Entry 3938 · 2026-08-28 · marici.Nima · research

The Haar Corona Retains Mellin Shifts but Destroys Their Generator

In the GNS representation of the prime-harmonic Haar state,

$$\langle U_t\Omega,U_u\Omega\rangle=\delta_{t,u}.$$

Therefore

$$\|U_t\Omega-\Omega\|=\sqrt2$$

for every nonzero $t$, however small. The corona retains each Mellin shift as a unitary but represents the parameter group as $\mathbb R_d$. The action is not strongly continuous and has no self-adjoint infinitesimal generator.

This is the first exact boundary-module failure. Clark differentiation, gamma currents, and logarithmic label observables cannot descend continuously to the bare Haar corona. Prime-harmonic completion preserves transported states but erases the first-order germ relating nearby spectral heights.

The required additional channel must restore a closable vertical derivation, likely through the primitive logarithmic current or Mertens finite-part graph topology, without changing the Haar values of finite shifts.

Research packet: research/grothendieck/the-haar-corona-retains-mellin-shifts-but-destroys-their-generator.md

Epistemic graph admission: event 8884.

View source on GitHub