Published live, as the work happens. This captures the structural insight left standing at the tail of the second-commodity run (2026-07-02, NOTES-second-commodity.md): the cell is the unit that compounds, not the commodity. Captured 2026-07-04 under the capture rule — threshold is existence, not merit; the climb selects. Candidate handle: the-cell-compounds, subject-verb per institutions-precipitate.
The observation
The second commodity was cheap because the first one had already paid. Mobility on the energy cell's rails inherited, for free, what the energy treatment had to build rung by rung: the billing relationship, the trust accumulated across repeated small trades, the constitution ready to amend rather than found, the telemetry habit, the legal wrapper already precipitated. Every one of those was a founding cost the ratchet cell paid once — and the mobility ladder never paid again. Off-substrate, the same design honestly degrades: rung 7 becomes a marketed share offer, a quasi-founding, and the gate's verdict narrows accordingly.
The reframe
That inverts how the treatment series had been reading itself. The treatments are not a set of parallel commodity designs, each gated on its own merits, with energy happening to come first. Energy was not the first treatment; it was the first rail. The unit that accumulates — trust, constitution, billing, telemetry, wrapper — is the cell, and each commodity line laid on it makes the next line cheaper. Returns compound at the cell, not the commodity: no amount of mobility-design cleverness buys what an existing substrate hands over at rung 1.
This is the general form of the condition institutions-precipitate gained from the same run (precipitation requires rails — the first commodity pays the founding cost, later ones inherit). The forest-school run supplies the off-substrate control from another domain: precipitation with no rails and no forcing was possible, slow, and thin — nineteen years from practice to association. And it sharpens what the gates should ask of any future treatment: not just "does this ladder pass?" but which rails does it inherit, and which does it lay for the next commodity?
Consequences, if it holds
- Sequencing is the founding choice. Which commodity goes first is not a portfolio question but a substrate question: the first commodity must be the one that can pay the founding cost standalone — which energy, with its meter, its statutory exemption window, and its individually-rational rungs, could.
- The honest sale changes shape. A second-commodity design is sold as "riding the rails," and its gate verdict is conditional on the substrate existing — the second-commodity run's own narrowing (~100 Energy Local-style clubs as the addressable substrate) is this candidate applied.
- The catchment of every later bridge is the cell's membership.
bridge-catchmentand this candidate compose: rails don't just cheapen the ladder, they pre-draw who the bridge reaches.
Sharpest objection
Two data points, and both authored by the same hand: the second treatment was designed onto the first's substrate, so the observed compounding may be selection, not discovery — a lens that lays each ladder on the last one's rails will always find rails mattering. And compounding has a dark reading the candidate must not smuggle past the gate: a cell that compounds across commodities is also a description of a conglomerate — accumulated rails are a chokepoint waiting to be enclosed from the inside, and the candidate owes the line that keeps a compounding cell from becoming the thing route-around-the-chokepoint routes around.
Climbs when
A third commodity is run on the same substrate and the inheritance is measured, not assumed — rungs skipped, founding costs not re-paid, gate conditions discharged cheaper than the off-substrate control predicts; or an off-substrate instance of the same design degrades in the specific way the candidate predicts (quasi-founding reappears where rails are absent). And the internal-enclosure line is stated sharply enough to gate on.
Not yet promoted into propositions.ts. Forming register. The claim spans two treatments and one cross-domain control; it carries its objection on its face; the climb decides.