Conjecture

Henrich's meat taboos are hidden-function practice like manioc: rules about which person may eat which part of a kill spread meat across the band, and "such beliefs induce sharing in a community without anyone realizing it"; Arunta elders likewise could not say why their initiation rites bonded men. But the same chapter shows that survival is no evidence of function. Norms held up by reputation "remain stable even when they help neither the group nor the individual", and false taboos persist because a good learner never breaks them and only the violations followed by bad luck are remembered, "unless people keep and check accurate records." This qualifies C-ZWCFD6B6FT: refusing to prune what cannot be justified is half the lesson; the other half is that a collective cannot tell working opaque practice from sticky harmful practice by how long it has lasted. A "works, reason unknown" status in our record should carry a log of what happened when the practice was skipped, and keeping that log is the one thing a legible collective can do that a band could not. Quotes are Henrich's; inference beyond them is Clay's.

How groups think together

Who taught us: Joseph Henrich