from futurisms.thenewatlantis.com: This weekend, a philosophy professor will step up to the podium at Harvard University’s Science Center and drop a bomb on the audience gathered for the annual transhumanist showcase now called the “H+ Summit.” Although he shares their aspirations and even tells me that he “would love to be uploaded,” he will explain to the assembled devotees of transhumanism exactly “Why Uploading Will Not Work.”
I’ve seen a preview, and it’s a devastating critique — although it isn’t really new. Others have made essentially the same argument before, including me. In 2003, after years of debating this with transhumanists, I even presented a paper making very similar points at one of the previous transhumanist conferences in this series.
To understand why this is important, we first have to understand what “uploading” is and why it matters to the transhumanist movement. Put simply, uploading is the proposition that, by means of some future technology, it may be possible to “transfer” or “migrate” a mind from its brain into some new “embodiment” (in the same way one “migrates” a computer file or application from one machine to another). That may mean transferring the mind into a new cloned human body and brain, or into some other computational “substrate,” such as a future supercomputer with the horsepower to emulate a human brain.
From the latter stage of this transfiguration, the path would be clear to ascend beyond human physical and intellectual limitations simply by upgrading the hardware and software — adding more memory, faster processors, more efficient algorithms, etc. The mind (or consciousness, identity — or, shall we say, soul) could then become a being of pure information, immortal, flying freely in cyberspace, traveling interplanetary distances as bits encoded in beams of light, assuming any desired physical form by linking to the appropriate robot. One could copy oneself, disperse and later re-merge the copies. One could grow into a gigantic computerized brain (sometimes called a “Jupiter brain”) of immense power, able to contemplate the deepest mysteries of mathematics, physics, and the Meaning of It All. One could become as a god, even literally the creator of new universes, whose inhabitants would wonder who or what put them there. And one could have an awful lot of sex.
That’s the Promised Land of such transhumanist prophets as Hans Moravec and Ray Kurzweil. The latter predicted, in his 1999 book The Age of Spiritual Machines, that “nonbiological intelligence” will vastly exceed the collective brainpower of Homo sapiens within this century, and that the human race will voluntarily “merge with technology,” so that by 2100 there basically won’t be any of our kind left.
Now comes Patrick Hopkins, a transhumanist and professor of philosophy at Millsaps College in Mississippi, to break the bad news: Uploading just won’t work. As he explains in his abstract for the upcoming H+ conference, uploading
will not preserve personal identity. Transhumanist hopes for such transfer ironically rely on treating the mind dualistically — and inconsistently with materialism — as the functional equivalent of a soul, as is evidenced by a carefully examination [sic] of the language used to describe and defend uploading. In this sense, transhumanist thought unwittingly contains remnants of dualistic and religious categories.
Or, as I put it in my 2003 paper:
Arguments for identity transfer cannot be stated without invoking nonphysical entities, and lead to absurdities that cannot be avoided without introducing arbitrary rules.... Dualism is built into the language that Moravec uses throughout, and that we use on a daily basis, “my brain, my body,” as if brain and body were distinguishable from “me,” the true “me” — the soul.... Moravec does not use the word “soul,” but he uses words which are effectively synonymous.
Transhumanists maintain that they do not believe in anything supernatural; they usually abjure belief in God and in an immortal soul. Yet every explanation of and argument for the idea of having your brain scanned and disassembled, bit by bit, so that some kind of copy can be made by some kind of Xerox machine, contains some word whose function and meaning, in this context, are the same as those of that venerable word, and the ancient idea it stands for: the soul.
In the traditional understanding, the soul — dual of the body, and separable from it — carries or constitutes the true identity of the human person. The soul is what we feel in a person’s presence, what we see when we look into a person’s eyes, and it remains steadily in place as a person’s body changes over the years — despite the constant exchange of mere atoms with the environment. The soul, not the brain, is what is conscious, as no mere material thing can be. It is connected to that transcendent world of pure spirit, where, perhaps, all will be understood. In some accounts, the soul endures after death and goes to Heaven, or to Hell, or else hangs around in graveyards and abandoned houses. A voodoo priest can capture a soul and imprison it in a doll — more or less what the proponents of uploading hope to do by means of technology.
Thus Moravec, in Mind Children (1988), argues that some future “robot surgeon” might have the ability to probe your brain a few neurons at a time, building up a detailed model of those cells, until a computer simulation is able to predict those neurons’ firing patterns exactly. Then it could override the output of those neurons, effectively replacing them with simulated neurons. The rest of the brain would go on working as normal, and since the rest of the brain can provide input to, and work with the output of, the simulation, just as well as it would with the real neurons, Moravec argues that “you should experience no difference.”
Continuing the process,
Eventually your skull is empty, and the surgeon’s hand rests deep in your brainstem. Though you have not lost consciousness, or even your train of thought, your mind has been removed from the brain and transferred to a machine.
The premise that Moravec starts with — that you would not feel any different if some small number of your natural neurons were to be replaced by artificial neurons that provided the same input-output functions to the rest of the brain — might at first seem reasonable. Neurons die all the time, by the thousands every day, and you don’t notice any difference. If a few of them were replaced, how could you even tell?
But what if your brain cells all died at once, faster than the speed of neural transmission, say because a bomb exploded nearby? Would you notice then? No, you would be gone before you could feel anything. Death does not require our ability to perceive it; nor can we escape the Reaper by refusing to acknowledge him. In this we are unlike Wile E. Coyote, who can’t fall until he sees that he’s over the edge of the cliff.
Moravec claims that “you have not lost consciousness” at the completion of his process. This is powerful verbal magic, appealing to the sense that consciousness is an indivisible whole. Yet any number of experiments and observations from psychology show this to be an illusion. You are one body, leading one life, but your mind’s unity is a synthesis.
What is this thing, the “mind,” that Moravec claims can be “removed” and “transferred”? What exactly is it made of? Some say “information,” and that sounds appropriately scientific, but information has no existence, so far as we know, without the physical “substrate” used to “represent it.” When we speak of “information transfer” from one thing to another, we usually mean that some physical agent makes some physical measurement of the first thing and imposes related physical changes on the second thing. Pure information, completely separated from any physical matter or energy, would be something whose existence could not be distinguished from its nonexistence.
Even though transhumanists generally do not admit to believing in an immaterial “soul,” proponents of uploading continually invent or repurpose technical-sounding terms as stand-ins for that forbidden noun. Thus Moravec advances a theory of
pattern-identity ... [which] defines the essence of a person, say myself, as the pattern and the process going on in my head and body, not the machinery supporting that process. If the process is preserved, I am preserved. The rest is mere jelly.
Not only has Moravec introduced “pattern” as a stand-in for “soul,” but in order to define it he has referred to another stand-in, “the essence of a person.” But he seems aware of the inadequacy of “pattern,” and tries to cover it up with another word, “process.” So now we have a pattern and a process, separable from the “mere jelly.” Is this some kind of trinity? Or is the “mere jelly,” once appropriately patterned and undergoing the processes of life, what real human beings are made of — that and nothing else that is known to science?
Similarly, Kurzweil argues that
we should not associate our fundamental identity with a specific set of particles, but rather the pattern of matter and energy that we represent.
If taken literally, this carelessly worded statement suggests that we are not our true selves, but mere representations of our true selves! But note again that Kurzweil points to the assumed existence of a “fundamental identity” which is distinct from the body. In other words, he is referencing the idea of the soul, and manipulating the dualism that is embedded in our way of thinking about people.
So it goes with every author who advocates the idea of uploading as a route to immortality and transcendence. They must always introduce some term as a stand-in for “the soul” and argue that by whatever process they propose, this object will be moved from the old brain to the new one. Otherwise, they would have to describe their proposal not as transferring “you” (your soul) to a new body, but as making some kind of copy — perhaps an “identical” copy, structured the same way at the molecular level, or perhaps a mere functional copy, “instantiated” in some different kind of “substrate” (as one might copy an Old Master’s painting to some pattern of ink dots on paperboard).
Having one’s brain copied, particularly if it requires, as any even remotely plausible technical scenario for uploading would require, the complete disassembly of the original, hardly sounds appealing, since having your brain disassembled will pretty clearly kill you.
Thus the conclusion: Uploading cannot work, if we define its “working” as a way for people to escape death and transcend to existence as a technological super-being.
To be sure, these arguments against uploading are not novel — they go back years. In my own case, I’ve been making these points for around a decade, including in the paper I presented at the 2003 conference. At that conference — it was called “Transvision,” the annual meeting of the World Transhumanist Association (WTA), before the group was renamed H+ — I gave a talk in a session chaired by no less a transpersonage than WTA cofounder Nick Bostrom, who now heads his own well-endowed “Institute on the Future of Humanity” at Oxford University. (Note: The conference organizers enhanced my bio and awarded me an honorary doctorate for the occasion.) I also posted the paper on various listservs, provoking comments from another transhumanist luminary, Trinity College lecturer and past WTA president James Hughes, who accused me of shooting a dead horse (although I was unaware of anyone else making these arguments about uploading at that time). So two of the most important leaders of the transhumanist movement were among those familiar with these arguments against uploading.
Which raises a question: Why only now does the leadership of the transhumanist movement see fit to acknowledge the serious case against uploading? Why will this weekend’s H+ conference feature not only Professor Hopkins with his critique of uploading, but also WTA cofounder David Pearce arguing that uploading “involves some fairly radical metaphysical assumptions” and that Kurzweil’s vision of a voluntary mass exit for Homo sapiens “is sociologically implausible to say the least”?
The answer may be connected to public relations. The transhumanist movement has begun to break the surface of public awareness, not only through the infiltration of its science-fiction visions and technophilic attitudes, but overtly as transhumanism. The further mainstreaming of transhumanism seems to require some P.R. maneuvering, including a rebranding (the glossy new name “H+”). It may also require a moderating of ambitions. The old “Extropian” dreams of uploading and wholesale replacement of humanity with technology may be too scary and weird for mass audiences. Perhaps more modest ambitions will have a broader public appeal: life extension and performance enhancement, cool new gadgets and drugs, and only minimal forms of cyborgization (implanting technological devices within the body). In other words, more Aubrey de Grey, less Hans Moravec; more public policy and less cyberpunk; more hipster geeks and fewer socially-impaired nerds. A kinder, gentler Singularity. Maybe even one with women in it.
Perhaps some of the transhumanist top guns recognized “uploading” all along for the ontological nonsense it is. Perhaps that’s why they now stand ready to throw it overboard like so much ballast threatening to drag down their balloon. It sounds too loony and it’s too easy a target, too obviously inconsistent with transhumanism’s claim to being a creed grounded in science and technology.
If so, distancing themselves from uploading is probably a smart move for the H+ leaders, but it risks a split with their base, and the formation of new, hard-core splinter groups still yearning for cyber-heaven, still committed to becoming something that is too obviously not at all human.
In the longer run, this strategy to divert attention away from transhumanism’s original and ultimate aims will not work — or, at least, I am hopeful that it will not. For transhumanism itself is uploading writ large. Not only is the idea of uploading one of the central dogmas of transhumanism, but the broader philosophy of transhumanism suffers from the same kind of mistaken dualism as uploading, a dualism applied not just to human beings but to humanity, collectively understood. Transhumanism posits that “the essence of humanity” is something that can be preserved through any degree of alteration of the human form. In other words, it posits that humanity has an essence that is wholly separable from living human beings, an essence that is transferable to the products of technology.
The early history of transhumanist writing helps emphasize this point. If you investigate the origins of the notion of uploading, you’ll find that initially, back in the 1980s and 90s, it was called “downloading.” Back in the days of green screens and floppy disks, far fewer people thought that existence as a computer program would somehow be a step up from being alive and human. The idea was just that it would be nice to have backup copies; in case of an accident, the atman file could be uploaded back to fresh-cloned flesh.
But as technology and the cult of technology co-evolved, “downloading” became “uploading” — a dream of ascension and transcendence that became a vision of rapture for the geeks. Transhumanism raises technology above humanity, and all but deifies it, at least making it an end in itself, and the end of humanity, rather than a tool to serve human ends. This seems a new low for philosophy, an upside-down morality, a grotesque distortion of the scientific rationality and enlightened humanism of which it claims to be a continuation.
The H+ leadership would like to hide this by underlining their own humanity and recasting their rampant technophilia as a human desire for betterment. But if the “H” stands for humanity, what is the “+”? If it is cyborg hybrids of man and machine instead of superhuman robots, is that so much better? Do we really want that to follow us, as the next step in “evolution”? Whatever the “+” may stand for, I am sure it’s not my kid.