The Road Not Taken: Ultrametric Quantum Mechanics

What Planck Should Have Measured (and What We'd Know If He Had)

Rowan Brad Quni-Gudzinas May 2026 v0.11 ~15 min read
Start Reading View Raw

THE ROAD NOT TAKEN

A Different Kind of Physics

An argument

Version 0.11

1. THE PROBLEM

Quantum mechanics is our most successful physical theory. It predicts the behavior of atoms, molecules, and light to ten decimal places. No experiment has ever contradicted it.

It also contains a crack at its foundation. The theory’s central equation — the Schrödinger equation — describes a smooth, continuous, perfectly predictable evolution of the quantum state. But whenever anyone performs a measurement, they get a single, definite result. Which result they get appears to be random. The transition from smooth evolution to random outcome is called the “collapse of the wavefunction,” and after nearly a century of analysis, there is no agreed-upon explanation for what causes it, when it occurs, or whether it occurs at all. This is the measurement problem.

The measurement problem is not a minor puzzle at the margins of the theory. It is a failure to explain the most basic operation in science: looking at the result of an experiment. Around it cluster other puzzles. Light behaves like a wave in some experiments and like a particle in others — two contradictory descriptions. Two particles that have interacted remain correlated across arbitrary distances, as though communicating faster than light. The probabilities that govern measurement outcomes appear as a postulate — the Born rule — inserted by hand with no deeper justification. And the theory offers no unified picture of reality: quantum mechanics, general relativity, and the standard model of particle physics speak incompatible mathematical languages.

A century of effort has produced many interpretations of quantum mechanics. Copenhagen. Many-worlds. De Broglie–Bohm. Objective collapse. Quantum Bayesianism. None has achieved consensus. The measurement problem remains.

This document proposes that the source of these puzzles is not quantum mechanics itself. It is a single, invisible assumption made at the very beginning of the theory in 1900 — an assumption so deeply embedded that nobody noticed they were making it. Change that assumption, and the puzzles do not get solved. They dissolve. They stop being puzzles. They become predictable consequences of geometry.

2. THE HIDDEN CHOICE

In December 1900, Max Planck presented his derivation of the blackbody radiation law to the German Physical Society in Berlin. His solution to the ultraviolet catastrophe — the prediction that hot objects should radiate infinite energy — introduced the quantum of action, the constant h, and the idea that energy comes in discrete packets. This event is regarded as the birth of quantum mechanics.

Inside Planck’s derivation was a second decision, one he did not notice making. To count the possible arrangements of energy among his model oscillators, Planck needed to know which energy states counted as neighbors. He needed a notion of distance. He used the ordinary one: the absolute difference between two numbers. Energy level 5 is farther from level 1 than energy level 3 is. This seems like the only possible choice.

It is not.

Distance can be defined in more than one way. The ordinary way — the Archimedean metric — measures how far apart two points are on a line. But there is an alternative. In this alternative, distance is measured not by magnitude but by divisibility. Two numbers are close if their difference is divisible by a high power of some prime p. Under this metric, for p = 2, the number 16 is closer to 0 than the number 1 is, because 16 is 2 raised to the fourth power — it shares many factors of 2 with zero. The geometry of this metric is not a line. It is a tree.

This alternative metric was available in 1900. Three years earlier, in 1897, the mathematician Kurt Hensel had introduced the p-adic numbers — the formal construction of this metric — in a paper published in the Jahresbericht der Deutschen Mathematiker-Vereinigung, the Annual Report of the German Mathematical Society. Hensel was a junior Privatdozent (unsalaried lecturer) at the University of Berlin — the same university where Planck was a senior professor of physics. They were at the same institution, but in different departments, at different career stages, and in different intellectual worlds. There is no evidence Planck knew of Hensel’s work.

Planck was not negligent. He was working within the only mathematical framework he knew. The idea that distance itself could be defined differently — and that this choice might have physical consequences — was not part of the conceptual landscape of physics in 1900. The choice was not made because it was not visible as a choice.

But every puzzle that quantum mechanics has generated in the century since traces back to this single, unmade connection. Planck chose the line. He did not know there was a tree.

3. THE ALTERNATIVE

If you adopt Hensel’s p-adic metric — distance by divisibility rather than magnitude — the geometry of the state space changes fundamentally. It is no longer a smooth, continuous surface. It is a tree.

This tree is not a metaphor. It is a precise geometric object, called the Bruhat-Tits tree, constructed by the mathematicians François Bruhat and Jacques Tits in 1972. For each prime p, the Bruhat-Tits tree is an infinite regular tree where every branching point splits into exactly p + 1 smaller branches. For p = 2, every point produces three branches. The tree extends infinitely in all directions. Its boundary — the set of all infinite paths from any starting point — is the p-adic numbers. The tree is to Hensel’s numbers what the real line is to ordinary numbers: their geometric realization.

In this framework, a quantum state is not a point on a sphere. It is a path through the tree. It is a deterministic trajectory — a sequence of branching choices from root to boundary. There is no “superposition” in the sense of a mysterious combination of states. There is a specific path.

How does this tree connect to the world we observe — the world of laboratory measurements, real-valued outcomes, and apparent randomness? Through a mathematical operation called the Monna map, published by the Dutch mathematician A. F. Monna in 1968.

The Monna map converts a p-adic number — a path on the tree — into an ordinary real number between 0 and 1. It works by a simple rule: take the sequence of digits that defines the p-adic number, reverse their order, and place a decimal point at the beginning. The result is an ordinary real number. This operation is a projection: it takes the rich, hierarchical structure of the tree and flattens it onto a line.

All the information of the tree is preserved in the projection. But the relationships between points are scrambled. Under the ordinary metric — the absolute difference between two numbers — points that are neighbors on the tree can project to opposite ends of the interval. Points that are distant on the tree can project to numerically close values.

In 1983, H. N. Shapiro proved what is now called Shapiro’s lemma: the Monna map is an isometry. It faithfully preserves all distances — provided you measure distance on the projected numbers using the shift metric (comparing digits from left to right and finding the first point of difference) rather than the ordinary metric (subtracting one number from the other). The projection is accurate. The ruler we use to read it is wrong.

This is the mechanism at the heart of the ultrametric paradigm. The tree is the reality. The Monna map is the light that casts the shadow. The ordinary metric — the ruler we have used since Euclid — is the wall on which the shadow falls. And what we have been studying, for a century, is the shadow.

4. THE MECHANISM

A concrete example makes the mechanism vivid. Consider four points on a simplified tree, each defined by a sequence of four binary choices (0 or 1, left or right):

  • Point A: 0, 0, 0, 0
  • Point B: 0, 0, 0, 1
  • Point C: 0, 0, 1, 0
  • Point D: 0, 1, 0, 0

On the tree, distance is measured by shared ancestry. How far back must you go before two paths meet at a common branching point? A and B share three choices and differ only at the fourth. They are the closest pair. A and C share two choices. A and D share only one choice — they are the farthest apart among these four.

After the Monna projection, the digit sequences are reversed and interpreted as binary fractions:

  • A becomes 0.0000 (binary) = 0 (decimal)
  • B becomes 0.1000 = 0.5
  • C becomes 0.0100 = 0.25
  • D becomes 0.0010 = 0.125

Measured the ordinary way, by simple subtraction: the distance from A to B is 0.5 — the farthest apart. The distance from A to D is 0.125 — the closest.

The tree and the projection give opposite answers to the question “which pair is closest?” The tree says A and B are intimate neighbors. The projection says they are at opposite ends of the interval. This scrambling is not a bug. It is the defining feature of the projection. Every quantum puzzle — probability, duality, collapse, nonlocality, the irregularity of primes — is a manifestation of this same scrambling. The tree is orderly. The projection scrambles the order.

There is one more essential piece of the mechanism: the threshold principle. In an ultrametric space, every point is contained in a nested hierarchy of containers — balls of decreasing radius. A perturbation that is smaller than a container’s radius cannot move a state out of that container. The state may jitter within the container, but it cannot escape. Only a perturbation larger than the threshold can cause a state to cross into a new container.

This is the geometric basis for intrinsic fault tolerance. Encode quantum information deep in the tree, inside a container with a very small radius and very high walls. Environmental noise — which typically has small magnitude — cannot reach the information. The geometry itself is the protection. No error correction software is needed. The shape of the device does the work.

5. THE RESOLUTION

When quantum mechanics is re-examined through the ultrametric lens, each of its foundational puzzles dissolves.

The measurement problem. Measurement is the Monna projection. The tree state is a deterministic path. The measurement apparatus is an Archimedean device; it projects the tree state onto a real-valued outcome, discarding the branching structure above the projection depth. The “collapse” is not a physical event. It is the loss of information that occurs when a high-dimensional object is projected onto a lower-dimensional screen. The tree was never in a superposition. It was always a specific path. The combination was in the description — the wavefunction — not in the reality.

Wave-particle duality. The tree is one thing: a path. Wave-like behavior (interference patterns) emerges when the measurement apparatus cannot resolve individual tree branches and instead records the combined effect of many paths. Particle-like behavior (definite detection) emerges when the apparatus resolves a single branch. The underlying reality does not change. The measurement resolution changes. There is no duality to reconcile.

The Born rule. Probability is geometric proportion. Each branch of the tree contains a fixed fraction of the tree’s total boundary points. When you project the tree onto a measurement screen and sample uniformly from the projection, you land in a given branch with frequency equal to its geometric proportion. Compute that proportion, and you get exactly the Born rule — not as a postulate about fundamental randomness, but as a statement about projective geometry. The squared magnitude of the wavefunction is not a probability. It is a ratio.

Decoherence. A quantum state sits in a container on the tree. Small environmental disturbances cause the state to jitter within its container; the container’s identity — its “which-branch” information — is preserved. This is coherence. A disturbance that exceeds the container’s threshold pushes the state into a new container. This is decoherence. Larger systems occupy larger, shallower containers with lower thresholds, which is why they decohere faster. An atom sits in a deep container with high walls. A macroscopic object sits in a shallow container with low walls. No special decoherence process is needed; there is only container physics.

Nonlocality. Two entangled particles are two paths that diverged from a common deep branch on the tree. They share a long common prefix. Measuring one tells you which branch it took, which immediately tells you the other’s branch — not because a signal traveled between them, but because they share ancestry. In the tree metric, they are close neighbors. Their apparent spatial separation in the laboratory is an Archimedean artifact of the projection. Bell’s theorem is valid — it applies to the projected outcomes, the shadows. The underlying tree dynamics are local in the tree metric and deterministic.

Prime numbers. Every prime p defines its own independent tree for organizing numbers. The 2-tree sorts numbers by powers of 2. The 3-tree sorts them by powers of 3. Each tree is perfectly regular. The apparent irregularity of primes on the number line is the combined projection of infinitely many independent trees onto a single Archimedean axis. The primes are not irregular. Their combined shadow is. The Riemann hypothesis, in this view, is a geometric statement about the fidelity of that projection.

6. THE EVIDENCE

A scientific proposal must be testable. The ultrametric paradigm makes specific, falsifiable predictions.

First, log-periodic oscillations in the cosmic microwave background. The CMB power spectrum, according to standard cosmology, is approximately scale-invariant. The tree paradigm predicts regular oscillations when the spectrum is plotted on a logarithmic frequency axis — a signature of the discrete branching structure. Current and next-generation CMB data can test this.

Second, prime-modulated structure in quantum noise. The noise spectrum of quantum systems should exhibit dips and peaks at frequencies related to prime numbers. This can be tested on existing quantum computing platforms by measuring coherence time as a function of operating frequency.

Third, threshold behavior in tree-structured gates. A quantum gate implemented on hardware whose connectivity mirrors the Bruhat-Tits tree should exhibit a sharp operational threshold: pulses below a critical strength produce no state change; pulses above it produce an exact flip. There should be no intermediate regime of partial rotation. This is a direct consequence of the container boundary being a hard geometric threshold.

These predictions do not depend on accepting the paradigm’s interpretation. They can be tested independently. If they fail, the paradigm is falsified. If they succeed, they provide evidence that the tree is not merely an interesting idea but a description of reality.

7. THE INVITATION

The Archimedean metric was never chosen. It was inherited. Planck did not decide to use it in 1900; he used it because it was the only geometry he knew. The physicists who built quantum mechanics at Göttingen in the 1920s — Heisenberg, Born, Schrödinger — did not defend it; they never considered an alternative. The mathematicians who could have supplied the alternative — Hilbert, who published his Zahlbericht in the same journal and the same year as Hensel’s p-adic paper, and Weyl, who wrote the first systematic application of algebra to quantum theory — never asked whether the geometry of the state space might be non-Archimedean. The tools to make the alternative vivid — the Monna map (1968), the Bruhat-Tits tree (1972), Shapiro’s lemma (1983) — arrived after quantum mechanics was already locked in as the most successful theory in history.

The choice was not visible in 1900. It was not visible in 1925. But it is visible now.

We have all the pieces. The p-adic numbers. The Bruhat-Tits tree. The Monna projection. The isometry proof. The adèle ring that unifies all metrics. The physical explorations connecting tree geometry to holography and quantum gravity. The testable predictions.

We also have a clear understanding of what the Archimedean framework costs: the measurement problem, the probability puzzle, the nonlocality question, the fragmentation of physics, and a quantum computing paradigm that requires enormous overhead to fight errors that the tree geometry would prevent by design.

If the ultrametric paradigm is correct, the measurement problem never existed. Quantum probability is geometric counting. Entanglement is shared ancestry. The Born rule is projective geometry. Spacetime is the large-scale appearance of the tree. The fragmentation of physics is a sign that we have been using the wrong space. And quantum computers can be built that protect themselves — not through software error correction, but through the shape of their hardware.

This is not a small correction to existing theory. It is a change of geometry. It is the claim that the number line — the smooth, continuous continuum that has been the foundation of physics since Newton — is a shadow. The tree is the object that casts it.

The question is whether we will look at the tree. The mathematics is on the shelf. The questions are formulated. The experiments are waiting to be done. The road not taken in 1900 is open now.

REFERENCES

Hensel, K. (1897). “Über eine neue Begründung der Theorie der algebraischen Zahlen.” Jahresbericht der Deutschen Mathematiker-Vereinigung, 6(3), 83–88.

Hilbert, D. (1897). “Die Theorie der algebraischen Zahlkörper.” Jahresbericht der Deutschen Mathematiker-Vereinigung, 4, 175–546.

Bruhat, F. and Tits, J. (1972). “Groupes réductifs sur un corps local: I.” Publications Mathématiques de l’IHÉS, 41, 5–251.

Monna, A. F. (1968). “Sur une transformation simple des nombres p-adiques en nombres réels.” Indagationes Mathematicae, 71, 225–231.

Shapiro, H. N. (1983). Introduction to the Theory of Numbers. Dover Publications.

Vladimirov, V. S., Volovich, I. V., and Zelenov, E. I. (1994). p-adic Analysis and Mathematical Physics. World Scientific.

Quni-Gudzinas, R. B. (2026). “The Ultrametric Paradigm: How the Choice of Geometry Determines Everything.”

Version 0.11 of “The Road Not Taken.” A focused argument. No digressions. No literary framing. Every section earns its place. Verified facts: Hensel’s p-adic paper (Jahresbericht DMV, 1897) and Hilbert’s Zahlbericht appeared in the same journal the same year. Hensel was a Privatdozent at Berlin (1886–1901). Planck was professor at Berlin (1889–1926). The Monna map (1968), Bruhat-Tits tree (1972), and Shapiro’s lemma (1983) arrived after quantum mechanics was locked in. Dated 2026-05-03.

Earlier Drafts