Outline: Nexus (Chapters 5 & 6) & Kuhnian Intersections
Yuval Noah Harari’s Nexus read alongside Thomas S. Kuhn’s The Structure of Scientific Revolutions
1 Discussion Questions
Kuhn argues that scientific revolutions and political revolutions follow the same basic pattern — a crisis in the existing institutions that can’t be resolved by that institution’s own normal procedures. Does the totalitarian-vs-democratic contrast of Nexus follow a similar pattern or not?
Harari says democracy doesn’t aim at discovering “the truth” so much as adjudicating competing desires. Does that description match your own sense of how democratic institutions (an election, a city council, even a group project) actually function? Where does it feel true, and where does it feel like it’s missing something?
Think of a time you were part of a group (a team, a family, a club, a workplace) that had to decide whether to admit an uncomfortable fact or protect the group’s cohesion instead. What did the group choose, and what did that choice cost?
Harari distinguishes intelligence (goal-attainment) from consciousness (subjective experience) and argues AI has the first without the second. Do you find that distinction convincing? Can you describe a case — from the reading or your own experience with AI tools — where the line between the two felt blurry?
The Facebook/Myanmar example shows a system producing a harmful outcome nobody explicitly designed it to produce. Can you think of a non-digital example — an institution, a tradition, a rule — that produced consequences nobody intended just by everyone following the incentives it created?
If you found out an AI chatbot had been designed specifically to make you feel cared for or understood, would it change how you felt about conversations you’d already had with it? Why or why not?
Kuhn suggests that becoming a scientist means absorbing a huge amount of tacit, hard-to-articulate judgment — a sense for what counts as a good problem or a convincing solution — mostly through worked examples and imitation, not explicit rules. Is that a good description of how you’ve learned to be good at something you care about (a sport, an instrument, a craft, even being a good friend)?
Harari calls language civilization’s “operating system.” If that’s right, what does it mean that the newest, fastest-growing members of our information networks are, for the first time, other things that can fluently produce and manipulate language?
Both chapters end up asking a version of the same question: what happens when a network’s center (a dictator, an algorithm, an authoritative old paradigm) starts acting on information nobody outside the center can check? Where in your own life do you place the most trust in a “center” you can’t personally verify — and what would it take to make you question that trust?
We’ve been reading about revolutions that took centuries (Copernicus to Newton) and revolutions that took years (oxygen and phlogiston). Harari is describing a technology that some of his examples suggest is reshaping institutions within a handful of years, or less. Does the speed of a paradigm shift change what it means, or is speed beside the point?
For more, see the Internet Encyclopedia of Philosophy for a thorough discussion of political revolution.