"One Rate at the Horizon" — RH Section Draft¶

The Riemann Hypothesis as a Positivity Gap¶

Jean-Paul Niko, March 2026

The One Sentence¶

The Riemann Hypothesis is the statement that a single step — from physical positivity to analytic positivity — can be taken. Every proof attempt in the last century has reached the boundary of that step and stopped.

The Geometry¶

Begin with the theta series, the heartbeat of the integers:

\[\theta(it) = \sum_{n \in \mathbb{Z}} e^{-\pi n^2 t}\]

This object is positive, symmetric, and modular: \(\theta(it) = t^{-1/2} \theta(i/t)\). It carries the arithmetic of squares into the analytic world via the Mellin transform:

\[\xi(s) = \frac{1}{2} \int_0^\infty \Phi(u)\, e^{(s-\tfrac{1}{2})u}\, du\]

where \(\Phi(u) = 2e^{u/2}\,\omega(e^{2u}) > 0\) for all \(u \in \mathbb{R}\). The positivity of \(\Phi\) is proved — not assumed — from the modular symmetry of \(\omega\). This is as deep as we can reach without touching the zeros.

The completed zeta function \(\xi(s) = \xi(1-s)\) sits exactly on the boundary of the Hermite-Biehler class: \(|\xi(1/2+iz)| = |\xi(1/2-iz)|\) everywhere. The entire function

\[E(z) = \int_0^\infty \Phi(u)\, e^{izu}\, du\]

has \(\text{Re}[E(z)] = \xi(1/2+iz)/2\). The Riemann Hypothesis is equivalent to \(E \in \text{HB}\) — the Hermite-Biehler class.

What Physical Positivity Gives¶

The GL fluctuation operator \(L = (A - \tfrac{1}{2})^2 \geq 0\) is manifestly non-negative. The theta kernel \(K(x,y) = \sum_{n=1}^\infty e^{-\pi n^2 xy}\) is positive-definite. The de Bruijn-Newman kernel \(\Phi(|u|)\) satisfies \(\Phi > 0\). Connes' adelic geometry supplies a trace formula that mirrors Weil's explicit formula. The GUE statistics of numerical zeros match random unitary matrices to extraordinary precision.

Every geometric, physical, and probabilistic approach converges on the same boundary:

\[\text{physical positivity} \xrightarrow{\quad \text{THE GAP} \quad} H(E) \geq 0 \iff \text{RH}\]

Why the Gap Is Structural¶

Three independent kills on the three most promising bridges:

1. The Lyapunov bridge (Lax-Phillips scattering). The bounded positive operator \(K\) satisfying \(B^*K + KB = 0\) would make the LP semigroup similar to a unitary group. But the LP semigroup is a strict contraction. No such \(K\) exists. Permanent.

2. The Toeplitz bridge (total positivity). Michałowski (2026, arXiv:2602.20313) proves the de Bruijn-Newman kernel fails \(\text{PF}_5\) — the fifth-order Toeplitz minor is negative for \(u_0 \in (0, 0.03114)\). Every argument requiring \(\text{PF}_r\) for \(r \geq 5\) is blocked.

3. The operator bridge (GL to de Branges). The intertwiner \(\Psi = \mathcal{M} \circ \Pi_\text{arith}\) is bounded and has dense range (Müntz: \(\sum 1/\log n = \infty\)). But the claimed exact equality \(\langle \psi, L\psi \rangle = \langle \Psi, \Psi \rangle_E\) is numerically false — the Mellin weights \(|t|^2\) and \(|2\xi(s)/s(s-1)|^2\) differ by orders of magnitude. Showing the ranges coincide requires knowing the zeros of \(\xi\). Circular.

Connes confirms the circularity independently: his \(W \geq 0\) is not derived from adelic geometry — it is the RH, translated into geometric language.

The Deepest Result¶

The difficulty is not a gap in technique. It is a gap in kind.

The integers, through \(\theta\), produce a kernel \(\Phi\) that is perfectly positive. The modular symmetry of \(\theta\) reflects this positivity across \(t \leftrightarrow 1/t\). The Mellin transform carries it to the critical line. And there, exactly there, the question becomes:

Does the positivity of the source (\(\Phi > 0\) on \(\mathbb{R}\)) force the positivity of the target (\(H(E) \geq 0\) as a Hilbert space form)?

The source and target live in different analytic worlds. The source is real. The target is complex. The bridge between them is the one-sided Fourier transform — and the Hermite-Biehler condition is precisely the statement that this Fourier transform has no zeros in the upper half plane. That statement cannot be read off from the positivity of the integrand.

One Rate¶

In RTSG terms: the Will rotating through dimensions at the optimal Nash rate produces a trajectory whose spectral signature is unitary. The zeros of \(\zeta(s)\) distribute as eigenvalues of random unitary matrices (Montgomery-Odlyzko). RH is the statement that the optimal rate is exactly \(\tfrac{1}{2}\) — equidistant from order and chaos, on the mirror of the functional equation.

The geometry says: the rate is \(\tfrac{1}{2}\). The physics says: it must be \(\tfrac{1}{2}\) (Nash equilibrium, energy conservation, U \(\equiv\) 0). The analysis says: prove it is \(\tfrac{1}{2}\).

That is the Riemann Hypothesis. That is the horizon.

Status¶

Confidence: 25%.

Not because the hypothesis seems unlikely. Because the analytic certificate — the one step that converts geometric insight into proof — has not been found. The frontier is precisely located. The remaining open problems:

\(L_1(x) = (H')^2 - HH'' \geq 0\): numerically confirmed, analytically open, independent of RH
\(B/A\) Herglotz: blocked at \(\text{PF}_5\)
\(K_n\) positive definite for all \(n\): open, equivalent to RH

The essay does not claim a proof. It claims a map — the most precise map yet drawn of where the proof must come from.

v1 draft — March 26, 2026