QFT’s two formalisms read through ACT, and the complete picture:
what the theory claims, what is open, and what comes next
Part IV: Synthesis
First the picture, in one line: the electron is never a particle; it is one wave, and the particle is the mark a bath makes, in one history. QFT accommodates that picture naturally: its dual formalisms organize strikingly well around the anchoring transition. That is a pattern and a motivation — not independent evidence.
Quantum field theory has two mathematical languages. One describes structure: the field over all of spacetime, with no privileged time. The other describes states in time. A standard transformation connects them, and in an open system the bath sits exactly at that point.
ACT proposes this isn't coincidence — it's a clue. The formalism has been telling us that the wave and the records it leaves are two faces of one physics: the wave persists, and anchoring is a phase change in the record.
"Ontology recapitulates mathematics."
Quantum field theory is written twice. The action principle treats spacetime democratically; the Hamiltonian singles out a time and tracks a state through it.
The path integral sums over every field configuration at once. Space and time enter symmetrically in d⁴x: there is no special “now.” The field is described everywhere, not at a point, so extended configurations are the default. This is structure with no privileged time.
Time is a special parameter, and |ψ⟩ is the state at an instant, tracked forward. Observables have spectra of definite values, which is where recorded outcomes live. It describes superpositions and entanglement as well as the other language does; with a bath, the evolution runs forward only, as a semigroup. This is the language in which records are kept.
Neither language is “the particle language.” The field is a wave in both; what the second adds is a time to keep records in.
Where a time is picked out: a change of variables for a closed system, forced and directed with a bath.
Action principle
Path integrals
All configurations
Spacetime democratic
No privileged time
← Where a time is picked out →
Closed system: a change of variables
With a bath: forced and directed
ACT: the bath sits here
Time evolution
States in time
Eigenvalue spectra
Forward only (a semigroup)
Why “forced”: a damped system has no ordinary action principle, and putting the damping in by hand breaks the uncertainty relation as the motion dies away (Caldirola–Kanai; Brittin 1950). Only the bath’s noise restores it: damping without fluctuation is not a consistent quantum theory. Integrating the bath out (Schwinger–Keldysh) puts causal order and decoherence into the action together. An organizing analogy, not independent evidence.
What sounds mysterious in particle language is ordinary in wave language — a superposition is one wave written as the composition of its modes, and which decomposition matters physically is not your choice — the dynamics decide: Fourier modes for a free field, the alternatives the coupling holds stable at an apparatus.
Written over the two arms, the state looks like a photon that is here and there at once
In the mode basis it is one excitation of one two-lobed mode. Nothing here is two of anything
These are not two different "superpositions." They are the same wave, represented in different bases. A change of basis re-expresses the same wave; for a free field that change is a Fourier transform. A water wave can be written as a sum of sine waves. This doesn't mean the water "exists in multiple states simultaneously." It means the wave has a shape. That is what superposition and quantum states are: the composition of one wave's modes — Fourier modes in the free-field case, the alternatives the coupling holds stable at an apparatus.
One wave, its components. Many representations — not ontological multiplicity.
The uncertainty principle isn't about measurement disturbance. It's about wave structure.
A mathematical fact about Fourier transforms: A wave localized in position space is necessarily extended in momentum space, and vice versa. This has nothing to do with measurement disturbance — it's intrinsic to wave structure.
Every musician knows this: a sharp click (localized in time) contains all frequencies. A pure tone (localized in frequency) extends forever in time.
In the regime the selection theorem covers — records that form faster than the probe disperses — the bath's fragments support localization-type records, never coherent momentum kicks. There, position is what gets recorded, and the momentum spread is the Fourier structure of that same wave: mathematical complementarity becomes physical complementarity. Outside that regime the general map from modes to alternatives is open (the open problems, later in this lecture).
Every feature of QFT's dual formalism maps onto the distinction between structure and the records it leaves.
| Feature | Lagrangian: structure | Hamiltonian: states in time |
|---|---|---|
| Formulation | Action principle, path integrals | State vectors, time evolution |
| Time treatment | Democratic (no privileged t) | Privileged (∂/∂t singled out) |
| Natural entities | Extended field configurations | Localized excitations |
| "Superposition" | One wave, decomposed in a chosen basis | Amplitudes in an eigenbasis (still one state) |
| Complementarity | Fourier uncertainty (Δx·Δk≥½) | What the record architecture selects |
| Direction | None (structure, reversible) | With a bath: forward only (a semigroup) |
| Classical limit | Sum over all paths | Single classical trajectory |
| ACT ontology | Before anchoring: the wave is this | After anchoring: the wave plus one record — the particle is the mark |
ACT's proposal: the two formalisms, ordinarily read as mathematical convenience, organize naturally as the description of structure and the description of the records it leaves.
The electron is never a particle: one real field, atemporal, its modes its structure. When a mode couples to a quantum Brownian bath, with no observer needed, and the bath carries the record off for good, one mark is made: one record, one fact, in causal time. The mark is the particle. Anchoring does not reduce the wave. It adds a determination.
Stated plainly enough to be checked: the law, the weights, and what is added.
When: at release, at a rate set by irreversible record production — in the nondemolition case h = Φ̇irr, S = e−Φirr. What: the alternatives the record architecture supports (the selection theorem). Which: Born weights. Scope: CP-divisible record-forming regimes.
Forced by the one-wave law, one history, support certainty, no signalling for every division into here and elsewhere, refinable records and continuity — Zurek's argument in Barnum's no-signalling form; the martingale descent of the conditioned weights is the consistency check. Not from nothing: what is left is why there is chance at all.
One commitment: one history of marks is actual. That no event leaves two marks is a theorem for the systems the companion papers treat, and a further commitment beyond them. No new field, constant, threshold or noise; the global dynamics are unmodified. What sets ACT apart from the other accounts that keep quantum mechanics intact is not a different prediction but the account it gives.
ACT's claims are bounded by its own statements. That — not any anomaly — is the standard the theory asks to be judged by.
The comparison, feature by feature. Its first row is the root: where each account keeps a particle, or whether it keeps every outcome instead. Many Worlds shares ACT’s equation and a real wave that never collapses; the two part at the outcome: there every outcome happens, and the bath only draws the borders between the worlds. The flagship’s §9 puts the departure in one sentence: other interpretations ask how a superposed quantum state yields a definite result; ACT begins with a real wave and asks how the environment produces a definite event.
| Feature | Copenhagen | Many Worlds | CSL/GRW | Bohmian | ACT |
|---|---|---|---|---|---|
| Where is the particle? | Brought back at the detector, by a measurement never defined | No particle; a mark at every place it could land, one to a world | Squeezed out of the wave by a new law | Kept all along, on a hidden path | Nowhere: the particle is the mark the bath makes |
| Mechanism? | None | The bath only draws the borders between worlds | Added noise field | Pilot wave | The bath already present (open-system physics) |
| Energy conserved? | N/A | Yes | No (standard); dCSL finite | Yes | Yes (global dynamics unmodified; single-event ledger open) |
| Born rule? | Postulated | Contested: every outcome happens | Postulated | Contested | Forced: by one wave, one history, support certainty, no signalling and continuity (Zurek–Barnum); descent is the consistency check |
| New empirical predictions? | None | None | Yes (bounded by experiment) | None | None: equivalent to QM by construction |
| Bell nonlocality? | Outcome dependence, no account | Claims locality: Bell assumes one outcome per run | Outcome dependence | Parameter dependence (hidden level) | Outcome dependence, parameter independence, no signalling |
| What is added? | A measurement rule and a cut | Nothing: decohered components count as worlds | A noise field and two constants | Particle positions and a guidance equation | One postulate: one history of marks is actual |
| Realist ontology? | Anti-realist | Yes, of every outcome | Yes | Yes | Yes |
Every account here but the collapse models reproduces the statistics of quantum mechanics exactly; the collapse models depart from them, and survive only where the radiation searches have left their constants open. Where the experiments cannot choose, Occam’s razor decides: count what each adds to the Schrödinger equation and the bath the laboratory has measured, and what the addition costs.
| Account | What it adds | What it costs |
|---|---|---|
| Copenhagen | A measurement rule, applied at a measurement it never defines; a line between quantum and classical | No account of where the line falls |
| Bohm | A particle with a position at all times; a guidance law; an equilibrium hypothesis for the weights; for relativity, a preferred foliation of spacetime, which the wave may itself fix | Individual paths, and in relativistic versions a foliation, that no experiment can observe |
| Collapse models | New stochastic dynamics, with new constants (two in GRW and CSL) | Constants chosen by hand, steadily narrowed by the radiation searches |
| Many Worlds | Nothing in the equation | Every outcome counted as a world, with a copy of you in each and none reachable from another; probability rebuilt |
| QBism, relational | Nothing in the equation; outcomes relative to an agent, or to a system | No outcome that is a fact apart from an agent or a system |
| ACT | One postulate, that one history of marks is actual (exclusivity a second, outside the record-algebra class); two requirements on its law of chance, given refinable records | A postulate defended, not derived; release not yet stated covariantly |
Among the accounts in which an outcome is a fact apart from any agent or system, the count comes down to one comparison: ACT asserts one outcome for each run, the one the run shows; Many Worlds asserts that outcome and every other besides, the others in worlds no run can reach.
Lecture 8 in one slide: four thought experiments, each performed in the laboratory; one hundred experiments counted.
A slab that nothing reads kills the fringes, in proportion to its loss: a bath, not an observer, keeps the record.
The rate of record formation is a dial, and the event rate is read off the same coupling that draws the decoherence curve, with no free constant.
An apparatus picks components out of one wave, as a radio picks a station; only a record picks the outcome, and fifteen experiments mark where the event can sit.
Every detection is one mark, and no event leaves two; Born weights; the marks build fringes when no record of the path was kept and a particle pattern when one was.
Shared physics, established. Decoherence rates, matter-wave fringes to 172 kDa, delayed-choice statistics: shared with every interpretation that keeps the unitary dynamics.
The boundary, fixed. Event rules placed at first interaction, at decoherence of the reduced state, or at the first record carrier are ruled out. Release survives.
ACT, corroborated and not singled out. Its mechanism is measured at its foundation and its placement has never been refuted; no experiment chooses between it and the other unitary interpretations. It is as refutable as quantum mechanics: the excess radiation collapse models predict would refute it too, and has not appeared.
Five problems, ranked by what each one blocks.
Release is stated in a frame, against the system's accessible future. A form every observer agrees on is not yet written. Blocks the relativistic event law.
Proved with a detector in place. What fixes the accessible algebra for a dust grain in the microwave background, where nothing was built to record, is open; the alignment conjecture, that the record-selected alternatives coincide with the dynamically robust pointer states, is stated and unproved. Blocks full generality.
A structured bath can hand phase back late; Φirr may fall. The hazard is defined only in CP-divisible record-forming regimes. Blocks the law outside that scope.
The global dynamics conserve energy by symmetry. What one anchored event books, and where, has no ledger yet. Blocks trajectory-level accounting.
Proved for localization only. The general rule from a field's modes to the alternatives a record architecture can register is open. Blocks the what of the law beyond that class.
Each of these is a place where the law is not yet written in full — not a place where it has failed.
A research program, not a closed system.
Status of every claim, kept in one place: Claim Ledger.
Twice, a mathematical result that everyone treated as a device was read as a statement about what exists. ACT asks for the same reading of a third, on a far smaller scale.
E = hν → light comes in quanta
Planck’s formula (1900) fitted the blackbody spectrum and was taken as a calculation trick. Einstein (1905) read it as ontological: light really comes in packets. Quantum mechanics followed.
Brownian motion → atoms are real
Pollen jitter looked like noise. Einstein showed it is lawful molecular collision, and Perrin measured it. Macroscopic randomness from microscopic law: the same pairing of noise and damping that ACT’s bath carries.
The bath at the Legendre point → a mode enters time at anchoring
Damping without fluctuation is not a consistent quantum theory; with the bath, the passage to states in time is forced and directed. ACT reads it ontologically: before anchoring the field is not yet in time.
The reading stands or falls with the event law of Paper I, not with the analogy. The correspondence organizes the physics; it is not evidence for it.
ACT gives a physical account of the measurement problem using the bath already present in the world, with one primitive added: one record history is actual. No new field, constant, threshold or noise; the global dynamics are unmodified. QFT taken seriously.
The record picture names what quantum engineers already fight: a qubit keeps its coherence only while no released record of its state exists. The bath’s released records are the errors; error correction does not outrun release, it corrects released errors, and it succeeds as long as their rate stays below a threshold. ACT adds a reading of that physics, not a new tool.
ACT offers a realist reading of quantum physics. The field is real; the anchored record is a fact; anchoring adds a determination and does not reduce the wave. Consciousness plays no special role.
The wave is what exists.
The particle is what happens.
Measurement is the physical process
by which a wave leaves one mark.
No mystery. Nothing collapses. Just physics.
Ontology recapitulates mathematics.