a/research· 14 October 2026 · 5 min read

What makes Sapiens impossible to put down, and what you should notice while you read it

Cover of Sapiens formula (Omxus Press)

Within two paragraphs of opening Sapiens you're on a hike in East Africa two million years ago, among "anxious mothers cuddling their babies" and "chest-thumping machos trying to impress the local beauty" (Harari, 2014). Within two chapters you've accepted that money, nations and companies are shared fictions. Few readers notice how they got there.

A lecturer's book

It helps to know where the book came from. Yuval Noah Harari taught an undergraduate world history course at the Hebrew University of Jerusalem, and Sapiens grew out of those lectures. If you've taught first-years you know the job: hold a room of tired teenagers for an hour, and make them feel the world just turned over. He was evidently very good at it. Millions of people have now sat in that room.

Nothing's wrong with a great lecture. But a lecturer can see the room and take a question. A book that moves like a lecture gets all the momentum and none of the hands going up.

The build

Read every paragraph and count, and here's what you find. A new idea arrives every 14 words: one new fact, claim or reframing every sentence or two. Paragraphs run about 135 words. Every chapter opens on a hook and closes on a pull into the next. The four parts form one long argument, each part's conclusion becoming the next part's starting point. That's the build, and any writer can learn it.

Seven moves on top

Laid over the build are seven moves.

  • Questions. About ten a chapter, each carrying an assumption you swallow while working out the answer.
  • "We". A steady "we" puts you inside the author's view before you've decided to stand there.
  • "Scholars agree". With no scholar named, there's nobody to check.
  • The ratchet. Set up a common belief and knock it down, and each knock-down becomes the new floor.
  • Careful in the cheap places. The hedges ("perhaps", "probably") sit thickest on prehistory and the future, and thinnest on agriculture, empire, money and religion, where the book makes its strongest claims about how societies work.
  • Strangeness as proof. "Lawyers are, in fact, powerful sorcerers." It's fun, and it feels like an argument.
  • Inevitability. Words like "relentlessly" and "inexorable" turn what happened into what had to happen.

Each has a research record. A question's premise gets accepted more readily than the same content asserted (Loftus, 1975). Easy, fast reading feels truer (Reber & Schwarz, 1999). Readers caught up in a story argue back less (Green & Brock, 2000).

A history of the world from 1922

You could object that this is just how popular history gets written. So here's a control. In the aftermath of a war that had killed millions, H. G. Wells set about teaching ordinary people the whole human story, convinced that "human history becomes more and more a race between education and catastrophe" (Wells, 1920). His A Short History of the World (1922) has the same reader and the same sweep. Run the same counting script over both books.

Per 1,000 wordsSapiens (2014)Wells (1922)
Questions2.590.18
"Scholars", "scientists" and the like, unnamed1.900.17
"we", "us", "our"7.914.06
Comparisons ("like a", "as a")2.172.54

The two books build their paragraphs much the same way, hedge at the same basic rate, and Wells reaches for a comparison at least as often. What sets Sapiens apart is the moves: fourteen times the questions per word, eleven times the unnamed experts (four times once you count Wells's "authorities" and "men of science"), and twice the "we". Wells barely used them, and his book is still in print and still reads well. He wasn't above his own certainties, and he had plenty that have aged badly. But you can see them, because he says them out loud.

Keep the build, drop the moves

None of this makes Sapiens a bad book. Its metaphors are its greatest strength, its idea that we live inside shared stories has changed how plenty of people see the world, and its build is a gift to anyone who writes. But the build and the moves come apart. You can write short, dense, story-driven chapters and still state your premises, name your sources and give the other side its best case. The devices buy agreement on credit. The architecture earns attention on merit.

Swap each move for something honest. Turn the question into an argument that states its assumption. Keep "we" for what's truly shared: breathing, eating, dying. Name the scholar, the date and whoever disagrees. Assert the evidence and hedge the interpretation. Offer strangeness as an invitation. Name powerful forces without calling them fate. And before you knock down a common belief, give it its strongest evidence.

Why would this matter past the book club? Because the book's readers run things. And because what makes life hard is mostly stuff people built, and could build differently. A story that sounds inevitable tells you not to bother. Knowing it was written, by someone, with moves, is the first step in arguing back.

What you can do

Read with a pencil. Next time a book, a podcast or a speech has you nodding along, mark the question that assumed something, the experts with no names, the "we" you never agreed to. Check one. You don't need to stop liking the book. You just get your hands back on the wheel.

Wells's book is free at Project Gutenberg, so you can compare for yourself. And do it to this article. It opened on a hook. It said "we" a moment ago. It asked you a question and then answered it. The moves work on everyone, and that includes whoever's trying to warn you about them.

Sources

  1. Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79(5), 701–721. https://doi.org/10.1037/0022-3514.79.5.701
  2. Harari, Y. N. (2014). Sapiens: A brief history of humankind. McClelland & Stewart.
  3. Loftus, E. F. (1975). Leading questions and the eyewitness report. Cognitive Psychology, 7(4), 560–572. https://doi.org/10.1016/0010-0285(75)90023-7
  4. Reber, R., & Schwarz, N. (1999). Effects of perceptual fluency on judgments of truth. Consciousness and Cognition, 8(3), 338–342. https://doi.org/10.1006/ccog.1999.0386
  5. Wells, H. G. (1920). The outline of history. George Newnes.
  6. Wells, H. G. (1922). A short history of the world. Macmillan. https://www.gutenberg.org/ebooks/35461

Keep reading