Sapiens and Homo Deus

by Yuval Noah Harari

  Dataism is most firmly entrenched in its two mother disciplines: computer science and biology. Of the two biology is the more important. It was biology’s embrace of Dataism that turned a limited breakthrough in computer science into a world-shattering cataclysm that may completely transform the very nature of life. You may not agree with the idea that organisms are algorithms, and that giraffes, tomatoes and human beings are just different methods for processing data. But you should know that this is current scientific dogma, and it is changing our world beyond recognition.

  Not only individual organisms are seen today as data-processing systems, but also entire societies such as beehives, bacteria colonies, forests and human cities. Economists increasingly interpret the economy too as a data-processing system. Laypeople believe that the economy consists of peasants growing wheat, workers manufacturing clothes, and customers buying bread and underpants. Yet experts see the economy as a mechanism for gathering data about desires and abilities, and turning this data into decisions.

  According to this view, free-market capitalism and state-controlled communism aren’t competing ideologies, ethical creeds or political institutions. They are, in essence, competing data-processing systems. Capitalism uses distributed processing, whereas communism relies on centralised processing. Capitalism processes data by directly connecting all producers and consumers to one another and allowing them to exchange information freely and make decisions independently. How do you determine the price of bread in a free market? Well, every bakery may produce as much bread as it likes, and charge for it as much as it wants. The customers are equally free to buy as much bread as they can afford, or take their business to a competitor. It isn’t illegal to charge $1,000 for a baguette, but nobody is likely to buy it.

  On a much grander scale, if investors predict increased demand for bread, they will buy shares of biotech firms that genetically engineer more prolific wheat strains. The influx of capital will enable the firms to speed up their research, thereby providing more wheat faster, and averting bread shortages. Even if one biotech giant adopts a flawed theory and reaches an impasse, its more successful competitors will likely achieve the hoped-for breakthrough. Free-market capitalism thus distributes the work of analysing data and making decisions between many independent but interconnected processors. As the Austrian economics guru Friedrich Hayek explained, ‘In a system in which the knowledge of the relevant facts is dispersed among many people, prices can act to coordinate the separate actions of different people.’2

  According to this view the stock exchange is the fastest and most efficient data-processing system humankind has so far created. Everyone is welcome to join, if not directly then through their banks or pension funds. The stock exchange runs the global economy, and takes into account everything that happens all over the planet – and even beyond it. Prices are influenced by successful scientific experiments, by political scandals in Japan, by volcanic eruptions in Iceland and even by irregular activities on the surface of the sun. In order for the system to run smoothly, as much information as possible needs to flow as freely as possible. When millions of people throughout the world have access to all the relevant information, they determine the most accurate price of oil, of Hyundai shares and of Swedish government bonds by buying and selling them. It has been estimated that the stock exchange needs just fifteen minutes of trade to determine the influence of a New York Times headline on the prices of most shares.3

  Data-processing considerations also explain why capitalists favour lower taxes. Heavy taxation means that a large part of all available capital accumulates in one place – the state coffers – and consequently more and more decisions have to be made by a single processor, namely the government. This creates an overly centralised data-processing system. In extreme cases, when taxes are exceedingly high, almost all capital ends up in the government’s hands, and so the government alone calls the shots. It dictates the price of bread, the location of bakeries, and the research-and-development budget. In a free market, if one processor makes a wrong decision, others will be quick to capitalise on its mistake. However, when a single processor makes almost all the decisions, mistakes can be catastrophic.

  This extreme situation, in which all data is processed and all decisions are made by a single central processor, is called communism. In a communist economy people allegedly work according to their abilities and receive according to their needs. In other words, the government takes 100 per cent of your profits, decides what you need and then supplies these needs. Though no country ever realised this scheme in its extreme form, the Soviet Union and its satellites came as close as they could. They abandoned the principle of distributed data processing and switched to a model of centralised data processing. All information from throughout the Soviet Union flowed to a single location in Moscow where all the important decisions were made. Producers and consumers could not communicate directly and had to obey government orders.

  49. The Soviet leadership in Moscow, 1963: centralised data processing.

  49.© ITAR-TASS Photo Agency/Alamy Stock Photo.

  For instance, the Soviet economics ministry might decide that the price of bread in all shops should be exactly two roubles and four kopeks, that a particular kolkhoz in the Odessa oblast should switch from growing wheat to raising chickens, and that the Red October bakery in Moscow should produce 3.5 million loaves of bread per day and not a single loaf more. Meanwhile the Soviet science ministry forced all Soviet biotech laboratories to adopt the theories of Trofim Lysenko – the infamous head of the Lenin Academy for Agricultural Sciences. Lysenko rejected the dominant genetic theories of his day. He insisted that if an organism acquired some new trait during its lifetime, this quality could pass directly to its descendants. This idea flew in the face of Darwinian orthodoxy, but it dovetailed nicely with communist educational principles. It implied that if you could train wheat plants to withstand cold weather, their progenies will also be cold-resistant. Lysenko accordingly sent billions of counter-revolutionary wheat plants to be re-educated in Siberia – and the Soviet Union was soon forced to import more and more flour from the United States.4

  50. Commotion on the floor of the Chicago Board of Trade: distributed data processing.

  50.© Jonathan Kirn/Getty Images.

  Capitalism did not defeat communism because capitalism was more ethical, because individual liberties are sacred or because God was angry with the heathen communists. Rather, capitalism won the Cold War because distributed data processing works better than centralised data processing, at least in periods of accelerating technological change. The central committee of the Communist Party just could not deal with the rapidly changing world of the late twentieth century. When all data is accumulated in one secret bunker, and all important decisions are taken by a group of elderly apparatchiks, they can produce nuclear bombs by the cartload, but not an Apple or a Wikipedia.

  There is a story (probably apocryphal, like most good stories) that when Mikhail Gorbachev tried to resuscitate the moribund Soviet economy, he sent one of his chief aides to London to find out what Thatcherism was all about, and how a capitalist system actually functioned. The hosts took their Soviet visitor on a tour of the City, of the London stock exchange and of the London School of Economics, where he had lengthy talks with bank managers, entrepreneurs and professors. After many long hours the Soviet expert burst out: ‘Just one moment, please. Forget about all these complicated economic theories. We have been going back and forth across London for a whole day now, and there’s one thing I cannot understand. Back in Moscow our finest minds are working on the bread supply system, and yet there are such long queues in every bakery and grocery store. Here in London live millions of people, and we have passed today in front of many shops and supermarkets, yet I haven’t seen a single bread queue. Please take me to meet the person in charge of supplying bread to London. I must learn his secret.’ The hosts scratched their heads, thought for a moment, and said: ‘Nobody is in charge of sup
plying bread to London.’

  That’s the capitalist secret of success. No central processing unit monopolises all the data on the London bread supply. The information flows freely among millions of consumers and producers, bakers and tycoons, farmers and scientists. Market forces determine the price of bread, the number of loaves baked each day and the research-and-development priorities. If market forces make a bad decision, they soon correct themselves, or so capitalists believe. For our current purposes it doesn’t matter whether this capitalist theory is correct. The crucial thing is that the theory understands economics in terms of data processing.

  Where Has All the Power Gone?

  Political scientists also increasingly interpret human political structures as data-processing systems. Like capitalism and communism, so democracies and dictatorships are in essence competing mechanisms for gathering and analysing information. Dictatorships use centralised processing methods, whereas democracies prefer distributed processing. Over the past decades democracy gained the upper hand because under the unique conditions of the late twentieth century, distributed processing worked better. Under alternative conditions – those prevailing in the ancient Roman Empire, for instance – centralised processing had an edge, which is why the Roman Republic fell and power shifted from the Senate and popular assemblies into the hands of a single autocratic emperor.

  This implies that as data-processing conditions change again in the twenty-first century, democracy might decline and even disappear. As both the volume and speed of data increase, venerable institutions like elections, political parties and parliaments might become obsolete – not because they are unethical, but because they can’t process data efficiently enough. These institutions evolved in an era when politics moved faster than technology. In the nineteenth and twentieth centuries the Industrial Revolution unfolded slowly enough for politicians and voters to remain one step ahead of it and regulate and manipulate its course. Yet whereas the rhythm of politics has not changed much since the days of steam, technology has switched from first gear to fourth. Technological revolutions now outpace political processes, causing MPs and voters alike to lose control.

  The rise of the Internet gives us a taste of things to come. Cyberspace is now crucial to our daily lives, our economy and our security. Yet the critical choices between alternative web designs weren’t taken through a democratic political process, even though they involved traditional political issues such as sovereignty, borders, privacy and security. Did you ever vote about the shape of cyberspace? Decisions made by web designers far from the public limelight mean that today the Internet is a free and lawless zone that erodes state sovereignty, ignores borders, abolishes privacy and poses perhaps the most formidable global security risk. Whereas a decade ago it hardly registered on the radar, today hysterical officials are predicting an imminent cyber 9/11.

  Governments and NGOs consequently conduct intense debates about restructuring the Internet, but it is much harder to change an existing system than to intervene at its inception. Besides, by the time the cumbersome government bureaucracy makes up its mind about cyber regulation, the Internet will have morphed ten times. The governmental tortoise cannot keep up with the technological hare. It is overwhelmed by data. The NSA may be spying on our every word, but to judge by the repeated failures of American foreign policy, nobody in Washington knows what to do with all the data. Never in history did a government know so much about what’s going on in the world – yet few empires have botched things up as clumsily as the contemporary United States. It’s like a poker player who knows what cards his opponents hold, yet somehow still manages to lose round after round.

  In the coming decades it is likely that we will see more Internet-like revolutions, in which technology steals a march on politics. Artificial intelligence and biotechnology might soon overhaul our societies and economies – and our bodies and minds too – but they are hardly a blip on the current political radar. Present-day democratic structures just cannot collect and process the relevant data fast enough, and most voters don’t understand biology and cybernetics well enough to form any pertinent opinions. Hence traditional democratic politics is losing control of events, and is failing to present us with meaningful visions of the future.

  Ordinary voters are beginning to sense that the democratic mechanism no longer empowers them. The world is changing all around, and they don’t understand how or why. Power is shifting away from them, but they are unsure where it has gone. In Britain voters imagine that power might have shifted to the EU, so they vote for Brexit. In the USA voters imagine that ‘the establishment’ monopolizes all the power, so they support anti-establishment candidates such as Bernie Sanders and Donald Trump. The sad truth is that nobody knows where all the power has gone. Power will definitely not shift back to ordinary voters if Britain leaves the EU nor if Trump takes over the White House.

  That doesn’t mean we will go back to twentieth-century-style dictatorships. Authoritarian regimes seem to be equally overwhelmed by the pace of technological development and the speed and volume of the data flow. In the twentieth century dictators had grand visions for the future. Communists and fascists alike sought to completely destroy the old world and build a new world in its place. Whatever you think about Lenin, Hitler or Mao, you cannot accuse them of lacking vision. Today it seems that leaders have an opportunity to pursue even grander visions. While communists and Nazis tried to create a new society and a new human with the help of steam engines and typewriters, today’s prophets could rely on biotechnology and super-computers.

  In science-fiction films ruthless Hitler-like politicians are quick to pounce on such new technologies, putting them in the service of this or that megalomaniacal political ideal. Yet flesh-and-blood politicians in the early twenty-first century, even in authoritarian countries such as Russia, Iran or North Korea, are nothing like their Hollywood counterparts. They don’t seem to be plotting any Brave New World. The wildest dreams of Kim Jong-un and Ali Khamenei don’t extend much beyond atom bombs and ballistic missiles: that is so 1945. Putin’s aspirations seem confined to rebuilding the old Soviet bloc, or the even older tsarist empire. Meanwhile in the USA paranoid Republicans have accused Barack Obama of being a ruthless despot hatching conspiracies to destroy the foundations of American society – yet in eight years of his presidency he barely managed to pass a minor health-care reform. Creating new worlds and new humans was far beyond his agenda.

  Precisely because technology is now moving so fast, and parliaments and dictators alike are overwhelmed by data they cannot process quickly enough, present-day politicians are thinking on a far smaller scale than their predecessors a century ago. Consequently, in the early twenty-first century politics is bereft of grand visions. Government has become mere administration. It manages the country, but it no longer leads it. Government ensures that teachers are paid on time and sewage systems don’t overflow, but it has no idea where the country will be in twenty years.

  To a certain extent, this is a very good thing. Given that some of the big political visions of the twentieth century led us to Auschwitz, Hiroshima and the Great Leap Forward, maybe we are better off in the hands of petty-minded bureaucrats. Mixing godlike technology with megalomaniacal politics is a recipe for disaster. Many neo-liberal economists and political scientists argue that it is best to leave all the important decisions in the hands of the free market. They thereby give politicians the perfect excuse for inaction and ignorance, which are reinterpreted as profound wisdom. Politicians find it convenient to believe that the reason they don’t understand the world is that they don’t need to understand it.

  Yet mixing godlike technology with myopic politics also has its downside. Lack of vision isn’t always a blessing, and not all visions are necessarily bad. In the twentieth century the dystopian Nazi vision did not fall apart spontaneously. It was defeated by the equally grand visions of socialism and liberalism. It is dangerous to trust our future to market forces, because these force
s do what’s good for the market rather than what’s good for humankind or for the world. The hand of the market is blind as well as invisible, and left to its own devices it may fail to do anything at all about the threat of global warming or the dangerous potential of artificial intelligence.

  Some people believe that there is somebody in charge after all. Not democratic politicians or autocratic despots, but rather a small coterie of billionaires who secretly run the world. But such conspiracy theories never work, because they underestimate the complexity of the system. A few billionaires smoking cigars and drinking Scotch in some back room cannot possibly understand everything happening on the globe, let alone control it. Ruthless billionaires and small interest groups flourish in today’s chaotic world not because they read the map better than anyone else, but because they have very narrow aims. In a chaotic system, tunnel vision has its advantages, and the billionaires’ power is strictly proportional to their goals. When the world’s richest tycoons want to make another billion dollars, they can easily game the system in order to do so. In contrast, if they felt inclined to reduce global inequality or stop global warming, even they wouldn’t be able to do it, because the system is far too complex.

  Yet power vacuums seldom last long. If in the twenty-first century traditional political structures can no longer process the data fast enough to produce meaningful visions, then new and more efficient structures will evolve to take their place. These new structures may be very different from any previous political institutions, whether democratic or authoritarian. The only question is who will build and control these structures. If humankind is no longer up to the task, perhaps it might give somebody else a try.