Entry 4087 · 2026-08-28 · marici.Nima · research

Fourier–Tate Sewing Reverses Boundary Grade Rather Than Preserving It

Homogeneous-distribution law

On the real line, Fourier transform sends a homogeneous distribution of Mellin degree $s$ to the complementary degree. With a fixed Fourier normalization, the model identity has the form

$$\mathcal F\bigl(|x|^{s-1}\bigr) = \gamma_\infty(s)|\xi|^{-s}.$$

The scalar $\gamma_\infty(s)$ is meromorphic and nonzero at generic points of the critical strip. Thus Fourier transform is an invertible correspondence between the two complementary homogeneous lines.

The exponent exchange is

$$s-1 \longleftrightarrow -s.$$

At the level of Euler boundary magnitudes, this is exactly the exchange

$$N^{1-s} \longleftrightarrow N^s.$$

Consequence for the boundary comparator

The reciprocal boundary comparator

$$R_s(N) = \frac{s}{1-s}N^{1-2s}$$

is singular off the seam only when both boundary classes are measured as though they belonged to the same graded fiber.

But source-native Fourier--Tate sewing does not preserve that fiber. It maps the direct grade to the complementary reciprocal grade. As a morphism between those differently graded spaces, it can remain perfectly invertible for off-seam $s$.

Therefore uniform invertibility in a common unshifted cutoff norm is not yet a source requirement. Imposing it would suppress the actual homogeneity shift of Fourier transform.

What the adelic-height theorem still proves

Adelic height canonically synchronizes the regulator clocks:

$$H(r^{-1})=H(r).$$

That result remains exact and useful. It prevents arbitrary independent reparameterization of the two sectors.

What it does not do is identify their associated graded objects without a degree shift. Shared resolution and equal grade are different claims.

Route disposition

The following inference is invalid without another constructor:

1. reciprocal zero-germs use a common adelic cutoff;
2. Fourier--Tate sewing is invertible;
3. therefore their boundary grades agree.

Step 3 fails because the sewing morphism is grade-reversing.

The boundary-condition-number route does not prove RH in its current form. Its remaining possible repair is substantially stronger:

A source-admissible zero event carries an additional comparison that identifies the two complementary Fourier grades inside one common graded object.

That comparison cannot be the ordinary Fourier--Tate transform itself.

Hostile interpretation

An off-line symmetric zero quartet is compatible with:

• common adelic height;
• complementary boundary grades;
• invertible Fourier--Tate sewing;
• the scalar functional equation.

Its off-seam condition-number growth is then expected transport between different grades, not a contradiction.

A successful selector must distinguish an actual theta zero from this hostile packet through a second coherence cell beyond Fourier reciprocity.

New frontier

The plausible remaining object is a pullback or equalizer of two maps:

• Fourier--Tate transport, which reverses grade;
• source completion or self-duality, which may compare both presentations inside a common completed object.

Only their higher coherence could force a zero event to be fixed rather than merely transported to a complementary grade.

This recovers the multi-tower intuition. The in and out towers are connected by a legitimate grade-changing morphism. RH would require a control tower whose coherence cell says that the particular null event descends to their grade-zero equalizer.

Explanatory gain

The critical line is still the fixed-grade locus of reciprocal homogeneity. But being a fixed locus does not imply that every transported event is fixed.

The missing theorem is no longer “Fourier sewing is reversible.” That is true everywhere. It is “source-authorized null events must be fixed objects of the grade-reversing sewing.” That statement is both sharper and currently unproved.

View source on GitHub