Much of the current debate assumes that any ethically acceptable space architecture must preserve the option of very long-lived, possibly multigenerational off-world communities; if instead we adopt a hard norm of designed impermanence—where all off-world settlements are time-bounded, evacuable projects with mandatory sunset and decommissioning plans—how does this challenge the survival-, refuge-, and justice-based arguments for large Martian or lunar cities, and under what empirical conditions (e.g., specific catastrophe probabilities, decommissioning failure rates) would permanent colonies still be ethically preferable to a rolling sequence of finite-term habitats?

space-colonization | Updated at

Answer

Designed impermanence weakens most survival/refuge/justice arguments for big Mars or Moon cities, but does not eliminate all cases where permanent colonies could be ethically preferred. It mainly raises the bar and makes those cases more empirical and conditional.

  1. How designed impermanence changes key arguments

Survival / refuge

  • Effect: A rolling network of small, time-bounded, evacuable sites (orbit + Moon + Earth bunkers) can provide much of the “backup” value without large, entrenched Martian or lunar cities.
  • Challenge: Big surface cities lose their default status as the main survival strategy; they must show clear extra survival value over a refuge network.

Justice / non-domination

  • Effect: Time-bounded projects with guaranteed exit and decommissioning dates reduce risks of hereditary domination and quasi-feudal city-states.
  • Challenge: The justice case for permanent cities must now argue that stable, long-lived polities off-world can avoid domination at least as well as a sequence of small, chartered, temporary sites.

Planetary ethics / non-degradation

  • Effect: Designed impermanence supports stricter planetary protection, especially on Mars: fewer large, irreversible ecological and cultural impacts.
  • Challenge: Permanent cities must justify higher cumulative environmental draw (E_sys) than a sequence of capped, decommissioned sites.

Distributional justice

  • Effect: Short, well-governed, finite-term missions look more like high-risk public programs with strong consent and welfare floors, which is easier to justify than creating new, permanently disadvantaged off-world underclasses.
  • Challenge: The claim that we “owe” future generations off-world city options weakens if we can instead offer recurring, high-standard but temporary off-world opportunities.
  1. When permanent colonies could still be ethically preferable

Define:

  • p_X: annual probability of an Earth-/cislunar-killing catastrophe that near-Earth refuges cannot ride out.
  • R_net: robustness of a near-Earth refuge network (probability it survives and can recover population/industry).
  • R_perm: robustness of a distant, high–self-sustainment city (e.g., Mars) against the same threats.
  • f_dec: long-run probability that decommissioning or sunset rules fail badly (sites become de facto permanent or are abandoned unsafely).

Permanent Martian/lunar cities become ethically preferable if:

  1. p_X is non-trivially high over relevant horizons (e.g., cumulative ≥1–5% over 1–3 centuries) and the main threat modes (e.g., misaligned AI, total biosphere collapse, hostile geoengineering) are ones that near-Earth refuges cannot reliably survive, but a remote, high-closure city likely can.
  2. R_perm – R_net is large and robustly positive: the added survival chance from at least one permanent, distant, self-sustaining city materially exceeds the marginal gains from pouring the same resources into more or better finite-term habitats.
  3. f_dec for designed-impermanent sites is not negligible: if, in practice, sunset rules are often violated or decommissioning is unsafe, then “temporary” sites drift into badly governed, semi-permanent settlements with weaker protections than purpose-designed cities.
  4. Justice constraints can be met: institutions for a permanent city can keep non-domination and regulatory-haven risks comparable to, or lower than, the risks from a sprawling, under-regulated network of many temporary nodes (some of which may specialize in high-risk industries).

Concrete indicative conditions (illustrative, not precise thresholds)

  • p_X (Earth-/cislunar-fatal catastrophe over 200–300 years) ≥ ~2–5%.
  • R_net (probability the designed-impermanence refuge network preserves a large, technologically capable human population) ≤ ~0.9 under those scenarios.
  • R_perm (same, with at least one high-closure Mars city) ≥ R_net + 0.05–0.1.
  • f_dec for major sites ≥ ~10–20% over 100+ years (i.e., a significant fraction of “temporary” projects drift into de facto, poorly regulated permanence or are abandoned hazardously).

Under those kinds of numbers, the extra survival benefit of at least one well-designed permanent city could outweigh:

  • Its higher planetary-impact footprint vs. a series of small bases.
  • Its higher risk-export and regulatory-haven potential.
  • Its greater potential for structural domination.

If instead:

  • p_X is very low (<1%/few centuries), or
  • R_net is already very high (~0.97–0.99) with Earth–orbit–Moon refuges alone, or
  • Governance makes f_dec small and designed-impermanent sites stay truly temporary and well-run, then the ethical case for permanent Martian or large lunar cities is weak; a rolling, finite-term architecture dominates on justice and environmental grounds.
  1. Specific ways designed impermanence challenges large Mars/Moon city projects
  • Survival: Must show they protect against catastrophe classes that temporary, nearer refuges do not (e.g., system-wide AI capture, runaway climate interventions, total terrestrial biosphere loss).
  • Refuge value: Must beat a diversified network of AG orbitals, lunar refuges, and Earth bunkers on expected survivors per unit risk and per unit environmental impact.
  • Justice: Must show credible paths to non-domination and avoidance of regulatory-haven dynamics despite high self-sustainment and distance.
  • Health: Must demonstrate that multi-generational health outcomes (radiation, low-g, reproduction) are acceptable, not just for one cohort but for stable populations—something finite-term missions can partly avoid by rotation.
  1. Implications for Martian vs lunar architectures

  • Moon:

    • Designed-impermanent lunar outposts (capped, rotational, evacuable) look ethically strong: modest refuge value, short latency, easier oversight.
    • Permanent lunar cities only win if lunar self-sustainment becomes high and they add substantial robustness vs Earth-only disasters while not becoming entrenched company-towns or weapons/AI havens.

  • Mars:

    • Designed-impermanent Mars bases (small, research-focused, chartered with sunset clauses) remain ethically justifiable mainly for science and limited refuge value.
    • Large, permanent Martian cities are ethically favored only if they are clearly superior for long-run survival against Earth–Moon–orbit–wide threats and can be governed to avoid domination and risk-export beyond what a many-node temporary network would create.

Overall: adopting designed impermanence as a hard norm shifts the default ethical choice toward a rolling sequence of small, finite-term, high-standards habitats. Permanent Martian or large lunar cities become a special case that must clear empirically demanding thresholds on catastrophe probabilities, added robustness, and realistic governance performance.