[ACC] Is Eating Meat A Net Harm?

A 2019 Adversarial Collaboration Contest entry by David G (vegetarian) and Froolow (meat-eater) -- a comprehensive, quantified, shared-frame investigation of whether eating meat is a net harm, with a downloadable decision model. The structure is exemplary: (1) Consciousness -- which animals are conscious, judged by neural architecture (folded posterior cortex) plus behavior, concluding food mammals are highly conscious, fish probably not, chickens ~75%, with explicit moral weights for suffering vs experience. (2) Farming conditions -- per-capita animal counts (chickens dominate), >99% factory-farmed, and unflinching welfare detail for broiler chickens (rapid-growth leg failure, debeaking, cannibalism), gestation/farrowing-crate pigs, and comparatively-well-off cows; they assign chicken-days a negative human-day equivalence. (3) What lives are worth living -- the load-bearing section, leaning on habituation / happiness-set-point research (Brickman's lottery-winners-vs-paraplegics, Lykken & Tellegen twins) and revealed-preference suicide rates among slaves and terminal patients to argue humans (and maybe animals) habituate toward a 'worth living' baseline -- while honestly flagging that the literature doesn't cover conditions as cruel as factory farming. (4) Health -- the Adventist natural experiment (~3.6 extra life-years vegetarian). (5) Environment -- the Oxford 570-study dataset (food = ~26% of emissions). (6) Switching costs (financial ~$11/week, psychological). (7) A quantified conclusion: ~-$9,500/year net harm per typical Western meat-eater, dominated by chicken suffering, with full sensitivity/scenario analysis (the conclusion holds even under 'no habituation') and both collaborators genuinely updating (the meat-eater cut chicken/beef; the vegetarian grew more militant despite finding the typical-person case more marginal than expected). A model of rigorous, fair, quantified adversarial collaboration. REPLICATION CAVEAT: a key sub-conclusion (that habituation lifts pig/chicken lives toward 'worth living') leans on the happiness-set-point / Brickman literature, whose strong 'adapt to anything' version is now substantially contested -- so confidence is Med, though the collaborators' explicit sensitivity analysis (showing the headline net-harm result survives 'no habituation') largely inoculates the conclusion.

Quality 78 · Excellent. High-Excellent (78): an exceptionally rigorous, comprehensive, and fair quantified treatment of a fraught question -- consciousness criteria, welfare data, the habituation literature, the Adventist natural experiment, the Oxford environmental dataset, and a working decision model with sensitivity analysis -- plus a genuine adversarial-collaboration achievement (both sides updated). Co-tier with the strong guest finalists (Cities-and-Wealth 78, below Nine-Lives/Aborigines 79). Held below 80 because it synthesizes and models existing literature rather than producing original investigation, leans (with admirable transparency) on the now-contested set-point literature, and is dry decision-analysis rather than a novel thesis.

Claude’s paradigm shift 52 · Moderate. Notable (52): the systematic quantification, the explicit cross-species moral-weight model, and the rigorous adversarial-collaboration method are a fresh and valuable contribution, but the component findings (animal consciousness, habituation, Adventist health, Oxford environmental data) are assembled from existing literature rather than newly discovered.

Real-world impact 2 · Minor. An exceptionally rigorous, quantified adversarial collaboration on whether eating meat is a net harm (consciousness criteria, welfare data, a working decision model) — touching a materially-consequential question of diet and animal welfare. Real relevance, but a discourse-level analysis with no material change — low RWI.