← Back to Home

One set of checks, four physics models

To Become a Stone with Six Birds: A Physics is A Theory

Applies the same three closure checks to four physics models: quantum dephasing, gas kinetics, turbulence filtering, and gravitational averaging.

Preprint - v1Not peer reviewedPublished Jan 29, 2026Open access

Plain-language overview

This paper treats physics theories as descriptive layers, each defined by what you keep (a lens), how you fill in what you discard (a completion), and how you verify the result (audits). It tests this view on four concrete models: quantum dephasing (removing quantum coherences), gas kinetics (the BGK collision model), turbulence filtering (large-eddy simulation), and gravitational averaging (backreaction). The same three checks recur everywhere: Does the compression stabilize? Does zooming out lose information monotonically? Do "compress then evolve" and "evolve then compress" agree? When they disagree, a correction term is forced.

A physics is a theory only when its closure is coherent under the chosen lens.

- Ioannis Tsiokos

At a glance

Closure coherence

Compress the data, then compress again. If the second pass changes nothing, the macro description is stable. The "idempotence defect" measures how far you are from this.

Information only flows downhill

Zooming out can only lose information, never gain it. This prevents false claims that a coarser description somehow knows more than the detailed one.

Evolution-vs-packaging mismatch

"Evolve then compress" vs. "compress then evolve" can give different answers. When they disagree, the macro theory needs a correction term to stay self-consistent.

Four instantiations

Quantum dephasing, BGK gas kinetics, turbulence filtering (LES), and gravitational averaging (backreaction) all tested with the same three checks.

Core lens

How the paper is structured

A small set of primitives, each with explicit audits and controls.

What you keep

Lens

Decides what information survives at the macro level. Everything the lens discards is gone.

How you fill

Completion

Fills in the discarded micro-detail with a default assumption (e.g., thermal equilibrium, spatial smoothness). This closes the description.

How you verify

Audits

Three checks: Does compressing twice give the same result? Does zooming out only lose information? Do "evolve then compress" and "compress then evolve" agree?

Highlighted results

What the paper establishes

Each claim is paired with a control or audit.

Quantum dephasing is perfectly stable

Removing quantum coherences in a fixed basis is exactly idempotent: doing it twice gives the same result as doing it once. Zero defect.

Gas kinetics improves with more collisions

The more frequently particles collide, the better the macro description (local equilibrium) closes. Failure modes are explicitly identified.

Turbulence and gravity force corrections

The mismatch between "evolve then compress" and "compress then evolve" grows with filter width or spatial variation, forcing structured correction terms.

Clean baselines stay near zero

When the lens is an exact projection or when there is zero spatial variation, mismatch and stability defects are near zero as expected.

Methods and reproducibility

How the results are supported

  • Reproducible Python harness that regenerates every figure, table, and metric from configuration files.
  • Machine-checked algebraic lemmas (Lean 4) anchoring the closure and audit identities.
  • Side-by-side null-vs-intervention comparisons for each of the four physics models.
  • Artifact bundles with configuration hashes, metrics, and provenance logs per run.

Sanity checks

  • - Clean baselines (exact projections, zero heterogeneity) produce near-zero stability defects and mismatch.
  • - Information monotonicity holds within numerical tolerance in all controlled tests.
  • - Interventions separate cleanly from baselines across all four models.

Media-ready

Figures and artifacts

Every plot in the paper is generated from deterministic scripts that produce run summaries, audit tables, and provenance logs.

- Stability defect and evolution-vs-packaging mismatch sweeps

- Quantum, kinetic, turbulence, and gravitational model runs

- Audit tables with hashed configurations

Regenerate figures from code

Limitations and scope

Read-this-first caveats

  • Status: research preprint, not peer reviewed.
  • Toy models demonstrate the closure logic; they are not full continuum physics simulations.
  • No analytic continuum-limit bounds or convergence proofs are claimed.
  • Results depend on the chosen lens, completion rule, and resolution scale.

Citation

How to cite

Ioannis Tsiokos (2026). To Become a Stone with Six Birds: A Physics is A Theory. Zenodo. https://doi.org/10.5281/zenodo.18412131

BibTeX

@misc{tsiokos2026become,
  title = {To Become a Stone with Six Birds: A Physics is A Theory},
  author = {Tsiokos, Ioannis},
  year = {2026},
  publisher = {Zenodo},
  doi = {10.5281/zenodo.18412131},
  url = {https://doi.org/10.5281/zenodo.18412131}
}

Press and contact

Talk to the author

For media inquiries, figures, or walkthroughs of the artifacts, reach out directly.

Ioannis Tsiokos

ioannis@automorph.io

Corresponding author - Press contact

Questions welcome about closure stability, information monotonicity, and evolution-vs-packaging mismatch.