Measurement: Forced Slot-Crossing and the Ontological Identity of the 4DD ρ-AND Closure Event
测量——强行越槽与 4DD ρ-AND 闭合事件的本体身份

Abstract

This paper gives an ontological characterization of quantum measurement: measurement is a 4DD ρ-AND closure event that forcibly gathers the ρ-OR capacity structure below the causal slot, drives it across the slot, and settles it into a single content record above. On this reading, "wavefunction collapse" is no longer an unexplained additional postulate but precisely this settling — irreversible along the causal axis. The paper distinguishes two irreversibilities that occur together in any measurement yet fall on different axes: decoherence belongs to the energy–thermodynamic axis and is in principle reversible at an energetic price; settling belongs to the information–causal axis and is irreversible because the causal arrow is one-way — no quantity of energy buys back its inverse. It then anatomizes the crossing as a span of operations — from scale, to direction, to memory, to closure — and on that basis reclassifies the quantum Zeno effect, Wigner's friend and Frauchiger–Renner, Kochen–Specker, and many-worlds as so many forms of a single confusion: mistaking pre-closure capacity (ρ-OR) for post-closure content (ρ-AND). As to the division of labor, the preceding paper supplied the Born weights of the alternatives at closure; this paper supplies the ontological identity of how a single content is generated; and the subsequent macroscopic stabilization of that content is left to later work. This paper changes none of the observable predictions of standard quantum mechanics, and offers no non-trivial prediction that would separate the framework from standard quantum mechanics in experiment; its weight lies at the ontological level — in identifying measurement as the bridge between the two sides of the causal slot.

Keywords: quantum measurement; wavefunction collapse; causal slot; 4DD ρ-AND closure; settling; decoherence; Born rule

---

Introduction

The measurement problem in quantum mechanics has long been treated as an intractable joint. This paper looks at it from a different angle: measurement is a bridge — a bridge spanning the two sides of the causal slot.

The thought can be compressed into two clauses: below the causal slot there is no causation; after the information lock there is information. Below the causal slot, multiple arrows of time coexist; there is no single before-and-after, hence no causation, and hence no information. Only when a single direction of time is locked in — which is what measurement does — do causation and information come to hold at once. The Information series of this work has already laid the groundwork for this logic: information is causation, and causation is the one-way arrow of 4DD. Measurement carries singular weight in the whole framework because the other papers treat, for the most part, what happens on one or the other side of the causal slot, whereas measurement treats how the two sides are joined — it is the crossing from below (no causation, the multi-tolerant ρ-OR) to above (causation, the single ρ-AND content).

The measure of this paper should be stated plainly at the outset. The empirical face of the measurement problem has been worked over for nearly a century; this paper changes none of the observable predictions of standard quantum mechanics, and erects no non-trivial prediction that would separate the framework from standard quantum mechanics in experiment. This is at once the reality of a mature field and the framework's own discipline of not manufacturing differences for their own sake — not a shortcoming. The weight of this paper lies one level down, in ontology: it re-grounds "wavefunction collapse," usually inserted into the theory as an unexplained postulate, as a causal-readout event with an ontological identity; it disentangles "decoherence" and "settling" as two things of different kinds; and, on this ontology, it reclassifies several paradoxes — Wigner's friend, Kochen–Specker, many-worlds — as so many forms of a single categorical confusion. As for the one possible empirical parting between the framework and standard quantum mechanics — a faint deviation of the discrete cell-tick at very short timescales — this paper names it, honestly, as a falsifiable commitment for the framework's future, and states plainly that it is not redeemed here.

The paper falls in three parts. Part I establishes the ontological identity of measurement: the causal slot as a boundary, the double influence the quantum is under, the closure event of forced slot-crossing, the internal anatomy of the crossing, and why settling is irreversible along the causal axis. Part II treats the mechanical face of measurement: what is enough to serve as a measuring apparatus, a tentative dynamical thread left for later, and this paper's neutrality, and its limits, toward the various interpretations. Part III reclassifies several well-known paradoxes by the ontology laid down in the first two parts.

---

Part I — The Ontological Identity of Measurement

1. Starting point: the two questions left by closure

The measurement problem can be split into two interlocking links. First: when closure occurs and the coexisting plurality of possibilities is gathered to a result, how does each result acquire its weight? Second: above those weights, why is it that a single definite result settles, rather than the plurality persisting? The preceding paper of this series, on the Born rule, treated the first link: when the 4DD ρ-AND closure occurs, the coexisting ρ-OR branches below the causal slot acquire weights according to $w_i = \psi_i^\sharp \psi_i$. This is the algebraic-projective face of closure — it answers "in what proportion do the results appear," but leaves two things outside the door.

First, what the closure event itself is. The weight formula characterizes the result of the projection, but not the event that makes the projection happen: what kind of event it is ontologically, why it is irreversible, and how it differs from the ordinary quantum evolution below the causal slot. Second, how a single result settles. The weights give the proportions of the several results, whereas what each actual measurement yields is one of them — not the proportion itself; from "a weighted set of possibilities" to "the one that settles," the preceding paper said nothing.

This paper continues from another conclusion of that one: closure unidirectionalizes the bilateral influence below the causal slot, selecting our side as the sole causal direction. This paper gives that unidirectionalization its ontological content — measurement is a causal-readout event that forcibly reads the ρ-OR capacity below the slot into a single ρ-AND content record; this event of "forced slot-crossing, settling the plurality into one" is what we shall call settling. Along this thread, the two opening questions can be answered together: what kind of event settling is; why it is irreversible and where its cost falls; and how it is distinguished from every operation that has not yet settled.

A word, at the outset, to mark this paper's jurisdiction. This paper treats the ontological identity of the closure event and how a single content record is generated; it does not treat how that record is thereafter amplified, redundantly copied, and stabilized into a macroscopically reliable record, nor how the environment-coupling-driven crossing unfolds dynamically — these belong to later work on decoherence and the classical limit. Nor does this paper explain "why it is precisely this result that settles": as set out below, this is fundamentally stochastic; what this paper gives is the mechanism by which a single result is generated, not a hidden law for which one is selected.

2. The causal slot and forced slot-crossing

To give measurement an ontological identity, one first needs a boundary. The paper of this series on the causal slot has set it up: below the causal slot lies the ρ-OR capacity structure — substrate plurality not yet formed into causal content; only across the slot is there content that can be transmitted, recorded, and entered into a causal chain. The quantum state therefore carries no content — superposition, entanglement, indeterminacy are all affairs of the substrate below the slot; never having crossed, they do not amount to causal content. This paper takes that as its ground, and does not restate it.

In what state, then, is the quantum below the slot? This must be seen from inside the layer where causal readout is active (written L₄ on the physical-quantity ladder, where 4DD resides). The work of this series on the four-fold pattern shows that such a layer divides internally into four steps: the emergence of the time-concept, the arrow of time, the direction of time, and information and causality (written L₄a through L₄d in order); and the causal slot falls precisely between the arrow (L₄b) and the direction (L₄c) — the arrow, below the slot, is multiple and coexisting, which is just the probability cloud; the direction, above the slot, is the single one that has been locked in. The fine anatomy of these four steps is left to the next section; here we take only the pair bearing directly on "what state the quantum is in": arrow and direction.

The quantum is thus under a double influence. In itself, below the causal slot it is governed by multiple arrows of time (L₄b) — there is no single before-and-after, hence no single causal direction, and hence no content. Yet at the same time it is overlaid, from above, by that single direction of time (L₄c), and carried forward with it — a wave packet's overall position drifts, ages along with the laboratory. This doubling is not two influences of equal standing, but a direct consequence of the one-way prior of the SAE dimensional sequence: the higher reaches the lower, the lower does not reach the higher. The settled direction L₄c overlays the quantum below the slot, while the multiple arrows of L₄b cannot reach up to disturb L₄c. The Schrödinger equation wears this doubling on its face: its dynamics are time-reversal symmetric (a direction with no preference — the L₄b side), yet it advances along a forward parameter $t$ (the side overlaid from L₄c). A long-standing puzzle — that Schrödinger evolution is definite and reversible, while the measurement result is indefinite — dissolves here as well: the definite-and-reversible side comes from the time-reversal symmetry of the evolution (the no-preference of L₄b) — a cloud is indeed translated bodily along the single direction set by L₄c, but its reversibility belongs to the time-symmetric evolution itself, not to the direction; the indefinite side is that the cloud's content at L₄b is many-arrowed to begin with, and has not yet settled. As for irreversibility, it is not in this translation; it enters only at settling.

Measurement, then, is the forcible settling-out of that below-slot side of the cloud: it takes the below-slot structure — many-arrowed by nature, unwilling to cross — forcibly locks it into a single one, drives it across the causal slot, and erects a single definite content above. To say "forcible" is not to say that someone exerts a force, nor to assert when the thing happens (the timing and triggering conditions are left for later); it is to say its directedness — it runs against the tendency of plurality to preserve itself, as friction runs against motion; one can say what it is without knowing when it begins. Placed back into this series' existing characterization of closure, this forced slot-crossing is a 4DD ρ-AND closure, an event of the swap class: it forcibly gathers a conjugate pair of coexisting orientations below the slot into one, and this is just what the preceding paper called unidirectionalization, at the microscopic level.

The confusion most apt to arise here is to take measurement and decoherence for one and the same — both are loosely called "quantum-to-classical," yet they belong to two distinct layers. Decoherence occurs naturally under environmental coupling: the system entangles with a vast number of environmental degrees of freedom, coherence is delocalized into the environment, and as far as a local readout of the system is concerned, the phase terms between superposed branches are erased, leaving a classical mixed state. But decoherence stops there — what it erases is the phase; what it leaves is a set of weighted possibilities; and it never picks out any one of them. In this paper's terms, it carries the system from the quantum ρ-OR (a coherent superposition) to a classical ρ-OR (a list, decohered yet still plural); and this step is itself already a kind of slot-crossing — the list thereby crosses the causal slot and is lifted into categorical information, no longer the quantum substrate below the slot but a discrete, classically treatable piece of information above it.

This brings out a distinction between two senses of "information." What decoherence makes by crossing the slot is categorical information — that classical list, still plural, undecided among its members; whereas what this paper repeatedly calls single content is another level, and is precisely the step proper to measurement — settling: forcibly gathering, out of that list, one definite content and recording it; this is the ρ-AND, the many into one. This distinction also joins this series' paper on the causal slot: that paper, from the standpoint of information theory, spoke of decoherence and slot-crossing as one affair — and for the purpose of laying down the boundary "above the causal slot there is information, below it there is none," that was enough; but the ontological identity of measurement demands a finer grain, and at this grain decoherence and slot-crossing are two distinct processes — decoherence is decoherence, slot-crossing is slot-crossing. So measurement is not a sped-up decoherence, nor decoherence a slowed-down measurement; what separates them is not speed, but the thing each does: one decoheres and leaves a list, the other settles one item and makes a content of it. What sets measurement apart from decoherence is the several things it has in addition — it draws the boundary of "what counts as a result," holds ready a place into which the settled content can enter a causal chain, and, as an active cell-aggregate coupling to the system, imposes that forcible closure on the system's ledger; this "grammar of readout" has nothing to do with any conscious intent. How decoherence spreads layer by layer in the environment, how the list forms, how the crossing is completed dynamically — all belong to later work on decoherence and the classical limit; this paper walks only the forcible road of measurement.

3. The anatomy of the crossing

Measurement is a bridge, and a bridge can be taken apart. The way it is taken apart follows the structure this series' work on the four-fold pattern repeatedly brings to light: first to fix a scale without constructing, then to accumulate a direction, to compound a history, and finally by closure to yield construct and remainder; this paper does not re-establish that structure, only puts it to use on this one crossing. Concretely, the layer where causal readout is active (the L₄ of the preceding section) unfolds internally along these four steps, and the process of crossing runs among them. A word first on the measure with which the paper uses it: whether some structure truly merits the name "four-fold pattern," the work on that pattern attaches several conditions that must each be met, and this paper has not verified them one by one; so the four-step division given here is registered only as a candidate, not as a settled conclusion. As for the complete fine-grained articulation of this division (which will involve the extension into the four-forces series), that too is left for later, and this paper does not fix its final placement.

The first step is the emergence of the time-concept (L₄a). It erects only the concept of "time" together with the Planck-time scale — it gives a "tick," but builds no structure and as yet has no direction; it marks without constructing.

The second is the arrow of time (L₄b). Upon the scale, addition accumulates time's pointing. But the arrow here is not single: below the causal slot it is multiple and coexisting — which is just why the probability cloud is a cloud, and just where the below-slot "no single before-and-after, hence no causation" resides.

The causal slot falls between the second step and the third: it locks the multiple arrows into a single direction. Before the locking there is no single before-and-after; after it, there is a definite before and after — and this single direction that is locked out is precisely the one that, in the preceding section's "double influence," overlays the quantum below the slot from above. What "after the information lock there is information" locks is just this: below the slot there is no single direction, hence neither causation nor information; once the direction is locked, causation and information come to hold at once.

The third step is the direction of time (L₄c). Once the direction is single, multiplication can, along this settled direction, compound and bind history layer upon layer into the present — and this is memory. Memory must depend on a single direction, for without a definite before-and-after there is no "past" to remember, nothing to recall; only along the locked direction does the prior state compound into the present and so become memory. That this step must come after the causal slot has its root just here.

The fourth step is information and causality (L₄d). This is closure: it yields at one stroke both "construct" and "remainder" — construct is information itself, a definite, established content; remainder is how this content is stored. With this, the settling is complete: it sits along the stretch from "the direction's being locked into one" to "closure's yielding information and remainder," and is, among these four steps, the one crossing that passes the causal slot and erects the content.

Looking back from here, what makes measurement the bridge between the two sides of the causal slot is just these four steps: the time-concept sets its tick, addition gives it pointing, the causal slot locks its direction, upon the direction multiplication compounds memory, and closure fixes its information and storage. The below-slot's no-causation and no-information, and the above-slot's causation and information, are joined by this one crossing. This also gathers the whole paper's thesis into a single sentence: below the causal slot there is no causation; after the information lock there is information — and measurement is just that lock.

4. Settling: the irreversible causal readout

Settling is irreversible; but to state this irreversibility rightly, one must first see that it is not of one source with another irreversibility. When measurement happens, decoherence and settling proceed together, each bringing a "no-going-back," and these two no-going-backs fall in different places.

The no-going-back of decoherence is thermodynamic. Coherence is delocalized into countless degrees of freedom as the system entangles with the environment; in principle, to gather these scattered phases back up and realign them is forbidden by no law — spin echo and photon echo are just instances of paying a price, on small systems, to pull coherence back. Only, to do this in a macroscopic environment is to make an already-spread correlation re-gather, a local decrease of entropy, which by the second law must be paid for by a greater increase of entropy elsewhere together with work; once the environment is large, this bill cannot in practice be paid. So decoherence's "irreversibility," put precisely, is: reversible in principle, but requiring an injection of net energy and a greater increase of entropy, hence not to be undone in practice. This is the irreversibility of the second-law kind, the thermodynamic arrow.

The no-going-back of settling is another matter, falling on the information–causal side, and is just the kind this paper claims. Settling reads the ρ-OR capacity below the slot into one ρ-AND content — it erects a causal connection. To invert it would be to invert a causal connection already in force, to send the effect back before the cause — and the arrow of causation is one-way. This no-going-back has nothing to do with energy: no quantity of energy assembles such an inverse operation. Why so, can be seen from two points. First, settling does not read out an answer already written below the slot, but generates that content (as detailed below); to invert it one would have to conjure back, out of nothing, a plurality that did not exist. Second, this series has already placed causation as a forward property of 4DD — information is causation, and causation is that one-way arrow — and settling is precisely the event that selects the single causal direction and erects this arrow; its irreversibility is therefore the irreversibility of the arrow itself.

With the two side by side, the cleanest way to tell them apart is to ask one question: can energy buy back its inverse? Coherence lives on the energy side; energy buys it back. Causal content lives on the causal side; energy does not. They are of different kinds, yet occur simultaneously in one and the same measurement — decoherence spreads out a list, settling picks one item from it; the former reversible at an energetic price, the latter irreversible and squared with energy not at all. What this paper claims is only the latter; the thermodynamic irreversibility of decoherence, and the finite time that process itself takes, are not things this paper sets out to explain or to shoulder.

This also disposes, in passing, of settling's cost. Below the slot there is no ready-made content; to obtain one definite causal content, one must pay this single irreversible causal readout itself — the cost lies just here, and it falls on the causal side, where energy does not redeem its inverse. This does not conflict with conservation: the remainder is not annihilated, only discharged from the ρ-OR form into the form of a single ρ-AND content together with a classical record; the total is unchanged. As to whether, and how, this event corresponds on the energy-and-heat ledger to some number — the Landauer heat for erasing one bit, say — that is a question on the other side; this series has a dedicated paper on it, whose conclusion is conditional (the bridge between statistical temperature and the SAE geometric quantity is not yet closed), and a macroscopic erasure of one bit is in fact the statistical aggregate of a great many substrate events spanning some thirty orders of magnitude, not the same thing as the single event of one settling. So the numbers at the energy level this paper leaves for later, and asserts here only one thing at the ontological level: content does not come free.

One must also distinguish settling's "whether-it-is-done" from that process's "how-long-it-takes." Settling is binary: a definite content has either been generated or not; there is no such thing as "half a content." This does not mean settling is timeless magic — this series' paper on the finite causal slot has shown that causal settling takes a finite time, which is the floor cost of a system's completing one settling, not an externally imposed cutoff. But duration is an affair of the process side, of the energy axis; settling's being all-or-nothing speaks of whether the settling is complete (whether the content has been generated), and how long completing it takes is another matter, which together with the dynamics of the crossing belongs to later work.

With this one can return and place the term "wavefunction collapse." Standard quantum mechanics calls the sudden jump after measurement collapse, and lodges it in the theory as an additional postulate without explanation. This paper does not deny collapse as a phenomenon; it supplies its ontology: what is called collapse is just this settling, irreversible along the causal direction — the remainder discharged from the ρ-OR form into a single ρ-AND content, the total conserved, the arrow of causation thereby fixed. Collapse is then no longer an externally imposed postulate, but a causal-readout event with an ontological identity.

5. Approach and settling

To set settling up as an irreversible event at once raises a class of apparent counterexamples: weak measurement, POVMs, premeasurement, even the quantum eraser — all couple to the measured system, yet are afterward often reversible, undoable, and need not destroy entanglement. If settling is truly irreversible, where are these to be placed?

They have not settled at all; they only approach settling. The coupling is too weak to gather that one content out and drive it across the causal slot, and the process halts short of settling. They are reversible precisely because there is as yet nothing settled to speak of — what has not come to pass can of course be undone. So far from contradicting "settling is irreversible," these phenomena confirm it: once there is real settling, once the content is really generated, reversibility is gone; the reversible, by its very meaning, is still short of settling. The boundary here is one of kind, not of degree — not "a weaker settling," but "not yet settled"; between settling and approach there is no continuous strong-weak spectrum on which to lodge a "half-settling."

As for the various anomalies read off within the approach region — weak values lying outside the eigenvalue spectrum, for instance — this paper passes no comment: those belong to the phenomenology of "what exactly is read where nothing has yet settled," bearing on the contested ontological status of weak values themselves, and may be left for later work and related studies. Here this paper sets only one boundary of kind: settling, and the approach toward it. Just where the approach must accumulate before it crosses, and how the crossing happens dynamically, likewise belong to later work.

6. Generation, not selection

Settling generates a content — and this "generates" matters. One must keep apart what lies below the slot from what stands above. Below is capacity, in the broad sense: real, the plural ρ-OR structure, yet not a definite answer. Above is content, in the narrow sense: a definite record that can enter a causal chain — and it is generated at the moment of crossing, not drawn out from something already written below and now merely retrieved. What measurement brings up is capacity read into content, not a hidden answer-sheet turned over.

From this, what the paper delivers must be set out clearly: what it gives is how a single content is generated — how closure gathers the plurality and erects exactly one definite record; it does not give a hidden law for "why precisely this one." Settling does generate one definite content, but which one it settles is ontologically undetermined, and this undetermination is fundamental, not a matter of our being uninformed. It is not some yet-to-be-discovered hidden variable working behind the scenes — before closure there is, below the slot, no single causal direction, hence no content written in advance awaiting disclosure; it is closure, that very event, that selects the direction and generates the content. Thus the randomness of the single result is randomness at the ontological level, fully compatible both with the conclusion that there are no factorizable local hidden variables and with the conclusion that one cannot assign a global definite value to the pre-closure state.

The point is therefore a pair, not a single thing: the whether of content is generated all at once by settling, definite and irreversible; the which one of content is fundamentally random, with no law to follow. To run these two together is just the source of the error "a hidden variable selected the result"; what this paper sets up is the generation of the former, not the selection of the latter.

7. What a readout can read

Since settling is a readout, it has a limit on what it can and cannot read. What is read lies below the causal slot, before closure; the event of reading is settling itself. What matters is to be clear about what this one reading can hand over.

What it hands over is one definite result — a content at the eigenvalue level, a token constrained by the weights. What it cannot hand over is the full ρ-OR complex distribution below the slot, that is, the state vector itself: a single 4DD readout cannot package the whole distribution and lift it up; it can only forcibly settle one item out of it. A single definite eigenvalue and "the whole distribution cannot be read" do not conflict — the former is one content made by this one settling, the latter says that no single readout can convert the whole plurality into content. As for the expectation value, it is not the object of any single settling's readout, but a statistical functional over the several results, by their weights, across a great many repetitions; it lives at the level of repetition, not in the single event.

Set back into the division of labor before and after, this is clearer still: the weights of the coexisting results were given by this series' paper on the Born rule; how a single content settles and is generated is this paper's affair; and how that content is thereafter amplified, made redundant, and stabilized into a macroscopically reliable record belongs to later work. What this paper stands on is the middle link — the generative identity of a single content.

8. No-signaling

Settling is a local event, and this bears on the compatibility of relativity with quantum mechanics, and must be set out on its own. Place it first in the standard form of measurement: let Alice and Bob share an entangled pair, and let Alice perform a measurement at her end. If her result is not selected (a non-selective measurement), Bob's reduced state is

$$\rho_B' = \mathrm{Tr}_A\!\Big[\sum_a (M_a \otimes I)\, \rho_{AB}\, (M_a^\dagger \otimes I)\Big] = \mathrm{Tr}_A\, \rho_{AB} = \rho_B,$$

independent of whether Alice measures or not, and of which measurement she makes. Bob's marginal statistics do not change with anything Alice does — this is the standard statement of no-signaling, which this paper takes over, proving no new theorem of.

What this paper adds is why it is so, ontologically. Alice's settling, at her end, reads capacity into content locally — a longitudinal, on-the-spot generative event; it sends Bob no below-slot content across. What could be sent would have to be content, and below the slot there is no content to send. What, then, is the unmistakable correlation between Alice and Bob? It is real, yet it lives at the level of capacity — a shared, non-factorizable ledger, not a message awaiting transmission. Being non-factorizable, it is not a hidden property local to each particle; being no message, it constitutes no transmissible signal. So two things keep their places: the correlation is non-local (one ledger borne by both ends), the signal is local (content generated on the spot at each end). The moment Alice settles, her end has content, while Bob's marginal does not stir a hair — the crossing is on-the-spot and longitudinal, and the transverse boundary of the causal slot still holds.

This joins precisely Einstein's unease over "spooky action at a distance": what is at a distance is only the correlation shown when the pre-shared capacity is read out at each end, not any content or action transmitted across the distance. Here the paper closes — it supplies the ontological face of no-signaling, and joins it to the existing formal statement, without erecting any new doctrine.

---

Part II — Apparatus, Mechanism, and Interpretation

9. What is enough to count as a measuring apparatus

It was said above that measurement draws the boundary of a result, holds a place for content, and imposes a closure on the system's ledger — which raises a question: what is enough to serve as a measuring apparatus?

The answer lies not in "consciousness," nor in "the macroscopic," but in a structure. A measuring apparatus is an active cell-aggregate — on this series' ladder, the kind of aggregate in which the mass layer and the causal-readout layer (L₃ and L₄) are active together. It can measure because it can at once draw, for the measured system, the boundary of "what counts as a result," hold ready a place into which the settled content can enter a causal chain, and, when coupled to the system, as an active aggregate trigger the L₄ reactivation at the system's end and impose that forcible closure on the system's ledger. Only so is measurement what it is.

Two common misidentifications can thereby be cleared. First, measurement is not brought about by consciousness: what imposes the closure is the apparatus as an active aggregate, not any observer's awareness; an instrument, a photographic medium, even a sufficiently coupled stretch of environmental structure — any of these, having this set of capacities, is enough to serve, with no one needing to "look." Second, measurement need not be a macroscopic large object: what matters is whether this set of capacities is in place, not size; between minute systems, so long as one constitutes such an active aggregate toward another, settling can occur.

This paper stops here at a single closure event — one apparatus imposing one settling on one system, erecting one content. How that content is thereafter amplified layer by layer along the apparatus, redundantly copied, and stabilized into a macroscopically readable pointer is a chain of subsequent closures, belonging to later work on macroscopic stabilization, into which this paper does not enter.

10. A thread left for later: the calibration remainder

Closure yields not only a construct but a remainder (§§3–4); does that remainder leave a discernible mark on the L₁ calibration? Here there is a thread, which this paper only hangs out, and does not redeem.

In this series' work on the L₁ calibration, closure is accompanied by a change in calibration. Carried over to measurement, one might conjecture: the occurrence of one settling may leave a remainder on the L₁ calibration — the erecting of a content must enter, somewhere, on the calibration ledger. But here one must halt several steps short. This paper derives from it no relation of the energy–time uncertainty form; asserts no claim that all measurements carry one and the same energetic kick; nor does it run this remainder together with the uncertainty relation given by this series' paper on ℏ and symplectic conjugation. It is at most a direction: should there in future be weak-measurement experiments of sufficient precision able to discern, within the approach region, an energy flow accompanying settling, only then would this thread be redeemable, at which point it could be rejoined to the work on L₁ calibration.

The reason for putting the matter this carefully is that this paper keeps to the limit of not over-drawing on its dynamical premises: the calibration remainder is a reasonable direction of inference, but to write it as a conclusion before a solid dynamical demonstration is in hand would overstep this paper's jurisdiction. So this paper sets it up only as a thread awaiting test, leaving its concrete dynamics, together with its possible bearing on weak-measurement anomalies, for later.

11. Neutrality, and its limits, toward interpretations

What this paper gives is the ontological level of measurement, which sits beneath what is usually called "the interpretation of quantum mechanics." Toward a fair range of interpretations, this paper is neutral: any that grant measurement an irreducible closure event — the Copenhagen interpretation, Bohmian mechanics, objective-collapse theories taken as physical process, Penrose's gravitational-collapse proposal — may, upon this deeper ontology, each unfold its own phenomenological description. This paper adjudicates not among them, only points out what the closure event they share is, ontologically.

This neutrality has its lineage. This series' paper on the causal slot has shown that SAE takes a classicality boundary aligned with Copenhagen, and is incompatible with the Everett-style many-worlds reading; this paper follows from that, and erects nothing of its own.

But neutrality has its limit, and past the limit it is no longer neutral. Should some interpretation hold that measurement contains no irreducible closure at all, that everything can be accomplished by unitary evolution plus an updating of knowledge — many-worlds carried through to the end, or quantum Bayesianism carried through to the end, is of this kind — then it is incompatible with the 4DD ρ-AND closure identity this paper sets up. It must be said plainly: this incompatibility belongs to the level of ontological structure, not to what experiment can adjudicate — "everything is really unitary, we just never see it" and "there is indeed a non-unitary closure" are the two faces of the measurement problem itself, and no experiment can decide between them. Here this paper does not pretend to hold an empirical criterion; it only states its ontological position outright — it requires that irreducible closure, and so parts ways with readings that abolish closure altogether.

---

Part III — Reclassifying Several Paradoxes

This series' ontological characterization of measurement loosens, in passing, several famous paradoxes of quantum mechanics. A declaration first: this paper does not claim to "solve" them — to solve is, often enough, to give a self-consistent account within some one interpretation; what this paper does is another thing, namely to point out that these paradoxes mostly arise from one and the same categorical confusion, and that once that confusion is exposed, the paradoxes dissolve of themselves. The common confusion is this: to take the pre-closure capacity (ρ-OR, plural, unitarily evolvable) for the post-closure content (ρ-AND, the one that has settled), or the reverse. The four below are each a form of this confusion.

12. The quantum Zeno effect: dissolving a pain point

The quantum Zeno effect says: to measure a system frequently drags out, even freezes, its evolution. It first teaches us something positive — that measurement is no passive looking-on. Were measurement only the reading-out of a value already fixed, frequent reading should not change the system's course; yet the Zeno effect shows that frequent measurement does change the course. This squares with this paper: each measurement is a forcible intervention on the system (a settling, or an approach toward settling), and to intervene frequently rewrites the evolution. In this respect the Zeno effect can be reclassified as the accumulation, in the system's evolution, of high-frequency settlings and approaches.

But one must halt here, and not take it as this paper's evidence. The qualitative phenomenon "frequent measurement drags out evolution" is accountable on almost every interpretation; to enlist a phenomenon everyone can account for as confirmation of some one ontology is the spurious support of affirming the consequent. Moreover, frequent measurement does not always drag out: in other circumstances it instead accelerates decay — the anti-Zeno effect. Whether the net effect is suppression or acceleration depends on the spectral details of the system's coupling to the environment, which are likewise common to all parties and have nothing to do with the ontology this paper sets up; this is left for later. Here this paper makes only one conservative reclassification, and asserts nothing about "frequent measurement necessarily freezing."

Were this paper's ontology truly to show a fingerprint differing from standard quantum mechanics in the Zeno region, the one possible place for it is a faint deviation left by the discrete cell-tick, at very short measurement intervals, in the relevant spectral overlap. This place this paper names as a falsifiable commitment for the framework's future, and states plainly that it is not redeemed here.

13. Wigner's friend and an impossibility theorem

Wigner's friend is this thought: the friend, in a sealed laboratory, makes a measurement on a system and obtains a result; Wigner, outside the room, describes "friend plus system" as a whole as a not-yet-collapsed unitary superposition. The question then is: does the friend's result count as a fact? Frauchiger and Renner showed further that, if one insists each observer apply quantum mechanics without reservation to other observers, several seemingly natural assumptions jointly lead to a contradiction.

In this paper's view, the root of the contradiction is to mistake an already-settled content for a still-unitarily-evolvable capacity. The friend made a measurement — that is, a closure occurred: a definite content (ρ-AND) has been erected in the friend's laboratory, and this is objective, unchanged by whether Wigner outside has interacted with it. Wigner's redescribing the friend as a coherent superposition amounts to treating this already-settled content, regressively, as the plural capacity (ρ-OR) before closure — and this step is just the categorical misplacement. Closure is one-way: once content has settled out of capacity, it is no longer capacity; no higher observer can unitarily "send it back" into a superposable capacity. This one-way boundary is the firewall this paper erects for the paradox.

Thus, what this paper judges to overstep in that impossibility theorem is precisely the assumption "apply unitary evolution without reservation to a completed measurement." It must be said that which of the Frauchiger–Renner assumptions is the culprit is still contested — there are readings that lay the blame on the assumption of consistency among observers instead; this paper's judgment follows the SAE ontology of one-way causation, and is one reading among several, not the sole verdict. On this reading, the paper keeps single outcomes, and keeps consistency among observers, restricting only that one assumption: unitary quantum mechanics describes the capacity before closure, not the content after. This dissolution does not clash with the various relational readings, yet need not appeal to "facts are only relative to observers" — here, content, once settled, is objective.

14. Kochen–Specker: the naturalness of contextuality

The Kochen–Specker theorem shows: in three dimensions or more, one cannot assign to all observables at once a set of definite values independent of measurement context. This is often read as a strange thing — that the quantum's several observables cannot all hold definite values in advance.

But here it is exactly to be expected. Before closure, below the slot there is only capacity, no content; and with no content, there is of course no set of globally definite values written in advance, awaiting disclosure. A value is not written in advance, but generated at the moment of closure, along with "which closure occurred" — and this generation depends on the context of settling, and is by its nature contextual. So Kochen–Specker contextuality is no anomaly of the quantum world, but another way of saying "there is no content written before closure" (which is just what §6 set up: what this paper delivers is the generation of content, not some hidden law for which one is selected). To read it as strange is still to presuppose that an answer-sheet awaiting disclosure is hidden below the slot; once one sees that there is no answer-sheet below the slot, the strangeness is gone.

15. Many-worlds: the branches are capacity, not content

The many-worlds interpretation holds: the universal wavefunction never collapses, every branch within the superposition is equally real, each a world of its own.

In this paper's view, those "branches" are the capacity before closure (the plurality of ρ-OR), not the content after (the one of ρ-AND). To take the branches for so many real worlds is to reify capacity into content — once more the misplacement of capacity and content. Here one further boundary, of causation, may be added: for a branch to qualify as a "world," it must have its own causation, its own content; yet before closure, below the slot there are multiple arrows of time coexisting and no single causal direction, hence no content to speak of. The single causal direction and that one content are generated only by closure, and closure occurs but once, settling but one item. So those branches, halting at the level of capacity, fall short of the threshold of "world."

It must be restated, by the boundary set earlier (§11), that this is the paper's ontological position, not a criterion that could adjudicate a win or loss against many-worlds in experiment — "everyone really did branch, the branches just never meet again" and "only one settled" are, empirically, no different. This paper only points out: in the SAE ontology, the branches are capacity, not content, and to read them as real worlds is to cross the boundary between capacity and content.

---

References

Self-as-an-End series

Other papers of the Self-as-an-End series referred to in this work, listed by the in-text handle used.

1. Qin, H. SAE Quantum Mechanics Paper VI (the Born rule). Zenodo, 2026. DOI: 10.5281/zenodo.20479658. — Cited in §1 and §7 as "this series' paper on the Born rule"; supplies the closure weights $w_i = \psi_i^\sharp\psi_i$. This is the immediately preceding paper.

1. Qin, H. SAE Information Theory III: The Causal Slot of Information — A 4DD Ontology from Quantum Fluctuation to Thermal-Floor Minimum. Zenodo, 2026. DOI: 10.5281/zenodo.19797457. — Cited in §2, §6, §11 as "this series' paper on the causal slot"; the causal-slot boundary, and the §2.4 / §6 reading of decoherence as a categorical lift across the slot.

1. Qin, H. Methodology Ten (the four-fold pattern). Zenodo, 2026. DOI: 10.5281/zenodo.20187591. — Cited in §3 as "this series' work on the four-fold pattern"; the marked-not-constructed → additive (direction) → multiplicative (memory) → closure (construct + remainder) structure and its adequacy conditions.

1. Qin, H. SAE Foundation v2: Systematic Restatement of the Physical-Quantity Ladder and Signature Discipline. Zenodo, 2026. DOI: 10.5281/zenodo.19361950. — Cited in §2 and §9 for the physical-quantity ladder L₀–L₅ and for L₄ as the causal-readout-active layer.

1. Qin, H. SAE Information Theory I (the 4DD ontology of information; information-conservation axiom). Zenodo, 2026. DOI: 10.5281/zenodo.19740019. — Cited in §4 for "information is causation, and causation is the one-way arrow of 4DD."

1. Qin, H. SAE Information Theory II (a structural derivation of Landauer's principle). Zenodo, 2026. DOI: 10.5281/zenodo.19780314. — Cited in §4 as the "dedicated paper" on the Landauer cost; its conditional status and the thirty-orders-of-magnitude span.

1. Qin, H. SAE Thermodynamics Paper VI (τ_dec, the finite time cost of causal settling). Zenodo, 2026. DOI: 10.5281/zenodo.19658595. — Cited in §4 as "this series' paper on the finite causal slot"; that causal settling takes a finite time.

1. Qin, H. SAE Quantum Mechanics Paper III: ℏ as the L₁↔L₂ Symplectic-Conjugation Closure Signature of the SAE Physical-Quantity Ladder. Zenodo, 2026. DOI: 10.5281/zenodo.20340595. — Cited in §10 as "this series' paper on ℏ and symplectic conjugation"; the uncertainty relation $\Delta x\,\Delta p \ge \hbar/2$.

1. Qin, H. SAE Quantum Mechanics Paper V (the L₁ calibration; the closure-accompanied calibration change). Zenodo, 2026. DOI: 10.5281/zenodo.20455398. — Cited in §10 as "this series' work on the L₁ calibration"; the basis of the tentative "calibration remainder" thread.

Framework lineage, presupposed but not directly cited: QM Paper I, DOI 10.5281/zenodo.20252029; QM Paper II, DOI 10.5281/zenodo.20277037; Mass-series convergence, DOI 10.5281/zenodo.19510868; Relativity Paper IV, DOI 10.5281/zenodo.20079718; Information Paper VII ("The Spark of Life"), DOI 10.5281/zenodo.20105883.

External references

Standard literature for the works named in §§5, 8, and 12–15. Volume/page details and final formatting are at the author's discretion.

• Born, M. (1926). Zur Quantenmechanik der Stoßvorgänge. Zeitschrift für Physik 37, 863.
• Einstein, A., Podolsky, B., Rosen, N. (1935). Can quantum-mechanical description of physical reality be considered complete? Physical Review 47, 777.
• Bell, J. S. (1964). On the Einstein Podolsky Rosen paradox. Physics 1, 195.
• Kochen, S., Specker, E. P. (1967). The problem of hidden variables in quantum mechanics. Journal of Mathematics and Mechanics 17, 59.
• Misra, B., Sudarshan, E. C. G. (1977). The Zeno's paradox in quantum theory. Journal of Mathematical Physics 18, 756.
• Wigner, E. P. (1961). Remarks on the mind–body question. In The Scientist Speculates (I. J. Good, ed.).
• Aharonov, Y., Albert, D. Z., Vaidman, L. (1988). How the result of a measurement of a component of the spin of a spin-½ particle can turn out to be 100. Physical Review Letters 60, 1351.
• Hahn, E. L. (1950). Spin echoes. Physical Review 80, 580.
• Joos, E., Zeh, H. D. (1985). The emergence of classical properties through interaction with the environment. Zeitschrift für Physik B 59, 223.
• Zurek, W. H. (2003). Decoherence, einselection, and the quantum origins of the classical. Reviews of Modern Physics 75, 715.
• Schlosshauer, M. (2007). Decoherence and the Quantum-to-Classical Transition. Springer.
• Kofman, A. G., Kurizki, G. (2000). Acceleration of quantum decay processes by frequent observations. Nature 405, 546.
• Frauchiger, D., Renner, R. (2018). Quantum theory cannot consistently describe the use of itself. Nature Communications 9, 3711.

引言

量子力学的测量问题，向来被当作一处难解的关节。本文换一个角度看它：测量是一座桥——一座架在因果槽两侧之间的桥。

这条思路可以收成两句话：因果槽下无因果，信息锁后有信息。因果槽以下，多重的时间箭头并存，没有单一的先与后，因而没有因果，也没有信息；唯有把单一的时间方向锁定下来——这正是测量所做的——因果与信息才一并得以成立。本系列论信息的诸篇已把这套逻辑的根基立好：信息即因果，而因果是 4DD 那条单向的箭头。测量之所以在整个框架里份量独重，是因为别的篇章讲的多是因果槽某一侧的事，而测量讲的，是两侧如何接通——它是从槽下（无因果、多容的 ρ-OR）到槽上（有因果、单一的 ρ-AND content）的那一跨。

须把本文的分寸先说明白。测量问题的经验一面，已被磨砺了将近一个世纪；本文不改动标准量子力学的任何一条可观测预言，也不另立一个能在实验上把本框架与标准量子力学分开的非平凡预测。这既是这一成熟领域的实情，也是本框架「不强行制造差别」的本分，而非短处。本文的分量在本体一层：它把通常作为一条无解释公设搁入理论的「波函数坍缩」，重新接地为一桩有本体身份的因果读出事件；它厘清「退相干」与「落定」本是两件种类不同的事；它也借这套本体，把维格纳的朋友、科亨–施佩克尔、多世界等几桩佯谬，重新归类为同一个范畴混淆的不同样式。至于本框架与标准量子力学之间唯一可能的经验分野——离散元胞节拍在极短时标上的微小偏离——本文据实命名为框架未来的可证伪承诺，明言不在本篇兑现。

全文分三部分。第一部分立测量的本体身份：因果槽这条边界、量子所受的双重影响、强行越槽这桩闭合事件、过槽的内部解剖，以及落定何以在因果的箭头上不可逆。第二部分谈测量的机制一面：什么东西足以充当测量装置、一处有待后续的动力学线索，以及本文对各种诠释的中立与不中立。第三部分把几桩著名的佯谬，按前两部分立下的本体逐一重新归类。

第一部分　测量的本体身份

一、起点：闭合留下的两问

测量问题可以拆成先后相扣的两环。其一，当闭合发生、并存的多重可能性收束为结果时，各结果如何获得自己的权重。其二，在这些权重之上，为什么最终落定的是一个确定结果，而非多重性继续维持。本系列关于玻恩规则的前篇处理了第一环：当 4DD ρ-AND 闭合发生，因果槽以下并存的诸 ρ-OR 分支依 $w_i=\psi_i^\sharp\psi_i$ 获得权重。这是闭合的代数投影面——它回答「各结果以何种比重出现」，却把两件事留在了门外。

其一，闭合事件本身是什么。权重公式刻画了投影的结果，没有刻画那桩使投影发生的事件：它在本体上属于何种事件，凭什么不可逆，又与因果槽以下寻常的量子演化区别何在。其二，单一结果如何落定。权重给出的是诸结果各自的比重，而每一次实际测量得到的，是其中一项，不是这份比重本身；从「一份带权重的可能性」到「落定的那一项」，前篇未论。

本文从前篇的另一结论接续：闭合把因果槽以下的双侧影响单向化，选定本侧为唯一的因果方向。本文为这一单向化给出本体内容——测量，是一桩把槽下 ρ-OR 容量强行读出为单一 ρ-AND content record 的因果读出事件；这一「强行越槽、将多重性落定为一项」的事件，下文称为落定。沿此线索，开篇两问可一并作答：落定是何种事件，它为何不可逆、其代价落在何处，以及它与一切「尚未落定」的操作如何区分。

末了须先划清本文辖区。本文论述闭合事件的本体身份，以及单一 content record 如何生成；不论述该 record 此后如何被放大、冗余复制、稳定为宏观可靠的记录，也不论述环境耦合所驱动的跨越在动力学上如何展开——这些属于后续关于退相干与经典极限的工作。本文亦不解释「为何落定的偏偏是这一个结果」：如后文所申，这一点是根本随机的；本文给出的是单一结果如何生成的机制，而非选定哪一个的隐藏规律。

二、因果槽与强行越槽

测量在本系列里份量独重，因为它是因果槽两侧之间的桥——从槽下到槽上的那一跨。要看清这座桥，先得有它所跨越的边界。本系列论因果槽的一篇已把这条边界立好：因果槽之下，是 ρ-OR 的容量结构，是尚未成形为因果内容的衬底多容；唯有越过此槽，才谈得上可传播、可记录、可进入因果链的 content。量子态因此并不携带内容——相干叠加、纠缠、不确定，都是槽下衬底的事，未曾越槽，也就谈不上是因果内容。本文以此为地基，不再重述。

那么槽下的量子，究竟处在怎样一种状态？这要到因果读出活跃的那一层（在物理量阶梯里记为 L₄，亦即 4DD 所在）的内部去看。本系列论四分形的工作表明，这样一层在内部又分作四道：时间概念的出现、时间的箭头、时间的方向、以及信息与因果律（依次记为 L₄a 至 L₄d）；而因果槽，恰落在箭头（L₄b）与方向（L₄c）之间——箭头在槽下，是多重而并存的，这正是概率云；方向在槽上，是锁成的单一一条。这四道的细致解剖留待下一节，此处先取其中与「量子处于何种状态」直接相关的一对：箭头与方向。

量子于是受着双重的影响。就其本己而言，它在 L₄b 受多重的时间箭头支配——没有单一的先与后，因而没有单一的因果方向，也就没有内容。但与此同时，它又被 L₄c 那条单一的时间方向自上而下地笼罩着，随之向前——一个波包的整体位置会平移，会与实验室一道「变老」。这双重并非两股对等的影响，而是 SAE 维度序列那条单向先验的直接后果：高层及于底层，底层不及于高层。L₄c 笼罩着槽下的量子，而 L₄b 的多重箭头上不去，干扰不到 L₄c。薛定谔方程恰把这双重写在脸上：它的动力学时间反演对称（无所偏好的方向，即 L₄b 那一面），却又沿着一个向前的参数 $t$ 推进（L₄c 笼罩下来的那一面）。一个由来已久的困惑——薛定谔演化明明确定而可逆，测量结果却不确定——也于此涣然：确定而可逆的那一面，来自演化的时间反演对称（L₄b 的无所偏好）——一朵云固然沿 L₄c 给定的单一方向被整体平移，可逆系于这时间对称的演化本身，而非系于方向；不确定的，是这朵云在 L₄b 的内容本就是多箭头、尚未落定。至于不可逆，并不在这平移里，要到落定才进来。

测量，就是对这朵云槽下那一面的强行了断：把本受多重箭头、不肯越槽的槽下结构，强行锁定为单一，顶过因果槽，在槽上立起一项确定的内容。说「强行」，不是说有谁在施力，也不是断言这件事何时发生（时机与触发的条件留给后续）；说的是它的方向——它逆着多容性自我保留的倾向而行，一如摩擦逆着运动，不必知其何时起，便可言其为何物。把这桩越槽放回本系列对闭合的既有刻画，它就是一次 4DD 的 ρ-AND 闭合，一桩 swap 类的事件：将槽下并存的一对共轭取向强行收束为一，而这正是前篇所谓「单向化」在微观上的所指。

至此最易生的混淆，是把测量与退相干当作一回事——二者都被笼统唤作「量子到经典」，实则分属两层。退相干是环境耦合下自然发生的：系统与浩繁的环境自由度纠缠，相干性随之离域到环境里，就系统的局域读出而言，叠加之间的相位被抹去，剩下一份经典的混合态。然而退相干至此即止——它抹去的是相位，留下的是一份各有权重的可能，却始终不曾从中点出哪一项。以本文的话说，它把系统由量子的 ρ-OR（相干的叠加），带到经典的 ρ-OR（失了相干、却依旧多容的一份清单）；而这一步本身已经是一种越槽——那份清单借此跨过因果槽、升为范畴义的信息，不再是槽下的量子衬底，而是槽上一份可作经典对待的分立信息。

这就引出两层意义上的「信息」之分。退相干跨槽所成的，是范畴义的信息——那份经典清单，多容仍在、未定其一；而本文反复说的「单一内容」，是另一层，也正是测量独有的那一步——落定：从这份清单里强行收束出一项确定的内容，并予记录，这才是 ρ-AND，由多而一。这一区分也接上本系列论因果槽的一篇：那一篇在信息论的视角下，把退相干与越槽并作一事来谈——就立定「因果槽之上有信息、之下没有」这条边界而言，如此已足；而测量的本体身份要求看得更细，在这一层上，退相干与越槽便是两个不同的过程——退相干是退相干，越槽是越槽。于是测量并非退相干的加速版，退相干也不是测量的缓行版；二者所差，不在快慢，而在所做之事本身：一个去其相干而留一纸清单，一个落其一项而成一则内容。能把测量从退相干中分出的，是它额外具备的几样——它为「何者算作结果」划出界限，为那项确定的内容备下一处可入因果链的位置，并以其作为活跃元胞集合之身，向被测系统的账目施加这一道强制的收束；这道「读出的语法」无关乎任何意识的意图。至于退相干如何在环境中层层铺展、那份清单如何形成、越槽在动力学上如何完成，皆属后续论退相干与经典极限的工作；本文只走测量这一条强行的路。

三、过槽的解剖

测量是桥，而桥身可以拆解。这一拆，循的是本系列论四分形的工作反复揭示的那个结构：先立定刻度而不建构，继而累加出指向、复合出历史，终以闭合给出构与余项；本文不重立此结构，只把它用在测量这一跨上。具体地说，因果读出活跃的那一层（前节的 L₄）在内部正按这四道展开，而越槽的过程，就行在这四道之间。须先说明本文用它的分寸：判定某一结构是否真当得起「四分形」，方法论十附有数条须逐一满足的条件，本文尚未一一核验；故此处给出的四层划分，只作一份候选的登记，而非已然坐实的定论。至于这套划分完整的细颗粒度层接（涉及四力系列的拓展），亦留与后续，本文不为它定死最终的归属。

第一道是时间概念的出现（L₄a）。它只立起「时间」这一概念连同普朗克时间的刻度——给出「一拍」，却不建任何结构，也还谈不上方向；它标而不构。

第二道是时间的箭头（L₄b）。在刻度之上，加法累出时间的指向。但此时的箭头不止一条：在因果槽以下，它是多重而并存的——这正是概率云之为概率云的缘由，也正是槽下「无单一前后、因而无因果」的所在。

因果槽就落在第二道与第三道之间：它把多重的箭头锁成单一的方向。锁定之前无单一前后，锁定之后方有确定的先与后——而这条被锁出的单一方向，恰是前节「双重影响」里自槽上向下笼罩着槽下量子的那一重。所谓「信息锁后有信息」，锁住的就是它：槽下本无单一方向，故无因果亦无信息；方向一经锁定，因果与信息才一并得以成立。

第三道是时间的方向（L₄c）。方向既已单一，乘法便得以沿这条已定的方向，把历史一层层复合、绑定进当下——这便是记忆。记忆必依赖单一方向，因为没有确定的先后，就没有「过去」可记、无从记起；唯在锁定的方向上，前态复合入现态，才成其为记忆。这一道之所以必排在因果槽之后，根由即在于此。

第四道是信息与因果律（L₄d）。这是闭合：它一举给出「构」与「余项」二者——构，就是信息本身，一项确定的、立定了的内容；余项，则是这内容如何被储存。至此，那桩落定方告完成：它正坐落在「方向被锁为单一」到「闭合给出信息与余项」这一段上，是这四道里跨过因果槽、并把内容立定的那一跨。

由是回看，测量之为因果槽两侧的桥，其桥身正是这四道：时间概念立其拍，箭头予其指向，因果槽锁其方向，方向之上乘出记忆，闭合定其信息与储存。槽下的无因果、无信息，与槽上的有因果、有信息，便由这一跨接通。这也把全文的主张收成一句：因果槽下无因果，信息锁后有信息——而测量，正是那一把锁。

四、落定：不可逆的因果读出

落定不可逆，但要说准这份不可逆，得先看清它与另一种不可逆并不同源。测量发生时，退相干与落定共时进行，各自带来一种「回不去」，而这两种回不去，落在不同的地方。

退相干的回不去，是热力学的。相干性随系统与环境的纠缠离域到无数自由度里去，原则上把这些四散的相位重新收拢、对齐，并不为任何定律所禁——自旋回波与光子回波，便是在小系统上付了代价、硬把相干拉回来的例子。只是在宏观的环境里做这件事，等于让一份已经摊开的关联重新聚起，是一处局域的熵减，按第二定律须由别处更大的熵增连同功来抵偿；环境一旦庞大，这笔账便实际付不起。所以退相干之「不可逆」，确切地说是原则上可逆、却需注入净能量并以更大的熵增为代价，因而在实践中回不去。这是第二定律那一路的不可逆，是热的箭头。

落定的回不去是另一路。它把槽下的 ρ-OR 容量读出为一项 ρ-AND 内容，立起了一条因果。要反演它，就是要反演一条已然成立的因果，让果退回到因之前——而因果的箭头是单向的。这一路的回不去与能量无涉：再多的能量也凑不出那样一个逆操作。何以如此？一来，落定并非从槽下读出一个早已备好的答案，而是把那项内容生成出来（详见后文），逆它便须凭空将一份原不存在的多容重新变回；二来，本系列早已把因果安放为 4DD 的前向属性——信息即因果，因果即那条单向的箭头——而落定恰是择定单一因果方向、把这条箭头立起来的事件，它的不可逆，也就是箭头本身的不可逆。

两路并立，最干脆的分辨是问一句：能不能用能量把它的逆买回来。相干性活在能量这一侧，买得回；因果内容活在因果那一侧，买不回。它们种类有别，却在同一桩测量里同时发生——退相干摊开一份清单，落定从中点出一项；前者可逆而吃能量，后者不可逆而与能量两清。本文认领的，只是后者；退相干那一路的热力学不可逆，以及那段过程本身要花的有限时间，都不是本文要去解释或担起的。

这也顺带交代了落定的「代价」。槽下原没有现成的内容，要得出一项确定的因果，就得付出这一次不可逆的因果读出本身——代价即在于此，且它落在因果这一侧，能量赎不回它的逆。这与守恒并不抵触：余项不曾被消灭，只是从 ρ-OR 的样式结清成 ρ-AND 的单一内容连同一则经典记录，总量未变。至于这桩事件在能量与热的账面上是否、又如何对得上某个数目——比如擦去一比特所需的那份 Landauer 热——则是另一侧的问题；本系列论此已有专篇，所得是有条件的结论，且一宗宏观的一比特擦除本是跨着约三十个数量级的众多衬底事件之统计聚合，与单次落定这一个事件并不是一回事。故能量层面的数目，本文留与后续，此处只在本体上认定一句：内容不是白来的。

还得把落定的「成没成」与那段过程的「要多久」分开。落定是要么成、要么不成的：一项确定的内容，或已生成，或尚未；并无「半项内容」可言。这并不是说落定是不费时间的魔法——本系列论有限因果槽的一篇已说明，因果的结清要占用有限的时间，那是系统完成一次结清的底限，而非外加的截断。但时长是过程那一侧、能量那一路的事；落定之为非此即彼，说的是结清成没成（内容生没生成），与它要花多久是两码事，后者连同跨越的动力学，一并归后续。

如此便可回头安置「波函数坍缩」这个词。标准量子力学把测量后的那一跳称作坍缩，并把它当作一条没有解释的附加公设搁进理论里。本文不否认坍缩这一现象，而是替它补上本体：所谓坍缩，正是这桩在因果上不可逆的落定——余项由 ρ-OR 结清为 ρ-AND 的单一内容，总量守恒，因果的箭头由此立定。坍缩于是不再是外加的公设，而是一桩有本体身份的因果读出。

五、逼近与落定

把落定立为不可逆的事件，立刻引出一类看似反例的现象：弱测量、POVM、预测量，乃至量子擦除——它们都与被测系统耦合，事后却往往可逆、可撤，也不必毁掉纠缠。若落定当真不可逆，这些又当如何安放？

它们根本没有落定，只是朝落定逼近。耦合弱到不足以把那一项内容收束出来、顶过因果槽，过程便停在落定之前。它们之所以可逆，恰因为还没有什么落定了的东西可言——没成的事，自然撤得回。于是这些现象非但不与「落定不可逆」相左，反倒是它的印证：一旦真落定、内容真生成，可逆便不复存在；可逆者，按其本义就还在落定之外。这里的界限是种类上的，不是程度上的——不是「弱一点的落定」，而是「尚未落定」；落定与逼近之间，并没有一条连续的强弱谱，可供安放一个「半落定」。

至于逼近区里读到的种种异常——例如弱值落在本征谱之外——本文不予置评：那属于「在尚未落定处究竟读到了什么」的现象学，牵涉弱值本身的本体地位之争，可留与后续及相关工作。本文在此只立一条种类上的分界：落定，与朝落定的逼近。逼近究竟累积到何处方才跨过、跨越在动力学上如何发生，同样归后续。

六、生成，而非选定

落定生成一项内容——这个「生成」要紧，须把两层意义上的「信息」分开。槽下那一层，是广义的容量：它真实存在，是 ρ-OR 的多容结构，却不是一个确定的答案。槽上这一层，才是狭义的内容：一条确定的、可入因果链的记录，而它是在越槽的当口被生成出来的，并非槽下早已写就、此刻被取出。测量顶上来的，是容量被读成了内容，而不是把一份藏好的答卷翻了出来。

由此须把本文交付的东西界定清楚：本文给出的，是单一内容如何被生成——闭合如何把多容收束、立起恰好一项确定记录；本文并不给出「为何偏偏是这一项」的隐藏规律。落定确乎生成了一项确定内容，但落定的是哪一项，在本体上是未定的，而这未定是根本的，不是我们不知情而已。它不是某个尚待发现的隐变量在背后做主——闭合之前，槽下并没有一个单一的因果方向，也就没有一份预先写好的内容等着被揭晓；恰是闭合这桩事件择定了方向、生成了内容。如此，单次结果的随机便是本体层面的随机，与「不携带可因子化的局域隐变量」、与「无法为闭合之前的态赋一个全局确定值」这两类结论，都相容无碍。

要点因而是一对，而非一桩：内容之「有无」，是落定一举生成的，确定而不可逆；内容之「为哪一项」，是根本随机的，无规律可循。把这两件混作一件，正是「隐变量选定了结果」之误的由来；本文所立，是前者的生成，而非后者的选定。

七、读出读得到什么

落定既是一次读出，便有它读得到与读不到的界限。被读的那一端，在因果槽以下、在闭合之前；读出这桩事件，则是落定本身。要紧的是说清这一次读出能交出什么。

它交出的，是一项确定的结果——某个本征值层面的内容，一枚受权重约束的代币。它交不出的，是槽下那份完整的 ρ-OR 复数分布本身，亦即态矢量本身：4DD 的一次读出，无法把整份分布打包提上来，它只能强行落定出其中一项。单次的确定本征值，与「读不出整份分布」，二者并不冲突——前者是这一次落定所成的一项内容，后者说的是没有哪一次读出能把多容整体兑作内容。至于期望值，它并不是任何单次落定所读出的对象，而是同一测量在大量重复之上、对诸结果按其权重所作的统计泛函；它住在重复的层面，不住在单次的事件里。

把这件事放回前后的分工便更清楚：并存诸结果的权重，由本系列论玻恩规则的前篇给出；单次的一项内容如何落定、如何生成，是本文之事；而这项内容此后如何被放大、冗余、稳定为宏观可靠的记录，则归后续。本文站定的，是中间这一环——单次内容的生成身份。

八、无信号

落定是一桩局域的事件，这一点关乎相对论与量子力学的相容，须单独交代。先把它放进标准的测量形式里：设 Alice 与 Bob 共享一个纠缠对，Alice 在己端施行一次测量。若不挑选她的结果（即非选择性测量），Bob 端的约化态为

$$\rho_B'=\mathrm{Tr}_A\!\Big[\sum_a\big(M_a\otimes I\big)\,\rho_{AB}\,\big(M_a^\dagger\otimes I\big)\Big]=\mathrm{Tr}_A\,\rho_{AB}=\rho_B,$$

与 Alice 测或不测、用何种测量皆无关。Bob 的边缘统计不因 Alice 的作为而改变——这是无信号的标准陈述，本文沿用，并不另证新的定理。

本文要补的，是它在本体上何以如此。Alice 那一端的落定，是在她本地把容量读成内容，是一桩纵向的、就地生成的事件；它并没有把什么槽下的内容横向送给 Bob。能送的得是内容，而槽下本无内容可送。那么 Alice、Bob 之间确凿的关联又是什么？它真实存在，却存在于容量这一层——是一份共享的、不可因子化的账目，而非一条待传的消息。正因不可因子化，它不是各粒子身上局域的隐藏属性；正因不是消息，它也不构成可供传递的信号。于是两件事各安其位：关联是非局域的（一份账目，两端共担），信号却是局域的（内容在每一端各自就地生成）。Alice 一落定，她那端便有了内容，Bob 的边缘却纹丝不动——越槽是就地的、纵向的，因果槽这条横向的边界依旧守着。

这正接上爱因斯坦对「鬼魅般超距作用」的不安：超距的，只是预先共享的容量在两端各自被读出时所显出的关联，而非任何越过距离传递的内容或作用。本文于此收口——它补全的是无信号的本体一面，并与既有的形式陈述相接，而不另立新说。

第二部分　装置、机制与诠释

九、什么东西算得上测量装置

前面说测量要为结果划界、为内容备位、并对系统的账目施加收束，这就引出一问：什么东西，才足以充当一具测量装置？

答案不在「意识」，也不在「宏观」，而在一种结构。一具测量装置，是一片活跃的元胞集合——在物理量阶梯里，是质量层与因果读出层（L₃ 与 L₄）一道活跃起来的那种聚集。它之所以能测量，在于它既能为被测系统划出「何者算作一个结果」的界限、又备有一处可把落定的内容送入因果链的位置，更能在与系统耦合时，以其作为活跃聚集之身，触发系统这一端的 L₄ 重激活，向系统的账目施加那道强制的闭合。唯其如此，才是「测量」之为测量。

由此可澄清两桩常见的误认。其一，测量不是意识所致：施加闭合的，是作为活跃聚集的装置，而非任何观察者的觉知；一台仪器、一片感光的介质、乃至一段足够耦合的环境结构，只要具备这套本事，便足以充当，无须有谁在「看」。其二，测量也不必是宏观的大物件：要紧的是这套本事是否齐备，而非尺寸大小；微小的系统之间，只要一方对另一方构成了这样的活跃聚集，落定便可发生。

本文在此只停在一桩单一的闭合事件上——一具装置对一个系统施加一次落定，立起一项内容。这项内容此后如何沿装置层层放大、冗余复制、稳定成宏观可读的指针，是一串后继的闭合，属于后续论宏观稳定化的工作，本文不入。

十、一处有待后续的线索：标定的余项

闭合在立起内容（构）的同时，也留下余项（§三、§四）；那么这余项，落到 L₁ 标定上，是否留有可辨的印记？这里有一条线索，本文只把它挂出，不在此兑现。

本系列论 L₁ 标定的工作里，闭合伴随着一桩标定上的变动。把它移到测量上，可以设想：一次落定的发生，或许在标定上落下一项余项——内容的立定，要在 L₁ 标定的账上记下一项。但必须就此打住几步。本文不由此推出任何形如能量—时间不确定关系的式子；不主张一切测量都伴有同一种能量冲击；也不把这一项余项，同本系列论 ℏ 辛共轭那一篇所给的不确定关系混为一谈。它至多是一处方向：倘若将来有足够精度的弱测量实验，能在逼近区里辨认出一种与落定相伴的能量流动，这条线索才谈得上兑现，届时可回头与 L₁ 标定的工作相接。

之所以在此郑重悬置，是因为本文恪守不透支动力学前提的界限：标定的余项是一处合理的推演方向，但在获得坚实的动力学明证之前便将它写成定论，将逾越本文的辖区。故本文只立其为待验的线索，其具体的动力学，连同它与弱测量异常的可能关联，一并留与后续。

十一、对诠释的中立与不中立

本文给的是测量的本体层，它坐落在通常所谓「量子力学诠释」之下。对相当一批诠释，本文是中立的：凡承认测量含有一桩不可约的闭合事件者——哥本哈根诠释、玻姆力学、作为物理过程的客观坍缩理论、彭罗斯的引力坍缩设想——都可在本文这一更深的本体之上，各自展开其现象学的描述。本文不在它们之间裁断，只指出它们所共享的那桩闭合事件，在本体上究竟是什么。

这一中立有其来历。本系列论因果槽的一篇已表明，SAE 采取一种与哥本哈根相协的经典性边界，并与艾弗雷特式的多世界读法不相容；本文承此而来，不另立。

但中立是有界限的，越过界限便不再中立。倘若有一种诠释主张测量根本不含任何不可约的闭合、一切尽可由幺正演化加上认知的更新来完成——彻底贯彻的多世界，或彻底贯彻的量子贝叶斯主义，便属此类——那么它与本文所立的 4DD ρ-AND 闭合身份，是不相容的。须说清的是，这一不相容属于本体结构的层面，而非经验所能裁判：「一切其实是幺正的、只是我们永远看不到」与「确有一桩非幺正的闭合」，本就是测量问题自身的两面，没有哪个实验能在它们之间裁夺。本文于此不假装握有经验的判据，只摆明自己的本体立场——它需要那桩不可约的闭合，因而与彻底取消闭合的读法分道。

第三部分　几桩佯谬的重新归类

本系列对测量的本体刻画，顺带松动了量子力学里几桩著名的佯谬。须先声明：本文并不声称「解决」它们——所谓解决，往往是在某一种诠释之内给出一套自洽的说法；本文做的是另一件事，即指出这些佯谬大多源于同一个范畴上的混淆，一旦把这层混淆点破，佯谬便自行消解。这个共同的混淆是：把闭合之前的容量（ρ-OR，多容、可幺正演化），当成了闭合之后的内容（ρ-AND，已落定的那一项），或反过来。以下四桩，各是这一混淆的一种样式。

十二、量子芝诺：一桩痛点的消解

量子芝诺效应说的是：对一个系统频繁地测量，会拖住乃至冻结它的演化。它先教给我们一件正面的事——测量绝非被动的旁观。倘若测量只是读出一个早已写定的值，频繁地读不该改变系统的走向；而芝诺效应表明，频繁的测量确乎改变了走向。这与本文相合：每一次测量都是一桩对系统的强行介入（一次落定，或朝落定的逼近），频繁地介入，自然改写演化。就此而言，芝诺效应可被重新归类为高频落定与逼近在系统演化上的累积。

但必须就此止步，不可把它当作本文的证据。频繁测量拖住演化这一定性现象，几乎每一种诠释都讲得通；拿一个人人都讲得通的现象去印证某一种本体，是以果证因的虚假支持。何况频繁测量未必总是拖住：在另一些情形里，它反而加速衰变，是为反芝诺效应。究竟净效果是抑制还是加速，取决于系统与环境耦合的谱学细节，这同样为各家所共有，与本文要立的本体无关，留与后续。本文于此只作一句保守的归类，不下任何关于「频繁测量必致冻结」的断言。

倘若本文的本体当真要在芝诺这一带显出与标准量子力学不同的指纹，那唯一可能的去处，是离散的元胞节拍在极短的测量间隔上、对相关谱交叠所留下的微小偏离。这一处，构成本框架未来的一项可证伪承诺，明言不在本篇兑现。

十三、维格纳的朋友与一则不可能定理

维格纳的朋友是这样一个设想：朋友在封闭的实验室里对一个系统做了测量，得到了一个结果；室外的维格纳却把「朋友加系统」整体当作一个尚未坍缩的幺正叠加来描述。于是要问：朋友那个结果，究竟算不算一桩事实？弗劳赫格尔与雷纳进一步表明，若坚持让每个观察者都把量子力学无保留地施于别的观察者，几条看似自然的假设会一道导出矛盾。

以本文观之，矛盾的根，是把已落定的内容错当成了尚可幺正演化的容量。朋友做了测量，便是发生了一次闭合：一项确定的内容（ρ-AND）已在朋友的实验室里立定，这是客观的，不因室外的维格纳是否与之交互而改变。维格纳把朋友重新描述成一个相干的叠加，等于把这项已经落定的内容，倒退着当作闭合之前那份多容的容量（ρ-OR）来对待——而这一步，正是范畴上的错置。闭合是单向的：内容一旦由容量落定而来，便不再是容量；没有哪个更高的观察者，能把它幺正地「退回」成一份可叠加的容量。这道单向的界，就是本文为这桩佯谬竖起的防火墙。

如此，那则不可能定理里被本文判为越界的，正是「把幺正演化无保留地施于一桩已完成的测量」这一条。须说明，弗-雷的诸假设中究竟哪一条是祸根，学界尚有争议——亦有读法归咎于观察者之间的一致那一条；本文的归判循的是 SAE 因果单向的本体，是诸读法中的一种，而非唯一定论。循此，本文保留单一结果、也保留观察者之间的一致，只把那一条限定为：幺正的量子力学描述的是闭合之前的容量，而非闭合之后的内容。这一消解与种种关系性的读法并不相抵，却无须诉诸「事实只相对于观察者」——在本文这里，内容一经落定即是客观的。

十四、科亨–施佩克尔：语境性之自然

科亨与施佩克尔的定理表明：在三维以上，无法给所有可观测量同时指派一组与测量背景无关的确定值。这常被读作一桩怪事——量子的诸可观测量，竟不能都预先持有确定的值。

但在本文这里，这正是意料之中。闭合之前，槽下只有容量，没有内容；既没有内容，自然也就没有一组预先写定、只待揭晓的全局确定值。值不是预先写好的，而是在闭合那一刻、随着「发生了哪一桩闭合」被生成出来的——这桩生成依赖于落定的背景，本就是语境性的。所以科亨–施佩克尔的语境性，不是量子世界的一桩反常，而是「闭合之前并无预写内容」的另一种说法（亦即第六节所立：本文交付的是内容的生成，而非某条选定哪一个的隐藏规律）。把它读成怪事，仍是预设了槽下藏着一份待揭晓的答卷；一旦认清槽下本无答卷，怪处便消失了。

十五、多世界：分支是容量，不是内容

多世界诠释主张：宇宙波函数从不坍缩，叠加里的每一支都同等真实，各自成一个世界。

以本文观之，那些「分支」是闭合之前的容量（ρ-OR 的多容），而不是闭合之后的内容（ρ-AND 的那一项）。把分支当成一个个真实的世界，就是把容量实体化成了内容——又一次容量与内容的错置。这里还可再添一道因果的界：一支要够格称作一个「世界」，它得有自己的一条因果、自己的内容；而闭合之前，槽下是多重时间箭头并存、并无单一的因果方向，也就没有内容可言。单一的因果方向与那一项内容，是闭合才生成的，而闭合只发生一次、只落定一项。于是那些分支，停在容量的层面，够不上「世界」的门槛。

须重申前文（§十一）的界定：这是本文的本体立场，不是一条经验上能与多世界裁判输赢的判据——「其实人人都分了叉、只是各支永不相见」与「只落定了一项」，在经验上并无分别。本文只是指出：在 SAE 的本体里，分支是容量而非内容，把它读成实在的诸世界，便越过了容量与内容之间那道界。

---

参考文献

自为目的（SAE）系列

本文所引本系列其他论文，按正文中所用的指称列出。

1. 秦汉，《SAE 量子力学 Paper 6》（玻恩规则）。Zenodo，2026。DOI：10.5281/zenodo.20479658。——正文 §一、§七引为「本系列论玻恩规则的一篇 / 前篇」，给出闭合诸结果的权重 $w_i = \psi_i^\sharp\psi_i$。此为紧接本文之前的一篇。

1. 秦汉，《SAE 信息论 III：信息的因果槽——从量子涨落到 thermal-floor minimum 的 4DD 本体论》。Zenodo，2026。DOI：10.5281/zenodo.19797457。——正文 §二、§六、§十一引为「本系列论因果槽的一篇」；因果槽这条边界，以及其 §2.4 / §6 把退相干读作跨槽的范畴提升。

1. 秦汉，《方法论十》（四分形）。Zenodo，2026。DOI：10.5281/zenodo.20187591。——正文 §三引为「本系列论四分形的工作」；标而不构 → 加法（方向）→ 乘法（记忆）→ 闭合（构与余项）这一结构及其须满足的诸条件。

1. 秦汉，《SAE Foundation v2：物理量阶梯系统性重述与签名纪律》。Zenodo，2026。DOI：10.5281/zenodo.19361950。——正文 §二、§九引，用其物理量阶梯 L₀–L₅，及 L₄ 作因果读出活跃的那一层。

1. 秦汉，《SAE 信息论 I》（信息的 4DD 本体论；信息守恒公理）。Zenodo，2026。DOI：10.5281/zenodo.19740019。——正文 §四引，用其「信息即因果，因果是 4DD 那条单向的箭头」。

1. 秦汉，《SAE 信息论 II》（Landauer 原理的结构性推导）。Zenodo，2026。DOI：10.5281/zenodo.19780314。——正文 §四引为论 Landauer 代价的「专篇」；其推导的有条件性与三十个数量级的跨度。

1. 秦汉，《SAE 热力学 Paper 6》（τ_dec，因果结清的有限时间代价）。Zenodo，2026。DOI：10.5281/zenodo.19658595。——正文 §四引为「本系列论有限因果槽的一篇」；因果结清需有限时间。

1. 秦汉，《SAE 量子力学 Paper 3：ℏ 作为 SAE 物理量阶梯 L₁↔L₂ 辛共轭闭合签名》。Zenodo，2026。DOI：10.5281/zenodo.20340595。——正文 §十引为「本系列论 ℏ 辛共轭那一篇」；不确定关系 $\Delta x\,\Delta p \ge \hbar/2$。

1. 秦汉，《SAE 量子力学 Paper 5》（L₁ 标定；闭合伴随的标定变动）。Zenodo，2026。DOI：10.5281/zenodo.20455398。——正文 §十引为「本系列论 L₁ 标定的工作」；「标定的余项」这条待验线索之所本。

框架谱系，正文未直接引用但为其前提： 量子力学 Paper 1，DOI 10.5281/zenodo.20252029；量子力学 Paper 2，DOI 10.5281/zenodo.20277037；质量系列收口，DOI 10.5281/zenodo.19510868；相对论 Paper 4，DOI 10.5281/zenodo.20079718；信息论 Paper 7（《生命的火花》），DOI 10.5281/zenodo.20105883。

外部文献

正文 §五、§八及 §十二–§十五所提及诸工作的标准文献。卷期页码与最终格式由作者定夺。

• Born, M. (1926). Zur Quantenmechanik der Stoßvorgänge. Zeitschrift für Physik 37, 863.
• Einstein, A., Podolsky, B., Rosen, N. (1935). Can quantum-mechanical description of physical reality be considered complete? Physical Review 47, 777.
• Bell, J. S. (1964). On the Einstein Podolsky Rosen paradox. Physics 1, 195.
• Kochen, S., Specker, E. P. (1967). The problem of hidden variables in quantum mechanics. Journal of Mathematics and Mechanics 17, 59.
• Misra, B., Sudarshan, E. C. G. (1977). The Zeno's paradox in quantum theory. Journal of Mathematical Physics 18, 756.
• Wigner, E. P. (1961). Remarks on the mind–body question. In The Scientist Speculates (I. J. Good, ed.).
• Aharonov, Y., Albert, D. Z., Vaidman, L. (1988). How the result of a measurement of a component of the spin of a spin-½ particle can turn out to be 100. Physical Review Letters 60, 1351.
• Hahn, E. L. (1950). Spin echoes. Physical Review 80, 580.
• Joos, E., Zeh, H. D. (1985). The emergence of classical properties through interaction with the environment. Zeitschrift für Physik B 59, 223.
• Zurek, W. H. (2003). Decoherence, einselection, and the quantum origins of the classical. Reviews of Modern Physics 75, 715.
• Schlosshauer, M. (2007). Decoherence and the Quantum-to-Classical Transition. Springer.
• Kofman, A. G., Kurizki, G. (2000). Acceleration of quantum decay processes by frequent observations. Nature 405, 546.
• Frauchiger, D., Renner, R. (2018). Quantum theory cannot consistently describe the use of itself. Nature Communications 9, 3711.