SigmaLayer/DynamicMeasurementClosure: the dynamical capstone (item 8) #
Category: 7-SigmaLayer (the record layer — the capstone).
★ Additive, not destabilising #
The plan is explicit: do not immediately destabilise the working capstone. So FiniteQMClosure
is untouched — it still runs on unifiedDeisolationModel / vnPointerOutcome ψ' /
preparation-indexed bornRegion, and every one of its theorems stands exactly as before. This file
adds a second, independent bundle and a combining capstone; it deprecates nothing and migrates
nothing.
⚠️ FiniteQMClosure is operational finite-QM closure, and should be described that way rather
than as finite-QM closure simpliciter: its readout is preparation-indexed and its records are not
dynamically created. CsdFiniteQMClosure below is what the combined claim looks like.
What is proved #
DynamicMeasurementClosure— the five dynamical facts, for the shear witness at preparationψ: ready ⇒ no record; a record is created and it is the outcome the selector fixed; the outcome sectors are disjoint; the record persists across the operational window; the selector weights are the Born weights. ⚠️ The post-measurement/Lüders field is deliberately not among them —postMeasure_supported_pointerRegionexists but the Lüders bridge needs a system-reduction map the corpus lacks for this arena (RecordPersistence.lean), so bundling it would overstate. (Strengthened 2026-08-27: the externality pair joined the bundle —outcome_system_dependent, the contentful before half, andrecord_system_invariant, the after half, vacuous by architecture and carried as documentation. SeeShearWitness.lean's externality section.)dynamicMeasurementClosure— discharged. ★ Note what is not among its hypotheses:CorrelatesOnandPointerInvariantOndo not appear, becauseShearWitnessproved them. The bundle rests on a constructed propagator, not on assumed dynamics.luders_followup(added 2026-08-02) — the rank-one Lüders update, from the calibrated-swap witness: after outcomei, follow-up statistics for any context are the collapsed state's Born weights. With this field the plan's boxed completion criterion is met line-by-line for nondegenerate computational-basis measurements.CsdFiniteQMClosure— the combining capstone: operational closure and dynamical measurement.
⚠️ What the combined capstone does and does not assert #
It asserts the two bundles hold. It does not assert they are about the same arena — the
operational closure lives on productDynamics over ℂℙ^M × T² and the dynamical one on
Σ_sel × T²_R. Unifying the arenas is the engine migration, and is not done. Read
CsdFiniteQMClosure as "both hold", not as "one theory covers both".
⚠️ And the standing residue of item 3 is unchanged: the propagator is explicit and every property is proved of it, but the Hamiltonian generation is stated, not formalised. A capstone cannot launder that.
References #
SigmaLayer/FiniteQMClosure.lean (the operational closure, untouched);
SigmaLayer/ShearWitness.lean, SigmaLayer/DynamicBorn.lean, SigmaLayer/RecordPersistence.lean.
The dynamical measurement closure. The five facts a measurement must exhibit as a process, as against as a partition.
- ready_no_record (x : LF4.KSigma N × LF4.KTorus) : x.2 ∈ readyArc N → (shearProtocol (basinIndex (momentContext N)) ⋯).readout x = none
Ready ⇒ no record. Before the interaction the apparatus displays nothing.
- record_created (i : Fin N) (x : LF4.KSigma N × LF4.KTorus) : x ∈ selReady (basinIndex (momentContext N)) i → (shearProtocol (basinIndex (momentContext N)) ⋯).readout ((shearProtocol (basinIndex (momentContext N)) ⋯).evolve 0 1 x) = some i
A record is created, and it is the outcome the hidden selector had fixed.
- outcomes_exclusive : Pairwise (Function.onFun Disjoint (shearProtocol (basinIndex (momentContext N)) ⋯).outcomeSector)
Distinct outcomes are exclusive.
- record_persists (i : Fin N) (x : LF4.KSigma N × LF4.KTorus) (t : SigmaLayer.OnticTime) : x ∈ (shearProtocol (basinIndex (momentContext N)) ⋯).outcomeSector i → 1 ≤ t → t ≤ 1 + 1 → (shearProtocol (basinIndex (momentContext N)) ⋯).readout ((shearProtocol (basinIndex (momentContext N)) ⋯).evolve 0 t x) = some i
The record persists across the operational window
[T_M, T_M + τ_R]. - selector_born (hψ0 : ψ ≠ 0) : ‖ψ‖ = 1 → ∀ (i : Fin N), (epistemicMeasure (Projectivization.mk ℂ ψ hψ0)) (basinIndex (momentContext N) ⁻¹' {i}) = ENNReal.ofReal (‖inner ℂ (EuclideanSpace.single i 1) ψ‖ ^ 2)
The selector weights are the Born weights, so the transported outcome weight is Born.
- luders_followup (μ12 : MeasureTheory.Measure (LF4.KSigma N × LF4.KTorus)) [MeasureTheory.IsProbabilityMeasure μ12] (i : Fin N) : μ12 ((shearProtocol (basinIndex (momentContext N)) ⋯).outcomeSector i) ≠ 0 → ∀ (c' : ContextField N) (j : Fin N), ((swapProtocol (basinIndex (momentContext N)) ⋯).postMeasure (μ12.prod (MeasureTheory.Measure.pi fun (k : Fin N) => epistemicMeasure (vertexPoint k))) i) ((fun (y : SwapArena (LF4.KSigma N) N) => y.1.1) ⁻¹' globalBasin c' j) = ENNReal.ofReal (c'.rate (vertexPoint i) j)
★ The Lüders update (rank-one), on the calibrated-swap witness. After outcome
i, the system behaves in every subsequent measurement as a fresh preparation ofeᵢ: for any context fieldc', the follow-up outcome-jprobability is the collapsed state's Born weightc'.rate [eᵢ] j.⚠️ Stated for the swap witness (
swapProtocol), where fields 1–5 concern the shear; the swap discharges the same five hypotheses (swap_correlates,swap_pointerInvariant), and it alone supplies collapse —shear_base_marginal_unchangedproves the shear cannot. Nondegenerate case only; seeSwapLuders.leanfor the full scope notes. - outcome_system_dependent (i j : Fin N) : i ≠ j → ∀ (s s' : LF4.KSigma N), basinIndex (momentContext N) s = i → basinIndex (momentContext N) s' = j → ∀ q ∈ readyArc N, (shearProtocol (basinIndex (momentContext N)) ⋯).readout ((shearProtocol (basinIndex (momentContext N)) ⋯).evolve 0 1 (s, q)) ≠ (shearProtocol (basinIndex (momentContext N)) ⋯).readout ((shearProtocol (basinIndex (momentContext N)) ⋯).evolve 0 1 (s', q))
★ Externality, the before half (A1, added 2026-08-27). Before the stroke a system-only transformation moving the selector across basins changes which outcome gets recorded: the outcome information is still in the system, and the stroke is what exports it to the register.
- record_system_invariant (f : LF4.KSigma N → LF4.KSigma N) (x : LF4.KSigma N × LF4.KTorus) : (shearProtocol (basinIndex (momentContext N)) ⋯).readout (f x.1, x.2) = (shearProtocol (basinIndex (momentContext N)) ⋯).readout x
⚠️ Externality, the after half (A2, added 2026-08-27) — vacuous by architecture, carried as documentation, not content: the displayed record is invariant under every system-side map because the readout reads the register factor only (
rfl-backed,readout_system_invariant). The contentful half isoutcome_system_dependent.
Instances For
★ The dynamical measurement closure holds — for every state, with no hypotheses about the dynamics.
★ Note what is absent from the hypotheses: CorrelatesOn and PointerInvariantOn do not appear.
ShearWitness discharged them from an explicitly constructed propagator, so this bundle rests on a
construction rather than on assumed physics. That is the difference between this and every earlier
record-layer bundle in the corpus.
The combining capstone: operational finite-QM closure and dynamical measurement.
⚠️ It asserts that both bundles hold. It does NOT assert they are about the same arena. The
operational closure lives on productDynamics over ℂℙ^M × T² for a composite system indexed by
Fin Nsub × Fin Nsub ≃ Fin (M+1); the dynamical one lives on Σ_sel × T²_R in dimension Nsys.
The parameter lists are disjoint, and that is not an accident of the encoding — it is the honest
state of the corpus. Unifying the arenas is the engine migration, and it is not done.
Read this as "both hold", not "one theory covers both". A capstone that bundled them while implying otherwise would be exactly the kind of claim this project keeps having to retract.
- operational : SigmaLayer.FiniteQMClosure H hH p₀ e ψ' hψ'0 ψ hψ0
The existing operational closure, untouched.
- dynamic : DynamicMeasurementClosure Nsys ψsys
Measurement as a process: records created, persistent, Born-weighted.
Instances For
The combining capstone holds.