Top Stories

21 Lessons for the 21st Century—A Review

A review of 21 Lessons for the 21st Century by Yuval Noah Harari. Spiegel & Grau (September 2018), 400 pages.

“[I]n a world in which everything is interconnected, the supreme moral imperative becomes the imperative to know.”1 With that sentence, Yuval Noah Harari, professor of world history at the Hebrew University of Jerusalem, sets the stage for his new book, 21 Lessons for the 21st Century. Previously, Harari has written about humanity’s development from scattered communities of hunter-gatherers to an international society capable of space travel and wireless communication in Sapiens, and the remarkable risks and potentials for advances in artificial intelligence and bioengineering over the next century in Homo Deus. Both of these books became New York Times bestsellers. In 21 Lessons, Harari turns his focus to the present day, addressing a litany of pressing concerns for the first half of the 21st century.

Harari’s new book is truly ambitious in scope. In 400 pages, he tackles the future of work and education, the ubiquitous and growing influence of data collection on people’s lives, the increasingly ominous cracks around the foundation of the European Union, the conflict between secularism and religion, and a trio of existential threats: climate change, nuclear war, and artificial intelligence. The sheer diversity of the topics discussed in 21 Lessons makes boiling the book down into a short summary that does justice to its scope impossible. Unlike Sapiens and Homo Deus, 21 Lessons is less a narrative and more a set of related but independent arguments. Nevertheless, there are two identifiable threads which tie Harari’s sprawling arguments together into a coherent view of the modern world.

First, as he says in what is arguably the book’s most important chapter, the world has become a de facto single globalized civilization but remains factionalized in a way that gravely endangers our ability to tackle its problems. Harari argues that the ‘clash of civilizations’ rhetoric—between the West and Islam, or liberal democracy and authoritarianism—is exaggerated. Every country in the world is united (to greater or lesser degrees) in the same network of economic and political systems. All countries construct their technology according to a single rubric, educated people in every country believe essentially the same scientific account about the origin of life and the universe, and almost all at least pay lip service to liberal democratic principles like free elections and universal suffrage, however much they may betray those principles in practice.

Second, traditional narratives which have helped unify society are quickly becoming inadequate to capture the complexity of our current situations. In his chapter on liberty, Harari warns that a combination of artificial intelligence, data mining, and biotech could effectively destroy the liberal narrative of the primacy of the individual over the next few decades. If systems are developed that can predict and recommend things like romantic relationships and career choices better than we can, people might be willing to cede increasingly large parts of their decision making to machines. But then what becomes of the traditional Enlightenment emphasis on the importance of the individual and free will? Harari believes that modern neuroscience has already discredited the notion of free will among intellectuals, but he fears that the widespread adoption of personalized data algorithms might lead to a popular erosion of the idea, further undermining the social fabric.

Harari’s contention that the world now constitutes a single civilization is central to his criticisms of more insular movements like religious traditionalism and nationalism. While this globalized civilization allows for effective cooperation in some areas, it has proven insufficient to address global-scale problems, and is likely to continue to do so. It might, Harari argues, even exacerbate them, because most parts of the world are still clinging to a factional worldview that fails to appreciate the scale of the problems confronting our species:

We now have a global ecology, a global economy, and a global science—but we are still stuck with only a national politics…We are trapped, then, between a rock and a hard place. Humankind now constitutes a single civilization, and problems such as nuclear war, ecological collapse, and technological disruption can only be solved on a global level. On the other hand, nationalism and religion still divide our human civilization into different and often hostile camps.2

As is often the case with jeremiads, 21 Lessons does a markedly better job at pointing out the weaknesses of nationalistic tendencies than in proposing a viable alternative. Global cooperation will certainly be necessary for addressing the ecological crisis and the potential for upgrading humans via bioengineering (a point discussed at length in Homo Deus), especially because at least some nations will likely have to sacrifice comparative advantages in the short-term. But Harari’s proposal of the European Union as the best example we have of successfully prioritizing international loyalties over those of the nation-state is unconvincing, to put it mildly, and frankly remarkable in light of his admission that “[a]t present, it is far from clear whether Europe can find a middle path that will enable it to keep its gates open to strangers without being destabilized by people who don’t share its values.”3

Yuval Noah Harari

Harari claims that the EU is a model for global cooperation because its constitution allows for the negotiation of things like Catalonian or Scottish independence within a peaceful political framework, unlike the pre-20th century scenario in which redrawing political boundaries often involved violence. But it hardly follows from this that the EU model would work on a worldwide scale. Its constitution binds together a group of people in a relatively small part of the world, who have operated within a common religious, geographic, and cultural framework for centuries. Indeed, the very concept of “Europe” as a continent separate from Asia, a division which has little to no geographical justification, further illustrates the fact that the tight cultural and political bonds shared by Europeans are not easily generalizable to the world as a whole. Even today, with Europe more diverse than perhaps ever before, Europeans arguably have more in common with each other than they do with even Americans (higher rates of secularism and a more diverse political scene being only two such examples), let alone the non-Western world. Harari is persuasive when he argues that “humankind will probably need some kind of global identity and loyalty,”4 but he fails to offer a compelling account of how such a thing might be invented and sustained.

But global divisions are not solely the result of conscious decisions on the part of nations to shut themselves out from the global process. In addition to legal and political attempts at isolation from the world system, the information needed to understand massive-scale problems is spread out across billions of people and likely to become increasingly inaccessible. It is in detailing this problem that Harari sounds most prophetic. Defenders of capitalism have long argued that disseminated market knowledge communicated through price-signalling is superior to centralized planning, in light of the impossibility of efficiently sorting through the complexity of a national economy. A grocery store manager doesn’t need to know the source of the higher wholesale potato prices to know that his store will make a smaller profit if it continues selling potatoes at current prices. But what if a system developed in which even individual signals were incomprehensible to human observers?

Adam Smith famously used the example of a pin factory to illustrate division of labor—a group of workers each specializing in one task can produce more pins than the same group performing all the steps individually. A similar analogy can be used to describe the division of knowledge throughout the industrialized world—no single person understands every part of the total system. Many parts are no longer performed by humans at all, but most parts are understood at least on some level by someone, even those that have been outsourced to machines. But what do we do when, as Harari warns is likely, substantial parts of technology and the economy are not merely outsourced to machines, but cease to be understandable to humans at all? A fully automated pin-producing factory may cost pin manufacturers their jobs, but each component of its commodity processing is fully comprehensible. But the inner workings of an algorithm developed in ten years for drone coordination or cancer diagnosis might not be understood by anyone, including those using it.

The removal of key technologies from beyond human attempts at understanding might well threaten our self-understanding and sense of control severely enough to further undermine confidence in the value of individual choice. Harari’s warnings on this point seem prescient in light of people’s willingness up to this point to hand over large amounts of personal data to large tech companies, the inner machinations of which they already do not understand. Some of Harari’s descriptions of what such a machine-driven world could look like might seem outlandish, with their visions of people letting machines choose their mates and dictators controlling the population through biometric sensors. But as he notes in chapter 19, “If somebody describes the world of the mid-twenty first century to you and it sounds like science fiction, it is probably false. But then again, if somebody describes the world of the mid-twenty first century to you and it doesn’t sound like science fiction, it is certainly false.”5

If I were pressed to name the most important flaw in an otherwise largely thoughtful book, it would be the lack of political and economic rigor devoted to criticizing nationalism and advocating globalism in its place. Given how central the notion of a global civilization is to the book’s arguments, this is more than a little disappointing. Harari’s suggestions for worldwide cooperation seldom go beyond a general call for the invention of new globalizing narratives that are—or so he says—the only way of creating the mass action needed to solve the problems outlined above. I am inclined to agree that a sense of global identity is the only way of coordinating these solutions. Nevertheless, whether or not internationalism is more effective than nationalism at solving global problems is a question that cannot be answered by a priori theorizing alone. Harari would have done well to borrow from Steven Pinker’s emphasis on quantitative data in Enlightenment Now and The Better Angels of Our Nature and include a broader selection of detailed historical statistics and case studies. Nevertheless, 21 Lessons remains an intriguing and often disturbing look at a bewildering world scene. Anyone seeking to understand a globalist perspective on the 21st century cannot afford to ignore it.


Spencer Hall is a statistician living in Athens, GA, who writes on science, philosophy, and religion. He graduated from the University of Georgia in 2016 and has worked for his alma mater in disease ecology ever since. His blog can be found here.


1 21 Lessons for the 21st Century, p. 231
2 Ibid, p. 126 and 139
3 Ibid, p. 156
4 Ibid, p. 121
5 Ibid, p. 268, emphasis original