism as forms of slave labour and enhancements of the lifestyle of biological humans - both sets of entity may have their own views on this. If uploaded enhanced humans and emergent AGI find themselves competing for computing space within the virtual world and biological humans and embodied AGI are competing for material resources in the world as we know it, then we have science fiction scenarios that make the Terminator series look thoroughly simplistic. It means three new sets of highly intelligent enhanced 'types' emerging out of technology in a situation where none of the three should underestimate the biological cunning of the root species in a material world of four-dimensional space that it understands well from hundreds of thousands of years of evolution. In the end, for all their advanced intellectual capacity, the three new proto-species may not stand a chance against the native humans ... plugs can get pulled.

We could go further down the line of our magical mystery tour and look at something which the proponents of brain emulation seem reluctant to discuss - the socio-politics of undertaking brain emulation in a society in which the first truly successful candidates (taking into account the possible horrors of existence for the experimental candidates) are likely to be represented by very rich people who can afford what they think is immortality but which is actually a post-human status that may give them enormous further personal advantages in the competition with the poor saps who remain homo sapiens sapiens. Science fiction has, with Battlestar Galactica and the Terminator series, now developed a corpus of work on the alleged dangers of one rival species, artificial general intelligence, but is only now trying to come to terms in its usual primitive and clumsy way with the theoretical threat of post-human brain transfer.

From this perspective, Luc Besson's Lucy was a far more interesting film than the predictable Transcendence simply because it tried to imagine the post-human without recourse to the standard trope of the genius-billionaire within a standard Hollywood adventure-love story that could as easily have been crafted at any time in Hollywood's history. The film failed, of course, but at least it tried. The interesting socio-politics lies not in the death-averse behaviour of billionaires - after all, the trope is as easily done with vampires as with technology as we have seen in The Strain - but the response of the masses and then of a minority of the masses to an existential threat to their identity as the dominant or most conscious of species. Extant covered this somewhat with a Kaczynski response to androids. Downloaded emulants would be highly vulnerable to deliberate warfare on their kind through destroying their new substrate as something dependent on energy and vulnerable to viruses and hacking. The terror of being stuck in a substrate and being sent to virtual hell and torment by vengeful hackers may make death the soft option. It is only (paradoxically) the lack of imagination of transhumanist billionaires and their Igors that allows them to continue with their mad struggle for immortality.

But we are falling back into the worst sort of speculative 'science'. Maybe Kurzweil is right that we will be 'digitally immortal' by 2045. I doubt it. What is more likely is that a form of digital consciousness that may or may not be zombie-like in practice and subject to its own programming will exist in some form at some stage, that it will be existentially different from us and may either be convergent with artificial intelligence or competitive with it. But the argument that 'we' can be downloaded as minds may be superficially attractive to the more autistic end of the spectrum that positions intellect ahead of emotion (yet why bother if AI can always trump us in intellect) but which strike me as speculative wish-fulfilment in the face of a classic existential anxiety over death - a sentimental and emotional response. The ultimate techno-maturbatory fantasy is to try to deal with two radical problems - that we are probably biologically incapable of reaching beyond the solar system in our current form (see our first Frontier posting) and that we are all going to die - by uploading ourselves into mechanistic starships that roam the universe near the speed of light, changing our perception of time without (apparently) going mad and yet remaining 'human'.

The argument that even if we could upload our minds, they would cease to be human minds (the new life form argument) is argued better from somewhere closer to the dear old 'bat' question of Nagel by Efstratios Filippidis in his analytical piece on mind uploading than here. He usefully summarises his position here. He argues quite simply from the differences in the perception of qualia that emerge from the formal substrate of the brain so that, although we have invented or discovered (according to Platonic taste) a scientific world of workable technologies based on the workings of matter, the mind, based on brain matter, works itself out with such massive variation between individuals that, even though we might reproduce the material basis for that mind, we may not so easily be able to reproduce the perception of qualia (sound, taste, smell, pleasure and so forth) that are based on the brain being embedded somatically in a total biological system. In short, reproducing minds or even brains in isolation from bodies is literally de-humanising a species that is not just intellectual but sentimental and emotional and has merged from within a processing system that is excellent not only at analysing situational data from several sources in motion but also, and this is key, selecting and forgetting. A mind detached from that system of mobility, sensory perception and somatic embodiment is a different from us as we are from the bat. A mind that cannot forget is never a human mind.

Of course, a mind might be uploaded with the memory of all sensations but, unable to live in the new present with mobility, the same sensory machinery and bodily structure of the past, the memory will be detached from any new experience under the new conditions of existence with very different mobilities or lack of mobilities and senses. In the end, the only true human upload is one into another (presumably improved and longer lasting) bio-mechanical body with the broadly similar sensory and mobility functions as the old. The true frontier technology of uploading is not really a matter of uploading minds but of transforming bodies in all their complexity. One may as well accept improvement to existing bodies and minds (the medical model) as a much more useful investment of human energy than sending isolated minds to go mad in energy-vulnerable autistic micro-worlds. Although much of Filippidis' essay is (frankly) absurdist science fiction and even unphilosophical moral valuation, this problem of perception of qualia does lend credence to the position that Nagel's 'bat' problem is also in play. As Filippidis puts it at one point:
"One of the powers that our virtual ones would have inside the virtual world of a computer [the model he is discussing] is to quickly transform themselves into whatever they want. By doing so, they could, soon after the uploading process finishes, rapidly or even immediately become something quite different from their original one. Consequently, their common identity with their original counterpart would be lost very quickly. We would not longer identify ourselves with our virtual counterpart and the entire idea of eternal permanence of ourselves would result in immediate failure."
This, of course, falls into the science fiction category of thought but the point underlying this is valid - that the relationship of a new uploaded identity and the old identity is not one (in fact) of identity but one of an intensification of the difference that exists even in ordinary humans between one moment and the next of being a 'self'. With humans, the transformation of self - whether circular re-invention or human progression and personal development or the leaps and bounds of punctuated equilibrium - operates at a steady pace in accordance with the underlying somatic and sensory apparatus. The transfer of a mind from one substrate to another is a far more radical shift of perception and embodiment and although, within the new body, the self will construct itself again within its steady time frame based on its sense data and type of embodiment, the distinction between the two selves will be a radical one in which only memory unconnected to the means of creating new memories of that type will exist. One may postulate that, just as the unconscious exists and is connected to neurosis and dysfunction, so the 'human' will exist as the unconscious of the newly uploaded mind and be its source of neurosis and dysfunction. Perhaps AI psychotherapists will be helping uploaded minds probe their 'humanity' for answers to dysfunctionality within the artificial world of a starship flight that takes 70,000 years ...

It is thus probably true that, at some stage in the future, minds will be uploaded in some form but it is unlikely that we will be able to call these minds human. We have already said that they will be post-human, evolved out of the human much as we are evolved from common ancestors to the chimpanzee. The mistake should not be made that evolution represents superiority or inferiority. It will simply be a better product for a new environment - whether a world created out of silicon within our world or a world of anti-biological space travel. For those who like their speculative science, Martine Rothblatt is a leading proponent of mindware but I do not think even she has quite thought through this post-human aspect of the case. She presumes (it would seem) that mind clones would be, well, clones, basically still 'us' in a different form rather than replacements for us in an environments where we cannot survive otherwise. This just does not seem plausible.

The post-human aspects of the case are made more explicit by the closing remarks of brain emulation guru Randal Koene in an excellent May 2014 Popular Science run-down of the more cultic, some might say parasitical, beliefs in brain emulation as a goal that (on the other hand) seems to be creative in pulling different scientific disciplines together in a way that reminds us of the early days of cognitive science:
"Brain uploading, Koene agreed, was about evolving humanity, leaving behind the confines of a polluted planet and liberating humans to experience things that would be impossible in an organic body. “What would it be like, for instance, to travel really close to the sun?” he wondered. “I got into this because I was interested in exploring not just the world, but eventually the universe. Our current substrates, our biological bodies, have been selected to live in a particular slot in space and time. But if we could get beyond that, we could tackle things we can’t currently even contemplate.”"
A lot of this falls down on one simple fact - unless we are dealing with a case of gradual uploading where one moves slowly and consciously from one substrate to another leaving nothing behind (that is we live consciously and aware through the transition, connected to both substrates and then jettisoning the first at the right time as a shell or first stage rocket), the transfer of a mind from one place to another is a copy. The reasoning is simple - if a transfer takes place and one mind remains in situ and another in the new substrate, then two versions (that will then diverge according to the dictates of the substrate) of the original exist. For the first to cease to be while the second version continues is to have one die and one live. Would the first observe their identical but diverging copy there and then freely extinguish themselves here in the belief that the copy was their very self. I think not - there would be a realisation that the first consciousness had not uploaded and was doomed to continue in the state into which it had been born whereas the brother or sister copy would go on to an entirely different existence albeit with a mental substrate of the first's history and memory and a belief that it was as continuous as the first. If belief makes the second into the first, then the second is the first but the first cannot become the second and remains behind ... or dead. Each upload will be, in fact, an existentially risky suicide and an upload to preserved a loved one in a virtual environment will be a simulacrum if only because existence in a virtual existence will transform the second into something different unless programmed otherwise and a loved one programmed by the lover is not a person but a toy.

The same applies to the fantastic notions of Koene and other transhumanists which are centred on species survival as if a) we are as individuals going to be satisfied with the abstract business of becoming representatives of the species (this, if anything, merely testifies to the obsessive and autistic aspects of the transhumanist mentality) and b), following on from the last paragraph, the starship persons in their silicon bodies may have descended from us but are not us, do not have our culture which is based on biology and a particular environment and at most may embody a certain sentimental relic of our human reality. If the best the transhumanists can offer us is an eternity of scientific curiosity in the wastes of space or on burning or frozen planets embodied in metal and code, then, again, it says something about the life-denying obsessions of men and women so frightened of death that they have forgotten how to live as humans.

Some of the speculative science involved here may be highly rational but it derives from assumptions about what it is to be human that are mere manifestations of the death instinct. The radical transhumanist of this type is little more than a figment of his own imaginative demise. No wonder there is a turn to religion - to Omega Points, the potty Christianity of Tipler and Teilhard de Chardin (with the occasional nod to the death instinct that is Buddhism) - in latter day transhumanism. This urge to the 'cod-spiritual' is the negation of the true transhumanism of Nietzsche (the overcoming of our socially determined nature and a return to an understanding of our biological drives and instincts as Life) and at one with the deeply sad loss of real science to the useless 'dialogue' with theology that infects the weaker minds of our culture. We can move on at this point and respect those practical transhumanists who hold to the Nietzschean vision of the enhancement of the human living in the world - lengthening lives to live well and not just longer, eliminating mental as well as physical disability, enhancing intellect, skill and the ability to love, pleasuring the senses and, yes, being curious about the world not as escapist speculation but as a process of command and control through the mastery of matter by mind.

Filippidis' own preferences are closer to mine but one still suspects that we would do better to spend our lives enjoying the time between birth and death and seeking to improve the environmental, somatic and mental conditions for this and future generations than expending energy and resources on a sub-Taoist search for immortality:
" ... my preferred methods for radical life extension and bio-techno immortality in the ( hopefully ) near future, are genetic ( and epigenetic ) engineering; continuous repair and elimination of all types of molecular garbage and structural defects accumulated by our metabolic processes, and, possibly, a progressive replacement of evolutionarily faulty cellular micro-structures and organs by artificial nano-structures or programmed nano-robots and synthetic improved organs."
So, returning to our frontier analogy, much of speculative science - starships and immortality - is really a distraction from the job in hand. The real frontier lies broadly in Filippidis' basic summary in the paragraph above. It is in the much more limited area of brain-machine interfaces (BMIs) that we see a real frontier rather than the hysterical almost cultic desires and claims of aspirant billionaire transhumanists. Sweep away the fruits and the nuts and we have some practical possibilities where technology can enhance human capability (which we are likely to return to in a later note). The first true BMI was probably the cochlear implant to improve hearing. Others could help victims of stroke or spinal chord injury walk again. Small electrode arrays might soon pick up neural signals from the motor areas of the brain, have them decoded by a computer and then re-transmit them to a prosthetic limb which then becomes integrated into the body-mind. The story we started the posting with is probably less interesting in that thre monkeys can communicate telepathically and more interesting in that the prosthetic limb is being manouvred by the mind of the monkeys whether one or three. There are still many technical challenges to be overcome, of course.

The bottom line here is that, with brain emulation and uploading 'minds', we are into a territory we have seen in previous postings and why the frontier analogy is so useful. There is a frontier but it is a frontier for human needs within sight of human resources. Just as entering into the inner solar system is part of the existential business of protecting all humans from asteroid impact and raising the possibility of strategic profit for some humans from resource mining, so investment in neuroscience is primarily about helping many humans and future generations overcome serious mental and physical diabilities. Just as hurtling outside our solar system to fly to the stars is scientific fantasy to all intents and purposes, at least as biological entities, so is cheating death by becoming post-human. The aspiration to the latter might motivate research to the former and feed young imaginative minds and perhaps public support for funding into the programmes but it could equally be a distraction that, if taken too seriously, could cut vital enthusiasm and funding with the first proof that the super-dream of the nerds was either certainly not feasible or too expensive for the alleged benefits or both. This would be tragic - as we saw with the first space programme which died too early because it was no longer useful in great power politics and lost us decades in defeating the asteroid threat. The speculative scientists always need managing and correcting if we are to bring people (especially intellectually lazy legislators) down to earth and return our efforts to what can be achieved if only we stick at it and do it for the right reasons. Becoming a post-human is up the with flying to Gliese 667 - second order to diverting asteroids and curing disability and we should perhaps start telling it like it is instead of treating ourselves like children (charming though that may be) with an excess of imaginative fantasy.