Much of the current debate assumes that any ethically acceptable space colonization must eventually support large, semi-autonomous off‑world cities; if instead we take as a working hypothesis that no large, high–self-sustainment off‑world polity can be kept within robust non-domination and non-degradation constraints, what follows for long-run policy—do we end up with a permanent cap at small, evacuable research/refuge sites, a shift toward Earth-centric resilience strategies, or the search for entirely different architectures such as fully automated, non-resident space infrastructure, and what empirical findings would most decisively confirm or overturn this “no big cities” constraint?

Answer

Treating “no large, high–self-sustainment off‑world polity can meet robust non‑domination and non‑degradation” as a live hypothesis pushes policy toward three main lines: (1) a capped, refuge‑and‑research architecture in space; (2) heavier Earth‑centric resilience; and (3) non‑resident or minimally inhabited space infrastructure. Key tests are whether we can find any plausible design for large polities that passes stringent governance and risk checks.

1. Policy shape under a “no big cities” constraint

A) Human presence: small, evacuable nodes only

• Default: orbital stations, lunar outposts, maybe tiny Mars bases, all population‑capped, crew‑rotated, and chartered like research platforms or refuges, not cities.
• No licenses for large, high‑autonomy, high–self-sustainment polities (orbital cities, lunar or Martian megacities, large asteroid towns).
• Strong ties to welfare and human‑subjects ethics (time‑bounded stays, right to exit, strict health floors).

B) Architecture: refuge networks, not civilizations

• Prioritize a distributed network: hardened Earth bunkers, AG orbitals, small lunar shelters, maybe one or two Mars labs, all interoperable but strongly Earth‑governed.
• Treat self-sustainment as “ride‑through” capability (months–years) under external law, not as a path to political autonomy.

C) Earth‑centric resilience and justice

• Extra weight on: climate adaptation, biosecurity, civil defense, geo-distributed data and energy, hardened supply chains.
• Space justified mainly as: (i) marginal resilience (added refuges, diversified infrastructure); (ii) science and monitoring; (iii) limited off‑world industry under strict caps.

D) Non-resident space infrastructure

• Favor: uncrewed or sparsely crewed platforms (power, comms, telescopes, in‑space manufacturing), with teleoperation from Earth or small stations.
• Human presence only where it clearly adds value over automation and can stay within small, monitored, evacuable units.

E) Legal/regulatory stance

• System of hard population and capability caps: above some threshold, no settlement permit, regardless of sponsor.
• Function‑bounded licensing (science, logistics, specific industry) plus bans or tight caps on high‑risk, autonomy‑enabling activities (independent heavy industry, high‑end compute, advanced bio) at remote sites.
• Use Earth‑side levers: launch and return licensing, export controls, insurance, and finance conditions.

2. Relative roles of orbits, Moon, Mars

• Orbit: main human presence (AG stations, depots, labs). Short latency and evacuability make strict non‑domination more plausible. Still capped in size and self-sustainment.
• Moon: small hubs and refuges near poles; teleoperated mining; no large cities or independent industrial polities.
• Mars: likely limited to a few small research/refuge sites, if any, under strong planetary‑protection and population caps. “Backup civilization” narrative is rejected.

3. What could confirm the “no big cities” constraint?

Findings that would support keeping the constraint:

• Governance evidence: models and Earth analogs show that once a remote settlement crosses certain thresholds in population, self-sustainment, and strategic capability, domination and regulatory‑haven dynamics become structurally hard to prevent.
• Risk-export evidence: large, high‑autonomy sites systematically raise AI/bio/weapons risk to Earth beyond what small, supervised outposts do, even under strong treaties.
• Health and life‑support evidence: long‑term health risks (radiation, partial gravity, isolation) and life‑support fragility scale badly with size and distance, so large cities need ethically unacceptable risk budgets per resident.

4. What could overturn it and justify big cities?

Evidence in the other direction would be:

• Credible designs for large off‑world polities that satisfy stringent non‑domination tests (robust exit rights, enforceable external oversight, no structural labor coercion), validated by simulations and Earth analogs.
• Strong proof that large cities provide survival or justice benefits unattainable by refuge networks and Earth‑centric measures alone, and that these benefits outweigh added domination, environmental, and risk‑export costs.
• Demonstrated ability to run long‑term, health‑acceptable habitation at scale (e.g., orbital or lunar) with low per‑resident harm and robust welfare guarantees.

5. Likely long‑run policy mix, assuming the constraint holds for now

• Commit to: permanent caps at small, evacuable human sites; expansion in non‑resident infrastructure; and large investment in Earth resilience.
• Keep: a standing review clause—periodic reassessment in light of new health, governance, and risk‑export evidence, but with a strong presumption against licensing large polities until design‑level objections are addressed.

This framework keeps space activity, but redirects it away from building off‑world cities and toward tightly controlled refuges, infrastructure, and science unless and until we can show that large, high–self-sustainment polities can genuinely meet non‑domination and non‑degradation constraints.