← All papers

The Trapped-Ion Ultrametric Testbed: A Falsifiability Register for Testing p-Adic Structure in Quantum Dynamics

DOI: 10.5281/zenodo.22013264
Published: 2026-08-19

1. Introduction

Between December 2025 and August 2026, a research program published sixteen records on

trapped-ion quantum computation, p-adic geometry, and the epistemology of physics. Read

individually, they look like three different projects: an experimental physics effort, a

mathematical physics effort, and a philosophy-of-science effort. This paper argues they are

one project, and that reading them as one project changes what a practitioner should do

next.

The unifying claim is structural, not metaphysical. A tree-shaped hierarchy -- the

Bruhat-Tits tree of p-adic geometry, the hierarchy of conditional quantum states under a

global constraint, the hierarchy of stabilizer and dissipative error correction, and the

hierarchy of claims ranked by evidential weight -- appears in all three movements

[@qnfo2026pwprotocol; @quni2026tnbt; @quni2025gkp; @quni2026falsifiability]. The claim made

here is that this appearance is testable on hardware that already exists, and that a set of

falsifiable observables with kill-conditions is the correct way to carry the claim forward.

This paper deliberately does NOT claim the tree structure has been confirmed. It claims

only that the structure is now testable, and it supplies the register by which it will be

tested.

Why a reader should care: near-term quantum hardware needs discriminating experiments, not

only gate-count benchmarks. A measurement that could rule out an entire class of

ultrametric interpretations of quantum dynamics -- or fail to rule it out -- is worth more

than another demonstration that a simulator can simulate. This paper hands an experimental

group a ready-to-run menu of such measurements, with predicted values, kill-conditions, and

the apparatus requirements already worked out in the cited records.

The premise-depth disclosure is stated here and formalized in Section 10. The derived

content of this paper is organizational: resource counts, predicted values, and the

register itself. Two named imported inputs carry the physical content: the Ultrametric

Bridge Theorem, which states that conditional quantum states under a global constraint

necessarily form ultrametric hierarchies [@quni2026bridge], and the Silent Parameter

Principle, which identifies the truncated SU(2) representation ring with the character ring

of the 2-adic units [@qnfo2026zbwadelic]. The premises END there: this paper neither

re-derives nor independently verifies those inputs, and no claim in this paper is stronger

than its inputs.

2. The Sixteen Records: Three Movements

The sixteen records sort into three movements.

The testability movement asks whether p-adic structure is measurable on trapped-ion

hardware. A protocol for testing ultrametricity via a Page-Wootters clock experiment

specifies a single trapped ion, laser-driven carrier and sideband transitions, conditional

state tomography, and an eight-week timeline on existing apparatus [@qnfo2026pwprotocol].

A companion analysis of 8,000 simulated Wheeler-DeWitt systems establishes the prediction

the protocol measures: generic clock-rest coupling violates the Parisi ultrametricity

condition at a 29-35 percent rate, while coupling diagonal in the clock eigenbasis yields

exact ultrametric structure [@quni2026pwdistances]. The Zitterbewegung is re-read as an

adelic observable with a 2-adic frequency of $\sqrt{2} \cdot 2mc^2/\hbar$, a 41.4 percent

deviation from the Archimedean value, and a single-ion timeline of 2028-2032

[@qnfo2026zbwadelic]. Two feasibility studies examine whether a vortex-enhanced

amplification mechanism can lift the sub-Compton-scale signal above the noise floor, and

conclude the question is open pending signal-to-noise work [@quni2026vortexD;

@quni2026vortexI]. A due-diligence assessment of the Innsbruck qudit program verifies the

measured light-shift gate fidelities (99.6 to 93.7 percent for dimensions 2 through 5) that

such experiments would inherit [@quni2026ringbauer; @ringbauer2022; @ringbauer2023;

@ringbauer2024]. A Monte-Carlo study extracts effective transient dimensions of 6.2 and 7.9

for diffusion on p-regular trees with p equal to 2 and 3, against the integer-line baseline

of 1.0 [@quni2026padicdiffusion]. An audit of a tunable quantum neural network demonstrates

the program's testing discipline: the strongest claim survives and the advantage claims do

not [@qnfo2026bqnnaudit].

The architectural movement asks what hardware built on the same structure looks like.

A room-temperature trapped-ion architecture stabilizes Gottesman-Kitaev-Preskill states

autonomously through engineered dissipation, converging to the GKP manifold when the

cooling rate exceeds the heating rate by more than $\pi$ [@quni2025gkp]. The Spin-Free

Substrate protocol simulates Posner-molecule nuclear spin dynamics with six global

Molmer-Sorensen gates per simulated second at 96.9 percent circuit fidelity, and names the

design principle connecting Kane-type, Posner-type, and trapped-ion architectures

[@quni2026spinfreeA; @quni2026spinfreeB]. Quantum Architectonics argues the passive path --

letting structure suppress error -- against the active-correction path that fights the

second law [@quni2026architectonics]. A p-adic metrology proposal claims sub-shot-noise

sensing from hierarchical noise correlations on a room-temperature photonic platform with

roughly ten entangled photons [@quni2026metrology]. A dimensionless reformulation of

fifty-three physics equations in Planck units supplies the place-democratic bookkeeping the

movement uses [@quni2026ostrowski].

The methodological movement supplies the discipline. The Falsifiability Crisis record

diagnoses five structural patterns by which contemporary fundamental physics has become

effectively unfalsifiable, and proposes the Bayesian delta-log-odds gate as the remedy

[@quni2026falsifiability]. The Reification record names the failure mode the other two

movements must avoid: mistaking a mathematical model for a mind-independent object

[@quni2026reification]. The Foundations record argues that Shor's algorithm is a

theoretical artifact defined under unrealizable conditions, and that error correction fails

to suppress errors beyond current scales due to correlated failures [@quni2026foundations].

The program's own archive is imperfect, and this paper records the imperfections rather

than hiding them. Two pairs of records are duplicate or near-duplicate deposits rather than

clean version chains (the Spin-Free Substrate pair [@quni2026spinfreeA; @quni2026spinfreeB]

and the vortex-feasibility pair [@quni2026vortexD; @quni2026vortexI]); version labels across

the set are inconsistent; and the archive's internal identifier table carries missing and

cross-wired DOI fields for several records (Section 12). A converged program should have a

converged archive; the disorder is disclosed here as a data-quality finding.

3. Lineage: How the Register Descends from Earlier Work

The sixteen records did not appear from nowhere. Each movement is the latest link in a

published chain, and the chain is the evidence that this paper is a continuation rather

than a restart. A reader can follow the citations backward from any register entry to

its oldest published premise.

The audit lineage. The critique of quantum-computing claims began with the

reassessment of quantum computation's foundations [@quni2026foundations], proceeded

through the energy-accounting benchmarks of the competitive landscape

[@qnfo2026jpcubcl], the qudit extension [@quni2026quditadv], and the

quantum-neural-network audit [@qnfo2026bqnnaudit], and reaches the trapped-ion

due-diligence record cited in this paper [@quni2026ringbauer]. The register's energy

discipline is inherited from this chain.

The ultrametric lineage. The claim that p-adic structure is physically testable

descends from the number-theoretic classification of error-correcting codes

[@qnfo2026uf], the ultrametric bridge theorem [@quni2026bridge], and the

conditional-state analysis that turned the theorem into a measurable prediction

[@quni2026pwdistances] before the trapped-ion protocol existed [@qnfo2026pwprotocol].

The Zitterbewegung entry of the register [@qnfo2026zbwadelic] continues the earlier

analysis of the effect's ultrametric readout and its bridge to anyon braiding [@quni2026zbwp1]. The

anomalous-diffusion entry [@quni2026padicdiffusion] descends from the dimensionless

reformulation program [@quni2026ostrowski; @ostrowski1916], the non-anthropocentric

units reformulation [@quni2026nonanthro], and the continuum analysis that bounds which

real-number structures carry physical content [@quni2026continuum1].

The architectural lineage. The passive-design claim descends from the

architectonics critique of active error correction [@quni2026architectonics], the

spin-free substrate protocol [@quni2026spinfreeA], and the autonomous GKP

stabilization proposal [@quni2025gkp], and is corroborated by the ultrametric

metrology proposal [@quni2026metrology]. The dissipation-first reading of the tree

hierarchy is this paper's own synthesis of those published claims; the chain is cited

so each link can be audited.

The methodological lineage. The falsifiability discipline applied here descends

from the falsifiability-crisis analysis [@quni2026falsifiability], the reification

analysis [@quni2026reification], the falsifiability protocol for discrete-continuum

signatures [@quni2026qfund], and the fifteen-question ignorance audit with its

companion case study [@uia2026; @iaps2026]. The register's kill-conditions are that

method applied to the program's own claims.

The consilience lineage. The synthesis of number theory and physics into one

tree-shaped structure was published in the five-pillar consilience paper

[@fivepillars2026] and the adelic core synthesis [@adeliccore2026]. The present paper

extends that line from structure to instrumentation: where the predecessor papers

established the shared geometry, this paper supplies the devices that measure it.

What the chain shows: the three movements share a published ancestry, and each

movement's latest record cites the earlier links. The chain of reasoning from the

oldest premise (the completions of the rationals [@ostrowski1916]) to the newest

instrument (a register of five kill-conditions, Section 4) runs through every one of

the sixteen records.

4. The Testability Movement: A Register of Five Falsifiable Observables

The register is the paper's core artifact. Each entry states the observable, the

pre-registered prediction, the kill-condition, and the apparatus.

R1 -- Ultrametricity violation rate in a Page-Wootters clock. Observable: the Parisi

ultrametricity violation rate (UVR) of conditional state overlaps, measured by conditional

state tomography on a single trapped ion with electronic states as the clock and motional

Fock states as the rest. Prediction: UVR equal to zero for diagonal clock-rest coupling;

UVR in the 29-35 percent band for nondiagonal coupling [@qnfo2026pwprotocol;

@quni2026pwdistances]. Kill-condition: a measured UVR indistinguishable between the two

coupling classes -- the theory predicts the split, and the split is what is tested.

Apparatus: existing trapped-ion capability; estimated eight weeks of beam time.

R2 -- The 2-adic Zitterbewegung frequency. Observable: the Zitterbewegung frequency

ratio in a Dirac-simulator implementation. Prediction: $\sqrt{2} \cdot 2mc^2/\hbar$, a

41.4 percent deviation from the Archimedean value [@qnfo2026zbwadelic]. Kill-condition: a

measured ratio consistent with 1 within error. Open constraint: the vanishing-signal

question -- whether the ZBW signal is observable at all in the Foldy-Wouthuysen frame --

and the amplification-feasibility question remain unresolved in the source records

[@quni2026vortexD; @quni2026vortexI]; R2 is therefore registered as a measurement whose

SNR budget must be closed before the experiment is meaningful, not as a settled test.

R3 -- Effective transient dimension on p-regular trees. Observable: the return

probability decay of a random walk on p-adic trees, extractable in simulation and, in

principle, in engineered hierarchical coupling graphs. Prediction: effective transient

dimensions near 6.2 (p=2) and 7.9 (p=3), against the Archimedean baseline near 1

[@quni2026padicdiffusion]. Kill-condition: dimension growth that flattens with p. This is

the register's simulation-first entry: it is executable today, with the code already

deposited with the source record.

R4 -- Dissipative break-even. Observable: logical bit-flip suppression of an

autonomously stabilized GKP state as a function of the cooling-to-heating rate ratio.

Prediction: convergence to the GKP manifold with exponential bit-flip suppression once the

ratio exceeds $\pi$ [@quni2025gkp]. Kill-condition: no break-even at the predicted ratio in

any ion species. Constraint: break-even on one mode is not a logical qubit; the

fault-tolerance step remains unclaimed.

R5 -- Tensor-network collapse under nonlocal perturbations. Observable: whether the

classical simulation advantage of tensor networks for local Hamiltonian dynamics persists

when the Hamiltonian acquires controlled nonlocal terms. The tensor-network record reads

the recent tensor-network simulation results as evidence that local-Hamiltonian dynamics

are ultrametric, because local structure is exactly where the Bruhat-Tits hierarchy lives

[@quni2026tnbt]. This is the register's retrodiction entry, graded honestly in the

Bayesian accounting (Section 6): the standard area-law explanation predicts the same

success, so R5 is registered as a discriminating test rather than credited as evidence.

Prediction under the ultrametric reading: the advantage degrades under nonlocal

perturbation faster than area-law extrapolation would predict. Kill-condition: advantage

persists at area-law-predicted levels. This entry exists precisely because the

tensor-network reading currently carries zero independent evidential weight and needs a

test to earn any.

5. Negative Results the Register Must Respect

A testable program is one that keeps its negative results. Five are load-bearing here.

First, generic clock-rest coupling does NOT produce ultrametricity. The 8,000-system study

found a 29-35 percent violation rate in the generic case [@quni2026pwdistances]. This

matters because an earlier, naive version of the program's expectation -- that

ultrametricity would emerge generically from any Page-Wootters construction -- is thereby

ruled out. The surviving claim is the sharp sufficient condition: diagonal coupling. The

trapped-ion protocol is designed around exactly that surviving claim. A reader who only

sees the positive claim without the ruled-out generic case would misprice the theory.

Second, trapped ions are the wrong economics play. The joules-per-solution competitive

landscape ranks trapped ions last among seventeen platforms (8.5-16.3 J/sol) because

50-100 microsecond gate times overwhelm their room-temperature power advantage

[@qnfo2026jpcubcl]. The testbed movement is therefore a physics claim, not a production

claim; the architectural movement (passive stabilization) exists in part to attack the

gate-time problem that produces this ranking. The qudit extension projects roughly

$10^{-5}$ J/sol for a p-adic-coded qudit platform and pre-registers its own disconfirmation

condition [@quni2026quditadv] -- a projection, not a measurement.

Third, the ZBW signal may not be observable at all in the natural readout frames, and the

amplification question is open. R2 is registered with this constraint attached, not buried.

Fourth, the audit discipline has produced verdicts against the program's own preferred

direction before: the quantum-neural-network audit sustained the methodology claim and

rejected the advantage claim [@qnfo2026bqnnaudit]. The register inherits that willingness.

Fifth, error correction at scale is diagnosed as failing beyond current scales due to

correlated failures rather than independent noise [@quni2026foundations] -- the constraint

that motivates the passive path in the first place.

6. Evidential Weight: What Carries Weight and What Does Not

Under a Bayesian accounting, the register's entries carry unequal weight.

R1 and R2 are pre-registered predictions: their values were published before measurement

[@qnfo2026pwprotocol; @qnfo2026zbwadelic]. A measured match would be surprising under the

null hypothesis of no ultrametric structure, so both carry potential positive evidential

weight -- and none until measured. R4 is likewise pre-registered [@quni2025gkp]. R3 is

simulation evidence: it constrains the modeling claim, not the physics claim.

The tensor-network reading (R5's target) carries ZERO current evidential weight. The

area-law explanation of tensor-network success is the incumbent, and it predicts the

observed success with no reference to p-adic structure; the ultrametric reading was

formulated after the results it explains. It is labeled here as retrodiction, and it is

included in the register only as a test to be run, not as evidence in hand

[@quni2026tnbt].

The falsifiability discipline applies symmetrically. The records that diagnose

unfalsifiability in incumbent frameworks [@quni2026falsifiability] are the same standard

under which the program's own claims are graded here: R5 is the demonstration that the

standard is applied inward.

7. The Architectural Movement: Passive Design as the Embodiment

The architectural claim is that the tree hierarchy is not only a geometry to be measured

but a design principle to be built. Three records carry the argument.

The GKP stabilization record replaces measurement-based feedback with engineered

dissipative reservoir dynamics: a mixed-species crystal couples the logical mode to a

coolant ion, exporting entropy continuously, with convergence to the GKP manifold when

cooling beats heating by more than $\pi$ [@quni2025gkp]. This is a hierarchy-respecting

design in a precise sense: stabilization is delegated to the environment's own structure

rather than imposed by an external controller -- the passive reading of the tree.

The Spin-Free Substrate protocol is the architectural bridge to simulation: it maps

Posner-molecule nuclear spin dynamics (J-couplings of 0.003-0.178 Hz; dipolar relaxation

treated as a Lindblad channel) onto 4-qubit trapped-ion circuits requiring six global

Molmer-Sorensen gates per simulated second at 96.9 percent fidelity [@quni2026spinfreeA].

The design principle named there -- the spin-free substrate -- is a MAP label: a design

pattern, not a new entity, and this paper keeps it at that status.

The metrology record extends the passive claim to sensing: hierarchical noise correlations

exploited for sub-shot-noise precision without active correction, claimed on a

room-temperature photonic demonstration [@quni2026metrology]. The demonstration claim is

reported as claimed by the source record; independent replication is not yet on record.

The architectonics record supplies the thermodynamic framing: active correction fights the

second law and pays energy for the fight; passive structure works with it

[@quni2026architectonics]. The dimensionless reformulation supplies the bookkeeping that

keeps the program's formulas place-democratic [@quni2026ostrowski].

8. The Methodological Movement: Discipline as Infrastructure

The third movement is the reason the first two do not float away. Its three records name

the failure modes: five structural patterns of unfalsifiability [@quni2026falsifiability];

the reification pattern by which models become mistaken for objects [@quni2026reification];

and the theoretical-artifact pattern by which idealized algorithms misdirect engineering

programs [@quni2026foundations].

Two instruments operationalize the discipline. The Bayesian delta-log-odds gate assigns

zero weight to explanations built after the observations they explain -- applied inward in

Section 6 to the program's own tensor-network reading. The joules-per-solution criterion

turns advantage claims into energy accounting, applied in the audit family

[@qnfo2026bqnnaudit; @qnfo2026jpcubcl; @quni2026quditadv]. A practitioner can adopt either

instrument without adopting any of the program's physics.

9. Practitioner Section: What Can Be Built Today

Nothing in this paper requires new theory to become useful. Four artifacts can be built by

an engineering team using the cited records alone.

Artifact 1 -- The decision-tool register (spec-sheet). Section 4 IS the spec-sheet.

An ion-trap team can take R1 directly to an experiment proposal: single ion, carrier and

sideband transitions, conditional state tomography, eight-week estimate, with the

predicted UVR split (0 percent diagonal versus 29-35 percent nondiagonal) as the

accept/reject criterion. The domain of validity is conditional: R1 holds for

clock-rest coupling engineered as specified; it says nothing about platforms without

clock structure. A superconducting team gains nothing from R1; a neutral-atom team might,

if a clock degree of freedom is available -- that is the conditional truth, stated plainly.

Artifact 2 -- The SNR budget template for R2. The ZBW test cannot be proposed until

the signal-to-noise question is closed. The two feasibility records contain the structure

of the budget (amplification mechanism, areal-rate observable, noise floor); a team can

turn them into a spreadsheet decision tool: parameterize amplification gain, measurement

time, and heating rate, and compute whether $\sqrt{2} \cdot 2mc^2/\hbar$ clears the floor

on available hardware [@quni2026vortexD; @quni2026vortexI]. The tool IS the deliverable;

the experiment is downstream of it.

Artifact 3 -- The reproducible simulation kit. R3 is executable today: the

Monte-Carlo code for anomalous diffusion on p-regular trees is deposited with the source

record [@quni2026padicdiffusion]. A team can re-run, extend to p equal to 5 and 7, and

publish the extended dimension table as a benchmark artifact. No hardware required.

Artifact 4 -- The energy-audit template. The joules-per-solution methodology is fully

specified across the competitive-landscape records [@qnfo2026jpcubcl; @quni2026quditadv].

Any hardware team can apply it to its own platform: the metric, the conservative-bound

discipline, and the published baseline values are all in the cited records. This artifact

works regardless of whether any ultrametric claim survives.

The practitioner-facing summary is one sentence: the program hands out one ready

experiment (R1), one spreadsheet problem (R2), one benchmark (R3), and one audit template

(Artifact 4) -- each usable without subscribing to the interpretation that motivated it.

10. Premise-Depth Disclosure

The derived content of this paper is organizational and arithmetic: the register, the

resource counts, the predicted values, and the evidential-weight grading. The physical

content is imported, and the imports are named.

Import 1: the Ultrametric Bridge Theorem -- conditional quantum states under a global

constraint necessarily form ultrametric hierarchies, with hierarchy depth fixing the radix

[@quni2026bridge]. Import 2: the Silent Parameter Principle -- the truncated SU(2)

representation ring is isomorphic to the character ring of the 2-adic units, from which

the $\sqrt{2}$ Zitterbewegung frequency follows [@qnfo2026zbwadelic]. Supporting imports:

Ostrowski's theorem as the place-democracy frame [@ostrowski1916; @quni2026ostrowski], and

the Page-Wootters conditional-state formalism [@pagewootters1983].

Where the premises END: R1's predictions end at the Bridge Theorem's sufficient condition.

R2's predicted value ends at the Silent Parameter Principle. R5's reading ends at the

identification of local-Hamiltonian dynamics with tree structure -- a MAP claim, not a

derived one. R3 and R4 are simulation and engineering claims whose premises end at the

stated physical models (regular-tree random walks; Lindblad dissipative dynamics). No

claim in this paper reaches below these inputs. A reader who rejects Import 1 can discard

R1 without touching the rest of the register; a reader who rejects Import 2 can discard R2.

That separability is deliberate: the register is designed so that its entries fail

independently.

11. Falsifiability Conditions

Formally, the paper is disconfirmed by any of the following measurements:

  1. UVR statistically indistinguishable between diagonal and nondiagonal clock-rest

coupling (kills R1 and, with it, the testable bearing of the Bridge Theorem).

  1. ZBW frequency ratio consistent with 1 (kills R2).
  2. Effective transient dimension growth flattening with p (kills R3's fractal-trap

reading).

  1. No dissipative break-even at cooling-to-heating ratio exceeding $\pi$ in any ion

species (kills R4).

  1. Tensor-network advantage persisting at area-law-predicted levels under nonlocal

perturbations (kills R5's ultrametric reading and confirms the incumbent).

A world in which all five conditions obtain is a world in which the sixteen records are

interesting but disconnected engineering notes, and this paper's organizational claim --

that they are one program -- is the thing that fails. The paper asserts no stronger

consequence.

12. Data-Quality Findings

Cross-system identifier audit of the input records surfaced seven findings: the two

vortex-feasibility records are cross-wired in the archive's identifier table (each row

carrying the other's DOI); three records lack DOI fields entirely; one record carries a

generic internal title; and an identifier-type field is systematically mislabeled. The

duplicate-record pairs (Section 2) are the archive-level expression of the same disorder.

These findings are published here so that the archive can be repaired, and because a

testable program that hides its bookkeeping errors invites the reification failure its own

methodology record diagnoses [@quni2026reification].

13. Conclusion

Sixteen records, three movements, one testable structure. The contribution is not the

claim that ultrametric structure exists in quantum dynamics; it is the claim that the

structure is now cheap to test, and the register that makes the tests concrete. R1 is an

eight-week experiment on existing apparatus. R3 runs on a laptop. The architectural

movement gives the design language; the methodological movement gives the discipline; the

register gives the practitioners something to build, measure, or reject. If the register

fills with null results, the program has still contributed: five kill-conditions executed

is five pieces of knowledge gained, and the falsifiability discipline the records advocate

will have been demonstrated on their own claims.

References