Published live, as the work happens. This captures a conversation (2026-07-02), thinking out loud in the forming register. The name is a scaffold — deliberately provisional, held until the concept has been triangulated from more directions; per the naming discipline, coining is the last step of concept formation, not the first.
The observation
The current global economic system has not factored in the true cost of living off our energy savings account. That much the canon already says (energy-income-inheritance: fossil energy is inherited savings, not income; the economy books the withdrawal as revenue). But there is a more general form of the same failure, and it is not about energy as such:
At the cosmic scale, the cost of re-establishing the configurations that we are re-configuring is not being accounted for.
Every act of economic consumption re-configures something — draws down a standing configuration (a coal seam, a soil profile, an aquifer, a fishery, a forest, a language community, a trust network) and disperses it into a lower-assembly state plus waste heat. The market records the price paid to whoever held the configuration. It records nowhere what it would take to put the configuration back.
The candidate accounting unit
Re-establishment cost: what it would cost, by the best currently-known path, to re-assemble the configuration being drawn down.
This is a borrow, and the mechanism transfers cleanly rather than just the vibe. Insurance has always known that market value and replacement cost are different numbers answering different questions — what someone will pay for the thing versus what restoring the thing actually takes. Ecological economics has used replacement-cost methods for ecosystem services for decades. The delta here: state it in configuration terms, at full generality (it applies to a hydrocarbon, a soil, a species, a language, an institution alike), and be honest about deep time — because for many configurations the original assembly path ran over timescales no balance sheet has a row for.
A barrel of oil illustrates the structure. Its market price reflects extraction cost plus rent. Its re-establishment cost can be honestly priced two ways: by the original path (~10⁸ years of photosynthesis, burial, and pressure — effectively unpayable, which is itself information), or by the best current path (synthesising equivalent hydrocarbons from air-captured CO₂ and renewable energy — finite, computable, and far above the market price; the specific numbers need a White-hat pass and are deliberately not asserted here). The gap between market price and either figure is the unaccounted drawdown, per unit, in currency a CFO recognises.
Why this might matter to the open node
option-space-measurability is the canon's one genuinely open problem: option space resists a global scalar. Re-establishment cost is not that scalar — but it may be a serviceable local, per-move proxy, which is the same move the canon makes everywhere else (local Δω comparison, never a global score). It is measurable, monetisable, and attributable per configuration drawn down. And attribution is the lever that makes an externality pullable at all: the general finding from the autonomous-fleet work was that legibility, not virtue, is usually the actual chokepoint. A drawdown becomes chargeable the moment it becomes measurable and attributable — this ledger is an attempt to make configuration-drawdown legible.
The consistency check
A proposed proxy should reproduce the canon's existing structure wherever they overlap; otherwise it is decoration. This one does, twice:
- For irreversible losses — an extinct species, a dead language — re-establishment cost diverges to infinity. That independently recovers
asymmetry-of-option-space-change(collapse fast, rebuild slow-to-never) and the no-regret shape ofviable-objective(some moves must be refused outright, not priced). - For reversible drawdowns, the cost falls as pattern-intelligence rises (better synthesis paths make re-establishment cheaper) — which is
pattern-intelligence-constraintshowing up in the ledger exactly where it should: the price of putting things back is a function of the patterns available, not of the atoms.
Where it is honestly limited
- It is a lower bound, not the value. A configuration's worth includes what it enables — its option space — not merely what re-assembling it costs. Two configurations with identical rebuild cost can differ enormously in what they open. The ledger prices the drawdown; it does not price the loss.
- The infinity-laundering danger. Assigning finite numbers to things is seductive, and for irreversibles it would licence exactly the wrong move: "we'll pay the rebuild cost" where no rebuild exists. The ledger must respect its infinities — carry them as refusals, not launder them into large-but-payable line items.
- Tech-dependence cuts both ways. Pricing at the best current path is honest (it is what re-establishment would actually cost today) but invites gaming via optimistic technology claims. The discipline: price at demonstrated paths, not promised ones.
- Why it isn't priced today is already in the canon —
displaced-costsandmoney-as-signal: the market prices what is legible to it, and configuration-drawdown has been kept illegible partly because the drawdown pays. This ledger does not dissolve that interest; it names what the interest keeps hidden.
Nearby lineage
Assembly Theory (Cronin) measures backward complexity — how much assembly a configuration embodies. Re-establishment cost is close kin: assembly, priced forward, by the paths we actually have. structural-memory says configurations are stored assembly; this ledger says: then keep the books.
Not yet promoted into propositions.ts. Forming register: the scaffold name, the two-path pricing structure, and the consistency checks are the claim; all specific figures await a White-hat pass. If the concept survives triangulation, it may earn a real name and a node.