Political Psychology Links, 2/5/2026
Vittorio on gossip at scale; Ayaan Hirsi Ali on Somali culture; Andrey Mir on polarization and media; Dan Williams on evolutionary mismatch
Social media is a consensus engine. You can see what everyone believes in real time. Disagreement is visible, measurable, and punishable at scale. The tribe used to be 150 people. Now it's everyone you've ever met plus a world of strangers watching.
…Women were roughly stable through the early 2000s. The acceleration starts around 2007-2008. The curve steepens through the 2010s as smartphones became universal and platforms became more sophisticated. Women are by nature more liberal, but the radicalization coincides with the rise in smartphones adoption.
…The marriage gap in voting is one of the most consistent predictors. And marriage rates have collapsed precisely during the period of divergence.
For a variety of reasons, you are likely to be less radical if you are married. I suspect that for declining marriage rates and female radicalization, causality runs in both directions.
Pointer from Jason Manning.
Somali society is organized around the clan. Loyalty is not abstract, nor is it civic. It is biological and binding. The individual exists only insofar as he serves the group. Protection, marriage, honor, silence, and punishment are governed by this code. Obligations flow inward, sanctions flow downward. The clan precedes the individual and outlives him.
This structure is pre-modern, but it is also anti-modern. It resists the very conditions that make liberal societies function: individual accountability, transparency, impersonal law, and trust beyond kin. Ernest Gellner warned that a modern nation-state cannot be built on tribal loyalty. Tribalism fragments authority and dissolves shared obligation. Where it persists, institutions decay.
The fundamental rule of social order is that you have to reward cooperators and punish defectors. But ironically, better execution of this at a small-unit level, such as a clan, makes it difficult to execute this at a large-unit level. The tightly-ordered clan or Mafia is able to act as a defector relative to the state.
After 2020, the erosion of trust in the media led to historically high levels of news avoidance, and now the broader erosion of trust in everything and growing anxiety over polarization is leading to engagement avoidance. It seems people are increasingly fed up with online rage and are trying to fence themselves off from it. How successful this is, and how it really changes the online behavior of the masses, is hard to say: outrageous events still drive—and likely will always drive—traffic. But some immunity to online rage may be forming.
He is responding to a study that showed that outrage-bait tactics by media companies increase engagement somewhat, but subscriptions fail to rise.
The scale and complexity of the modern environment that bears on political debate are mind-boggling. Hundreds of millions of strangers are enmeshed in interacting economic, political, and institutional forces that bear no resemblance to the small-scale worlds we evolved in.
…First, the modern world radicalises our reliance on social learning. When forming beliefs about topics relevant to modern politics, we almost always lack the ability to cross-check what we’re told against our experience, either because it is too distant in space and time or because the topics concern abstract phenomena (GDP, inflation, demographic trends, economic growth, etc.) that no one can directly experience.
Second, the intuitions most people bring to understanding modern societies are systematically misleading.
We evolved to be highly skilled at forming alliances, reading intentions, tracking reputations, and playing local status games. In contrast, neither our evolutionary endowment nor first-hand experiences prepare us to understand large-scale systems characterised by emergent properties,
…Third, the modern information environment through which people attempt to learn about this strange, new world and overcome their default ignorance and confusion is more of a hindrance than a help.
substacks referenced above:
@
@
@





"But ironically, better execution of this at a small-unit level, such as a clan, makes it difficult to execute this at a large-unit level."
I think you're essentially referencing how multi-level selection (MLS) works. The body doesn't want its heart doing its own thing, rebelling against the body's needs. The corporation doesn't want its Des Moines sales office to be too independent. A nation doesn't want each city to go off and do their own thing
The higher level organization (body, corporation, nation) functions better when the parts align their actions to the higher goals. If survival depends on the success of that higher level, i.e., if the corporation goes bankrupt then the Des Moines sales team loses their jobs, then higher level entities that keep their parts in line will have longer term success
I don't mean this in an exclusive sense. There are other models, but you see this one repeated over and over.
If a Somali tribe wanted greater long term wealth and security (for instance), it probably has to subordinate some of its desires to a larger, more internationally competitive organization
Dan Williams has earned my official "Underrated" designation.