The audited vacuum.

1 · One axiom: the vacuum keeps books, and reads only the books.

The vacuum maintains a locally conserved budget. Matter is expensive: existing consumes budget, and what is consumed must be drained from elsewhere under a per-site sum rule — a Gauss law, not as an assumption about fields but as the meaning of conservation. The vacuum monitors violations of its own sum rule and nothing else: it never reads states. That single restraint — audit the ledger, not the world — turns out to generate the framework's every distinctive consequence: forces, superselection, the decoherence knee, and the protection of quantum superpositions at rest.

2 · The audited sector is electromagnetism.

Price the drainage by the vacuum's own information metric — the only positive functional a vacuum owns, since its energy is provably concave in its couplings (a theorem) — and the conserved sector delivers, with nothing declared: exact Coulomb form in one, two and three dimensions; two charge classes with the electrostatic sign rule (like repels like, machine-exact); emergent neutrality — isolated charges cannot exist on compact space; a potential that is provably invisible to clocks, which is precisely what gauge invariance demands; charge superselection derived from the audit — a fact of nature that quantum theory ordinarily postulates; a short-range neutral residual (r⁻³·⁵, the sector's van der Waals); and a coupling 1/S whose measured running screens — QED's direction — with α⁻¹ landing within one to two orders of 137, untuned. We spent a year calling this sector gravity and fighting its "anti-gravity problem"; the theorems were telling us its name the whole time. The sector that refuses to touch clocks was never gravity.

3 · Gravity is the active monitor's embrace.

Gravity lives in the monitored medium itself: masses couple multiplicatively to a local mediator field, the monitor's back-action induces the interaction, and three measured facts fix its character. First: the sign. For every physically passive bath the induced force is repulsive — passivity owns the sign (the Casimir/KMS structure), and it owns the wrong one. Attraction demands an active vacuum: a monitored sector held in inversion, a gain medium — which is not a knob but measurement thermodynamics: a vacuum that audits continuously is a working measuring device, and a demon needs a battery. Measured: the active medium is dynamically stable, its idle bill is exactly zero at Gaussian order (where quantum field theory overshoots by 120 orders), and it adds no noise — the sign of gravity is dispersive, the fluctuations are absorptive, and only the former flips (the Caves attack died on a measured zero). Second: the pairing. In the attractive branch, clocks near mass run slow — and the clock sector is sign-rigid: forces could flip, rates cannot. General relativity's signature arrives whole. Third: the wiring. Classical drift rates see only tides — a theorem, exact to machine precision — while quantum phases read the potential linearly, universally, at long range: ω = E/ħ, and the interaction energy is in the Hamiltonian. Gravitational redshift is an intrinsically quantum phenomenon. Every real clock is a phase clock; every energy-derived rate (pulsar rotation included) redshifts identically; the framework agrees with every existing test and ranks the equivalence principle asymmetrically: force-EP may break someday, clock-EP structurally cannot.

4 · The theorems that carved the map.

The concavity theorem (n45): vacuum energy can never price a rearrangement — it destabilizes; only the information metric can. The dichotomy theorem (n51a): in flux-conservation media, the architecture that attracts has no linear clock response for substrate observables, and the one with linear clocks repels — the theorem that evicted gravity from the drainage and identified the local potential field as the assumption secretly carrying the metric in every emergent-gravity program. The gradient theorem (n72): gradient-coupled monitors give tidal-only classical clocks, exactly — the dichotomy's ghost, and the proof that redshift needed ħ. The passivity theorem (n73): the sign of monitored-bath gravity is fixed by the bath's spectrum; attraction requires activity. Four knives, four walls, one corridor — the corridor is the theory above.

5 · The locks: what the theory refuses to let float.

The medium's observables are chained: γ_knee/G = κ/|Δ| — the vacuum's quality factor, one dimensionless number tying the decoherence knee to the strength of gravity (verified to four digits across the parameter space). Consequences, each measured or exact: the weakness of gravity is the overdamping of the audit — the rate-half of the hierarchy problem collapses into Q; a measured knee makes the hierarchy an output. The PPN reading survives as γ = (1−2a)/(1+2a) with a the audit's state-dependence: Cassini's six digits read as |a| < 5.8×10⁻⁶ — the ledger's clock does not care what it prices. And β and Λ are locked to arrive together: both live at the first anharmonic order beyond a measured exact zero — no Einstein perihelion without dark energy, no dark energy without the perihelion.

6 · The knee: the experiment that owns the whole house.

A mass superposition at rest decoheres not at all — measured through eighteen orders, floor-free even in the active branch. Modulated at frequency ω, it decoheres along the exact Lorentzian (dT₀₀/dt)²/(γ²+ω²), knee at the audit rate. General relativity predicts nothing at any frequency; collapse models predict a floor already present at rest; this framework predicts zero, then the knee, then more than the collapse scale above it — no overlap with any rival at any frequency. And the concession extracted by our own final assault stands at the top of the ledger: this is the framework's existence gate. A confirmed flat null kills the audit, the budget, and every sector at once. Everything on this page is a wager on one frequency sweep. The experiment page → · the ten dated forecasts →

7 · What is not built, said plainly.

Three organs, each with named watchers. The one-substrate junction (n81): the two sectors live in two toys never yet shown to share a lattice; the registered bet is that they coexist because charge and energy audits commute, and the failure locus is named in advance. The tensor junction: β = 1 (Mercury), g₀ᵢ (LARES), the quadrupole (binary pulsars), the optical γ and emergent Lorentz invariance must all arrive together through year one's formal chain (TT → unimodular → Bianchi → Einstein) — five experiments watch one door. The pump: the microphysics of the inversion, which owns the sign's origin, Λ's value, the floor question, and the G-drift correlation with dark energy. Until these are built, this framework is a weak-field, two-sector account with its dynamical completion owed — and it says so on every page.

8 · The theory in one breath.

The vacuum keeps a conserved ledger and reads nothing but the ledger. The audited sector is electromagnetism — Coulomb's law is the bookkeeping, charge superselection is the audit, gauge invariance is the books' indifference to clocks. Gravity is what the auditing costs: an active monitor's embrace, attractive because a measuring vacuum cannot be passive, weak because the audit is overdamped, readable only by quantum phases because only energies feel the ledger — and Einstein's γ equals one because the ledger's clock does not care what it prices. The proof that any of this touches our universe fits in one apparatus: sweep a mass superposition in frequency, and find the knee — or bury the whole house with the twenty-seven ideas it buried first.

Every claim above links to a pre-registration, a script that runs in minutes, or a proof — and the failures are published at the same prominence at every step. That is not a rhetorical flourish; it is the only reason to believe any of it.