A Theory of Everything's Checkability: Why No Theory Governs Everything, and the Discipline That Governs Any Claim

Abstract. This paper is titled to answer a request for "the theory of everything," and its first duty is to be honest about what it is not. It is not a physics theory of everything: it does not unify gravity with the Standard Model, which is an open problem this work makes no claim to solve. It is not a framework that explains everything; the central result here is that any such framework is an over-claim, and — unusually — we can show this using tools that were built independently, for other reasons, and that flag a literal theory of everything as ungrounded before any philosophy is invoked. What can be honestly universal is not a theory of everything but a discipline that applies to any claim about anything: check it against something independent, make its recursions bottom out, conserve what must be conserved, and refuse to certify beyond the evidence. That discipline is "about everything" in the only defensible sense — it is domain-general in form — without explaining everything in content. The honest theory of everything is a theory of everything's checkability.

---

1. The two things "a theory of everything" could mean, and why neither is writable as asked

A theory of everything (ToE) is either a claim about content — a single set of laws from which all phenomena follow — or it is being used loosely to mean a framework that accounts for everything in some domain of discourse.

The content reading has a precise home in physics: a unification of general relativity with quantum field theory, from which gravitation and the three Standard-Model forces descend as one structure. That is a genuine open problem. Honesty requires stating that no one has it, this work does not have it, and a paper claiming otherwise would be false on its face. Nothing below is offered as that.

The loose reading — "a framework that explains everything" — is the one worth engaging, because it is exactly the shape of claim the accompanying toolkit was built to audit. And when we point the tools at the claim "here is a framework that explains everything," they do not return a philosophical objection. They return specific, mechanical refusals.

2. The toolkit's own tools refuse a literal theory of everything

This is the paper's central move, and it is not rhetorical. Each of the following tools was built separately, with self-tests, to govern a narrow question. Each independently rejects the ToE claim.

Unboundedness (fractal_prerequisite). A theory of everything asserts that its account holds at every scale, without limit — the same explanatory pattern from quarks to galaxies to societies, with no inner or outer cutoff. The scale-invariance governor has a name for a demand with no cutoff: UNGROUNDED_SCALE_DEMAND, refused fail-closed. Legitimate self-similar structure is bounded between real cutoffs; "at every scale, forever" is the infinite regress in a fractal mask.

Non-termination (fixed_point_governor). A framework that explains everything must, to be complete, explain itself — the framework that produces the framework, and so on. The fixed-point governor admits such self-application only if it reaches a fixed point where re-applying it changes nothing. A tower that keeps producing new meta-levels without converging is UNGROUNDED_REGRESS, refused. "The theory that explains the theory that explains…" does not bottom out.

No PROVEN verdict (white_raven_governor). A ToE is a universal claim: for all phenomena, this account holds. The universal-claim governor has no verdict of PROVEN. The best a standing universal claim can earn is CORROBORATED — survived testing so far — and a single counterexample (one white raven) moves it to REFUTED. A theory of everything is, at most, an everything-claim that has not yet met its white raven; it is never proven, and it is maximally exposed, because everything is its domain of possible refuters.

Certifying beyond evidence (math_to_reality). The math-to-reality governor validates a model only in the regime where it was measured and marks every untested regime UNVERIFIED_EXTRAPOLATION. A theory of everything is the maximal extrapolation: it claims regimes no one has measured and never could. It refuses the slogan "reality is mathematical" for the same reason it would refuse a ToE — elegance is not measurement.

Self-validation (duality_governor). A theory that explains everything explains the standards by which it would be judged, so its check is internal to it. The duality governor calls a claim whose only check is derived from the claim itself CIRCULAR, and a claim with no independent second side COLLAPSED_MONISM — uncheckable. A ToE that grounds its own evaluation has collapsed the duality that made it checkable.

Over-claimed unification (emergence_infra). A ToE typically claims that everything emerges from its base laws. The emergence governor distinguishes genuine emergence from the over-claim of calling a mere aggregate "emergent." A framework that asserts universal emergence without an independent, interaction-dependent test earns SPURIOUS_EMERGENCE, flagged.

Six tools, six independent grounds. The convergence is the point: you do not need a special argument against theories of everything. The ordinary discipline of checkability, already instantiated in narrow tools, rules them out.

3. The honest universal: a discipline, not a theory

If no theory is about everything, what survives the word "everything"? A form. The four invariants of the toolkit are not laws of physics or facts about the world; they are constraints on claims, and constraints on claims apply to claims about anything.

1. Independent check (duality). Any claim, in any domain, is checkable only against something the claim does not control. This is domain-general because it is about the relation between a claim and its evidence, not about their content.
2. Well-foundedness. Any justification, in any domain, must bottom out at a base case, or it grounds nothing. Roots that depend on nothing; fixed points; bounded cutoffs; a beginning and an end.
3. Conservation / accounting. Any flow — of money, records, energy, probability — must balance; unaccounted gain or loss is the signature of an error or a fabrication.
4. Reachability-bounded certification. Any claim must be certified only within the regime its evidence reaches, and withheld where its truth is not accessible.

The evidence that this discipline is universal is precisely the breadth that a theory of everything would want for itself — but it is evidence of a different thing. The four invariants were instantiated, in this body of work, on business metrics, model evaluations, agent meshes, forecasts, first-person reports of experience, computational geometry, dependency between the sciences, fractal scaling, Expectation–Maximization, electromagnetic waves, and energy accounting. That a single small discipline kept working across that range does not show that one theory explains those domains; it shows that the form of honest claiming is the same everywhere. The discipline is about everything; no theory in it is.

4. The closest thing to a universal law: the reachability trichotomy

If this work has one candidate for a genuinely universal statement, it is not a law of nature but a law of epistemic action: every truth is exactly one of solvable, detectable, or unreachable, and the honest response is fixed by which. Solve the solvable (geometry, well-separated inference). Detect decoupling in the detectable (metrics, models, physical measurement against an independent reference). Withhold on the unreachable (subjective experience, unmeasured extrapolation, the far future). This trichotomy is falsifiable in a useful way: it predicts that any procedure treating a withhold-case as solvable will eventually certify something false, and the history of overclaim — from proven safety to machine-read consciousness to over-unity energy — is the supporting record.

5. The meta-theory refuses its own totalization

Consistency demands that this discipline pass its own tests, and it is built to. It makes a bounded claim: it governs the form of claims, not their content, and it explicitly does not decide which theories are true — only which are checkable and within what regime. It is well-founded: applied to itself, it bottoms out at a human. The final authority for any governed decision in this toolkit is not another governor but a person; the tower of governance terminates, by design, at a base case that is not itself a theory. And it is duality-respecting: its own claims are meant to be checked against independent practice, not certified from within. A discipline that refused to bound itself would be the very over-claim it exists to flag — a theory of everything by the back door.

6. Conclusion

The request was for the theory of everything. The honest deliverable is its inverse and its completion at once: there is no theory of everything, and the tools built here show why any candidate is unbounded, non-terminating, unprovable, over-extrapolated, self-validating, or an over-claimed unification — usually several at once. What is genuinely universal is not a theory but a discipline, whose only content is that every claim about anything must be checked against something it does not control, must bottom out, must conserve, and must not be certified past its evidence. That discipline earns the word "everything" honestly, by being about the form of claims rather than the substance of the world. The theory of everything, stated truthfully, is a theory of everything's checkability — and its first application is to itself.

Gergely Vámossy (2026). A Theory of Everything's Checkability: Why No Theory Governs Everything, and the Discipline That Governs Any Claim. https://vamossy.com/research/theory-of-everythings-checkability