Boundary Loss and the Born Rule

This paper addresses a specific gap in the foundations of probability. The Born rule gives numerical quantum probabilities, but it does not have to be the first appearance of probability-like structure. Before numbers are assigned, a boundary-readable description may first lose guarantee. When a deterministic pre-boundary structure is read through a lossy boundary, the boundary residue may no longer determine which admissible alternative obtains.

The paper develops this prior layer as boundary-unresolved quotient status. It uses the phrase pre-numerical probability-status only in a restricted technical sense: admissible alternatives remain live under the same boundary-readable residue, but no numerical probability, Born weight, frequency, stochastic law, collapse mechanism, or measurement theory has yet been supplied.

Plain-language summary

Boundary loss does not first produce numbers. It first destroys guarantee. If the same boundary-readable residue is compatible with more than one admissible alternative, the boundary description cannot determine which alternative obtains. That unresolved status is not yet probability. Numerical probability appears only after a local scalarization rule supplies nonnegative weights and normalization. In quantum contexts, the Born rule is the Hilbert-space scalarization of that prior unresolved status.

Download PDF Zenodo record

1. What Problem Does This Paper Solve?

The paper separates three things that are often compressed together: loss of guarantee, numerical probability, and the Born rule.

A deterministic system may contain enough structure to fix what occurs before readout. But if that system is passed through a boundary map, projection, quotient, coarse-graining, measurement interface, export, horizon, or other lossy readout, the resulting boundary-readable residue may no longer preserve enough information to recover which admissible alternative obtains.

That failure is not automatically randomness. It is also not merely subjective ignorance. It is a structural failure of guarantee at the boundary-readable level. The boundary residue itself no longer carries enough recoverable structure to decide the relevant alternative.

2. The Boundary Map

The basic setup is simple. Let a pre-boundary state space be mapped into a boundary-readable residue space:

B : Spre → SB

Here Spre is the pre-boundary state space, and SB is the space of residues available after boundary readout. A residue is written:

r = B(x)

The residue r may represent what remains after a coarse-graining, projection, export, measurement readout, quotient, horizon transition, or other recoverability-limiting operation. The paper uses “boundary” in this abstract recoverability sense. It does not require a physical surface, spacetime horizon, metric boundary, or measurement device.

3. Compatibility Classes

The boundary map identifies pre-boundary states that produce the same residue. This creates a compatibility class:

B⁻¹(r) = { y ∈ Spre : B(y) = r }

Every state in this class is compatible with the same boundary-readable residue. If the states in the class differ in a way the residue cannot recover, then the boundary description has lost guarantee for that distinction.

This is the first key distinction. Generic many-to-one mapping is not enough. The lost information must be relevant to the proposition, classification, attribution, recovery claim, prediction, or outcome being tested.

4. Boundary Guarantee, Exclusion, and Undetermination

A proposition P is boundary-guaranteed at residue r when it holds for every pre-boundary state compatible with that residue:

GuaranteedB(P, r) iff ∀y ∈ B⁻¹(r), P(y)

It is boundary-excluded when it fails for every compatible pre-boundary state:

ExcludedB(P, r) iff ∀y ∈ B⁻¹(r), ¬P(y)

It is boundary-undetermined when compatible pre-boundary states disagree:

UndeterminedB(P, r) iff ∃y, z ∈ B⁻¹(r) such that P(y) and ¬P(z)

Boundary undetermination is the formal loss of guarantee. It does not yet define probability. It only says that the boundary-readable residue cannot decide the claim.

5. Boundary-Loss Non-Guarantee

The first theorem proves the minimal structural result:

If compatible pre-boundary states under the same residue differ with respect to an admissible proposition, then the residue neither guarantees nor excludes that proposition.

This result is intentionally elementary. Its importance is not that non-injectivity is surprising. Its importance is that it isolates the first layer before probability enters. Boundary loss can destroy guarantee without yet producing numerical probability, a measure, a stochastic rule, a frequency interpretation, or a Born-rule assignment.

6. Boundary-Unresolved Quotient Status

The second layer adds an admissible scalar-neutral resolution relation. Let D be a declared domain inside the compatibility class:

D ⊆ B⁻¹(r)

Let K be an admissible boundary-relevant resolution context, and let ∼K be a scalar-neutral resolution relation over D. The quotient is:

FK := D / ∼K

The elements of this quotient are admissible unresolved alternatives. They are not weighted events. They are not probabilities. They are resolution-eligible alternatives that remain live under the same boundary residue.

The central object is:

PreProbContextB(r, K, D) := [FK]pres

This denotes a representative-invariant boundary-unresolved quotient status. It is called pre-numerical probability-status only in the restricted technical sense used by the paper.

7. What Scalar-Neutral Means

Scalar-neutrality is the firewall that protects the theorem from overclaiming. A scalar-neutral quotient status contains no:

scalar weights;
ranking;
likelihood values;
confidence values;
frequency rule;
measure structure;
normalization rule;
stochastic transition law;
Born-form assignment.

This is why the paper does not claim that boundary loss directly derives probability. Boundary loss creates the unresolved status. Numerical probability requires additional structure.

8. Local Scalarization

Numerical probability enters only when a local realization supplies scalar residues for the unresolved alternatives.

Given a finite representative:

F = {A1, ..., An}

a local scalarization rule assigns nonnegative scalar residues:

sF : F → R≥0
si := sF(Ai)

If the scalar residues are nonnegative and have finite nonzero total:

SF := Σi si > 0

then formal normalized residues can be defined:

ni := si / SF

These normalized residues are still not automatically empirical probabilities, stochastic probabilities, Born probabilities, or objective chances. They become actual probabilities only inside a declared probability regime.

9. The Layer Structure

The paper’s corrected layer structure is:

boundary loss
→ loss of guarantee
→ admissible unresolved alternatives
→ boundary-unresolved quotient status
→ local scalarization
→ normalized residues
→ declared probability regime
→ numerical probability

This prevents boundary loss from being asked to do work it cannot do. Boundary loss does not create numbers. It creates the condition under which assigning numbers may become meaningful, if a later scalarization rule is supplied.

10. The Born Rule as Local Hilbert-Space Scalarization

The Born rule enters only at the quantum scalarization layer. A Hilbert-space regime supplies additional structure not present in the abstract boundary-loss theorem:

a complex Hilbert space;
a normalized state vector Ψ;
projectors or effects;
amplitudes;
orthogonal additivity;
normalization;
regularity or noncontextuality conditions where required.

Inside that local quantum regime, the standard Born assignment is:

pi = ⟨Ψ, PiΨ⟩

For a basis representation, this becomes:

pi = |⟨ϕi, Ψ⟩|²

The paper therefore does not replace the Born rule. It relocates it. The Born rule is treated as the local Hilbert-space numerical scalarization of boundary-induced probability-status, not as the origin of probability-status itself.

11. Why Boundary Loss Alone Cannot Derive Born Numerics

Boundary loss supplies a quotient-level result. It gives a residue, a compatibility class, and unresolved admissible alternatives. It does not supply Hilbert space, amplitudes, projectors, inner products, orthogonal additivity, or a normalization rule.

Therefore Born-rule numerics cannot be derived from generic boundary loss alone. They require the narrower Hilbert-space scalarization regime.

Firewall

Boundary loss gives loss of guarantee. Scalar-neutral quotient structure gives unresolved alternatives. Local scalarization gives numbers. Hilbert-space scalarization gives Born weights. These are separate layers.

12. Deterministic Compatibility

The result is compatible with deterministic pre-boundary structure. A deterministic system may produce a boundary-readable residue that fails to preserve enough recoverable structure to determine which admissible alternative obtains.

That does not prove determinism. It also does not refute indeterministic theories. It establishes a narrower compatibility result: probability-status at the boundary does not by itself require stochastic dynamics at the pre-boundary level.

A deterministic pre-boundary system can yield boundary-relative non-guarantee when the boundary readout is lossy in a recoverability-relevant way.

13. What This Paper Does Not Claim

The paper is deliberately restricted. It does not claim to derive numerical probability from boundary loss alone.

It does not replace the Born rule.
It does not derive the Born rule from generic boundary loss.
It does not derive a universal numerical probability law.
It does not claim that every boundary-loss regime admits scalarization.
It does not claim that all uncertainty is ordinary ignorance.
It does not prove that the physical universe is deterministic.
It does not derive collapse, measurement outcomes, empirical frequencies, stochastic dynamics, or objective chance.
It does not require Entanglement Compression Theory as a premise.

The claim is narrower: recoverability-relevant boundary loss, paired with an admissible scalar-neutral resolution relation, can support boundary-unresolved quotient status. Numerical probability requires additional local scalarization. Born probabilities require Hilbert-space scalarization.

14. Relation to Entanglement Compression Theory

This paper is mechanism-neutral. It does not depend on ECT, the Primordial Wave Equation, compression geometry, or any specific physical realization theory.

For ECT, the paper clarifies the burden of a later realization. Entanglement compression should not be treated as directly deriving Born numerics merely by invoking information loss or energy partition. Instead, ECT must supply or preserve the relevant boundary map, admissible unresolved alternatives, local scalar residue, normalization regime, and Hilbert-space scalarization conditions where Born weights are claimed.

This is a stronger architecture. It separates the abstract probability-status layer from the realization layer. ECT may enter later as a candidate mechanism that physically realizes the boundary-loss and scalarization structure, but the structural theorem itself stands independently.

15. Why It Matters

The paper repairs a common overcompression in probability foundations. It prevents numerical probability from being treated as the first and only meaningful probability-like object.

The important prior object is loss of guarantee under boundary readout. When a boundary residue no longer determines which admissible alternative obtains, the system has crossed into an unresolved quotient condition. That condition is not probability yet, but it is the right pre-numerical place for probability to begin.

This gives a cleaner sequence: not randomness first, and not Born weights first, but recoverability loss first. Numbers enter only when the local structure supports them.

16. Source and Formal Paper

This webpage is an explanatory guide. The formal argument, theorem stack, definitions, non-claims, scalarization rules, Born-rule discussion, appendices, and references are in the Zenodo preprint.

Download PDF Zenodo record

Source: Lawrence, W.A. (2026). Boundary Loss and the Born Rule: Pre-Numerical Probability-Status in Deterministic Systems. Zenodo. https://doi.org/10.5281/zenodo.20102172