“In the 21st century,” Yuval Noah Harari writes, “the next big project of humankind will be to acquire for us the divine powers of creation and destruction, and upgrade Homo sapiens into Homo deus.” According to the bestselling Israeli historian, humankind will turn itself into God by using three new types of engineering. Bio-engineers will “take the old sapiens body, and intentionally rewrite its genetic code, rewire its brain circuits, alter its biochemical imbalance, and even grow entirely artificial limbs”. Cyborg engineers will “go a step further, merging the organic body with non-organic devices such as bionic hands, artificial eyes, or millions of nano-robots that will navigate our bloodstream, diagnose problems and repair damages”. The third kind of engineering will dispense with organic parts altogether. “Neural networks will be replaced by intelligent software, which could surf both virtual and non-virtual worlds, free from the limitations of organic chemistry.” Once the human mind has been separated from its biological base, “human history will come to an end and a completely new process will begin”.
Homo Deus: a Brief History of Tomorrow is a companion volume to Harari’s Sapiens: a Brief History of Humankind (first published in Hebrew in 2011), a provocative and thought-stirring volume in which he mounted a powerful challenge to the prevailing faith that humankind can shape its future. That book concluded with the seemingly paradoxical suggestion that the extinction of the species could come about as a result of the expansion of human power. Through ongoing advances in artificial intelligence a vast, post-human mind could come into being that would attach no importance to human survival or the values that make our lives worth living. In Homo Deus Harari spells out what this shift to a post-human future means. “[M]ore and more individuals, organisations, corporations and governments are taking very seriously the quest for immortality, happiness and godlike powers . . . This is not what most people will do in the 21st century. It is what humanity as a collective will do.” Homo sapiens is transforming itself into a divine being.
It’s an arresting vision. If you think about it, however, the prospect that the human species could “upgrade itself” to godhood melts away. “Humanity” can’t become God, because “humanity” does not exist. All that actually exists is the multifarious human animal, with its historic enmities and intractable divisions. The idea that the human species is a collective agent, setting itself “big projects” and realising them throughout history, is a humanist myth. Surfacing here and there in Sapiens, it didn’t do very much damage. In Homo Deus, which shows signs of having been written hastily, it muddles and undermines much of the argument.
A telling example occurs when, discussing how accelerating technical change frightens many people, the author considers the possibility that “somebody will hit the brakes and slow us down”. This won’t happen, he argues, because “nobody knows where the brakes are”. But this misses the point. “Humanity as a collective” can’t put the brakes on change, because there is no such agent. Even if someone could find the brakes, the process of change wouldn’t slow down
for long. Pursuing internecine human rivalries – geopolitical and ideological – somebody else would soon speed it up again.
Harari pooh-poohs these conflicts. Liberalism has triumphed, he assures us:
After decades of defeats and setbacks, liberalism won a decisive victory in the Cold War . . . Liberalism still sanctifies individual liberties above all, and still has a firm belief in the voter and the customer. In the early 21st century, this is the only game in town. As of 2016, there is no serious alternative to the liberal package of individualism, human rights, democracy and a free market.
Questionable in the autumn of 1989, when Francis Fukuyama first announced the end of history in the National Interest, this assertion of the victory of liberalism sounds dated and weirdly surreal today.
Liberal societies everywhere confront problems for which they have no solutions. The socially disruptive effects of globalisation are fuelling powerful anti-liberal forces in continental Europe and in the United States. The transatlantic liberal alliance that emerged from the Second World War is facing increasing military competition from Russia and China. In Homo Deus as in Sapiens, Harari – citing Steven Pinker’s intensely fashionable speculations – tells us that warfare is becoming obsolete. But armed conflict isn’t fading away, only changing shape. Hi-tech wars will be fought with cyber attacks on air-traffic-control systems, nuclear power stations and the like, while physical attacks will be carried out by drones. At the same time terrorists will continue to attack using low-tech tools such as trucks and kitchen knives.
Harari tells us not to worry too much about these threats: because liberal societies are the most productive and innovative, they will prevail in any competition with their enemies. It is true that Nazi and communist totalitarianism were defeated partly as a result of the technical superiority of America and Britain. Yet how can we be sure that the authoritarian order that Xi Jinping is consolidating now in China will continue to lag behind in technical
development? Or that some new regime will not appear elsewhere, perhaps less overtly repressive, but also fundamentally illiberal, in which science and technology will advance faster than in liberal societies? There is no law of history which says this cannot happen.
The threat to liberalism, Harari believes, comes not from its enemies but from the advance of science. Rightly, he argues that secular liberalism is based on ideas inherited from religion, such as the soul and free will. Neuroscience undermines the belief that human beings have “a single and indivisible self”, and “once we accept that there is no soul, and that humans have no inner essence” the idea of free will no longer makes sense:
. . . when a neuron fires an electric charge, this may be either a deterministic reaction to external stimuli, or it might
be the outcome of a random event such as the spontaneous decomposition of a radioactive atom. Neither option leaves any room for free will . . . The sacred word “freedom” turns out to be, just like “soul”, an empty term that carries no discernible meaning. Free will exists only in the imaginary stories we humans have invented.
In the end, human beings aren’t so different from vending machines: “A much more complicated algorithm than the vending machine, no doubt, but still an algorithm . . . The algorithms controlling vending machines work through mechanical gears and electric circuits. The algorithms controlling humans work through sensations, emotions and thoughts.” Like all other living things, human beings are embodied algorithms, programmed by natural selection to produce “copies of themselves”.
Many readers may find this a disheartening vision. But consolation is at hand: according to Harari, a new religion is emerging to reflect the shift that is occurring in our sense of ourselves. Dataism is the belief that “the universe consists of data flows, and the value of any phenomenon or entity is determined by its contribution to data processing”. This may sound a slender basis for a religion, but Harari insists that Dataism is not a fringe cult. It combines two of the most powerful intellectual developments of modern times: the neo-Darwinian view that organisms are biochemical algorithms and Alan Turing’s theory of artificial intelligence. Melding these factors produces “the scientific holy grail that has eluded us for centuries: a single overarching theory that unifies all the scientific disciplines from literature and musicology to economics and biology”.
The appeal of the new religion isn’t exclusively intellectual. It offers its followers ever longer life and eventually immortality, as the human mind detaches itself from its fleshly shell and is uploaded into cyberspace. Like Christianity, Dataism could attain “enormous popularity and power”.
In fact, it is hard to envision Dataism having an influence anywhere near comparable with that of traditional religions. Its chief supporters are the visionary entrepreneurs and billionaire investors of Silicon Valley – a powerful group, but also a narrow one. It’s not Dataism that is spreading at the present time: it is the old-time faiths – Christianity in China and Islam in Africa, for instance. Putting the human story at the centre of things, these religions give meaning to the lives of believers. Because it leaves human beings stranded on the margins, or consigns them to extinction, that is something Dataism cannot do.
In an interesting chapter, Harari contrasts Dataism with the more popular cult of techno-humanism. Whereas Dataists believe that humankind is obsolete, the techno-humanists think that technology can be used to fashion a superior human model. Using genetic engineering, nanotechnology and brain-computer interfaces, human beings can fashion a higher version of themselves. But who decides what a higher version of the human animal would be?
Harari writes that whereas “Hitler and his ilk” wanted to create a superhuman species “by means of selective breeding and ethnic cleansing . . . 21st-century techno-humanism hopes to reach the goal more peacefully”. However, techno-humanists are not pursuing the same objective as the Nazis by different means. They have different ideas about what constitutes a superior species, and this points to a crucial fact – there are likely to be as many types of super-human and post-human being as there are human groups with the technology to create them.
Harari is right in thinking of human development as a process that no one could have planned or intended. He fails to see that the same is true of the post-human future. If such new species appear, they will be created by governments and powerful corporations, and used by any group that can get its hands on them – criminal cartels, terrorist networks, religious cults, and so on. Over time, these new species will be modified and redesigned, first by their human controllers, then by the new species themselves. It won’t be too long before some of them slip free from their human creators. One type may come out on top, at least for a while, but there is nothing to suggest this process will end in a godlike being that is supreme over all the rest. Like the evolution of human beings, post-human evolution will be a process of drift, with no direction or endpoint.
Will our post-human successors be sentient? Harari thinks not. “Intelligence is decoupling from consciousness.” Algorithms – step-by-step mathematical operations – already control the book recommendations on Amazon, much stock-market investment, the working lives of Uber drivers and the selection of potential partners on dating websites. It probably won’t be long before medical diagnosis and legal counsel are routinely provided by robots. As Harari notes, none of this requires conscious awareness. But it is hard to see why he takes it for granted that consciousness cannot develop in machines controlled by algorithms. In his view, after all, we too are such machines. And yet, somehow – presumably through a Darwinian process of natural selection – self-awareness has appeared in human beings. Why should it not also develop in machines that human beings have made?
Perhaps what Harari means to say is that science has shown that consciousness – like free will – is an illusion. Maybe so, but a robot that labours under the illusion of being conscious is very different from one that is flatly insentient. However superhumanly clever, self-aware robots might be very much like human beings in other respects: inwardly divided, plagued by conflicts and incapable of controlling the future.
If Harari fails to take these possibilities into account, one reason is that he shrinks from giving up the idea that we can shape our lives. Questioning free will, he writes, is “not just a philosophical exercise. It has practical implications. If organisms indeed lack free will, it implies we could manipulate and even control their desires using drugs, genetic engineering or direct brain stimulation.” This is “the only way to ensure lasting contentment . . . Forget economic growth, social reforms and political revolutions: in order to raise global happiness levels, we need to manipulate human biochemistry.” Yet who are “we”, exactly? An elite of benevolent scientists, perhaps? If choice is an illusion, however, those who do the manipulating will be no freer than those who are being manipulated.
The future will be shaped by the same purposeless processes as the ones that have shaped humankind to date. Any realistically imaginable post-human future will be a continuation of human history by other means. Perhaps Harari backs off from reaching this conclusion because of a marketing imperative: if there is to be any emotional pay-off from reading Homo Deus, the book needs to hold out some glimmer of hope, however faint or delusive. Nearly everything written on the human prospect today belongs in the genre of the self-improvement manual, and this is no exception.
A deeper reason may be the formative impact of monotheism on Western thinking. The chief legacy of monotheistic religion isn’t a fading belief in a creator-God: it is the belief that humankind is some kind of universal subject, striving throughout history to realise common ends. No such conception can be found in polytheistic faiths, which see people as essentially disparate in their goals and values. The idea that the human species could act as a conscious agent is a
relic of monotheism, one that – like so many people today – Harari is unable to shake off.
Homo Deus oscillates between envisaging humankind playing God and human beings becoming like gods. These are very different futures, and only the latter has any plausibility. If it ever comes about, a post-human world won’t be one in which the human species has deified itself. More like the cosmos as imagined by the Greeks, it will be ruled by a warring pantheon of gods. Will these new deities follow those of ancient times in seeking to relieve the tedium of immortality by stirring up turmoil and destruction in the human world? Or will they be absorbed in their own rivalries and sublimely indifferent to any remaining human beings? We cannot tell. But if you want a glimpse of what a post-human future would be like, read Homer.
John Gray is the New Statesman’s lead book reviewer. His latest book is “The Soul of the Marionette: a Short Inquiry into Human Freedom” (Penguin)
This article appears in the 15 Feb 2017 issue of the New Statesman, The New Times