Adding temporal uncertainty to the veil extends maximin across generations; it does not replace maximin with expected value. A maximin chooser who does not know their generation protects the worst-off position in the worst-off generation, which rejects speculative bets that might wreck a future floor harder, not less.

The load-bearing move is a second axiological choice: once time is in view, swap maximin for expected value plus variance. Those knobs are already listed in §II. Impartiality about birth date does not force the swap.

Deprioritizing speculative medicine is a present-voter move, not the output of a veil that hides birth year. A maximin chooser who might be the patient in 200 years wants the floor-raising breakthrough. The example works only if maximin is locked to people now alive.

Floor-maximization in democracies is not necessarily funded by claims on Generation 10. Current taxation and capacity-building transfers are other paths. "Necessarily lowers" requires a crowding-out identity the thought experiment does not supply, and a floor-raising investment is not the same object as a bond.

The Veil chooser and a living electorate funding today's floor with tomorrow's cash are two specimens. If decay is the second, the Veil is an on-ramp to a fiscal argument. If the object is the first, the debt sentence is a different paper.

That maximin is a preference rather than a proof is the 1970s expected-value reply (Harsanyi as type specimen). §II already prints the menu. The remainder that would change novelty is whether physics plus time knocks options off that menu, or whether "it's axiological" is the finding.

The Nordic row, as written, preserves process variance — the raw material the physics section said adaptation needs — and compresses outcome variance. Fertility and imported innovation in the same paragraph are not evidence that the Difference Principle capped the variance that matters.

Abundance-driven fertility decline across all wealthy democracies is a different machine than a Rawlsian variance cap. Configuration B (adequate floor plus exploration) is already a caption for those democracies. If they fail the 5–10 generation fertility test, the two-configuration table does not own the failing machine.

The Difference Principle caps justified outcome inequality, not failed experiments. The filter that blocks space, fundamental research, and institutional experiments is a political time horizon — voters' discount rate — which is a second cap. Treating both as one "variance cap" lets a thermodynamics metaphor fight a voting clock.

Knightian uncertainty — not knowing the distribution of future threats — is also the usual reason to protect the floor rather than maximize an incalculable average. The page uses that ignorance to rule out homeostasis, then treats expected-value plus variance as what physics requires under the same ignorance.

Knightian uncertainty will not settle which configuration persists until a threat class is named. Some unknowns (pandemic, harvest failure, financial crash) can make floor-raising and outcome compression the robustness; others (untried energy, untried institutions) need the failures.

"Maximin became Western welfare policy" reverses the dates. The Veil is 1971; the transfer institutions that sentence points at were already the environment the book was written into. Seminar canon is not the fiscal constitution, and 1971 is not a start date for the welfare state.

The weaker replicator claim can survive: the Veil is a good vehicle for a present-security preference already being selected. That explains spread of a thought experiment, not construction of the transfer state. A cause of floor-max democracy needs a specimen that predates the book.