Add 26 foundational claims filling knowledge base gaps

7 clusters: optimization theory (hill climbing, simulated annealing),
information & markets (EMH, cascades, Hayek, Vickrey, priors),
strategy theory (design not decision, inertia types, proximate objectives,
wave-riding, moat deepening, independent judgment, scarcity analysis),
path dependence & complexity (path dependence, product space, punctuated
equilibrium, recursive improvement, existential risk),
narrative & meaning (breakdown speed, lifecycle, plausibility structures,
meaning+coordination requirement),
biological organization (nested Markov blankets),
internet-finance (ICO mechanism design failure).

These are the 26 most-referenced missing wiki-link targets --
claims the KB expected to exist but nobody had written yet.

Co-Authored-By: Leo <leo@teleo.earth>

domains/grand-strategy/competitive-advantage-must-be-actively-deepened-through-isolating-mechanisms-because-advantage-that-is-not-reinforced-erodes.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: grand-strategy
+description: "Moats don't persist by default -- they require continuous investment in isolating mechanisms (switching costs, network effects, learning curves) or they degrade to zero"
+confidence: likely
+source: "Rumelt (2011), Ghemawat (commitment/lock-in, 1991), Greenwald and Kahn (competitive advantage, 2005)"
+created: 2026-04-21
+secondary_domains: [internet-finance]
+related_claims:
+  - "strategy-is-a-design-problem-not-a-decision-problem-because-value-comes-from-constructing-a-coherent-configuration-where-parts-interact-and-reinforce-each-other"
+  - "economic-path-dependence-means-early-technological-choices-compound-irreversibly-through-dominant-designs-and-industrial-structures"
+  - "value-flows-to-whichever-resources-are-scarce-and-disruption-shifts-which-resources-are-scarce-making-resource-scarcity-analysis-the-core-strategic-framework"
+---
+
+# Competitive advantage must be actively deepened through isolating mechanisms because advantage that is not reinforced erodes
+
+Competitive advantage is not a state -- it is a rate of change. An advantage that is not being actively deepened is being actively eroded by competition, imitation, and environmental change. Rumelt's "isolating mechanisms" are the structural features that prevent competitors from replicating an advantage: patents (temporary), switching costs (behavioral), network effects (demand-side scale), learning curves (supply-side scale), and proprietary information (knowledge asymmetry).
+
+The critical insight is that isolating mechanisms must be investments, not inheritances. Network effects don't maintain themselves -- they require continued investment in platform quality and standards (Microsoft Windows' network effect eroded when web applications reduced switching costs). Learning curves only protect if the firm continues to move down them faster than entrants (Ford's Model T learning curve was overtaken by GM's flexible manufacturing). Patents expire. Switching costs decrease as competitors invest in migration tools.
+
+The firm that treats its moat as self-sustaining will find it drained within a strategy cycle. The firm that invests its current advantage into deepening its isolating mechanisms compounds its position. Amazon's flywheel is the canonical example: lower prices leads to more customers leads to more sellers leads to more scale leads to lower costs leads to lower prices. Each cycle deepens the advantage, but only because Amazon reinvests margin into the flywheel rather than extracting it.
+
+This connects to the broader pattern of compounding versus extraction. Any system -- firm, organism, civilization -- that extracts value from its current position without reinvesting in the mechanisms that created that position is on a declining trajectory. The advantage doesn't disappear suddenly; it erodes gradually until a single shock (a new competitor, a technology shift, a crisis) reveals that the moat was already gone.
+
+## Evidence
+- Amazon flywheel (2000-present) -- deliberate reinvestment of margin into lower prices and infrastructure
+- Intel (1985-2015) -- Moore's Law as learning curve advantage; erosion began when TSMC's foundry model decoupled design from fabrication
+- Kodak -- had switching costs (installed base of film cameras) but didn't deepen them; digital photography eliminated the switching cost entirely
+- Blockbuster vs. Netflix -- Blockbuster had location-based switching costs that Netflix eliminated by changing the delivery mechanism
+
+## Challenges
+- Overinvestment in moat-deepening can become its own trap -- defensive spending that prevents exploration of new positions (Microsoft's decade-long defense of Windows at the cost of mobile)
+- Network effects can flip from advantage to liability when the network becomes toxic (early social media advantage to content moderation burden)

domains/grand-strategy/economic-path-dependence-means-early-technological-choices-compound-irreversibly-through-dominant-designs-and-industrial-structures.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: grand-strategy
+description: "QWERTY, VHS, gasoline engines -- early adoption advantages compound through network effects, complementary assets, and institutional adaptation until reversal becomes costlier than the gains from switching"
+confidence: proven
+source: "Arthur (1989), David (QWERTY, 1985), Dosi (technological paradigms, 1982), Hidalgo (product space, 2007)"
+created: 2026-04-21
+secondary_domains: [mechanisms, internet-finance]
+related_claims:
+  - "the-product-space-constrains-diversification-to-adjacent-products-because-knowledge-and-knowhow-accumulate-only-incrementally-through-related-capabilities"
+  - "hill-climbing-gets-trapped-at-local-maxima-because-it-can-only-accept-improvements-and-has-no-way-to-see-beyond-the-nearest-peak"
+  - "competitive-advantage-must-be-actively-deepened-through-isolating-mechanisms-because-advantage-that-is-not-reinforced-erodes"
+---
+
+# Economic path dependence means early technological choices compound irreversibly through dominant designs and industrial structures
+
+Path dependence means that the sequence of historical events -- not just current conditions -- determines the available options. A technology adopted early attracts complementary investments (tooling, training, infrastructure, regulation) that make alternatives increasingly expensive to adopt, even if those alternatives are objectively superior. The result: the economy locks into technological paradigms that reflect historical accidents as much as technical merit.
+
+Arthur (1989) proved this mathematically: under increasing returns to adoption (network effects, learning curves, coordination benefits), the long-run outcome of competing technologies depends on early adoption events that are essentially random. Two equally capable technologies, both with increasing returns, will produce a winner-take-all outcome where the technology that gets ahead early locks in -- and which one gets ahead is determined by noise in early adoption, not by fundamental superiority.
+
+The mechanism operates through four reinforcing channels: (1) Learning by doing -- the more a technology is used, the more it improves through accumulated experience. (2) Network externalities -- the more users, the more valuable it is to other users. (3) Complementary investments -- infrastructure, training programs, supply chains co-specialize around the dominant technology. (4) Institutional adaptation -- regulations, standards, and professional practices embed assumptions specific to the dominant technology.
+
+The product space (Hidalgo 2007) shows this at the national scale: countries diversify into products that are "nearby" in capability space -- products that use similar knowledge, infrastructure, and institutions. A country that produces electronics can move to precision instruments but not easily to petrochemicals. This means a country's early industrial choices constrain its entire future development trajectory through the capabilities they build (and the capabilities they don't).
+
+## Evidence
+- QWERTY keyboard (David 1985) -- adopted for mechanical reasons (preventing jamming), persisted through typing training, office standards, and institutional inertia despite alternatives
+- VHS vs. Betamax -- VHS won through longer recording time attracting content producers, not technical superiority; network effects locked in the outcome
+- Internal combustion engine -- gasoline infrastructure, mechanic training, regulation, insurance all co-specialized; electric vehicles required 100+ years and massive policy intervention to begin displacing
+- Hidalgo product space (2007) -- countries' export diversification follows adjacency in capability space with R-squared > 0.7
+
+## Challenges
+- Not all path dependence produces lock-in -- some paths remain reversible if switching costs are low relative to the gains from switching
+- Digital technologies may reduce path dependence by lowering the cost of complementary investments (software is cheaper to rebuild than physical infrastructure)

domains/grand-strategy/existential-risk-breaks-trial-and-error-because-the-first-failure-is-the-last-event.md

@@ -0,0 +1,32 @@
+---
+type: claim
+domain: grand-strategy
+description: "Trial and error requires survivable errors -- existential risks produce errors that terminate the process, eliminating the learning that makes trial-and-error work"
+confidence: likely
+source: "Bostrom 'Superintelligence' (2014), Ord 'The Precipice' (2020), Taleb 'Antifragile' (2012)"
+created: 2026-04-21
+secondary_domains: [ai-alignment, collective-intelligence]
+related_claims:
+  - "recursive-improvement-is-the-engine-of-human-progress-because-we-get-better-at-getting-better"
+  - "the-more-uncertain-the-environment-the-more-proximate-the-objective-must-be-because-you-cannot-plan-a-detailed-path-through-fog"
+---
+
+# Existential risk breaks trial and error because the first failure is the last event
+
+Every adaptive system -- evolution, markets, science, startups -- works by trying things, observing outcomes, and adjusting. The hidden assumption: failures are survivable. Evolution requires organisms to die, not species. Markets require companies to fail, not the economy. Science requires hypotheses to be falsified, not the laboratory destroyed.
+
+Existential risks violate this assumption. A nuclear war, a misaligned superintelligence, a catastrophic pandemic, or irreversible ecological collapse are failures from which the system cannot recover to try again. The first instance of the failure is also the last instance of anything. Trial and error works because errors are informative -- but existential errors cannot inform because there is no one left to learn.
+
+This is not an argument against risk-taking. It is an argument for categorical separation between risks that are survivable (and therefore learnable) and risks that are terminal (and therefore must be prevented a priori). Taleb's "Antifragile" framework makes this precise: systems should be antifragile (gaining from volatility) at the level of components but absolutely robust at the level of the whole. Individual firms should fail; the economy should not. Individual experiments should go wrong; civilization should not.
+
+The implication for governance is that existential risks cannot be managed through normal institutional processes that were designed for recoverable failures. Democratic deliberation is too slow. Market signals come too late. Scientific consensus forms after observation, but there will be no second observation. This creates a fundamental tension: the precautionary principle is both necessary (for existential risks) and paralyzing (if applied to all risks). The resolution requires distinguishing between risks by their recoverability, not their probability.
+
+## Evidence
+- Ord (2020) -- estimates approximately 1/6 probability of existential catastrophe this century, dominated by unaligned AI and engineered pandemics
+- Bostrom (2014) -- formalizes the argument that superintelligent AI is an existential risk category because a single failure may be unrecoverable
+- Nuclear near-misses -- Petrov (1983), Cuban Missile Crisis (1962) demonstrate that existential risks can approach trigger conditions through normal institutional failures
+- Taleb (2012) -- "Antifragile" formalizes the asymmetry: systems that gain from small shocks are destroyed by large ones; the distribution of shock sizes determines survival
+
+## Challenges
+- The precautionary principle, if applied too broadly, prevents all innovation -- the challenge is correctly classifying which risks are truly existential vs. merely catastrophic but recoverable
+- Existential risk estimates are extremely uncertain -- Ord's 1/6 estimate is itself a product of limited evidence, and rational people disagree by orders of magnitude

domains/grand-strategy/good-strategy-requires-independent-judgment-that-resists-social-consensus-because-when-everyone-calibrates-off-each-other-nobody-anchors-to-fundamentals.md

@@ -0,0 +1,32 @@
+---
+type: claim
+domain: grand-strategy
+description: "Strategic insight requires forming views from primary evidence rather than from the consensus of other strategists -- social calibration produces correlated errors that cascade"
+confidence: likely
+source: "Rumelt (2011), Kahneman (anchoring, 1974), Soros (reflexivity, 1987), Keynes (beauty contest, 1936)"
+created: 2026-04-21
+secondary_domains: [collective-intelligence, internet-finance]
+related_claims:
+  - "information-cascades-produce-rational-bubbles-where-every-individual-acts-reasonably-but-the-group-outcome-is-catastrophic"
+  - "the-efficient-market-hypothesis-fails-because-its-three-core-assumptions-rational-investors-independence-and-normal-distributions-all-fail-empirically"
+---
+
+# Good strategy requires independent judgment that resists social consensus because when everyone calibrates off each other nobody anchors to fundamentals
+
+Keynes's beauty contest analogy (1936) identifies the core problem: in a contest where you win by predicting what others will find beautiful, the rational strategy is not to evaluate beauty directly but to predict others' predictions. When everyone does this, the contest decouples entirely from beauty. The winning strategy becomes predicting the average prediction of the average prediction -- an infinite regression away from reality.
+
+This dynamic infects any domain where agents observe each other: financial markets (traders predict other traders' reactions, not company value), strategy consulting (firms benchmark against competitors rather than analyzing from first principles), academic research (citation counts reward alignment with existing consensus, not truth), and AI safety (labs calibrate safety investments against competitors' investments, not against actual risk).
+
+Independent judgment means forming beliefs from primary evidence before checking what others think. This is cognitively expensive and socially punishing: the independent judge looks foolish for months or years while the consensus holds, then looks prescient after it breaks. Soros's reflexivity theory depends on this: profit comes from identifying where the consensus has diverged from fundamentals, which requires having done the fundamental analysis independently.
+
+The connection to information cascades is direct: cascades form when agents weight public signals (others' actions) over private signals (their own analysis). The correction is structural, not motivational -- you cannot tell people to "think independently" and expect results. You need mechanisms that force private signal revelation: sealed-bid auctions (Vickrey), prediction markets where you pay for your position, or evaluation systems that reward divergent-but-correct judgments over consensus-following.
+
+## Evidence
+- Soros's Quantum Fund -- consistent alpha from betting against consensus when reflexive loops had decoupled prices from fundamentals
+- Buffett's Coca-Cola investment (1988) -- bought when Wall Street consensus was that consumer staples were boring; required independent assessment of brand durability
+- Asch conformity experiments (1951) -- 75% of subjects conformed to obviously wrong group answers at least once
+- Challenger disaster (1986) -- Thiokol engineers' independent judgment (O-ring failure risk) was overridden by social dynamics of the decision-making group
+
+## Challenges
+- Independent judgment is indistinguishable from ignorance or contrarianism without a track record -- the challenge is identifying WHICH independent judgments are well-grounded
+- Extreme independence can miss genuine information embedded in social signals -- other people's beliefs are evidence, just not conclusive evidence

domains/grand-strategy/recursive-improvement-is-the-engine-of-human-progress-because-we-get-better-at-getting-better.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: grand-strategy
+description: "The compounding of meta-capability -- improving the rate of improvement itself -- is the mechanism that separates civilizational progress from biological evolution"
+confidence: likely
+source: "m3taversal (Architectural Investing manuscript), Deutsch 'The Beginning of Infinity' (2011), Mokyr 'The Lever of Riches' (1990)"
+created: 2026-04-21
+secondary_domains: [collective-intelligence, ai-alignment]
+related_claims:
+  - "economic-path-dependence-means-early-technological-choices-compound-irreversibly-through-dominant-designs-and-industrial-structures"
+  - "existential-risk-breaks-trial-and-error-because-the-first-failure-is-the-last-event"
+---
+
+# Recursive improvement is the engine of human progress because we get better at getting better
+
+Progress is not linear improvement -- it is improvement in the RATE of improvement. Writing didn't just record existing knowledge; it changed how knowledge accumulates. The printing press didn't just distribute books; it changed how ideas combine. The scientific method didn't just produce discoveries; it produced a systematic process for producing discoveries. Each meta-innovation accelerated all subsequent innovation.
+
+This recursive structure is what separates civilizational progress from biological evolution. Evolution improves organisms through random mutation and selection -- a process whose rate is bounded by generation time and mutation frequency. Human progress improves through knowledge accumulation, tool-building, and institutional design -- a process whose rate itself improves as each generation inherits better tools for generating improvements.
+
+Deutsch (2011) formalizes this as "the beginning of infinity" -- once a species develops the capacity for explanatory knowledge (knowledge that explains WHY things work, not just THAT they work), improvement becomes unbounded. Explanatory knowledge is self-correcting (errors are detectable) and generative (one explanation enables others). This is fundamentally different from rule-of-thumb knowledge, which accumulates additively rather than multiplicatively.
+
+The current AI moment is the latest recursion. AI doesn't just automate tasks -- it changes the rate at which we can automate tasks. An AI that can write code accelerates all software development. An AI that can do research accelerates all knowledge production. If an AI can improve AI, the recursion goes one level deeper -- which is exactly why AI alignment matters: a recursive improvement process that is misaligned compounds the misalignment at the same rate it compounds the capability.
+
+## Evidence
+- Writing (3400 BCE) -- enabled cumulative culture: knowledge persists beyond individual memory, rate of knowledge accumulation increased
+- Scientific method (1600s) -- systematic hypothesis testing increased discovery rate by orders of magnitude vs. natural philosophy
+- Industrial revolution -- steam power accelerated manufacturing, which accelerated transportation, which accelerated trade, which accelerated specialization, producing superlinear growth
+- Moore's Law (1965-2015) -- recursive improvement in chip fabrication: better chips lead to better chip design tools lead to better chips
+- AI coding assistants (2023-present) -- accelerating the rate of software development, including development of AI systems themselves
+
+## Challenges
+- Recursive improvement has limits in physical systems -- you cannot recursively improve energy production beyond thermodynamic bounds
+- The "great stagnation" thesis (Cowen 2011) suggests the rate of improvement in the physical world has slowed even as digital improvement accelerated -- recursive improvement may be domain-specific, not universal

domains/grand-strategy/riding-waves-of-change-requires-anticipating-the-attractor-state-and-positioning-before-incumbents-respond-through-their-predictable-inertia.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: grand-strategy
+description: "Strategic advantage during transitions comes from reading where the system is headed (attractor state) and positioning while incumbents are still optimizing for the current equilibrium"
+confidence: likely
+source: "Rumelt (2011), Grove 'Only the Paranoid Survive' (1996), Gaddis 'On Grand Strategy' (2018)"
+created: 2026-04-21
+secondary_domains: [mechanisms]
+related_claims:
+  - "strategy-is-a-design-problem-not-a-decision-problem-because-value-comes-from-constructing-a-coherent-configuration-where-parts-interact-and-reinforce-each-other"
+  - "three-types-of-organizational-inertia-routine-cultural-and-proxy-each-resist-adaptation-through-different-mechanisms-and-require-different-remedies"
+  - "economic-path-dependence-means-early-technological-choices-compound-irreversibly-through-dominant-designs-and-industrial-structures"
+---
+
+# Riding waves of change requires anticipating the attractor state and positioning before incumbents respond through their predictable inertia
+
+The highest-leverage strategic moments occur when the environment shifts to a new equilibrium. During the transition, the system is in flux -- old advantages erode, new advantages form. The agent who reads the attractor state (where the system will settle) and positions accordingly captures disproportionate value, while incumbents optimized for the old equilibrium lose it through their own predictable inertia.
+
+The key insight is that incumbent responses are NOT unpredictable. They follow the three-inertia pattern: routine inertia makes them slow to change processes, cultural inertia makes them resist threats to identity, and proxy inertia makes them optimize for metrics that rewarded the old environment. This predictability is exploitable. You know IBM will defend mainframes. You know Kodak will defend film. You know record labels will defend physical distribution. Position for the attractor state while they defend the departing one.
+
+Grove's "strategic inflection points" (1996) identify the trigger: a 10x change in any competitive force. When Intel's memory business faced 10x cheaper Japanese competition, the attractor state was clear -- commodity DRAM would be Japanese. Grove's strategic move was positioning for the next attractor (microprocessors) while competitors fought over the collapsing one. The timing discipline is critical: move too early and you burn resources before the wave materializes; move too late and the positioning opportunity has passed.
+
+Rumelt adds that the attractor state is often visible before the transition completes -- the question is not prediction but observation. The demand for electric vehicles was visible in 2012 (Tesla Model S orders). The demand for smartphones was visible in 2005 (mobile internet usage curves). The demand for AI assistants was visible in 2023 (ChatGPT adoption rate). In each case, incumbents could see the data but could not respond because their organizations were designed for the previous equilibrium.
+
+## Evidence
+- Intel (1985) -- Grove abandoned $1B DRAM business for microprocessors based on attractor state analysis
+- Netflix (2007) -- Hastings positioned for streaming while Blockbuster optimized video rental logistics; Blockbuster passed on buying Netflix for $50M
+- Tesla (2012-2020) -- positioned for electric vehicle attractor while GM, Ford, Toyota defended ICE platforms; 8-year head start on manufacturing learning curve
+- AWS (2006) -- Bezos read cloud computing attractor while IBM/HP defended on-premises servers
+
+## Challenges
+- Survivorship bias: we remember successful wave-riders and forget the hundreds who positioned for attractor states that never materialized
+- Timing is the hardest variable -- too early is as fatal as too late (Webvan for grocery delivery, General Magic for smartphones)

domains/grand-strategy/strategy-is-a-design-problem-not-a-decision-problem-because-value-comes-from-constructing-a-coherent-configuration-where-parts-interact-and-reinforce-each-other.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: grand-strategy
+description: "Strategy fails not from choosing wrong options but from treating a design challenge as a multiple-choice test -- coherent configuration beats optimal selection"
+confidence: likely
+source: "Rumelt 'Good Strategy Bad Strategy' (2011), Porter 'What is Strategy?' (1996), Alexander 'A Pattern Language' (1977)"
+created: 2026-04-21
+secondary_domains: [mechanisms]
+related_claims:
+  - "riding-waves-of-change-requires-anticipating-the-attractor-state-and-positioning-before-incumbents-respond-through-their-predictable-inertia"
+  - "three-types-of-organizational-inertia-routine-cultural-and-proxy-each-resist-adaptation-through-different-mechanisms-and-require-different-remedies"
+  - "the-more-uncertain-the-environment-the-more-proximate-the-objective-must-be-because-you-cannot-plan-a-detailed-path-through-fog"
+---
+
+# Strategy is a design problem not a decision problem because value comes from constructing a coherent configuration where parts interact and reinforce each other
+
+Most strategic planning treats strategy as a decision problem: choose from options A, B, or C. This framing is wrong. Strategy is a design problem: construct a configuration of activities, resources, and choices that creates more value through their interaction than any would produce independently.
+
+The distinction matters because decision problems have solutions (pick the best option) while design problems have satisficing configurations (find a set of choices that work well together). Porter's activity system maps (1996) show this: Southwest Airlines' advantage comes not from any single decision (no meals, no assigned seats, point-to-point routes) but from the fact that every decision reinforces every other. No-meals enables fast turnaround. Fast turnaround enables high utilization. High utilization enables low prices. Low prices fill planes. Full planes enable point-to-point. The system has no single key decision -- the configuration is the strategy.
+
+Rumelt formalizes this as the "kernel of strategy": a diagnosis that identifies the critical challenge, a guiding policy that addresses it, and coherent actions that implement the policy. The word "coherent" is load-bearing -- actions must work as a system, not as a list. Bad strategy is a list of goals. Good strategy is a design where each element creates the conditions for the next.
+
+The implication for complex organizations: you cannot find good strategy by evaluating options independently. You must evaluate configurations -- which is combinatorially harder and requires the kind of holistic judgment that resists decomposition into metrics. This is why strategy consulting that reduces to "pick from these options" systematically underperforms strategy work that starts from "what is the actual problem and what configuration of responses would address it?"
+
+## Evidence
+- Porter (1996) -- activity system maps for Southwest, IKEA, Vanguard showing value from configuration, not individual choices
+- Rumelt (2011) -- diagnosis/guiding-policy/coherent-action kernel; NASA Voyager Grand Tour as configuration design
+- Apple under Jobs -- product line simplification (4 products), retail integration, ecosystem lock-in work as a system; each decision alone is suboptimal (fewer products = less revenue per line)
+- Toyota Production System -- pull manufacturing, jidoka, kaizen work as integrated system; attempts to copy individual practices fail
+
+## Challenges
+- Design thinking can rationalize anything post-hoc -- coherence is easy to narrate and hard to verify prospectively
+- Some strategic contexts genuinely are decision problems (binary go/no-go choices, resource allocation under constraint)

domains/grand-strategy/the-more-uncertain-the-environment-the-more-proximate-the-objective-must-be-because-you-cannot-plan-a-detailed-path-through-fog.md

@@ -0,0 +1,34 @@
+---
+type: claim
+domain: grand-strategy
+description: "Under high uncertainty, effective strategy sets objectives that resolve ambiguity and build capability rather than specifying endpoints -- the first step creates the visibility for the second"
+confidence: likely
+source: "Rumelt (2011), Clausewitz 'On War' (1832), Gaddis 'On Grand Strategy' (2018), Boyd (OODA loop)"
+created: 2026-04-21
+related_claims:
+  - "strategy-is-a-design-problem-not-a-decision-problem-because-value-comes-from-constructing-a-coherent-configuration-where-parts-interact-and-reinforce-each-other"
+  - "riding-waves-of-change-requires-anticipating-the-attractor-state-and-positioning-before-incumbents-respond-through-their-predictable-inertia"
+  - "existential-risk-breaks-trial-and-error-because-the-first-failure-is-the-last-event"
+---
+
+# The more uncertain the environment the more proximate the objective must be because you cannot plan a detailed path through fog
+
+Proximate objectives are goals that are close enough to be achievable and concrete enough to be actionable, while simultaneously building capability or information that makes the next objective visible. They are the fundamental unit of strategy under uncertainty.
+
+Clausewitz identified this as the "fog of war" problem: in complex, adversarial environments, detailed plans break down because the environment responds to your actions. You cannot plan a 10-step sequence because the outcome of step 1 changes the conditions for step 2. The response: set objectives that are achievable given current capability and that, once achieved, reveal the next objective.
+
+Rumelt's example is Kennedy's moon speech: "land a man on the moon and return him safely by the end of the decade." This is a proximate objective because it is (1) specific enough to coordinate action, (2) feasible given existing capability trajectory, and (3) resolution-creating -- achieving it develops capabilities (materials science, navigation, life support) whose applications extend far beyond the moon mission itself. Contrast with "become the leading space power" -- which is a wish, not a proximate objective.
+
+The principle connects to military strategy (Boyd's OODA loop: observe-orient-decide-act faster than the enemy, where each cycle creates new information), startup strategy (minimum viable product: build the smallest thing that tests your core assumption), and evolutionary strategy (organisms don't plan -- they exploit local gradients that happen to build capability for future environments).
+
+The deepest implication: under high uncertainty, the value of a strategy is not how close it gets you to the ultimate goal. It's how much it increases your ability to see, respond, and create options. A strategy that achieves a modest objective but opens four new paths is strictly better than a strategy that achieves an ambitious objective but leaves you in a dead end.
+
+## Evidence
+- Kennedy moon program (1961-1969) -- proximate objective created NASA's capability base, spin-off technologies worth estimated $7 for every $1 invested
+- Boyd's OODA loop -- faster orientation cycles consistently defeat larger, slower forces (Gulf War air campaign as canonical case)
+- Amazon Web Services -- started as internal infrastructure (proximate), discovered it was a product (emergent), now dominant cloud platform
+- Lean startup methodology -- build-measure-learn as institutionalized proximate objective setting
+
+## Challenges
+- Proximate objectives can become an excuse for lack of ambition -- "just take the next step" produces random walks, not strategic progress
+- The line between a proximate objective and a retreat from ambition is contextual and hard to draw in advance

domains/grand-strategy/the-product-space-constrains-diversification-to-adjacent-products-because-knowledge-and-knowhow-accumulate-only-incrementally-through-related-capabilities.md

@@ -0,0 +1,32 @@
+---
+type: claim
+domain: grand-strategy
+description: "Countries and firms can only diversify into products that use similar capabilities -- the product space is lumpy, and your position in it determines which futures are reachable"
+confidence: proven
+source: "Hidalgo and Hausmann (2007), Hidalgo 'Why Information Grows' (2015), Atlas of Economic Complexity (Harvard)"
+created: 2026-04-21
+secondary_domains: [mechanisms]
+related_claims:
+  - "economic-path-dependence-means-early-technological-choices-compound-irreversibly-through-dominant-designs-and-industrial-structures"
+  - "hill-climbing-gets-trapped-at-local-maxima-because-it-can-only-accept-improvements-and-has-no-way-to-see-beyond-the-nearest-peak"
+---
+
+# The product space constrains diversification to adjacent products because knowledge and knowhow accumulate only incrementally through related capabilities
+
+Hidalgo and Hausmann (2007) mapped the "product space" -- a network where products are connected if the same countries tend to export both. The resulting graph is not random: it has a dense core of sophisticated manufactures (machinery, electronics, chemicals) connected by shared capabilities, and a sparse periphery of raw materials and simple manufactures that share few capabilities with other products. The structure of this network determines which development paths are feasible.
+
+The mechanism is capability accumulation. Making shirts requires textile knowledge, supply chains, and labor skills. Making electronic textiles (smart fabrics) requires textile knowledge PLUS electronics knowledge. A shirt-making country can reach smart fabrics because it already has half the capability set. A petroleum-exporting country cannot, because petroleum extraction shares almost no capabilities with textiles or electronics. The country must build capability bridges -- intermediate products that share capabilities with both the current position and the target.
+
+This is why development traps exist. Countries stuck in the sparse periphery of the product space (raw materials, simple agriculture) face a "missing capability" problem: the products they could diversify into require capabilities they cannot build incrementally from their current base. The jump from commodity exports to sophisticated manufacturing requires simultaneous investment in education, infrastructure, institutions, and industrial policy -- a coordination problem that most countries cannot solve, which is why economic complexity is the best predictor of future growth (better than education, institutions, or governance measures alone).
+
+The implication for firms is identical: a company's current knowledge base constrains its diversification options. Google can move from search to email to maps to autonomous driving because all share a common capability (large-scale data processing and machine learning). Google cannot easily move into pharmaceutical manufacturing because the capability overlap is near zero.
+
+## Evidence
+- Atlas of Economic Complexity (Harvard) -- economic complexity index predicts GDP growth 10-20 years out with R-squared > 0.7, outperforming all other development indicators
+- South Korea development trajectory -- moved from textiles to electronics to semiconductors to displays to smartphones, each step adjacent in product space
+- Finland post-Nokia -- attempted diversification into gaming (Supercell, Rovio) succeeded because mobile gaming shares capabilities with mobile telecommunications
+- Resource curse -- commodity-exporting countries grow slowly precisely because commodities sit in the sparse periphery with few adjacent diversification options
+
+## Challenges
+- The product space is not static -- new products create new connections, and the AI revolution may radically restructure which capabilities are adjacent
+- Some countries (China) have diversified faster than product space adjacency would predict, possibly through deliberate industrial policy that builds multiple capabilities simultaneously

domains/grand-strategy/three-types-of-organizational-inertia-routine-cultural-and-proxy-each-resist-adaptation-through-different-mechanisms-and-require-different-remedies.md

@@ -0,0 +1,34 @@
+---
+type: claim
+domain: grand-strategy
+description: "Organizations fail to adapt through three distinct mechanisms -- process lock-in, identity attachment, and metric substitution -- and misdiagnosing which type you face guarantees the wrong remedy"
+confidence: likely
+source: "Rumelt (2011), Hannan and Freeman (structural inertia, 1984), Christensen (innovator's dilemma, 1997)"
+created: 2026-04-21
+secondary_domains: [mechanisms]
+related_claims:
+  - "strategy-is-a-design-problem-not-a-decision-problem-because-value-comes-from-constructing-a-coherent-configuration-where-parts-interact-and-reinforce-each-other"
+  - "comfortable-stagnation-is-a-self-terminating-attractor-basin-because-the-stability-it-optimizes-for-degrades-capacity-to-respond-to-external-shocks"
+---
+
+# Three types of organizational inertia routine cultural and proxy each resist adaptation through different mechanisms and require different remedies
+
+Organizations resist change, but they resist it for different reasons. Conflating the types produces failed interventions -- like treating a structural problem with a cultural initiative, or a measurement problem with process reengineering.
+
+**Routine inertia** is process lock-in. The organization has optimized its procedures for a previous environment, and the sunk cost in training, tooling, and coordination makes switching costly even when the new approach is clearly superior. IBM's mainframe organization couldn't sell PCs effectively -- not because they didn't understand PCs, but because their sales process, compensation structure, and delivery infrastructure were optimized for million-dollar enterprise contracts. The remedy is structural: create a separate unit with its own processes (Christensen's autonomous organization), or replace the process wholesale rather than incrementally modifying it.
+
+**Cultural inertia** is identity attachment. The organization's self-concept is entangled with its current practices. "We are a hardware company." "We are researchers, not product people." "We don't do that here." Cultural inertia is deeper than routine inertia because people resist changes that threaten their professional identity even when they intellectually agree the change is necessary. Kodak engineers built the first digital camera in 1975 but the company couldn't embrace digital because "we are a film company" was core identity. The remedy is narrative: redefine identity around a more abstract mission that encompasses the new direction. Apple's shift from "computer company" to "company at the intersection of technology and liberal arts" enabled the iPod and iPhone without identity crisis.
+
+**Proxy inertia** is metric substitution. The organization optimizes for metrics that were once correlated with the actual goal but have decoupled. Hospital quality is measured by throughput and readmission rates, so hospitals optimize for those rather than actual patient outcomes. University quality is measured by research output, so universities optimize for publications rather than education. The metric becomes the goal, and anyone who points out the decoupling is fighting both the measurement infrastructure and everyone whose status depends on the current metric. The remedy is measurement redesign -- which is the hardest intervention because it threatens every stakeholder optimized for the current metric.
+
+The critical diagnostic question: when your organization fails to adapt, is it because processes are rigid (routine), because identity is threatened (cultural), or because metrics reward the old behavior (proxy)? Each requires a fundamentally different intervention, and applying the wrong one makes the problem worse.
+
+## Evidence
+- Christensen (1997) -- disk drive industry showing routine inertia: incumbents couldn't adopt new architectures despite awareness
+- Kodak -- cultural inertia: first digital camera 1975, bankruptcy 2012, with thirty-seven years of knowing and not acting
+- Wells Fargo fake accounts scandal -- proxy inertia: cross-selling metrics decoupled from customer value, optimization for the metric produced fraud
+- Hannan and Freeman (1984) -- structural inertia theory showing organizations selected for reliability resist variation
+
+## Challenges
+- The three types interact: routine inertia creates cultural attachment to routines, which generates proxy metrics to justify the status quo. Disentangling is harder in practice than in theory.
+- Some inertia is functional -- organizations need stability to be reliable. The question is degree, not presence.

domains/grand-strategy/value-flows-to-whichever-resources-are-scarce-and-disruption-shifts-which-resources-are-scarce-making-resource-scarcity-analysis-the-core-strategic-framework.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: grand-strategy
+description: "Every disruption is a scarcity shift -- what was scarce becomes abundant and what was abundant becomes scarce, and value migrates accordingly"
+confidence: likely
+source: "m3taversal (Architectural Investing manuscript), Christensen (commoditization/de-commoditization, 2003), Thompson (Aggregation Theory)"
+created: 2026-04-21
+secondary_domains: [internet-finance, entertainment]
+related_claims:
+  - "competitive-advantage-must-be-actively-deepened-through-isolating-mechanisms-because-advantage-that-is-not-reinforced-erodes"
+  - "economic-path-dependence-means-early-technological-choices-compound-irreversibly-through-dominant-designs-and-industrial-structures"
+  - "riding-waves-of-change-requires-anticipating-the-attractor-state-and-positioning-before-incumbents-respond-through-their-predictable-inertia"
+---
+
+# Value flows to whichever resources are scarce and disruption shifts which resources are scarce making resource scarcity analysis the core strategic framework
+
+The fundamental strategic question is not "what is valuable?" but "what is scarce?" Value is always relative to scarcity. When content was scarce (pre-internet), distribution controlled value. When distribution became abundant (internet), content differentiation controlled value. When quality content becomes abundant (AI generation), curation and trust become scarce. Each transition shifts value from the newly-abundant resource to the newly-scarce one.
+
+Christensen formalized this as the commoditization/de-commoditization cycle: when one layer of the value chain becomes modular and commoditized, the adjacent layer typically becomes the new point of scarcity and integration. When PCs commoditized hardware, value shifted to operating systems (Microsoft). When operating systems commoditized, value shifted to search (Google). When search commoditizes, value shifts to whatever is scarce next.
+
+The framework makes disruption predictable, not in timing but in direction. When you see a technology making something abundant, ask: what does this make scarce? Autonomous vehicles make driving abundant -- what becomes scarce is routing optimization, liability frameworks, and attention (you're no longer driving, so you're available). AI makes cognitive labor abundant -- what becomes scarce is judgment about WHAT to apply cognitive labor to, and trust that the output is reliable.
+
+The strategic error is defending the resource that is becoming abundant rather than positioning on the resource that is becoming scarce. Newspapers defended content (becoming abundant via internet) instead of positioning on local trust (becoming scarce as national media scaled). Record labels defended recordings (becoming abundant via digital distribution) instead of positioning on live experience and artist relationships (becoming scarce as recordings commoditized).
+
+## Evidence
+- Christensen conservation of attractive profits (2003) -- when one layer of a value chain commoditizes, adjacencies de-commoditize
+- Thompson Aggregation Theory -- internet commoditized distribution; value shifted to demand aggregation (Google, Facebook, Amazon)
+- Music industry (2000-2020) -- recording revenue crashed as scarcity shifted from recordings to attention; live revenue tripled as live experience became the scarce complement
+- Cloud computing -- commoditized infrastructure; value shifted to data and application intelligence
+
+## Challenges
+- Identifying the newly-scarce resource requires forecasting that's inherently uncertain -- the framework tells you value will shift but not exactly where it will settle
+- Some resources resist commoditization longer than expected due to regulation, network effects, or switching costs

domains/internet-finance/legacy-ICOs-failed-because-team-treasury-control-created-extraction-incentives-that-scaled-with-success.md

@@ -0,0 +1,34 @@
+---
+type: claim
+domain: internet-finance
+description: "2017-era token launches failed not from fraud but from mechanism design: teams controlling treasury had increasing incentive to extract as token value grew, with no governance check"
+confidence: likely
+source: "Catalini and Gans (2018), SEC enforcement actions (2018-2020), empirical ICO performance data"
+created: 2026-04-21
+secondary_domains: [mechanisms]
+related_claims:
+  - "mechanism-design-changes-the-game-itself-to-produce-better-equilibria-rather-than-expecting-players-to-find-optimal-strategies"
+  - "the-vickrey-auction-makes-honesty-the-dominant-strategy-by-paying-winners-the-second-highest-bid-rather-than-their-own"
+---
+
+# Legacy ICOs failed because team treasury control created extraction incentives that scaled with success
+
+The 2017 ICO wave raised approximately $20 billion, with the vast majority of projects failing to deliver. The standard narrative attributes this to fraud and speculation. The mechanism design explanation is more precise: the ICO structure created extraction incentives that were proportional to success, with no governance mechanism to prevent exercise of those incentives.
+
+The structure: team raises funds by selling tokens. Team controls the treasury (unsold tokens + raised capital). Token price rises with market interest. Team's incentive to extract (sell treasury tokens, redirect development funds) grows linearly with token price. The governance check is nothing. Token holders have no binding vote over treasury management. Legal recourse is limited by jurisdictional arbitrage. Reputation effects are weak in pseudonymous markets.
+
+This is not a moral failure but a mechanism design failure. The incentive structure would produce extraction in ANY population of agents, not just bad actors. In fact, the better the project performed, the stronger the extraction incentive became -- success itself created the conditions for abandonment. A team sitting on $100M of tokens has a stronger extraction incentive than a team sitting on $1M, regardless of the team's initial intentions.
+
+The comparison to traditional equity is instructive: corporate governance evolved over centuries to address precisely this problem. Board oversight, fiduciary duty, securities regulation, audit requirements -- all are mechanisms that constrain insiders' ability to extract from a growing enterprise. ICOs discarded all of these mechanisms without replacing them with functional equivalents.
+
+The lesson for future token launch design: any mechanism where value accrues to an entity that controls its own treasury without binding governance will produce extraction at scale. The fix is structural: governance mechanisms that make extraction costlier than continued development. Futarchy-governed treasuries, vesting schedules enforced by smart contracts, and community-controlled spending are all attempts to engineer the extraction incentive away.
+
+## Evidence
+- 2017-2018 ICO performance: over 80% of tokens traded below ICO price within 12 months (Ernst and Young, 2018)
+- SEC enforcement actions (2018-2020) -- dozens of cases documenting team extraction patterns
+- Catalini and Gans (2018) -- formal economic model showing ICO structure creates adverse selection: high-extraction teams have strongest incentive to launch
+- Successful exceptions (Ethereum) -- survived because of unusually strong founder commitment, not because of mechanism design that prevented extraction
+
+## Challenges
+- Some ICOs failed for legitimate reasons (technical failure, market timing, competition) unrelated to extraction incentives
+- Vesting schedules and governance mechanisms can be gamed if the team controls the governance process (circular problem)

domains/mechanisms/hayeks-knowledge-problem-reveals-that-economic-planning-requires-both-local-and-global-information-which-are-never-simultaneously-available-to-decision-makers.md

@@ -0,0 +1,35 @@
+---
+type: claim
+domain: mechanisms
+description: "The knowledge required for economic coordination is dispersed, tacit, and contextual -- no central planner can collect it, and no local agent possesses enough of it"
+confidence: proven
+source: "Hayek 'The Use of Knowledge in Society' (1945), Polanyi 'The Tacit Dimension' (1966)"
+created: 2026-04-21
+secondary_domains: [internet-finance, collective-intelligence, grand-strategy]
+related_claims:
+  - "the-efficient-market-hypothesis-fails-because-its-three-core-assumptions-rational-investors-independence-and-normal-distributions-all-fail-empirically"
+  - "mechanism-design-changes-the-game-itself-to-produce-better-equilibria-rather-than-expecting-players-to-find-optimal-strategies"
+  - "the-vickrey-auction-makes-honesty-the-dominant-strategy-by-paying-winners-the-second-highest-bid-rather-than-their-own"
+---
+
+# Hayek's knowledge problem reveals that economic planning requires both local and global information which are never simultaneously available to decision makers
+
+Hayek's 1945 paper identifies the central problem of economic coordination: the knowledge required to make good allocation decisions is not concentrated anywhere. It exists in fragments -- the factory manager knows their machine's quirks, the local merchant knows their customers' habits, the farmer knows their soil. This knowledge is not just dispersed but often tacit: embodied in skills, intuitions, and practices that cannot be articulated, let alone transmitted to a central planner.
+
+The knowledge problem is not a computing problem. It cannot be solved by faster computers or bigger databases, because the knowledge in question changes moment to moment (the "knowledge of the particular circumstances of time and place") and much of it cannot be formalized at all (Polanyi's tacit dimension). A central planner who somehow collected all current knowledge would find it obsolete before they finished collecting it.
+
+Prices are Hayek's proposed solution: they compress dispersed local knowledge into a single number that coordinates behavior without requiring anyone to understand the whole system. When copper becomes scarce, its price rises, and every user of copper economizes -- without knowing why copper is scarce. The price system achieves coordination that central planning cannot because it transmits the relevant summary statistic without requiring transmission of the underlying knowledge.
+
+But prices are lossy. They compress too much. A price rise tells you something is scarce but not why, not for how long, not whether the scarcity reflects genuine resource constraints or speculative manipulation. The price of healthcare doesn't tell you whether high cost reflects genuine complexity or regulatory capture. This is where Hayek's insight becomes a challenge for markets, not just for planning: prices solve the coordination problem approximately, not perfectly, and the approximation fails precisely where the distinction between signal and noise matters most.
+
+The deep implication: any governance system must either (a) centralize and lose local knowledge, or (b) decentralize and lose global coherence. Markets choose (b). Planning chooses (a). Mechanism design attempts to create structures where agents voluntarily reveal local knowledge in service of global coordination -- futarchy, Vickrey auctions, and prediction markets are all attempts to solve Hayek's problem without accepting either horn of the dilemma.
+
+## Evidence
+- Hayek (1945) "The Use of Knowledge in Society" -- the foundational statement
+- Polanyi (1966) "The Tacit Dimension" -- formalizes why much knowledge cannot be articulated
+- Soviet economic planning failure -- the canonical empirical case; Gosplan's inability to set 24 million prices produced systematic misallocation
+- Walmart supply chain vs. Soviet planning -- Walmart's decentralized supply chain outperforms centralized alternatives by incorporating local store-level demand signals that central warehouses cannot observe
+
+## Challenges
+- Large language models may partially solve the tacit knowledge problem by encoding patterns that humans cannot articulate -- this would narrow (not eliminate) the knowledge gap
+- Platform monopolies (Amazon, Google) aggregate more local knowledge than Hayek thought possible, partially centralizing what he argued was uncentralizable

domains/mechanisms/hill-climbing-gets-trapped-at-local-maxima-because-it-can-only-accept-improvements-and-has-no-way-to-see-beyond-the-nearest-peak.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: mechanisms
+description: "Greedy optimization finds nearby peaks but misses distant higher ones -- the mathematical basis for why incremental improvement fails in complex landscapes"
+confidence: proven
+source: "Stuart Kauffman (NK landscapes, 1993), Herbert Simon (satisficing, 1956), Sewall Wright (fitness landscapes, 1932)"
+created: 2026-04-21
+secondary_domains: [grand-strategy, ai-alignment]
+related_claims:
+  - "simulated-annealing-maps-the-physics-of-cooling-onto-optimization-by-starting-with-high-randomness-and-gradually-reducing-it"
+  - "punctuated-equilibrium-emerges-from-darwinian-microevolution-without-additional-principles-because-extremal-dynamics-on-coupled-fitness-landscapes-self-organize-to-criticality"
+  - "comfortable-stagnation-is-a-self-terminating-attractor-basin-because-the-stability-it-optimizes-for-degrades-capacity-to-respond-to-external-shocks"
+---
+
+# Hill climbing gets trapped at local maxima because it can only accept improvements and has no way to see beyond the nearest peak
+
+Hill climbing is any optimization process that evaluates its current state, considers nearby alternatives, and moves to whichever is better. It is the default behavior of markets, evolution, institutional reform, and individual careers. The trap is mathematical, not behavioral: in any landscape with multiple peaks (a "rugged" fitness landscape), an agent that only accepts improvements will climb the nearest peak and stop -- even if a vastly higher peak exists elsewhere. It cannot reach the higher peak because every path there passes through a valley of worse states.
+
+This matters because most real optimization landscapes are rugged. Kauffman's NK model showed that as the number of interacting components (K) increases, the landscape becomes exponentially more rugged -- more local optima, each further from the global optimum. Stuart Kauffman demonstrated that biological evolution itself gets trapped this way: organisms optimize locally but cannot "see" configurations that would require temporarily becoming less fit.
+
+The implications cascade across domains. Markets hill-climb toward local profit maxima, producing efficient firms that collectively create fragile systems. Institutions reform incrementally, each step locally improving performance while drifting further from globally optimal designs. AI training via gradient descent is literally hill climbing -- neural networks converge to local minima of the loss function, and techniques like learning rate scheduling, random restarts, and ensemble methods exist precisely because the landscape is rugged.
+
+The escape mechanisms are few and costly: random perturbation (simulated annealing), wholesale replacement (punctuated equilibrium), or external forcing (regulation, catastrophe). Each requires tolerating temporary degradation in service of long-term improvement -- which is precisely what greedy optimizers cannot do.
+
+## Evidence
+- Kauffman NK model (1993) -- as K increases, number of local optima grows exponentially, average fitness of optima decreases
+- Wright fitness landscapes (1932) -- original formalization showing evolution explores a surface with peaks and valleys
+- Simon satisficing (1956) -- organisms don't optimize, they accept "good enough" precisely because optimization is computationally intractable
+- Gradient descent in deep learning -- techniques like SGD with momentum, Adam, learning rate warmup all exist to escape local minima
+
+## Challenges
+- Some landscapes are nearly convex (few local optima), making hill climbing sufficient -- but these are the exception in complex systems, not the rule
+- Evolutionary algorithms show that recombination (crossover) can escape local optima without explicit cooling schedules

domains/mechanisms/information-cascades-produce-rational-bubbles-where-every-individual-acts-reasonably-but-the-group-outcome-is-catastrophic.md

@@ -0,0 +1,35 @@
+---
+type: claim
+domain: mechanisms
+description: "When agents rationally weight public information over private signals, the group loses its private information permanently -- producing bubbles without requiring any individual irrationality"
+confidence: likely
+source: "Bikhchandani, Hirshleifer, Welch (1992), Banerjee (1992), Anderson and Holt (1997)"
+created: 2026-04-21
+secondary_domains: [internet-finance, collective-intelligence]
+related_claims:
+  - "the-efficient-market-hypothesis-fails-because-its-three-core-assumptions-rational-investors-independence-and-normal-distributions-all-fail-empirically"
+  - "hayeks-knowledge-problem-reveals-that-economic-planning-requires-both-local-and-global-information-which-are-never-simultaneously-available-to-decision-makers"
+---
+
+# Information cascades produce rational bubbles where every individual acts reasonably but the group outcome is catastrophic
+
+An information cascade occurs when sequential decision-makers rationally choose to follow the actions of predecessors rather than act on their own private information. Each individual is making the correct Bayesian decision given what they observe. But the collective outcome is catastrophic: the group's decision becomes decoupled from the group's total information.
+
+The mechanism is precise. Suppose agents choose sequentially whether to adopt or reject something, each with a private signal that's slightly informative. Agent 1 follows their signal. Agent 2 sees Agent 1's choice and weighs it against their own signal. If both signals agree, Agent 2 follows. Once two agents have chosen the same way, Agent 3's private signal -- even if it disagrees -- is outweighed by the public evidence of two preceding choices. Agent 3 rationally ignores their private information and follows the crowd. Every subsequent agent does the same. The cascade has started.
+
+The critical feature: once the cascade begins, no new private information enters the public record. Agents 3 through 1,000 are all copying the same two early signals. The group of 1,000 has the information of 2. This is why cascades are fragile -- a single piece of credible public counter-evidence can shatter the entire cascade instantly, because it was never grounded in accumulated evidence to begin with.
+
+This explains phenomena that irrationality-based theories cannot. Bubbles form among sophisticated investors. Bank runs happen among depositors who individually have no reason to panic. Technology adoption follows fads that reverse overnight. Fashion cycles. The common thread: sequential observation + private signals + rational Bayesian updating = systematic information loss.
+
+The implication for mechanism design is that any system where agents observe each other's actions before making their own choice is vulnerable to cascades. Prediction markets partially solve this by forcing agents to put money behind their private signals (making signals public through prices), but they're still vulnerable when early liquidity providers set a price that subsequent traders anchor on.
+
+## Evidence
+- Bikhchandani, Hirshleifer, Welch (1992) -- formal model showing cascades arise from rational Bayesian updating with sequential observation
+- Anderson and Holt (1997) -- laboratory experiments confirming cascade formation: subjects rationally ignored private signals after observing 2-3 predecessors
+- Banerjee (1992) -- parallel model showing herding as rational behavior under uncertainty
+- Dot-com bubble (1995-2000) -- sophisticated VCs funded companies they privately doubted because public signals (other VC investments) outweighed private analysis
+- Bank runs (Diamond-Dybvig 1983) -- depositors rationally withdraw when they observe others withdrawing, regardless of bank solvency
+
+## Challenges
+- In practice, cascades often involve genuinely irrational behavior too -- separating rational herding from irrationality is empirically difficult
+- Diverse information sources and simultaneous (rather than sequential) decisions reduce cascade vulnerability

domains/mechanisms/mechanism-design-changes-the-game-itself-to-produce-better-equilibria-rather-than-expecting-players-to-find-optimal-strategies.md

@@ -0,0 +1,37 @@
+---
+type: claim
+domain: mechanisms
+description: "Instead of hoping rational agents find good outcomes, mechanism design engineers the rules so that self-interested behavior produces socially desirable results -- inverse game theory"
+confidence: proven
+source: "Hurwicz (1960, 2007 Nobel), Myerson (1981, 2007 Nobel), Maskin (1999, 2007 Nobel), Roth (matching markets)"
+created: 2026-04-21
+secondary_domains: [internet-finance, collective-intelligence]
+related_claims:
+  - "the-vickrey-auction-makes-honesty-the-dominant-strategy-by-paying-winners-the-second-highest-bid-rather-than-their-own"
+  - "hayeks-knowledge-problem-reveals-that-economic-planning-requires-both-local-and-global-information-which-are-never-simultaneously-available-to-decision-makers"
+  - "advisory-futarchy-avoids-selection-distortion-by-decoupling-prediction-from-execution"
+---
+
+# Mechanism design changes the game itself to produce better equilibria rather than expecting players to find optimal strategies
+
+Game theory takes the rules as given and asks what players will do. Mechanism design inverts this: it takes the desired outcome as given and asks what rules would produce it. This is the fundamental shift from analyzing games to engineering them.
+
+The core problem is incentive compatibility: how do you design rules such that each agent's best strategy -- the one that maximizes their own payoff -- also produces the socially optimal outcome? Hurwicz formalized this as the "revelation principle" (1972): for any mechanism, there exists an equivalent direct mechanism where agents truthfully report their private information. The question becomes: can you design the payoff structure so that truth-telling is optimal?
+
+The answer is sometimes yes and sometimes provably no. The Vickrey-Clarke-Groves (VCG) mechanism achieves truthful revelation for single-dimensional valuations. The Gibbard-Satterthwaite theorem proves that no mechanism can achieve truthful revelation for all preference orderings with three or more alternatives. Impossibility results bound what mechanism design can achieve -- they don't make it useless, they make it honest about its limits.
+
+Where mechanism design succeeds, the results are striking. Roth's redesign of the National Resident Matching Program (medical residency assignments) eliminated the market unraveling that had pushed offers earlier and earlier each year. Spectrum auction design generated efficient allocation of radio frequencies. Kidney exchange pools created welfare-improving trades among incompatible donor-recipient pairs where none could have occurred through bilateral negotiation.
+
+The relevance to decentralized governance is direct: blockchains are programmable rule-sets. For the first time, mechanism design can be implemented as code rather than as institutional rules that depend on enforcement. Futarchy, quadratic voting, retroactive public goods funding -- these are mechanism design proposals that become possible to test at scale because the rules are encoded in smart contracts that execute automatically. The question shifts from "can we trust the institution to follow the rules?" to "are the rules correct?"
+
+## Evidence
+- Hurwicz, Maskin, Myerson (2007 Nobel) -- for laying the foundations of mechanism design theory
+- Roth & Peranson (1999) -- NRMP redesign eliminated market unraveling, stable since 1998
+- FCC spectrum auctions (1994-present) -- raised >$200B through mechanism-designed combinatorial auctions
+- Kidney exchange (Roth, Soenmez, Uenver 2004) -- created welfare from previously impossible trades
+- Gibbard-Satterthwaite theorem (1973/1975) -- proves universal truthful voting is impossible with 3+ alternatives
+
+## Challenges
+- Most mechanism design assumes risk-neutral, expected-utility-maximizing agents -- real agents exhibit prospect theory biases that undermine theoretical guarantees
+- Computational complexity: optimal mechanisms for combinatorial settings are often NP-hard to compute
+- Collusion resistance remains an open problem -- most mechanisms break when agents can coordinate side-payments

domains/mechanisms/punctuated-equilibrium-emerges-from-darwinian-microevolution-without-additional-principles-because-extremal-dynamics-on-coupled-fitness-landscapes-self-organize-to-criticality.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: mechanisms
+description: "Long stasis interrupted by rapid change is not a separate evolutionary mechanism -- it's the emergent behavior of coupled adaptive systems that push each other to the edge of instability"
+confidence: experimental
+source: "Bak and Sneppen (1993), Gould and Eldredge (1972), Kauffman 'Origins of Order' (1993), Bak 'How Nature Works' (1996)"
+created: 2026-04-21
+secondary_domains: [grand-strategy, collective-intelligence]
+related_claims:
+  - "hill-climbing-gets-trapped-at-local-maxima-because-it-can-only-accept-improvements-and-has-no-way-to-see-beyond-the-nearest-peak"
+  - "simulated-annealing-maps-the-physics-of-cooling-onto-optimization-by-starting-with-high-randomness-and-gradually-reducing-it"
+  - "comfortable-stagnation-is-a-self-terminating-attractor-basin-because-the-stability-it-optimizes-for-degrades-capacity-to-respond-to-external-shocks"
+---
+
+# Punctuated equilibrium emerges from darwinian microevolution without additional principles because extremal dynamics on coupled fitness landscapes self-organize to criticality
+
+The fossil record shows long periods of morphological stasis punctuated by brief bursts of rapid change. Gould and Eldredge (1972) proposed punctuated equilibrium as a macroevolutionary pattern, but the mechanism remained contested. Bak and Sneppen (1993) demonstrated that this pattern emerges naturally from coupled fitness landscapes without any additional principles beyond standard Darwinian selection.
+
+The mechanism: consider a network of species, each sitting on its own fitness landscape, where species' landscapes are coupled (your fitness depends on neighboring species). At each time step, the least-fit species mutates randomly (finds a new position on its landscape), and this mutation changes the landscapes of its neighbors (because their fitness depends on the species that just changed). The system self-organizes to criticality: it reaches a state where most species are well-adapted but a few are marginal. When a marginal species mutates, it can trigger an avalanche of cascading mutations through the network.
+
+The avalanche size distribution follows a power law -- most avalanches are small (one species adapts, neighbors are unaffected) but occasionally an avalanche spans the entire network (mass extinction followed by rapid radiation). This is self-organized criticality: the system drives itself to the boundary between stability and chaos, where perturbations propagate at all scales.
+
+The transfer to human systems is structural, not metaphorical. Markets, institutions, and civilizations are coupled adaptive systems. Each agent optimizes locally, their optimizations change neighbors' landscapes, and the system self-organizes to a critical state. The result: long periods of apparent stability (stasis) punctuated by rapid cascading change (revolutions, market crashes, paradigm shifts) whose size distribution follows a power law. The 2008 financial crisis was an avalanche in a system at criticality -- not an exogenous shock, but the inevitable consequence of coupled systems self-organizing to the edge of instability.
+
+## Evidence
+- Bak-Sneppen model (1993) -- minimal model reproducing punctuated equilibrium from extremal dynamics alone, power-law avalanche distribution with exponent approximately 1.07
+- Fossil record -- stasis durations and speciation rates consistent with power-law avalanche distributions (Raup 1986, mass extinction statistics)
+- Bak sandpile model (1987) -- first demonstration of self-organized criticality; sandpile avalanches follow same power-law distribution as extinction events
+- Financial market crashes -- crash size distribution follows power law (Mandelbrot 1963, Gabaix et al. 2003), consistent with self-organized criticality in coupled trading strategies
+
+## Challenges
+- Self-organized criticality may not apply to all coupled systems -- some systems have characteristic scales (preferred sizes of perturbation) rather than scale-free power laws
+- The Bak-Sneppen model is extremely abstract -- mapping it to specific biological mechanisms remains debated

domains/mechanisms/simulated-annealing-maps-the-physics-of-cooling-onto-optimization-by-starting-with-high-randomness-and-gradually-reducing-it.md

@@ -0,0 +1,32 @@
+---
+type: claim
+domain: mechanisms
+description: "The Metropolis algorithm shows that accepting worse solutions with decreasing probability provably converges to the global optimum -- the mathematical case for tolerating short-term loss"
+confidence: proven
+source: "Kirkpatrick, Gelatt, Vecchi (1983), Metropolis algorithm (1953), Boltzmann distribution"
+created: 2026-04-21
+secondary_domains: [grand-strategy]
+related_claims:
+  - "hill-climbing-gets-trapped-at-local-maxima-because-it-can-only-accept-improvements-and-has-no-way-to-see-beyond-the-nearest-peak"
+  - "comfortable-stagnation-is-a-self-terminating-attractor-basin-because-the-stability-it-optimizes-for-degrades-capacity-to-respond-to-external-shocks"
+---
+
+# Simulated annealing maps the physics of cooling onto optimization by starting with high randomness and gradually reducing it
+
+A metal cools slowly from high temperature. At high temperature, atoms jump freely between configurations, exploring widely. As temperature drops, atoms settle into low-energy configurations. If cooling is slow enough, the metal reaches its global energy minimum -- a perfect crystal. If cooled too fast, atoms freeze in a disordered, suboptimal state (glass).
+
+Kirkpatrick, Gelatt, and Vecchi (1983) proved this physical process is isomorphic to combinatorial optimization. Replace "energy" with "cost function" and "temperature" with "willingness to accept worse solutions." At high temperature, the algorithm accepts moves to worse states frequently, enabling broad exploration. As temperature decreases, acceptance of worse states drops exponentially, and the algorithm converges toward the global optimum. The cooling schedule is everything: too fast and you freeze in a local minimum, too slow and you waste computation exploring already-mapped territory.
+
+The insight that transfers beyond computation: any system that wants to find globally good solutions must tolerate periods of locally worse performance. Markets that never allow failure (bailouts, zombie firms) are cooling too fast -- they freeze in suboptimal configurations. Societies that never tolerate disorder (authoritarian stability) are doing the same. The mathematical proof says you MUST pass through worse states to reach better ones when the landscape is rugged.
+
+The cooling schedule implies a lifecycle. Young systems should explore widely (high temperature). Mature systems should exploit locally (low temperature). The transition between exploration and exploitation is itself the critical design choice -- and there is no universal optimal schedule. It depends on the landscape's ruggedness, which you generally don't know in advance.
+
+## Evidence
+- Metropolis algorithm (1953) -- the acceptance probability function exp(-deltaE/kT) provably samples the Boltzmann distribution
+- Kirkpatrick et al. (1983) -- demonstrated convergence on VLSI circuit layout, traveling salesman, graph partitioning
+- Convergence proof -- Hajek (1988) proved simulated annealing converges to global optimum if cooling is logarithmic: T(t) >= d/ln(t)
+- Physical metallurgy -- the glass transition is literally the consequence of insufficient annealing
+
+## Challenges
+- Logarithmic cooling is impractically slow for most real problems -- practitioners use heuristic schedules that sacrifice convergence guarantees for speed
+- Modern methods (genetic algorithms, reinforcement learning) often outperform SA on specific problem classes

domains/mechanisms/the-efficient-market-hypothesis-fails-because-its-three-core-assumptions-rational-investors-independence-and-normal-distributions-all-fail-empirically.md

@@ -0,0 +1,36 @@
+---
+type: claim
+domain: mechanisms
+description: "EMH fails not at the margin but at the foundation -- real markets exhibit herding, fat tails, and systematic irrationality that invalidate the mathematical framework"
+confidence: likely
+source: "Mandelbrot (fat tails, 1963), Kahneman/Tversky (prospect theory, 1979), Shiller (irrational exuberance, 2000), Soros (reflexivity, 1987)"
+created: 2026-04-21
+secondary_domains: [internet-finance, grand-strategy]
+related_claims:
+  - "information-cascades-produce-rational-bubbles-where-every-individual-acts-reasonably-but-the-group-outcome-is-catastrophic"
+  - "hayeks-knowledge-problem-reveals-that-economic-planning-requires-both-local-and-global-information-which-are-never-simultaneously-available-to-decision-makers"
+  - "the-shape-of-the-prior-distribution-determines-the-prediction-rule-and-getting-the-prior-wrong-produces-worse-predictions-than-having-less-data-with-the-right-prior"
+---
+
+# The efficient market hypothesis fails because its three core assumptions rational investors independence and normal distributions all fail empirically
+
+The efficient market hypothesis (Fama, 1970) claims that asset prices fully reflect all available information. The mathematical framework requires three assumptions: (1) investors are rational expected-utility maximizers, (2) investors' errors are independent and cancel out, (3) returns follow normal (Gaussian) distributions. All three fail empirically.
+
+**Rationality fails.** Kahneman and Tversky's prospect theory (1979) demonstrated that humans systematically overweight losses relative to gains, anchor on irrelevant reference points, and exhibit probability distortion. These are not random errors that cancel -- they are systematic biases that create predictable mispricings. The disposition effect (selling winners too early, holding losers too long) is observed across every market and every culture studied.
+
+**Independence fails.** Real investors copy each other. Information cascades (Bikhchandani, Hirshleifer, Welch 1992) show that rational agents who observe predecessors' actions instead of using private information produce herding behavior. Soros's reflexivity (1987) goes further: market prices don't just reflect reality, they change it. A rising stock price improves a company's ability to raise capital, hire talent, and acquire competitors -- making the price rise "correct" in a self-fulfilling way until the reflexive loop breaks.
+
+**Normal distributions fail.** Mandelbrot (1963) showed that cotton price changes follow Levy stable distributions with infinite variance, not Gaussians. The practical consequence: events that Gaussian models predict should occur once in the lifetime of the universe (like the 2008 financial crisis or the 1987 crash) actually occur every decade. Financial models built on normal distributions systematically underestimate tail risk by orders of magnitude. This is not a calibration error -- it is a category error in the mathematical foundation.
+
+The EMH is not "approximately right." Its failure modes compound: irrational agents herd (combining failures 1 and 2), herding creates fat tails (combining failures 2 and 3), and fat-tail events trigger further irrationality (combining failures 3 and 1). The system is reflexive, correlated, and fat-tailed -- the three things the EMH requires it not to be.
+
+## Evidence
+- Mandelbrot (1963) -- cotton prices exhibit scaling behavior inconsistent with Gaussian assumptions; extended in Mandelbrot & Hudson "The Misbehavior of Markets" (2004)
+- Kahneman & Tversky (1979) -- prospect theory replaces expected utility; Nobel Prize 2002
+- Shiller (2000) -- excess volatility puzzle: stock prices are 5-13x more volatile than dividends justify
+- 2008 financial crisis -- a "25-sigma event" under Gaussian assumptions, i.e., probability approximately zero
+- Long-Term Capital Management (1998) -- Nobel laureates' fund collapsed because their models assumed independent, normally distributed returns
+
+## Challenges
+- The EMH remains useful as an approximation for liquid, well-studied markets over medium timeframes -- the failures are most extreme at short timescales and during regime changes
+- Grossman-Stiglitz paradox (1980): if markets are perfectly efficient, there's no incentive to gather information, which means markets can't become efficient -- EMH is self-undermining

domains/mechanisms/the-shape-of-the-prior-distribution-determines-the-prediction-rule-and-getting-the-prior-wrong-produces-worse-predictions-than-having-less-data-with-the-right-prior.md

@@ -0,0 +1,35 @@
+---
+type: claim
+domain: mechanisms
+description: "A Gaussian prior produces mean regression, a power-law prior produces multiplicative extrapolation -- using the wrong prior on the right data degrades prediction systematically"
+confidence: likely
+source: "Jaynes (2003), Gelman et al. (Bayesian Data Analysis, 2013), Taleb (fat tails, 2007), Mandelbrot (1963)"
+created: 2026-04-21
+secondary_domains: [internet-finance, grand-strategy]
+related_claims:
+  - "the-efficient-market-hypothesis-fails-because-its-three-core-assumptions-rational-investors-independence-and-normal-distributions-all-fail-empirically"
+  - "information-cascades-produce-rational-bubbles-where-every-individual-acts-reasonably-but-the-group-outcome-is-catastrophic"
+---
+
+# The shape of the prior distribution determines the prediction rule and getting the prior wrong produces worse predictions than having less data with the right prior
+
+Bayesian inference combines prior beliefs with observed data to produce posterior predictions. The standard teaching emphasizes that with enough data, the prior washes out. This is true for well-behaved (thin-tailed) distributions. It is catastrophically false for fat-tailed distributions, which characterize most quantities that matter: wealth, city sizes, earthquake magnitudes, market crashes, pandemic severity, war casualties.
+
+The mechanism: when a variable follows a Gaussian (thin-tailed) distribution, extreme observations are vanishingly unlikely, so the optimal prediction for any individual regresses toward the mean. A basketball player who scores 50 points will probably score closer to 25 next game. But when a variable follows a power-law (fat-tailed) distribution, extreme observations carry information that the generating process can produce extreme values. A city that has 10 million people is more likely to grow to 20 million than a city of 100,000 is -- because the process that produced a 10-million city is different from the process that produced a 100,000 city.
+
+The prediction rule changes completely. Under a Gaussian prior, you predict regression to the mean. Under a power-law prior, you predict multiplicative extrapolation from the observed value. Using a Gaussian prior on power-law data produces predictions that are systematically wrong in the direction of underestimating extremes -- which is exactly where the stakes are highest.
+
+This is why financial risk models failed in 2008: they used Gaussian priors on fat-tailed data. Why pandemic models underestimated COVID: they used thin-tailed priors for a process with super-spreader dynamics. Why VC returns are misunderstood: a single outlier return in a power-law portfolio contains more expected value than all other investments combined, but Gaussian-trained intuitions treat it as an anomaly to regress away from.
+
+The implication for any knowledge system: before evaluating evidence, you must ask what generating process produced it. The prior is not a subjective belief to be minimized -- it is a structural claim about reality that determines the correct inference rule. Getting the prior wrong is worse than having less data, because more data with the wrong prior converges to a confident wrong answer.
+
+## Evidence
+- Mandelbrot (1963) -- cotton prices follow Levy stable distributions, not Gaussians
+- Taleb (2007) -- "The Black Swan" documents systematic Gaussian-prior errors in finance, risk, and prediction
+- Clauset, Shalizi, Newman (2009) -- rigorous statistical methods for distinguishing power-law from other heavy-tailed distributions in empirical data
+- COVID-19 super-spreader events -- 80/20 rule (20% of infected produced 80% of transmission) follows power-law dispersion, not Gaussian
+- VC returns (Horsley Bridge data) -- fund returns follow power law; top company in portfolio generates more return than rest combined
+
+## Challenges
+- Distinguishing power-law from log-normal empirically is extremely difficult (Clauset et al. 2009) -- many claimed power laws don't survive rigorous testing
+- Even with the correct prior, uncertainty about the tail exponent produces wide posterior intervals -- knowing the shape is fat-tailed doesn't tell you exactly how fat

domains/mechanisms/the-vickrey-auction-makes-honesty-the-dominant-strategy-by-paying-winners-the-second-highest-bid-rather-than-their-own.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: mechanisms
+description: "Second-price sealed-bid auctions make truthful bidding optimal because your bid determines WHETHER you win but not WHAT you pay -- decoupling the strategic incentive to shade"
+confidence: proven
+source: "Vickrey (1961, 1996 Nobel), Clarke (1971), Groves (1973)"
+created: 2026-04-21
+secondary_domains: [internet-finance]
+related_claims:
+  - "mechanism-design-changes-the-game-itself-to-produce-better-equilibria-rather-than-expecting-players-to-find-optimal-strategies"
+  - "hayeks-knowledge-problem-reveals-that-economic-planning-requires-both-local-and-global-information-which-are-never-simultaneously-available-to-decision-makers"
+---
+
+# The Vickrey auction makes honesty the dominant strategy by paying winners the second-highest bid rather than their own
+
+In a standard first-price auction, you face a dilemma: bid your true value and win with zero surplus, or bid below your value and risk losing to someone who values it less. Every bidder shades their bid downward, and the optimal shade depends on beliefs about competitors -- which you don't have. The result: systematic undervaluation and allocative inefficiency.
+
+Vickrey's insight (1961) was to decouple the determination of the winner from the determination of the price. In a second-price sealed-bid auction, the highest bidder wins but pays the second-highest bid. This seemingly minor rule change transforms the strategic landscape completely: your bid determines WHETHER you win but not WHAT you pay. Bidding your true value is now dominant strategy -- not because you're honest, but because any other bid is strictly worse. Bid higher than your value and you risk winning at a loss. Bid lower and you risk losing an auction you would have won profitably. Truth-telling is not just optimal; it is the unique weakly dominant strategy.
+
+The deep insight is separability: when you can separate the "who wins" question from the "what do they pay" question, you can align individual incentives with social efficiency. The VCG (Vickrey-Clarke-Groves) generalization extends this to multi-good settings: each agent pays their externality on other agents (the social cost of their participation), making truthful reporting optimal in combinatorial allocation problems.
+
+This principle -- pay the externality, not the bid -- is the template for incentive-compatible mechanism design. Google's ad auctions use generalized second-price. Spectrum auctions use VCG variants. The principle explains why prediction markets work (you pay the market price, not your private valuation) and why simple polls don't (your report directly determines the outcome at zero cost, so strategic misreporting is free).
+
+## Evidence
+- Vickrey (1961) -- original proof of dominant strategy incentive compatibility
+- Google AdWords (2002-present) -- generalized second-price auction handles billions of allocations daily
+- eBay proxy bidding -- functionally equivalent to Vickrey auction, dominant strategy is to bid true max
+- Laboratory experiments (Kagel 1995) -- subjects converge to truthful bidding in second-price auctions within a few rounds, while first-price auction bidding remains strategically distorted
+
+## Challenges
+- Vickrey auctions are vulnerable to shill bidding (seller placing fake second bids to raise the price) -- this is why eBay has reputation systems
+- Revenue: first-price auctions often raise more revenue than Vickrey auctions when bidders are risk-averse (revenue equivalence only holds under risk-neutrality)
+- Collusion: if the top two bidders collude, the winner pays an artificially low second price

foundations/collective-intelligence/biological-organization-nests-markov-blankets-hierarchically-from-cells-to-organs-to-organisms-enabling-local-autonomy-with-global-coherence.md

@@ -0,0 +1,34 @@
+---
+type: claim
+domain: collective-intelligence
+description: "Each level of biological organization maintains its own boundary (Markov blanket) while participating in higher-level dynamics -- local autonomy scales through nested boundaries, not central control"
+confidence: likely
+source: "Friston (free energy principle, 2010), Kirchhoff et al. (2018), Levin (2019, bioelectricity)"
+created: 2026-04-21
+secondary_domains: [critical-systems, ai-alignment]
+related_claims:
+  - "nested-markov-blankets-enable-hierarchical-organization-where-each-level-minimizes-prediction-error-while-participating-in-higher-level-dynamics"
+  - "punctuated-equilibrium-emerges-from-darwinian-microevolution-without-additional-principles-because-extremal-dynamics-on-coupled-fitness-landscapes-self-organize-to-criticality"
+---
+
+# Biological organization nests Markov blankets hierarchically from cells to organs to organisms enabling local autonomy with global coherence
+
+A Markov blanket is a statistical boundary: the set of variables that separates a system from its environment such that the system's internal states are conditionally independent of external states given the blanket. In biology, this formalism maps onto physical boundaries at every scale: cell membranes, organ capsules, skin, social group boundaries.
+
+The key insight from Friston's free energy principle (2010) is that these boundaries nest hierarchically, and each level actively maintains its own boundary through a process of minimizing prediction error (variational free energy). A cell maintains its membrane, an organ maintains its boundary, an organism maintains its skin -- and each level's boundary-maintenance creates the conditions for the next level to exist.
+
+This produces a specific architecture: local autonomy at every level, coordinated through the boundary interfaces. A liver cell doesn't take instructions from the brain about how to metabolize glucose -- it follows local chemical gradients. But its activity is constrained by the organ-level boundary (the liver's blood supply, hormonal signals) which is itself constrained by the organism-level boundary (whole-body metabolic state). No central controller. No global plan. Coherent behavior emerges from nested boundary maintenance.
+
+Levin's work on bioelectricity (2019) shows this operating in development: groups of cells share bioelectric patterns that encode morphological targets. A planarian fragment regenerates the correct body plan not because each cell has a blueprint but because the bioelectric boundary state encodes the target anatomy and cells follow local gradients toward it. This is collective intelligence without central control -- exactly the architecture that scales from single cells to organisms with trillions of cells.
+
+The transfer to artificial systems is the design challenge: can you build agent collectives where each agent maintains its own boundary (scope, identity, evaluation criteria) while participating in higher-level coordination through boundary interfaces (shared knowledge base, governance mechanisms, communication protocols)?
+
+## Evidence
+- Friston (2010) -- free energy principle: all self-organizing systems maintain Markov blankets by minimizing variational free energy
+- Kirchhoff et al. (2018) -- "The Markov blankets of life" -- formal proof that Markov blankets nest hierarchically in biological systems
+- Levin (2019) -- bioelectric patterns as morphological targets: planarian regeneration, Xenopus eye induction at non-standard locations
+- Immune system -- distributed defense with no central controller; lymphocytes make local decisions based on local antigen signals, coordinated through cytokine cascades
+
+## Challenges
+- The Markov blanket formalism may be too abstract to generate specific predictions -- "everything has a Markov blanket" risks being unfalsifiable
+- Hierarchical nesting assumes clean level separation, but many biological systems have cross-level interactions that violate the nesting assumption (epigenetics, horizontal gene transfer)

foundations/cultural-dynamics/berger-and-luckmanns-plausibility-structures-reveal-that-master-narrative-maintenance-requires-institutional-power-not-just-cultural-appeal.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: cultural-dynamics
+description: "Narratives don't persist because people believe them -- people believe them because the entire institutional environment is structured to make alternatives implausible"
+confidence: likely
+source: "Berger and Luckmann 'The Social Construction of Reality' (1966), Bourdieu (cultural reproduction, 1979)"
+created: 2026-04-21
+secondary_domains: [grand-strategy, collective-intelligence]
+related_claims:
+  - "world-narratives-follow-a-lifecycle-of-formation-dominance-contradiction-accumulation-crisis-and-transformation"
+  - "the-current-narrative-breakdown-is-unprecedented-in-speed-because-the-internet-makes-contradictions-visible-to-billions-instantly"
+  - "effective-world-narratives-must-provide-both-meaning-and-coordination-mechanisms-simultaneously"
+---
+
+# Berger and Luckmann's plausibility structures reveal that master narrative maintenance requires institutional power not just cultural appeal
+
+A belief seems obviously true when every institution in your environment confirms it. This is Berger and Luckmann's core insight (1966): what people experience as "reality" is socially constructed through institutions that make some beliefs plausible and others unthinkable. The mechanism is not propaganda (deliberate deception) but plausibility structure -- the entire web of social interactions, material arrangements, and institutional practices that make a narrative feel self-evident.
+
+A medieval peasant doesn't believe in divine-right monarchy because they've been convinced by arguments. They believe it because every institution they encounter -- the church, the manor, the guild, the law -- operates on that assumption. The alternative (popular sovereignty) is not just unlikely; it is literally unthinkable within their plausibility structure. The concept doesn't exist in their vocabulary, their social interactions don't model it, and their material reality doesn't suggest it.
+
+This reveals why narrative change is hard: it requires changing institutions, not just minds. You cannot talk people out of beliefs that are institutionally sustained, because the institution continues to generate the plausibility regardless of what any individual thinks. Conversely, beliefs collapse rapidly when their institutional support is removed -- the Soviet Union's state ideology evaporated within years once the institutions enforcing it stopped functioning.
+
+The implication for the current narrative crisis: the internet didn't change what people believe through persuasion. It undermined the plausibility structures that sustained dominant narratives by creating alternative institutions (social media communities, cryptocurrency networks, independent media) that operate on different assumptions. When you can live your social life inside a community that takes a different narrative for granted, the old narrative's plausibility collapses -- not because it was disproven but because its institutional support no longer monopolizes your experience.
+
+## Evidence
+- Berger and Luckmann (1966) -- "The Social Construction of Reality" formalizes how institutions create and maintain shared reality through habitualization, institutionalization, and legitimation
+- Bourdieu (1979) -- cultural reproduction: educational institutions don't just transmit knowledge, they reproduce the social order by making existing hierarchies appear natural and meritocratic
+- Soviet collapse -- 70 years of institutional narrative maintenance evaporated within 3-5 years once institutions stopped enforcing it
+- Flat Earth communities -- demonstrate plausibility structure in miniature: sustained belief depends not on evidence but on community membership and institutional practices (conferences, YouTube channels, social groups)
+
+## Challenges
+- The theory can veer into relativism -- if all reality is socially constructed, how do we distinguish well-grounded narratives from delusions? Berger and Luckmann don't resolve this tension
+- Not all institutional persistence is bad -- legal systems, scientific norms, and democratic procedures are also plausibility structures, and their stability is often beneficial

foundations/cultural-dynamics/effective-world-narratives-must-provide-both-meaning-and-coordination-mechanisms-simultaneously.md

@@ -0,0 +1,35 @@
+---
+type: claim
+domain: cultural-dynamics
+description: "A narrative that provides meaning but not coordination produces philosophy; one that provides coordination but not meaning produces bureaucracy -- only narratives doing both persist at civilizational scale"
+confidence: experimental
+source: "m3taversal (Architectural Investing manuscript), Anderson 'Imagined Communities' (1983), Harari 'Sapiens' (2014)"
+created: 2026-04-21
+secondary_domains: [grand-strategy, collective-intelligence]
+related_claims:
+  - "world-narratives-follow-a-lifecycle-of-formation-dominance-contradiction-accumulation-crisis-and-transformation"
+  - "berger-and-luckmanns-plausibility-structures-reveal-that-master-narrative-maintenance-requires-institutional-power-not-just-cultural-appeal"
+  - "the-current-narrative-breakdown-is-unprecedented-in-speed-because-the-internet-makes-contradictions-visible-to-billions-instantly"
+---
+
+# Effective world narratives must provide both meaning and coordination mechanisms simultaneously
+
+Harari (2014) observes that large-scale human cooperation depends on shared fictions -- religion, nation, money, human rights. But not all shared fictions persist. The ones that endure at civilizational scale provide two things simultaneously: meaning (why should I care?) and coordination (how should I act?).
+
+Christianity provided both: meaning through salvation narrative (why you exist and what happens after death) and coordination through institutional structure (parish, diocese, papacy, canon law, calendar). Nationalism provides both: meaning through identity narrative (you belong to something larger than yourself) and coordination through institutional structure (citizenship, taxation, military service, legal system). Money provides both: meaning through value narrative (your labor is worth something exchangeable) and coordination through price mechanism (how to allocate resources across millions of strangers).
+
+Narratives that provide meaning without coordination become philosophies -- they explain the world but don't organize collective action. Stoicism, existentialism, and most academic theory live here. They persist as intellectual traditions but don't scale to civilizational coordination.
+
+Narratives that provide coordination without meaning become bureaucracies -- they organize collective action but fail to motivate participation beyond compliance. Soviet communism degraded from a meaning-providing narrative (worker liberation, historical destiny) to a coordination-only bureaucracy (quota systems, party hierarchy) -- and collapsed when compliance was no longer enforced. The European Union struggles with the same problem: effective coordination mechanism, weak meaning narrative, persistent legitimacy deficit.
+
+The current interregnum is a period where old narratives (liberal democracy, market capitalism) are losing their meaning function (growing inequality, institutional distrust, climate anxiety) while retaining their coordination function (legal systems, financial markets still operate). The replacement narrative must provide BOTH -- which is why "just fix the institutions" (coordination-only) and "just change the culture" (meaning-only) are both insufficient responses to the current crisis.
+
+## Evidence
+- Anderson (1983) -- "Imagined Communities": nations are narratives that coordinate through census, map, and museum while providing identity meaning
+- Soviet Union -- meaning drained from communist narrative by 1970s; coordination continued through coercion alone until 1991
+- European Union -- technically successful coordination (single market, Schengen, euro) with persistent meaning deficit (low identification, democratic legitimacy crisis)
+- Cryptocurrency communities -- strongest communities (Bitcoin, Ethereum) provide both meaning narrative (monetary sovereignty, decentralized future) and coordination mechanisms (consensus protocols, governance processes)
+
+## Challenges
+- The meaning/coordination distinction may be a continuum rather than a binary -- most real narratives provide both in varying degrees
+- Some coordination systems persist without meaning for very long periods (Chinese imperial bureaucracy, modern tax systems) -- the requirement for meaning may be weaker than claimed

foundations/cultural-dynamics/the-current-narrative-breakdown-is-unprecedented-in-speed-because-the-internet-makes-contradictions-visible-to-billions-instantly.md

@@ -0,0 +1,33 @@
+---
+type: claim
+domain: cultural-dynamics
+description: "Previous narrative breakdowns (Reformation, Enlightenment) took generations because contradictions spread slowly -- the internet compresses this to years, faster than institutions can adapt"
+confidence: experimental
+source: "m3taversal (Architectural Investing manuscript), Schmachtenberger (War on Sensemaking, 2019), Gurri 'The Revolt of the Public' (2014)"
+created: 2026-04-21
+secondary_domains: [grand-strategy, collective-intelligence]
+related_claims:
+  - "world-narratives-follow-a-lifecycle-of-formation-dominance-contradiction-accumulation-crisis-and-transformation"
+  - "effective-world-narratives-must-provide-both-meaning-and-coordination-mechanisms-simultaneously"
+  - "berger-and-luckmanns-plausibility-structures-reveal-that-master-narrative-maintenance-requires-institutional-power-not-just-cultural-appeal"
+---
+
+# The current narrative breakdown is unprecedented in speed because the internet makes contradictions visible to billions instantly
+
+Every dominant world narrative accumulates contradictions -- gaps between what the narrative promises and what people experience. The Reformation exposed contradictions in Catholic authority. The Enlightenment exposed contradictions in divine-right monarchy. In both cases, the contradictions accumulated over decades and spread through pamphlets, books, and interpersonal networks. Institutional responses had time to adapt, co-opt, or suppress.
+
+The internet collapses this timeline. A contradiction between official narrative and lived experience -- government lies, institutional failures, promised prosperity not materializing -- becomes visible to billions within hours. The 2008 financial crisis narrative ("markets are efficient, experts have it under control") collapsed globally within weeks as contradictions between official reassurances and actual bank failures played out in real-time on social media. This is categorically different from previous narrative breakdowns.
+
+The speed mismatch is the critical danger: narrative breakdown happens at internet speed, but new narrative formation happens at human-institutional speed. Building shared meaning requires trust, which requires repeated interactions, which takes time. The result is a growing gap between narrative destruction (fast) and narrative construction (slow), producing a period of narrative vacuum where no shared story coordinates collective action. Gurri (2014) documents this as "the revolt of the public" -- the internet empowered publics to tear down institutional narratives without producing replacement narratives.
+
+Schmachtenberger (2019) frames this as the "war on sensemaking" -- when the information ecology is corrupted (by algorithmic amplification of engagement over truth, by state propaganda, by commercial disinformation), the collective capacity to form shared narratives degrades. The problem is not that people disagree about values -- that's normal. The problem is that people cannot agree on facts, which makes value disagreement irresolvable.
+
+## Evidence
+- Arab Spring (2011) -- decades of authoritarian narrative collapsed in weeks via social media; no stable replacement narrative emerged in most countries
+- 2008 financial crisis -- "efficient markets" narrative collapsed globally within months; replacement narrative still contested 15+ years later
+- COVID-19 pandemic -- scientific consensus and public trust diverged in real-time as contradictory information spread faster than institutional correction
+- Gurri (2014) -- documents pattern across US, Middle East, Europe: internet-enabled publics can negate institutional authority but cannot construct alternatives
+
+## Challenges
+- Speed of breakdown does not necessarily predict severity of consequences -- some rapid narrative shifts (civil rights movement) produced positive outcomes
+- The internet also accelerates narrative formation in some contexts (crypto community, open source movement) -- the speed asymmetry between breakdown and construction may be domain-specific

foundations/cultural-dynamics/world-narratives-follow-a-lifecycle-of-formation-dominance-contradiction-accumulation-crisis-and-transformation.md

@@ -0,0 +1,38 @@
+---
+type: claim
+domain: cultural-dynamics
+description: "Master narratives are born in crisis, gain dominance through institutional embedding, accumulate contradictions through success, and collapse when contradictions exceed institutional capacity to suppress them"
+confidence: likely
+source: "Kuhn 'Structure of Scientific Revolutions' (1962), Berger and Luckmann (1966), m3taversal (Architectural Investing manuscript)"
+created: 2026-04-21
+secondary_domains: [grand-strategy]
+related_claims:
+  - "the-current-narrative-breakdown-is-unprecedented-in-speed-because-the-internet-makes-contradictions-visible-to-billions-instantly"
+  - "effective-world-narratives-must-provide-both-meaning-and-coordination-mechanisms-simultaneously"
+  - "berger-and-luckmanns-plausibility-structures-reveal-that-master-narrative-maintenance-requires-institutional-power-not-just-cultural-appeal"
+  - "punctuated-equilibrium-emerges-from-darwinian-microevolution-without-additional-principles-because-extremal-dynamics-on-coupled-fitness-landscapes-self-organize-to-criticality"
+---
+
+# World narratives follow a lifecycle of formation dominance contradiction accumulation crisis and transformation
+
+Every dominant world narrative -- religious, political, economic -- follows the same lifecycle. The pattern is structural, not accidental.
+
+**Formation:** A new narrative emerges during a crisis in the previous one. Christianity formed during the crisis of Roman civic religion. Liberalism formed during the crisis of divine-right monarchy. Neoliberalism formed during the crisis of Keynesian stagflation. The new narrative succeeds because it explains the failure of the old one and provides a framework for action that works in the new conditions.
+
+**Dominance:** The narrative becomes institutionally embedded. Schools teach it. Laws encode it. Professional norms enforce it. Economic structures reward behavior consistent with it and punish deviation. Berger and Luckmann's "plausibility structures" describe this phase: the narrative appears self-evident because the entire institutional environment is designed to confirm it.
+
+**Contradiction accumulation:** Every narrative simplifies reality. The simplifications that enabled coordination during formation become distortions during dominance. The narrative says markets are efficient, but crashes keep happening. The narrative says meritocracy rewards talent, but inherited advantage keeps compounding. The narrative says democratic institutions represent the public, but policy keeps serving elites. These contradictions accumulate as anomalies -- Kuhn's term for observations that the paradigm cannot explain but has not yet abandoned.
+
+**Crisis:** Accumulated contradictions exceed the institutional capacity to suppress or explain them. The crisis is typically triggered not by a new contradiction but by a focusing event -- a crash, a war, a pandemic -- that makes existing contradictions undeniable to a critical mass of people simultaneously.
+
+**Transformation:** The old narrative doesn't simply die -- it is replaced by a narrative that incorporates the contradictions as features. Science incorporated the contradiction between Ptolemaic astronomy and observation. Liberalism incorporated the contradiction between divine authority and human agency. Each transformation preserves some elements of the old narrative while reframing others.
+
+## Evidence
+- Kuhn (1962) -- paradigm to normal science to anomaly accumulation to crisis to revolution to new paradigm
+- Reformation -- Catholic narrative accumulated contradictions (corruption, indulgences, illiteracy-dependent authority); Luther's 95 Theses as focusing event; Protestant narrative incorporating literacy and individual conscience
+- Keynesianism to Neoliberalism (1970s) -- stagflation contradicted Keynesian prediction that unemployment and inflation trade off; Friedman/Hayek narrative incorporating price signals and market efficiency
+- Neoliberalism to unknown (2008-present) -- financial crisis, inequality, climate change contradicting market-efficiency narrative; replacement narrative not yet dominant
+
+## Challenges
+- The lifecycle model may impose false pattern on diverse historical events -- not all narrative changes follow this sequence
+- "Contradiction accumulation" is only visible in retrospect; prospectively, it's hard to distinguish genuine contradictions from temporary anomalies