What this covers

Moshe Koppel, an Israeli computer scientist and Talmud scholar, discusses with Russ Roberts why traditions and social norms should be understood as emergent orders—akin to language rather than law—that encode adaptive wisdom societies cannot fully articulate. Koppel argues that because long-surviving societies have independently evolved similar restrictive practices around food, sex, ritual, and transactions, these traditions likely confer real advantages even when their exact rationale remains opaque. He contends that such norms build social capital and train deferred gratification in ways that benefit even those outside the tradition, and that they should be maintained by self-selecting communities, evolve slowly through accumulated practice, and never be coerced by the state.

The conversation ranges across why legislation cannot replace tradition, how codification should follow established practice rather than precede it, and why both utilitarianism and deontology secretly depend on moral intuition rather than formula. Koppel illustrates his thesis through the Tucanoans' cassava-preparation rituals—which protect against cyanide poisoning through mechanisms no individual noticed—and explores how communities adapt through self-selection and fission while states freeze adaptive capacity through coercion. He defends tribalism and family partiality as natural and efficient rather than morally indefensible, argues that rapid ideological movements like cancel culture are ersatz religions lacking the equilibrium of evolved tradition, and warns that Epistemic Humility should guide how we treat ancient practices. The discussion also touches on why Peter Singer's attempts to overcome human nature prove futile and whether Hayek's extended order requires both intuition and formal rules.

Sharpest takeaway

Koppel argues that tradition and social norms are best understood as Hayekian emergent orders—like language, not legislation—that encode hard-won wisdom societies can't articulate, so they should be maintained by self-selecting communities, evolve slowly, and never be coerced by the state.

  • Long-surviving societies all independently evolved similar restrictive norms (food, sex, ritual, transaction), suggesting they confer adaptive advantages even when their rationale is unknown
  • Both utilitarianism and deontology secretly rely on moral intuition, so no formula can replace the intuition-plus-codification process that tradition embodies
  • Communities increase social capital and adapt through self-selection and split-offs, whereas state coercion freezes and decreases that adaptive capacity

The argument · threads4 threads · 26 claims
0.85

Coercion should codify tested norms, not invent rules; practice should precede policy design.

2 pointscentrality 5/5
  • Coercion (top-down legislation) should be adopted only when a norm has been well-tested by a bottom-up process of trial and error, and even then carefully and not overzealously—mirroring Hayek's view that a judge's job is to discover and codify the expectations people brought into a transaction rather than to invent rules.

    “you're arguing that coercion should only be adopted when it's well tested by a bottom up approach of trial and error, and then should be done on a careful basis and not to overzealous. Is that a fair summary?”

  • Codification should follow practice the way a campus should let people walk and then pave wherever the grass is worn down, rather than paving first based on a theory of shortest distances and then forbidding deviation—because the actual paths people pioneer reflect emergent needs that planners cannot foresee.

    “you let people walk around for a while, and wherever the grass gets worn down, that's what you pave. As opposed to: We have a theory about shortest distance between two points... So, we'll pave first, and then we'll put signs up, 'Don't walk here.'”

0.70

Expertise reveals complexity that justifies cautious, gradual change rather than novel interventions.

1 pointcentrality 5/5
  • Per Robert Conquest (via Scruton), in areas we know deeply we are conservative: a midwife who has delivered 500 babies, or a master pilot, resists novel changes not out of dogmatism but because deep expertise reveals the complexity and danger involved—so the rational response to that complexity is to change slowly and cautiously, which is the essence of a conservative approach to a complicated world.

    “Conquest argues that in areas that we know a lot about we're conservative... 'I really know this area deeply, this area of childbirth. I know how complicated it is... It's not a place to try new stuff.'”

0.56

Everyday interaction norms enable strangers to cooperate and should not be deliberately redesigned.

1 pointcentrality 4/5
  • The unnoticed everyday norms governing how we interact—how close to stand, what to specify in a contract versus leave unspoken, what one can rely on others to do—are what allow strangers, even people across an ocean with different traditions, to cooperate and trust each other; this near-miraculous cooperation, not abstract 'good for society' arguments, is the real foundation tradition serves, and we should not try to design these institutions in advance.

    “It's one thing to say that I trust my child, my spouse, my parent, but to trust not just the neighbor... but the person on the other side of the ocean to develop the institutions and norms that allow that to happen are not things we want to, I think, design in advance.”

0.29

Disrupting foundational norms like physical proximity risks lasting harm to social development.

1 pointcentrality 2/5
  • COVID risks creating in children a lasting suspicion of physical closeness to others, a disruption of fundamental interaction norms that constitutes an unplanned social-science experiment likely to have unpleasant long-term consequences.

    “One of the things I worry about COVID is that creating among children a suspicion of getting physically close to people. I don't think that's a good thing.”