Listen Get

The Planetary Constitutional Convention

Stress-test and improve the current instruction manual until humanity has something less terrible to use

Abstract

The Planetary Constitutional Convention is the deliberative process that precedes implementation of the Earth Optimization Prize141. Before anyone recruits a single referendum voter or sells a single bond, there is a prior question: is the current Earth Optimization manual actually the one humanity should use? We think it is terrible. We said so in the title. The honest first step is to improve the manual until it is less terrible. The protocol lives on GitHub. Anyone can fork it. The adversarial review process is a podcast series where domain experts critique the manual and state what changes would be required for them to endorse it, plus published critiques and pull requests. The ratified revision becomes the current manual the referendum, the bonds, and the lobbying campaign execute. This runs before everything else because committing billions to an unstress-tested manual is how your species normally operates, and that is not a compliment.

Keywords

war-on-disease, 1-percent-treaty, medical-research, public-health, peace-dividend, decentralized-trials, dfda, dih, victory-bonds, health-economics, cost-benefit-analysis, clinical-trials, drug-development, regulatory-reform, military-spending, peace-economics, decentralized-governance, wishocracy, blockchain-governance, impact-investing

The Earth Optimization manual is the current instruction manual for a complete governance reform package. We think it is terrible. We said so in the title.

This constitutional convention exists to produce something less terrible. Fork the current manual, fix it, and improve it continuously. The ratified revision becomes the current manual the Earth Optimization Prize141 follows.

The protocol lives on GitHub. Anyone can fork it. The adversarial review process is a podcast series where domain experts critique the manual and state what changes would be required for them to endorse it. Each episode produces publishable content and a concrete list of required changes tracked as issues against the current manual. Pull requests are how improvements get proposed and ratified. The constitutional constraints below are the filter; formal ranking only matters if multiple serious revised manuals emerge at the same time.

ImportantDisqualification Rule

If your proposal requires politicians, bureaucrats, investors, regulators, voters, or interest groups to become less selfish than they currently are, you have described a pleasant world rather than a mechanism for reaching it. “Requires selflessness” is not the same as “makes self-interest coincide with better outcomes.” We have been observing your species since 1945 and have not yet seen selfishness decline on request.

The Current Manual

What follows is our homework. The thing to improve is the integrated set of required functions implied by the companion papers.

Required function Default implementation in the Earth Optimization Protocol v1 What an improved manual still has to solve
Large initial reallocation wedge The 1% Treaty142 Must redirect at least as much low-NSV spending to higher-NSV use
Medical throughput and evidence generation Ubiquitous Pragmatic Trial Impact Analysis143 + The Continuous Evidence Generation Protocol144 Must produce faster, cheaper, more reliable treatment discovery
Regulatory-delay removal The Invisible Graveyard145 + Right to Trial & FDA Upgrade Act + Drug Development Cost Increase Analysis76 Must reduce efficacy lag and development cost without increasing net harm
Political financing and adoption engine Incentive Alignment Bonds146 (funds the lobbying campaign; PRIZE deposits fund the prize pool separately) Must make passage at least as incentive-compatible for selfish actors
Citizen preference aggregation Wishocracy147 Must recover public preference at equal or lower cognitive cost
Waste and opportunity accounting The Political Dysfunction Tax48 + United States Efficiency Audit148 Must identify low-NSV pools at least as well
Policy recommendation engine Optimocracy149 + The Optimal Policy Generator150 Must generate better recommendations under real-world constraints
Budget recommendation engine The Optimal Budget Generator151 Must allocate public-goods spending better on the terminal metrics
Legal and institutional implementation path Right to Trial & FDA Upgrade Act + treaty/statutory tools Must create equal or stronger binding force
Narrative, coalition, and sequencing wrapper How to End War and Disease Must coordinate adoption at equal or lower cost and failure risk

That is the Minimum Acceptable Governance. Ratification requires covering every row or replacing any row with something strictly better.

Constitutional Constraints

Wishocracy147 selects the ratified version. But Wishocracy picks from a filtered pool, because crowds are good at preferences and bad at evaluating whether a 50-page systems engineering document is mechanistically sound. A beautifully written plan that skips the hard parts should not outrank an ugly plan that solves them. And your species is deeply status-quo biased: you have been programmed your entire lives to treat the Federal Reserve, the congressional budget process, the IRS, and representative democracy as laws of nature rather than design choices that are producing catastrophic results. Left unconstrained, crowds will pick plans that preserve these institutions with cosmetic reforms, which will recreate every problem the prize exists to solve.

So: eleven hard constraints that no proposed revision can violate, regardless of how many humans vote for it. The first four ensure the plan is real. The next two ensure it survives. The last five ensure the new system does not become the old one.

The plan must be real:

  1. The selfishness rule. If your plan requires any participant to become less selfish, it is disqualified. This is the single most powerful filter. Most policy proposals fail here.
  2. The two metrics. Must demonstrably improve median healthy life years and median real after-tax income. Not “probably.” Show the mechanism.
  3. The completeness rule. Must cover every required function in the Minimum Acceptable Governance table, or prove one is unnecessary. No skipping the hard parts.
  4. The mechanism rule. Must describe HOW, not just WHAT. “Universal healthcare” is a destination, not a route. Disqualified.

The plan must survive:

  1. The capture resistance rule. Must explain why concentrated interests cannot co-opt the system. If your plan has a board, explain why the board cannot be bought. If it does not have a board, explain what prevents capture anyway.
  2. The ratchet rule. Must be self-sustaining once started. No plans that depend on continued goodwill, because your species’ goodwill has a half-life of approximately one news cycle.

The old failure modes must die:

  1. No capturable intermediary in public goods allocation. Budget decisions must flow from citizen preferences to resource allocation without a layer of representatives who can be lobbied, bribed, or captured. This does not prescribe any specific allocation mechanism. It rules out “we’ll fix Congress.” Your species has been trying to fix Congress for 248 years. Congress is performing as designed. The design is the problem.
  2. Monetary neutrality. Any new money created by the system must enter at the population level, not through politically connected intermediaries. The Cantillon effect (whoever gets new money first benefits; whoever gets it last pays) means new money enters through politically connected intermediaries, not the population. This does not prescribe any specific monetary system. It rules out preserving a money printer that funds wars without consent and steals 2% of purchasing power per year by design.
  3. No single point of rollback. No individual, committee, or institution can unilaterally reverse the system’s progress. A treaty that one president can exit by tweet is not a treaty. A fund that one board can liquidate is not durable. If your plan has a kill switch, it will be used, because your species has never built a kill switch it didn’t eventually pull.
  4. Rights firewall. The allocation mechanism is constitutionally restricted to bounded resource questions. No collective mechanism can override individual bodily autonomy, criminalize personal choices that harm no one else, or restrict access to proven-safe treatments. You cannot vote away rights. No plan may improve its terminal metrics by externalizing costs onto populations excluded from the allocation mechanism: future generations, foreign populations outside the treaty, and nonhuman sentient beings. Allocation is democratizable. Rights are not.
  5. Full transparency. Every allocation, transaction, and decision must be publicly auditable by any participant. No classified budgets. No hidden flows. Opacity is how capture hides.

These are constitutional constraints, not preferences. A proposed revision that fails any one of these never reaches the voters, the same way a law that violates the First Amendment never reaches the statute books (in theory; your species’ track record on this is mixed).

How It Works

Constitutional constraints are the filter. The eleven constraints above are binary: a proposal either satisfies them or it does not. Anyone can check compliance. No expert panel needed; the constraints are designed to be verifiable by inspection.

Public adversarial review. Domain experts critique the protocol in podcast conversations and state what changes they would require to endorse it. Published critiques, red-team memos, and GitHub issues track every objection. This is peer review, except it happens in public, produces listenable content, and generates a concrete punch list of required changes rather than a rejection letter.

GitHub forks and pull requests. Anyone can fork the current manual and propose improvements. Pull requests are how improvements get proposed, debated, and ratified. The version history is the audit trail.

Wishocratic ranking (when needed). If multiple serious revised manuals emerge at the same time, Wishocracy activates: citizens compare two plans at a time (“Which of these two would you rather live under?”), about 20 comparisons, 5 minutes. The crowd does not need to understand mechanism design. The crowd needs to understand “which of these two futures sounds less terrible?” Humans are surprisingly good at that question and surprisingly bad at everything else. This infrastructure is not needed until the scaling problem actually exists.

The ratified revision becomes the current Earth Optimization manual. If no revision improves the current manual, the current manual remains in force by default, which means we were right that it is terrible but wrong that nobody could improve it, and that is a more depressing outcome than we are prepared to contemplate.

The minimum bar for scoring. Any proposal that cannot beat the risk-adjusted treaty benchmark on cost per DALY does not get scored:

Intervention Cost per DALY averted
Malaria bed nets (current gold standard)

$89 (95% CI: $78-$100)
1% Treaty (conditional, if adopted)

$0.00177 (95% CI: $0.000715-$0.00412)
1% Treaty (risk-adjusted, 1% (95% CI: 0.1%-10%) success)

$0.177 (95% CI: $0.029-$3.2)

What the Current Manual Claims (and Where To Improve It)

The prize asks one question: who can cause verified implementation of a complete package that dominates on welfare per dollar? We wrote the current manual because conventions need a starting draft. We expect it to be improved. Here is what the current manual claims, paper by paper, and the specific benchmarks any revision still has to clear:

Paper What the prize uses it for Key benchmark any improved manual still has to clear
The 1% Treaty142 First large transfer from low-value military spending to high-value medical discovery $27.2B/year; $0.00177 (95% CI: $0.000715-$0.00412)/DALY conditional; 10.7 billion (95% CI: 7.4 billion-16.2 billion) deaths averted
Ubiquitous Pragmatic Trial Impact Analysis143 ROI case for medical evidence generation as destination for redirected capital 12.3x (95% CI: 4.2x-61.4x) capacity increase; queue from 443 (95% CI: 324-712) to 36 (95% CI: 11.6-77.1) years; $0.842 (95% CI: $0.242-$1.75)/DALY
Incentive Alignment Bonds146 Financing and political-adoption engine 272% annual return; 230 (95% CI: 186-284) mechanism BCR
Wishocracy147 Preference aggregation with intensity Binding at >=2% participation; 10-30 pairwise comparisons per citizen
The Political Dysfunction Tax48 Master ledger of value left on the floor $101T (95% CI: $83.3T-$191T)/year globally
The Invisible Graveyard145 Mortality proof that delay kills 8.2 (95% CI: 4.85-11.5) years delay; 102 million (95% CI: 36.9 million-214 million) deaths; 7.94 billion DALYs (95% CI: 4.43 billion DALYs-12.1 billion DALYs)
The Price of Political Change152 Budget ceiling for buying legal democratic pressure $1B; >259k (95% CI: 110k-419k) benefit-cost ratio
United States Efficiency Audit148 Concrete waste map $4.9T (95% CI: $3.62T-$6.5T)/year; $2.45T (95% CI: $1.81T-$3.25T) recoverable
Optimocracy149 Cross-jurisdiction recommendation engine Causal policy comparison across thousands of jurisdictions
OPG150 Law-level enact/replace/repeal engine Policy Impact Score; 5-15% GDP welfare gains
OBG151 Spending-level reallocation engine Budget Impact Score; 20-40% misallocation correctable
Continuous Evidence Generation Protocol144 Medical evidence machine $0.1 (95% CI: $0.03-$1)/patient Stage 1; $929 (95% CI: $97-$3K)/patient Stage 2
Drug Development Cost Increase Analysis76 Proof current regulation is catastrophically expensive 105x (95% CI: 90.6x-119x) cost increase; $2.6B (95% CI: $1.5B-$4B)/drug
How to End War and Disease Integrated narrative and coalition wrapper $1B total cost; 1% (95% CI: 0.1%-10%) success probability