Yuval Noah Harari is the most read living historian. Sapiens alone has sold something on the order of 25 million copies in 65 languages. He gives commencement addresses, advises governments, sits on Davos panels, writes for the Guardian and the Financial Times, and has been the gateway author through which a substantial fraction of the global reading public encountered the long arc of human history for the first time.

He is also one of the most contested public intellectuals of the last decade. Professional historians have been sharp about Sapiens in particular, citing simplifications and occasional factual errors. Critics on the left have been sharp about his Davos-friendly framing of structural problems as questions of cognitive adjustment. Critics on the right have been sharp about his treatment of religion and tradition. Critics on the post-rationalist circuit have been sharp about what they read as a sometimes-cavalier inevitability around the transitions he describes.

All of these critics have a point. They also do not, between them, refute the work. Harari is at his best a synthesist of the first rank, capable of making readable arguments out of evidence drawn from biology, anthropology, history, technology, and political economy. He is not at his best in every page he has written. The job of a serious reader is to take the parts where he is genuinely useful and use them, and to be precise about the parts where he is not.

This piece is an attempt to do that.

Who he is

Yuval Noah Harari, born 1976 in Israel, doctorate from Oxford in medieval military history, faculty at Hebrew University of Jerusalem. His academic work was on medieval and early-modern military memoirs and was respectable but unremarkable. The books that made him a global figure are popular history. Sapiens: A Brief History of Humankind (Hebrew 2011, English 2014). Homo Deus: A Brief History of Tomorrow (2016). 21 Lessons for the 21st Century (2018). Nexus: A Brief History of Information Networks from the Stone Age to AI (2024).

He is openly gay, married to Itzik Yahav (also his manager), vegan, and a long-term Vipassana meditator in the S.N. Goenka tradition. He typically takes two-month silent retreats annually. He is, by reputation, courteous, careful, and considerably more politically engaged than the Davos-tour persona suggests. He has been publicly critical of the Israeli occupation, of the Netanyahu government, and of the conduct of the war in Gaza, in terms that are unusual for a high-profile Israeli figure with international audience.

The thesis that does the most work

The single most useful idea in Harari's catalogue is the "shared fictions" thesis. Human beings cooperate at scale, the argument runs, because we are uniquely capable of believing in things that do not exist in the physical world. Money. Nations. Corporations. Religions. Constitutions. Human rights. None of these are physical objects. They have no biological basis. They exist because enough humans collectively agree they do, and they organise our behaviour to a degree that makes the collective agreement, in practice, indistinguishable from physical reality.

This framing, sometimes labelled inter-subjective reality, is not original to Harari. It runs through Benedict Anderson's Imagined Communities, John Searle's institutional facts, and a long sociological tradition. What Harari did was make it accessible to a mass audience. The example most readers carry away from Sapiens, that a 50-euro note has no intrinsic value and only works because everyone in the chain believes it does, is the kind of small-scale demonstration of a large idea that good popular history is built on.

The framing is genuinely useful for political analysis. Once you see that nations, parties, money, and corporations are stories people collectively maintain, certain political phenomena become legible. The slow unravelling of public trust in established institutions is a story about a shared fiction losing adherents. The rise of new political movements is a story about a competing fiction recruiting adherents. The peculiar power of money over modern life is a story about an extraordinarily robust collective agreement that almost everyone is part of, with very little ability to opt out. The Irish 12.5 percent corporate tax rate is a story Ireland tells the world to attract capital, and the entire institutional architecture supporting it is what makes the story credible. None of this requires conspiracy. It requires only the recognition that humans coordinate through stories, and that the stories shape what gets built.

This is the Harari frame at its best. It does real political-analytical work.

The supporting argument

Harari's broader historical narrative, across the books, runs roughly as follows.

Early Homo sapiens developed, sometime around 70,000 years ago, a capacity for symbolic and fictional thinking that gave the species an unprecedented ability to cooperate in large groups. This is what Harari calls the Cognitive Revolution. It enabled the spread of Homo sapiens across the globe and the displacement of every other hominid species. From that point on, human history is the story of progressively larger and more powerful cooperative fictions: tribes, kingdoms, religions, empires, nations, economies, scientific communities.

The Agricultural Revolution, around 12,000 years ago, made these fictions necessary. Larger populations required larger coordinating stories. Harari's contested claim is that the Agricultural Revolution was a Faustian bargain. It produced more humans, who lived in worse material conditions than their hunter-gatherer ancestors, in service of fictions (kings, priests, debt-records) that benefited a small ruling class. The argument is provocative and is contested by archaeologists, but it forces a useful question: who benefits from a particular shared fiction, and what does it cost the people who believe it?

The Scientific Revolution, from the 17th century onwards, generated a new set of shared fictions (progress, capitalism, secular humanism) that displaced the older religious and feudal ones, with results that Harari treats as both liberating and catastrophic. The 20th century is, in his account, the high-water mark of the humanist fictions: liberal democracy, market economics, individual rights, all of which are stories we collectively maintain and which structure modern life.

The 21st century, in Homo Deus and Nexus, is the period in which artificial intelligence and biological engineering may generate new actors that are not bound by the human cognitive infrastructure that produced our shared fictions. This is the part of Harari's work that is most speculative and most contested. The argument is that AI-mediated information networks, biological augmentation, and the increasing capacity of states and corporations to predict and shape human behaviour at scale, may produce a future in which the humanist fictions no longer organise life and something else does. He is more careful in Nexus than in Homo Deus about presenting this as one possible trajectory rather than as an inevitable one, but the trajectory remains the central worry of his recent work.

Where he is right

Three places where I think the work earns its reputation.

The shared-fictions thesis is genuinely useful for political analysis. Once you see that the institutions of modern political life are collective stories rather than facts of nature, you stop being surprised when they fail and start asking what story is replacing them. This is more analytically powerful than the conventional treatment of institutional decline as if it were weather.

The recent work on information networks, particularly Nexus, is a serious contribution. Harari's argument that the relationship between information technology and human cooperation has been the load-bearing variable across human history, and that AI-mediated networks change the equation in ways earlier information revolutions did not, is correct in broad outline and well-argued. He has been earlier and clearer than most public intellectuals on the specific risks of AI-driven information warfare, algorithmic radicalisation, and the erosion of shared empirical reality.

He has reached an audience that mainstream political and historical writing largely could not. A book that begins on a beach in 70,000 BC and ends with thoughts about consciousness uploading is not a book most people would pick up if it were written by a normal academic. Harari's prose, narrative pacing, and willingness to be speculative are all assets in this sense. The question of whether deep historical and structural thinking remains accessible to ordinary readers in the 21st century is not trivial. Harari has done substantive work on the access side of that problem.

Where he is vulnerable

Worth being honest about.

The empirical history is sometimes shaky. Professional historians have catalogued specific errors and oversimplifications across his books, particularly in Sapiens. He works at a level of generalisation that any specialist will find unsatisfying. Some of the generalisations are defensible as the necessary cost of a popular synthesis. Some are simply wrong. A reader who treats his historical claims as authoritative without checking specialist responses will be misled in some places.

The treatment of agriculture as a mistake, while provocative, oversells what is actually a more complicated picture. The hunter-gatherer-vs-agriculture comparison depends heavily on which gatherer society and which agricultural society you compare. Recent archaeology, particularly the work emerging from sites like Göbekli Tepe and the broader reconsideration of pre-agricultural complexity, suggests the transition was less monolithic and the trade-offs more variable than Harari describes.

The futurology in Homo Deus is the weakest part of his catalogue. The argument that technology and biology are converging on a post-human horizon was already contested when published and has not aged well. Harari hedges enough to maintain the position, but the breathlessness of the original presentation reads, in 2026, as more confident than the evidence warranted. He has, to his credit, been more careful in Nexus.

The Davos-friendliness is a real critique. Harari has been a regular fixture at the World Economic Forum and gives a particular kind of address that many serious readers find suspect. The framing of structural problems (climate, AI, inequality) as cognitive or coordination challenges that elites can solve through better thinking has the effect of depoliticising problems that have specific beneficiaries and specific resisters. Harari is not naive about this and has been more politically explicit in his recent writing, but the criticism stands as a tendency in the earlier work.

There is, finally, an occasional contradictory quality across the books. The same Harari who argues in Sapiens that all stories are equally fictional sometimes writes in 21 Lessons and Nexus as if some stories (liberal democracy, scientific rationality, human rights) are particularly worth defending. The contradictions are arguably honest engagement with a difficult position rather than carelessness, but they do create slippage in how the work is read.

These are not fatal. They are the things to track if you want to use the work seriously rather than as decoration.

How it lands in Ireland

The shared-fictions frame translates directly to several Irish political problems.

The Republic itself is a story. The 26-county Irish state is held together by a collective agreement that this particular geographical-political-cultural arrangement is what "Ireland" means. The story has been remarkably successful for a hundred years. It is not metaphysically necessary. The Northern question, both the unification argument and the unionist counter-argument, is a fight between competing shared fictions about what "Ireland" or "the United Kingdom" actually consists of. Harari's frame makes this legible without taking sides on the underlying question. The argument is between stories. Whichever story enough people commit to is the one that organises the political reality.

The corporate tax model is a story Ireland tells the world. The 12.5 percent rate works because international capital believes that Ireland will defend it across changes of government, that the institutional commitments behind it are robust, and that Ireland's status as a stable European jurisdiction adds value to companies routing capital through it. None of this is built into the physical world. It is a collective expectation, enforced by a long series of policy and political signals. If the story stops being credible, the model collapses. This is exactly what international tax-coordination pressure is testing.

The housing debate is a fight over which story we tell ourselves about what housing is for. The Anglo-Saxon story is that housing is an investment-asset class that incidentally provides shelter. The continental story is that housing is shelter that incidentally has financial value. The Nordic story is that housing is a public good that should be substantially decommodified. Ireland has nominally retained the second story while institutionally implementing the first. The current crisis is partly the consequence of this contradiction becoming visible.

The Irish media-and-information ecosystem is degrading along the lines Harari describes in Nexus. Concentration of ownership, social-media platform substitution for traditional journalism, the financial collapse of regional and investigative reporting, and the algorithmic amplification of polarising content are all elements of the broader information-network shift Harari has been describing. Ireland is a small market and is therefore particularly exposed to these shifts because its native media has less margin to absorb them.

The bigger point is that Ireland, like most modern democracies, is currently in the middle of a slow renegotiation of which collective stories organise political life. The old stories (the Catholic-nationalist consensus, the FF-FG civil-war framing, the FDI-led economic narrative, the neutral-but-aligned defence posture) are losing adherents. The new stories that will replace them are still being written. Harari's frame is useful for tracking which stories are gaining and which are losing, and what that means for institutions that depend on the older stories continuing to be believed.

Where to start

If you have an evening: pick up Sapiens in a bookshop, read the introduction and the first hundred pages. The Cognitive Revolution chapter and the early Agricultural Revolution material is the strongest section and is what made the book famous. You can decide on the rest of the book based on how you feel about the first hundred pages.

If you have a week: Nexus (2024) is, in my view, his strongest book and the most useful for current political and technological argument. It is more careful and more serious than the earlier work and addresses the questions that matter now. Read it with attention.

If you have read Sapiens and want the politically more engaged Harari, 21 Lessons for the 21st Century (2018) is the more topical book. It is uneven, but the chapters on nationalism, work, education, and post-truth are useful.

I would not recommend Homo Deus as a starting point. It is the most speculative and the least supported by what has actually happened since publication. Read it after the others if you want to understand the trajectory of Harari's thinking, not as an introduction.

For the critics: David Graeber and David Wengrow, The Dawn of Everything (2021), is the most substantial book-length argument against the historical picture in Sapiens. It is also a major work in its own right and is essential reading for anyone serious about the deep-historical question Harari is addressing. Reading the two together is more productive than reading either alone.

The thing Harari does, that almost no other public intellectual reaches a comparable audience doing, is take the long view of human cooperation and ask what makes it work. The frame is imperfect. The reach is real. The questions he raises are the right ones, even when his answers need correcting.

Related in the Political Literacy series

  • Daniel Schmachtenberger — civilisational-systems thinker working adjacent ground at higher rigour
  • Andrej Karpathy — the technical AI side of the information-network thesis Harari develops in Nexus
  • Iain McGilchrist — closely related work on cognition, attention, and modern cultural drift
  • Sir Roger Penrose — the consciousness-and-reality questions Harari brushes against without engaging at depth
  • What Is Nationalism? — Harari's "imagined communities" frame applied directly to nationalism

Plus the framing piece, What Do Ireland's Parties Actually Stand For?, and the full Political Literacy archive.