Digital Brains in Silicon Vats
Transhumanism and the dictatorship of the dead
What is transhumanism?
Max More, a leading figure in the movement, has said that the term refers to:
Philosophies of life … that seek the continuation and acceleration of the evolution of intelligent life beyond its currently human form and human limitation by means of science and technology, guided by life-promoting principles and values.
Transhumanists want humanity to take conscious control of its own evolution. For instance, they want to overcome or eliminate ageing, postpone or abolish death, and radically enhance human physical and cognitive capacities. They see homo sapiens as a staging-post on the way to a new species, perhaps involving a fusion between biological and silicon intelligence.
I want to reflect on one aspect of this movement: the intergenerational issues to which it gives rise, and the resulting ethical pitfalls. I draw on two writers from utterly different perspectives: C.S. Lewis, and Martine Rothblatt.
C.S. Lewis wrote about transhumanism in the 1940s, long before the term itself had been coined. He saw it coming and was vehemently opposed. This is a central theme in “That Hideous Strength” (1945), the final volume of his science fiction trilogy. The clearest and most forceful statement of his position is in the third chapter of “The Abolition of Man”, a short non-fiction work published in 1943.
Lewis foresaw that technological developments might enable one generation deliberately to shape the minds, bodies and dispositions of its successors. He referred to eugenics, pre-natal conditioning, and applied psychology. He asked whether all this would constitute the triumph of humans over nature. He said no, for two reasons.
First, he thought that the humans who embarked on this project would lack any objective moral framework, and hence would ultimately be guided only by their own subjective desires: but since those desires would themselves be the product of whatever natural forces happened to be operating at that time on these engineers of human souls, the outcome of their work could just as well be described as the triumph of nature over humans. And secondly, he pointed to the radical inequality between successive generations that such a project would involve. The relationship between generations would no longer be that of ancestors to successors, but instead one of manufacturers to products. If one generation has the effective technological power to shape its successors, then the triumph of humans over nature turns out to mean the triumph of one (present) group of humans over all other (future) groups of humans.
Martine Rothblatt is the polar opposite of Lewis. If what follows seems outlandish, please bear in mind that Rothblatt is a substantial and influential figure: a lawyer, a pharmaceutical entrepreneur, and a modern thought leader who boasts both a TED talk and a profile in Forbes magazine as one of The World's 100 Greatest Living Business Minds.
Rothblatt’s approach to transhumanism is summarised in an essay entitled Mind is Deeper than Matter: Transgenderism, Transhumanism and the Freedom of Form (included in The Transhumanist Reader, a 2013 collection edited by Max More and Natasha Vita-More). Rothblatt (who has a trans identity) begins with an explicit parallel between transgenderism and transhumanism. Transgenderism affirms that the mind (and particularly its gender identity) takes priority over the body’s material form and should therefore be free to reshape that form as it sees fit. Transhumanism goes one stage further: it envisages that the mind can not only modify its biological form, but escape it altogether, abandoning a flesh-based container for a digital container.
For Rothblatt, a human person is primarily mind rather than matter - and mind is essentially information, that can be stored in more than one form. Rothblatt therefore envisages “mindfiles” - replicating the contents of a human mind - that will be uploaded and stored on “mindware”, which may itself be located within “bodywork” (anatomically realistic robots). The resulting digital-based persons - or transhumans - will need in due course to be recognised as bearers of legal and political rights, including the right to vote, to own property, and to marry: one could summarise this position as “transhumans are humans”. Rothblatt imagines a young woman who wants to marry a digital person, and who encounters parental opposition, saying to her father: “Dad, the trouble is that you see yourself as a flesh person and I see myself as a person”.
Rothblatt recognises that there is room for debate about whether computer-based intelligences can attain consciousness, and suggests that the solution is for psychologists to interview them and decide if they pass (this is a version of the well-known Turing Test). At this point, Rothblatt draws an explicit parallel with the medical gatekeeping that precedes gender reassignment surgery.
The suggested relationship between transgenderism and transhumanism, and the proposed solution to the problem of consciousness, themselves deserve extensive discussion. For present purposes, though, what interests me about Rothblatt’s proposal is that it is, explicitly, an immortality project. Rothblatt envisages that before death a person might upload the contents of his or her mind into one or more digital copies, and that these could be recognised after physical death as constituting the same person continuing in a different form, just as an adult is continuous with their former child self:
Death laws can be amended to provide that a person whose higher brain functions continue to be performed by information technology, such that there is a continuity of identity and consciousness to the satisfaction of psychiatrists, is not legally dead even if their heart has stopped beating.
The clear implication is that the digital version of the person would retain whatever property and status was held by the previous, flesh-based version of that same person.
At this point, I return to Lewis’s central insight that the apparent triumph of humans over nature can readily turn into the tyranny of one generation over another. Lewis feared that one generation might become the manufacturer of its successors. Instead, Rothblatt points us towards a world in which one generation might exercise a more direct technologically mediated power, of indefinite duration.
The sort of quasi-immortality that Rothblatt is envisaging, if it ever came to pass, would be of particular appeal to two (overlapping) groups: the wealthy, and the politically powerful. Great empires - including commercial or financial empires - often do not long survive the death of their founders. This partly modifies extreme inequality; death shakes the kaleidoscope and offers the chance for a reset. But if a billionaire, or a dictator, can use digital means to evade death, what then? Genghis Khan, founder of the Mongol Empire, died in 1227. By 1294 (at the death of his grandson Kublai Khan) his Empire had fragmented into four competing khanates. What followed was a long period of disintegration and decline. If Rothblatt’s technology had been available to Genghis Khan, would we now be living under a unified Mongol Empire, as subjects of Genghis Khan’s continuing digital rule? Or what if the same technology had been available on the death of Stalin in 1953, or Kim Il Sung in 1994? Imagine a digital avatar with the power to launch nuclear weapons.
An obvious retort is that the technology Rothblatt discusses is altogether fanciful: there is no way that a human mind can ever be digitally copied and saved on a computer system. At most you would get a kind of digital replica of a real human being, a zombie without self-awareness.
Even if all of that is right, it won’t necessarily save us. A fake digital immortality could have alarming real-life consequences, if the digital avatar was able to wield effective social or economic power; and its ability to do so would depend (in part) on socially-accepted beliefs about the avatar’s significance.
“Silicon persons are persons”, “digital identities are valid”: it’s not hard to imagine such slogans becoming popular. They could lead us to some dark places.
"The relationship between generations would no longer be that of ancestors to successors, but instead one of manufacturers to products. If one generation has the effective technological power to shape its successors, then the triumph of humans over nature turns out to mean the triumph of one (present) group of humans over all other (future) groups of humans."
I love this piece. It gets right to the root of the gender industry (business): "Gender rights" are really just the gestational phase of AI rights. When people finally understand this, perhaps they will stop arguing about sex role stereotypes (as if governments, the most powerful banks, corporations and law firms in the world would care about the bodily distress of a minuscule part of the population), and they will engage the real threat these techno fascists present.
People often tell me, "but they'll never succeed." I keep telling them, it doesn't matter. What matters is the havoc they wreak in the meantime. The many billions of dollars going into these modern eugenics projects could feed the global poor. The animal torture and use of vital planet resources, going toward the obliteration of our species as we know it, is nothing short of insanity.
Bravo, Timothy!
Stay human.
This is an extremely old fantasy, back to at least Plato’s cave.
Vast portions of consciousness are subverbal, interconnected with body sensory and regulatory systems. Consciousness presents itself as operating like a camera and microphones. It is isn’t, it is an active hallucination which uses sensory inputs to error correct constantly, by measuring the gap between the figment and the sensory inputs and adjusting the figment until the gap is small to zero for areas of attention.
It’s not recording, there is no mechanism in the brain to “record”, because of that there isn’t really a thing which can be copied.
That’s the fatal flaw. It’s a very large nonlinear system worn “hidden variables” with no known way of accessing them, and I suspect no known mathematical way to reverse engineer them.
From that comes the essential failure in the trans human project. The premise of “the matrix” doesn’t work beyond movies.
We can’t move faster than the speed of light, you can’t go backwards in time, you can’t know the position and momentum of a particle, heat never flows from cold to hot, and you can’t create an accurate copy of a nonlinear system from its output.