Two-axis tradeoffs (happiness vs longevity, consumption vs investment) don’t have a stance-independent optimum — only viable distributions of preferences whose holders don’t destroy themselves over time. The Platonic move is to find THE answer; the Aristotelian move is to find a process that keeps multiple answers viable. Markets fit the second move; democracy bets on the first. Adaptability is the meta-metric that picks among candidate processes — and it picks differently than the substantive metric does.
Links: Structure vs Adaptability, The Weighting Problem, Scope Confusion, The Constitutive/Elective Distinction, Opposing Forces, The Gödel Governance Problem, Civilizational Cycles, Newcomb’s Paradox, Directional Truth and the Discrete Trap
Should you have a short happy life or a long miserable one? Economics frames this cleanly: indifference curves over (happiness, longevity); a Production Possibilities Frontier defines what’s feasible; the optimum sits at the tangency of the highest indifference curve with the PPF.
Two complications disrupt this clean picture, and chasing both produces the rest of this note:
Each new axis reproduces the first complication. People differ on time preference. People differ on intergenerational obligation. People differ on externality weighting. The dimensionality is unbounded — and that’s a clue, not a frustration.
There is no stance-independent answer to “what’s the right tradeoff.” This is the Weighting Problem almost verbatim: objective measurements don’t yield objective aggregates. Adding axes doesn’t produce a unique optimum; it just multiplies the dimensions of the same aggregation problem.
The mistake is looking for THE optimum. There isn’t one. What can exist is a stable distribution of strategies whose holders don’t destroy each other or themselves over time. Convergence isn’t to a point; it’s to the space of viable strategies. The viable space gets carved out by selection: strategies that always pick “short happy life” leave fewer descendants; strategies that always pick “long miserable life” select against themselves through unhappiness reducing reproductive success or cultural transmission. The set of stable answers is constrained by what can persist.
This is Plato vs Aristotle re-litigated. Plato wanted THE Forms; Aristotle wanted essences-in-things. The Platonic move (universal optimum) generates conflict because two universal claims can’t both be right. The Aristotelian move (particular optima at particular scales) lets multiple correct answers coexist.
The principle (tangency of indifference curve and PPF) must operate at one scale at a time. Individual indifference + individual PPF = personal optimum. National indifference + national PPF = national optimum. Mixing scales produces the policy disasters that look like “good intentions, bad outcomes.” This is exactly scope confusion: most political disagreements aren’t about values, they’re about which scale the question is properly answered at.
A corollary: cigarettes are a consumption decision (effects primarily on the chooser); the death penalty is a constitutive decision (what kind of polity are we). Markets handle consumption beautifully — they’re explicitly designed for “what does each person want, given their resources.” Constitutive decisions answer “who are we together,” and that question doesn’t decompose into individual preferences. Markets aren’t worse for constitutive questions; they’re answering a different question.
If no static optimum exists and the environment keeps changing, the substantive choice mechanism matters less than the speed of correction. Both markets and democracy have failure modes — bubbles, manias, sectoral over-investment on the market side; voting-yourselves-into-oblivion on the democratic side. Both make the same kind of mistakes. What differs is the half-life of mistakes:
This is Hayek’s “Use of Knowledge in Society” argument extended to time. Hayek’s insight was that markets aggregate dispersed spatial knowledge — no central planner can know what each producer/consumer knows. The temporal version: markets aggregate dispersed time preferences and update faster than any central body can. Democracy bets on a single time-slice of preference aggregation being roughly right for years; markets re-aggregate every transaction.
The right framing isn’t “markets find better answers” — it’s “markets find approximately-as-bad answers, but the half-life of bad answers is shorter.”
Anarcho-capitalism is, by this metric, the most-adaptive design — competing providers, no central choke point, continuous correction. So why does it converge to states historically (Iceland’s commonwealth, medieval merchant law, the early American frontier)?
Because a maximally-adaptive system is, by construction, vulnerable to being voted out of adaptability by people who’d rather have stability. Security has natural-monopoly characteristics: largest coalition wins disputes; smaller coalitions aggregate to compete; aggregation produces a single coalition. Ancap has no mechanism to defend its own meta-rules from a coalition that prefers a less-adaptive system.
This is the Gödel Governance Problem in this frame: complete adaptability + maintaining adaptability = inconsistent. You can have one or the other, not both. Constitutional pre-commitment (Newcomb’s Paradox territory) trades adaptability for stability; market anarchy trades stability for adaptability; states are an unstable mixture of both. There’s no escape from one of those failure modes, only choices about which one you’d rather live with.
The cycle theorists (Khaldun, Spengler, Toynbee, Strauss & Howe, Glubb) all assume societies should avoid collapse and analyze why they failed. But if no static optimum exists AND the environment keeps changing, some collapse is structurally unavoidable. The metric isn’t “did this society last forever” (impossible) but “did it adapt for as long as the environment permitted before changes outpaced its adaptive capacity, AND did it leave viable seeds for what comes next.”
Crops are the cleanest analogy. We don’t say wheat “failed” because it gets harvested. The harvest is the structure exhausting its viable trajectory in that field-season; rotation puts in something different that can build different soil. Civilizations might be the same: each cycle exhausts what its institutional configuration can extract, then rotates. The “got it wrong” frame is borrowed from engineering (where systems should keep running indefinitely if correctly designed). Ecology might be the better source domain. Ecologies don’t “fail” when species go extinct — extinction IS the mechanism by which the ecology updates.
This connects to Opposing Forces: concentrations generate counter-forces; trying to eliminate counter-forces produces brittleness; embracing them produces resilience-AND-eventual-change. Adaptability isn’t a defense against change; it’s the willingness to be the structure that gets rotated next.
Two people with different correct-at-their-scale optima both insisting “this is THE optimum” produces all conflict. Drop the universalization — accept “this is MY optimum at MY scale” / “this is OURS at OUR scale” — and most of the heat dissipates.
The vault’s constitutive/elective distinction is doing this same work: there ARE constitutive facts (what makes coordination possible at all), AND elective ones (how each agent fills in the rest). Trying to make the elective constitutive is the universalization error. The political disputes that look intractable are mostly people trying to constitutionalize an elective preference, OR mistaking a constitutive fact for an elective preference.
The actual answer to “convergence to what” is: convergence to a viable adaptive process, not a viable end-state.
The conventional Platonic answer to “happiness vs longevity” is to find the right point on the line. The Aristotelian answer is that the line itself is moving, the points are person-specific, and the only stable thing is the process by which civilizations and people keep updating their answers as conditions change.