Why a framework that never fails proves nothing — and where Paradoxical Interactions actually stop.
"We ought then to regard the present state of the universe as the effect of its anterior state and as the cause of the one which is to follow. Given for one instant an intelligence which could comprehend all the forces by which nature is animated and the respective situation of the beings who compose it — an intelligence sufficiently vast to submit these data to analysis — it would embrace in the same formula the movements of the greatest bodies of the universe and those of the lightest atom; for it, nothing would be uncertain and the future, as the past, would be present to its eyes."
— Pierre-Simon Laplace, Essai philosophique sur les probabilités, 1814
A framework that applies everywhere has no domain — only a claim that keeps confirming itself. Foucault's Pendulum passed the test: what looked like decoupling from the rotating Earth turns out, on closer inspection, to be coupling of its own — frame-dragging included. No fixed point found. The singularity at the center of a black hole fails the test. No system, no elements, no relation between them, no inside, no outside — the category Paradoxical Interactions works with has nothing left to grip there. Not because PI is too weak to reach that far. Because physics itself stops asserting anything there is to talk about at that point. A test that shows where a theory ends is worth more than ten that confirm it.
28. August 2026
The Setup
1965, Birkbeck College, London. Roger Penrose publishes a theorem physics has not shaken off since: under very general conditions, gravitational collapse ends inevitably in a singularity — not a place in space, but the breaking-off of a geodesic, the point where spacetime itself stops being describable. He wins the Nobel Prize for it in 2020. No physicist disputes the theorem. Almost none will say its real consequence out loud: that there is a point where the question "what happens here" stops having an answer — not for lack of data. For lack of a subject.
That is exactly the point where any framework claiming to be universal should submit itself for testing.
The Test That Holds
Foucault's Pendulum seemed, in 1851 at the Panthéon, to demonstrate a fixed point outside the Earth — the fixed stars as a rigid frame against which Earth's rotation could be read. A clean case of decoupling: the pendulum doesn't follow the Earth, and that is exactly what makes the Earth's rotation visible.
General relativity destroys the premise. Rotating masses drag spacetime around with them — frame-dragging, the Lense-Thirring effect. Gravity Probe B measured it directly in 2011: a gyroscope in Earth orbit precesses minutely against a guide star, because the rotating Earth drags its surroundings along. There is no pure outside. Only graded coupling — to a larger frame instead of a smaller one. That is von Foerster's point, written in metric: no position outside the system from which to observe without distortion.
PI didn't pass this test because it wanted to. It passed because the attempted refutation, on closer inspection, wasn't one.
What Self-Reference Needs
A point has no parts. Self-reference needs at least two elements capable of relating to each other, and a boundary separating inside from outside. Luhmann's autopoiesis works because a system has enough internal complexity to reproduce itself as distinct from its environment. At a true point singularity, that complexity doesn't exist.
It goes deeper than "nothing inside." Under the Penrose-Hawking theorems, the singularity is not a location within spacetime carrying any property at all — open, closed, coupled, self-referential. It is geodesic incompleteness: the point where the manifold itself stops being defined. There is no system left to predicate anything of, because the differentiable structure that "system" means in this context ends exactly there.
That is structurally closer to "what lies north of the North Pole" than to "is the system open or closed." Not unanswerable for lack of data. Unanswerable because the question presupposes a domain that no longer exists there.
The Rescue That Isn't
The obvious move to save the theory is this: “But the physicists talking about the singularity are PI.”
True.
And the trap anyway.
If every case where PI fails at the object level is automatically reclaimed one level higher, PI stops being falsifiable. A theory that absorbs its own refutation has not become stronger. It has become immune.
“PI is whatever PI describes” would not be a framework.
It would be dogma.
The real PI sits elsewhere: not in the singularity itself, but in how the discipline handles its own methodological limit.
Cosmic censorship, proposed by Penrose in 1969 and unproven to this day, claims that singularities always remain hidden behind event horizons. Never naked. Never observable.
Known counterexamples — overspinning or overcharged black holes in exact solutions of the field equations — are often treated as “unphysical.”
Not because the methodological reflex is irrational.
Because accepting them as physically relevant would threaten causality and predictability — the operating basis of physics as a predictive discipline.
The Cosmic Censorship PI:
Physics displaces the unsolvable case behind a horizon instead of leaving it standing.
Everyone acts rationally:
- Every individual physicist — clean, methodical, cautious about unproven edge cases
- The field as a whole — defends its own survival as a predictive discipline
- Outcome — an unproven conjecture gets treated like a law of nature, structurally, not by agreement
All are guilty. None is at fault.
The Boundary as Finding
At the object itself: no PI. No system, no self-reference, no category of coupling applies. That has to stand — and stay standing, or it becomes exactly the move the previous section just described.
A theory that never fails has no test result to show — only an application that keeps confirming itself. Foucault's Pendulum and the singularity aren't a contradiction, then. They're a pair: one case where the attempted refutation, on closer look, wasn't one. The other where the category simply has nothing left to grip. Together they're worth more than any single proof that PI applies "everywhere."
Laplace's demon would need only one thing: all the forces, all the positions, a formula large enough. At the singularity there are no positions left for it to know, and no formula that still applies. The demon doesn't fail there for lack of computing power. It fails for lack of a subject.
The singularity knows no PI. That's exactly why it holds.
Related Posts
Why a pendulum proves what a genius was condemned for — and why indifference is the only witness that counts.
Why a sixty-year-old model of black hole evaporation just got corrected by a fiber-optic bench in Paderborn — and why the correction was never about black holes, but about what a boundary is allowed to do to the thing that draws it.
Why String Theory didn't fail physics — and why the structure made sure it couldn't.
Why every attempt to close mathematics for good has expanded it instead — and why an AI that treats it as closed cannot learn the one thing that matters about it.
On piinteract.org:
- ["Accept Incompleteness"] — Gödel and Penrose-Hawking are saying the same thing on two levels: incompleteness isn't a gap to be closed, it's the condition under which a system stays meaningful at all.
- ["Shared Blindspot"] — A theory confirmed everywhere shares the blind spot of a collective that can't see its own premise: it can't find its own limit because it never looked for one.
- ["Galileo's Paradox"] — Not every structure that resists a conclusion has thereby refuted it. Cosmic censorship is the modern case: defended, not proven.
- ["Academic Escape Reflex"] — The unsolvable edge case doesn't get solved, it gets methodically declared unphysical. That exact reflex has kept cosmic censorship standing for over fifty years.
See also (external links):
Gravitational Collapse and Space-Time Singularities — R. Penrose, Phys. Rev. Lett. 14 (1965) — The original theorem: gravitational collapse leads, under general conditions, inevitably to geodesic incompleteness, not to a describable location. The Nobel Prize in Physics 2020 — Roger Penrose — Institutional confirmation that the theorem is treated as proven and foundational, not a footnote in the literature. Golden Oldie: Gravitational Collapse — The Role of General Relativity, R. Penrose, General Relativity and Gravitation (2002 reprint of 1969) — The original source of the cosmic censorship conjecture, unproven to this day and still the field's working assumption. Gravity Probe B Results — Stanford / Einstein.Stanford.edu — The direct measurement of frame-dragging that confirms the Foucault-Pendulum analogy — and disproves its naive reading at the same time. A Philosophical Essay on Probabilities — Pierre-Simon Laplace, Project Gutenberg — Full text of the source for Laplace's demon, readable in the context of his original argument.
Paradoxical Interactions (PI): When rational actors consistently produce collectively irrational outcomes — not through failure, but through structure.
All are guilty. None is at fault.
Peter Senner Thinking beyond the Tellerrand
contact@piinteract.org
https://piinteract.org
Co-created with Claude (Anthropic) & ChatGPT — three incomplete systems making each other's gaps visible.