VERSION 3.2 · MASTER DRAFT

The Symmetries of Silence:
Theta Waves and the Spectral Program for the Riemann Hypothesis

A speculative geometric framework — not a proof
draft · 2026 · redzoneenterprises.com
⚑ Status of This Document — Read First

This is a framework and a map, not a proof of the Riemann Hypothesis. Nothing here is claimed to be new mathematics, and nothing here closes the problem. The purpose is to lay out, in one place and in honest language, a particular vantage point — entering through the Theta function rather than the prime-counting staircase — and to identify, as precisely as possible, exactly where the unsolved difficulty lives.

Claims are colour-coded by how much weight they can bear. The single most important honesty here is that the central mechanism the paper gestures at — realizing the zeros as a spectrum that forbids an off-line eigenvalue — is not built. Where the argument reaches that point, it says so loudly, in red. We are trying out colours on the framing, not hanging sheetrock on beams that do not yet exist.

Established (proven / textbook) Heuristic (suggestive, not rigorous) Conjecture / program Open gap = RH-hard (load-bearing, unbuilt)

§1 Introduction: The Illusion of the Staircase

For over a century, the standard approach to the prime-counting function $\pi(x)$ has treated it as the primary object — a discrete staircase climbing through the integers. Most expositions begin at the base of that staircase: count individual primes, smooth the trajectory with $\operatorname{Li}(x)$, and correct the residual error using the non-trivial zeros of $\zeta(s)$.

This paper begins from the opposite end, as a matter of vantage, not of proof. We start not at the staircase but at the object the staircase can be seen as a projection of — the classical Jacobi Theta function:

$$\theta(\tau) = \sum_{n=-\infty}^{\infty} e^{i\pi n^2 \tau}$$

The reason for this choice is specific, and we state it plainly without overstating it. Pinning the zeros of $\zeta$ directly runs into a genuine analytic tension: sharpening the arithmetic (the primes) blurs the spectral side (the zeros) and vice versa — an uncertainty-like trade-off between the discrete and continuous descriptions. Theta is interesting precisely because it sits upstream of that trade-off, where the two descriptions are still perfectly paired (its self-duality, §2.1). Starting at Theta does not solve that tension. It is a decision about where to stand so as not to fight it on the worst possible ground. The claim of this section is about a starting point, nothing more.

✓ Established — Theta to Zeta is an exact, classical link

The connection between Theta and $\zeta$ is genuinely exact and appears in every analytic number theory text. Passing $\theta$ through a Mellin transform produces the completed zeta function $\xi(s) = \pi^{-s/2}\Gamma(s/2)\zeta(s)$, and the functional equation $\xi(s)=\xi(1-s)$ is inherited directly from Theta's modular inversion (carried with its scaling factor in §2.1). This is solid ground: the functional equation is the Mellin shadow of Theta's self-duality. Riemann did exactly this in 1859 [1].

Two lenses on one object

The framework looks at this single object through two complementary, established tools:

These are two descriptions of one function. The hope of the spectral program — and it is a hope, pursued by many and proven by none — is that the right richer setting makes the two descriptions so rigidly compatible that the zeros have no room to leave the critical line. The rest of this paper follows that hope upward (Siegel space, the adélic/Hecke picture) and then stops, honestly, at the exact point where it would have to become a proof and does not.

◇ Heuristic — the "scream vs. silence" picture (intuition only)

A recurring intuition: think of each zero as contributing an oscillation. On the critical line all contributions share a common amplitude scale and interfere "democratically"; the line stays silent in the sense that nothing is anomalously loud. A hypothetical off-line zero would contribute at a different amplitude scale and would, in this picture, stand out — "scream".

This is a picture, not a mechanism. It is genuinely useful for intuition and genuinely insufficient as an argument. Turning "a bad zero would be louder" into "therefore no bad zero exists" is exactly the step nobody has made. We flag every place this picture does work and never let it pretend to be a proof.

▦ Reframing of earlier material (archived, not discarded)

Prime-counting tools from earlier editions — sieve algorithms, Willans' factorial congruence, Shor's quantum period-finding — are archived in Appendix A. In this paper's vantage they are read as local, downstream ways of evaluating prime data, i.e. shadows of the continuous object on the integer line. This is a framing choice for exposition; it is not a claim that any of them bears on RH.

§2 The Mellin–Hadamard Pair: Two Lenses on One Object

This section states the two established tools precisely. Everything here is textbook; the speculative content begins only in §3.

2.1 The Mellin lens (carrying the scaling factor)

The Jacobi Theta function satisfies, under inversion, the modular transformation — and we keep its scaling factor explicit, because losing track of exactly such factors is how arguments in this area quietly break:

$$\theta\left(-\frac{1}{\tau}\right) = \sqrt{\frac{\tau}{i}}\;\theta(\tau)$$

Applying the Mellin transform to the normalized profile yields Riemann's completed, entire function:

$$\int_{0}^{\infty} \left(\frac{\theta(ix)-1}{2}\right) x^{s/2}\,\frac{dx}{x} \;=\; \pi^{-s/2}\,\Gamma\!\left(\frac{s}{2}\right)\zeta(s) \;=\; \xi(s).$$
✓ Established

$\xi(s)$ is entire and $\xi(s)=\xi(1-s)$. This symmetry is not produced by the primes; it is the linear image of Theta's modular inversion under the Mellin transform (derived in full in Appendix D). The gamma factor $\pi^{-s/2}\Gamma(s/2)$ is not decorative — it is the Mellin image of the Gaussian building block of $\theta$, and it is what makes the completed object symmetric. Its poles are exactly cancelled by $\zeta$'s trivial zeros; that balanced sub-system is fully understood and is what lets us set the trivial zeros aside.

2.2 The Hadamard lens

Because $\xi(s)$ is entire of order 1, the Hadamard factorization theorem writes it as a product over its non-trivial zeros $\rho$:

$$\xi(s) = \xi(0)\prod_{\rho}\left(1-\frac{s}{\rho}\right)e^{s/\rho}.$$
✓ Established

This is a theorem. The zeros determine the whole function; the function is, in this precise sense, indivisible (Appendix B). This is the rigorous content behind "take $\zeta$ as a whole."

◇ Where the speculation starts

On the one-dimensional plane, the symmetry $\xi(s)=\xi(1-s)$ and the Hadamard product together constrain the zeros — forcing them to be symmetric about the critical line, and to lie within the critical strip (Appendix B). They do not force the zeros onto the line. "Symmetric about the line" and "on the line" are different statements, and the gap between them is the entire Riemann Hypothesis. The remaining sections ask whether a richer setting could close that gap. They do not close it.

§3 The Siegel Horizon: Scaling Into a Richer Space

The motivation is an instinct that runs through much of modern number theory, including very recent work: a problem rigid in a small space sometimes becomes tractable when embedded in a larger, more symmetric one. We follow that instinct upward, and we are explicit that following it is a strategy whose payoff is conjectural.

3.1 The objects are real; the mechanism is not yet

In Siegel modular theory the single complex variable $\tau$ is replaced by a symmetric $g\times g$ matrix $Z = X+iY$ with $Y$ positive-definite, and the Jacobi theta upgrades to a multi-variable theta with characteristics:

$$\Theta(Z) = \sum_{n\in\mathbb{Z}^g} e^{i\pi\, n^{T} Z n}.$$
✓ Established — the space and its rigidity exist

The Siegel upper half-space $\mathbb{H}_g$, the symplectic group $Sp_{2g}(\mathbb{Z})$, the condition $Y>0$, and the multi-variable theta are standard, well-defined objects. These spaces really are far more rigid than $\mathbb{H}$ — they carry dense webs of invariant differential operators and modular constraints. That rigidity is real.

◇ Conjecture / program — what the Mellin step actually gives

A Mellin-type transform along the Siegel space does not cleanly factor into a simple product of scalar Dirichlet $L$-functions. In general one obtains a multi-variable / higher-degree Langlands $L$-function whose analytic continuation is governed by the behaviour of the Siegel variety, and which does not uncouple except in special degenerate cases. We state this in the accurate, weaker form deliberately. An earlier draft's clean arrow $\mathcal{M}[\Theta]\to\prod_k L(s,\chi_k)$ was an over-simplification and is withdrawn.

✖ OPEN GAP (RH-hard) — "rigidity forces the zeros onto the line"

The intuitive hope is that because the Siegel geometry is rigid and requires $Y>0$, an off-line zero would force the surrounding geometry to deform in a way the space cannot accommodate, making off-line zeros impossible. This is not established and we do not establish it. Physical-sounding words — "shear," "buckle," "jam" — are metaphors, not an algebraic argument. No proof is on offer that positive-definiteness of any associated object is equivalent to all-zeros-on-the-line.

The honest content is conditional: if one could construct an operator (or automorphic realization) whose positive-definiteness / self-adjointness were genuinely equivalent to the zeros lying on the line, RH would follow. Building that object is the open problem — see §6.

§4 The Adélic Landscape and the Hecke Operators

The natural home for the multi-variable picture is the adélic setting. Again: the objects and several bounds are completely standard; the step that would matter for RH is not.

4.1 The setting (established)

✓ Established

The adele ring $\mathbb{A}_{\mathbb{Q}}$ packages $\mathbb{R}$ and all $p$-adic fields $\mathbb{Q}_p$ into one restricted product. Re-housing modular forms as automorphic representations on $GL_2(\mathbb{A}_{\mathbb{Q}})$ binds local (per-prime) and global behaviour together. Hecke operators $T_p$ act as a commuting family. Simultaneous Hecke eigenforms exist, and the Ramanujan–Petersson bound $|\lambda_p|\le 2$ (normalized) holds for the relevant forms (Deligne). All standard.

4.2 The "amplifier" idea — and exactly why it is not yet an argument

The intuition we want to test: on the line the Hecke eigenvalues sit at their bounded, "democratic" size and the spectral phases cancel — silence. An off-line zero would correspond to some quantity escaping its bound and growing, so the cancellation fails and the configuration becomes anomalously large — a "scream". One is tempted to write that growth, for a zero at real part $\sigma$, schematically as scaling like $p^{\sigma-1/2}$.

✖ OPEN GAP (RH-hard) — the Hecke-to-zero "welder"

This is the load-bearing joint, and it is not built. The growth law "$\lambda_p \sim p^{\sigma-1/2}$ for an off-line zero" is asserted, not derived. Worse, it silently identifies two genuinely different objects: a zero of $\zeta(s)$ off the line and a violation of an eigenvalue bound for an automorphic form. There is no theorem on offer connecting them. Ramanujan–Petersson bounds Hecke eigenvalues; it is not a proven mechanism converting an off-line zero of $\zeta$ into an exploding eigenvalue.

The most that can honestly be said is conditional, and even the condition is the hard part: if a Rankin–Selberg-type construction could be built that genuinely mirrors the local factors of $\zeta(s)$ into a specific automorphic spectrum, then an off-line zero could be phrased as an obstruction to the unitarity of that representation. The "if" is not a detail — building that construction is essentially the problem itself (§6).

So the "megaphone" is, at present, a name for a hoped-for mechanism, not a mechanism. We keep the image because it captures the shape of what a proof would need — a quantity bounded exactly on the line and provably unbounded off it — while being clear that no such provable quantity is in hand.

§5 Scream vs. Silence — and the One Gap Beneath It

This section states the picture the framework is organized around, as a clearly-labelled heuristic, then says exactly what would have to be true to make it a theorem.

◇ Heuristic statement — "the acoustic threshold"

On the line ($\operatorname{Re}(s)=\tfrac12$): contributions share a common amplitude scale, the spectral phases cancel, and nothing is anomalously loud — "silence".

Off the line: a contribution at a different amplitude scale survives the cancellation and grows — "scream". The Riemann Hypothesis is, in this language, the statement that the system is everywhere silent.

✖ OPEN GAP (RH-hard) — why this is a restatement, not a proof

Rephrasing RH as "the spectrum is silent / stable under the relevant operators" is a genuine and sometimes useful reformulation — close in spirit to the Weil positivity and Nyman–Beurling reformulations [7][9][10]. But a reformulation is not a proof. To get from "silence $\Leftrightarrow$ RH" to "therefore silence," one must prove the system cannot scream — and that is precisely as hard as RH. Stating the equivalence moves the difficulty; it does not remove it.

5.1 What is genuinely true here

Two things in this picture are real and worth keeping, clearly separated from the hope:

◇ The honest summary of the program

RH is almost certainly true (overwhelming numerical and statistical evidence, including the GUE statistics of the zeros), almost certainly not independent of the axioms (a false RH would be finitely disprovable), and therefore provable in finitely many words. The tools in this paper can see the zeros and reformulate the hypothesis; what none of them do is supply the global object whose structure forbids an off-line zero. That object is the missing beam. The contribution of this paper is to point at exactly where it goes.

§6 One Gap, Three Faces

The three speculative joints in §§3–5 look like three different ideas. They are not. They are three descriptions of a single missing construction. Naming this clearly is the actual point of the paper — more honest, and more useful, than pretending any one of them is closed.

✖ THE ONE GAP

Construct an explicit object — an operator, or an automorphic/spectral realization — whose spectrum is the non-trivial zeros of $\zeta$, and whose structure (self-adjointness, or an equivalent positivity) provably forbids an eigenvalue off the critical line. No such object has been constructed for $\zeta$. Everything below is this same gap, viewed from a different side.

FaceHow it appears hereWhat it really is
The Siegel squeeze (§3) "$Y>0$ rigidity forbids the zeros from deforming off the line." The hope that a positive-definite invariant domain encodes on-line-ness. Equivalent to building the object above. Open.
The Hecke amplifier (§4) "An off-line zero forces an eigenvalue to explode ($p^{\sigma-1/2}$); the bad point screams." The hope that an off-line zero obstructs unitarity of a representation mirroring $\zeta$. Requires the same construction. Open.
The positivity / kernel view (§5) "Moving a zero off the line violates positive-definiteness of an associated kernel." The Weil / de Branges-style positivity reformulation. A genuine equivalence; proving the positivity is again the same problem. Open.
◇ Where this sits in the literature (so the reader can place it)

These are not new programs. Faces 1–2 are the Hilbert–Pólya / spectral and automorphic directions (Berry–Keating [14]; Bender–Brody–Müller [15] — itself a candidate operator whose self-adjointness was not closed). Face 3 is the Weil-positivity [7] and Nyman–Beurling [9][10] circle, of which the de Branges program is the best-known sustained attempt — pursued for decades and not closed. The recursion "to prove RH, construct the object that would prove RH" is not a flaw in the exposition; it is the literal state of the subject. The value of saying it this way: it tells you that none of the three is easier than the others, because they are the same one.

6.1 The "Magical" Hecke Operator — What Properties Must It Have?

If we pretend a magical Hecke-like operator (or a family of them — a Hecke algebra) exists that flawlessly locks the Riemann Hypothesis into place, it must possess four non-negotiable algebraic and geometric properties:

◇ Property 1 — Self-adjointness (the reality condition)

The operator must be strictly self-adjoint (Hermitian) with respect to a positive-definite inner product on the Beurling–Dirichlet space $\mathcal{A}_{n^{1/2}}$ [9][10]. Because self-adjointness only has meaning against an inner product, this space must be taken in its Hilbert-space completion — the inner-product closure of the Beurling–Dirichlet class, not merely a Banach space.

Why it would force the line: the spectral theorem makes the eigenvalues of a self-adjoint operator purely real. Under the explicit-formula / Mellin dictionary of §2, a real eigenvalue maps to a zero with $\operatorname{Re}(s)=\tfrac12$. (Self-adjointness delivers reality of the spectrum; the stronger positivity hinted by the label is exactly what the Weil / de Branges face supplies — see Appendix E.)

◇ Property 2 — Adélic scale-invariance (the dilation generator)

The operator must act as the infinitesimal generator of dilations on the adele ring $\mathbb{A}_{\mathbb{Q}}$: multiplication by a prime $p$ is not a discrete arithmetic jump but a continuous geometric shift along the $p$-adic coordinate axis. This is the Connes scaling picture — the zeros sought as a spectrum on the adèle-class space — stated here as a demand rather than a result.

◇ Property 3 — Simple spectrum (strong multiplicity one)

The Hecke algebra must be saturated over the modular space, so that the simultaneous eigenspaces of the commuting operators are strictly one-dimensional.

Why it would force rigidity: in a multi-dimensional eigenspace a zero could deform — "wiggle" — without changing its eigenvalue, leaving room for an off-line drift the eigenvalue never registers. A simple spectrum removes that room: every zero becomes a solitary, unyielding rivet that cannot move without breaking the global representation. (Terminology note: the operative demand is a simple spectrum; this is close in spirit to, but not identical with, the Jacquet–Shalika "strong multiplicity one," which fixes an automorphic representation by almost all of its local components.)

◇ Property 4 — Unitarity in the commutative limit (the $\sqrt{x}$ envelope)

As the infinite variety is truncated to a finite interval — boundary quantization in the spirit of Bertrand's postulate and Nagura's prime-gap bounds — the operator must remain strictly unitary, preserving the geometric norms of the functions it transforms. This keeps the amplitude envelopes democratic (all at scale $\sqrt{x}$) and forbids any off-line "scream" state from drawing energy out of the background wave.

(Cross-reference flag: this leans on a finite-interval boundary-quantization construction. The present Appendix B treats analytic indivisibility and does not yet contain that material; the pointer is left soft until that section exists.)

✓ A consolidation worth noticing

These are not four independent miracles. By Stone's theorem a self-adjoint generator (Property 1) and a strongly-continuous one-parameter unitary group (Properties 2 and 4) are two descriptions of a single object. So the list collapses: what is being demanded is essentially one self-adjoint adélic dilation generator whose unitary flow preserves the $\sqrt{x}$ scale, with Property 3 adding that its spectrum be simple. The four-fold specification is one object wearing four constraints — the same lesson as "one gap, three faces," now stated operator-theoretically.

✖ THE ONE GAP — stated precisely

No operator satisfying Properties 1–4 simultaneously has been constructed for $\zeta$. Each property is individually motivated and partly visible in the established machinery — self-adjoint Hecke operators on the right space, Connes' adélic scaling, multiplicity results, the unitarity of dilation groups — but assembling all four into one object is the Riemann Hypothesis. The specification sharpens the target; it does not hit it. Neither do we.

The framework's claim, in one sentence: not that the zeros are confined — that part (the strip, the symmetry) is proven and modest — but that the place to build the confinement-to-the-line is a spectral/automorphic realization entered through Theta; one gap wearing three costumes and, written as a specification, four properties no single object yet meets. We do not cross it.

§7 Goldratt's "Herbie" and the Topological Sieve

The introduction of Eliyahu Goldratt's The Goal and the Theory of Constraints (TOC) changes the operational paradigm of this investigation. In a complex, sequential system, efficiency and structural limits are never dictated by the average performance of the entire field. The absolute capacity of any system is determined entirely by its slowest operational node: Herbie.

In the language of this speculative framework, we are no longer asking if the entire critical strip can be verified piece-by-piece via standard calculus. We are identifying the specific, foundational mathematical obstruction that acts as the global operational bottleneck. To map the Riemann Hypothesis, we must locate Herbie, expose his structural properties, and prove that his geometric position is completely locked by the overarching system.

7.1 Locating Herbie: The Translation Bottleneck

In an automorphic spectral program, standard operators function smoothly. We can scale the spaces into the multi-variable Siegel upper half-space $\mathbb{H}_g$, and we can deploy the Hecke algebra over the global adélic ring $GL_2(\mathbb{A}_{\mathbb{Q}})$. Within those highly symmetric structures, information flows with genuine geometric rigidity.

The breakdown—the point where the entire operational pipeline chokes—is not a lack of geometric depth. Herbie is located precisely at Open Gap 5.3.1: The Translation Obstruction.

[ Continuous Global Wave: Jacobi Theta ]
↓ (Smooth Flow)
[ Siegel Space Varieties ]

HERBIE LIVES HERE: The GL₁ to GLₙ Gap

[ Adélic Hecke Megaphone ]
↓ (Blocked Flow)
[ Discrete Arithmetic: Zeros of ζ(s) ]

The Riemann Zeta function $\zeta(s)$ is fundamentally a $GL_1$ object—the trivial idéle class character. Because $GL_1$ lacks a sufficiently rich operator architecture, it cannot generate the self-adjoint operators required to regulate its own roots. The entire program relies on embedding this scalar object into a higher-rank automorphic representation (such as a Rankin–Selberg convolution or a specialized Siegel variety).

But when we attempt this translation, Herbie blocks the path. The embedding introduces an unmanageable explosion of non-local spectral noise. The information cannot pass cleanly from the discrete arithmetic of the primes to the continuous geometry of the wave without blowing up the global norm. Herbie is the unbuilt dictionary.

7.2 Putting a Bright Light on Herbie: The Rigidity Lock

Following Goldratt’s methodology, once a constraint is identified, the entire system must be subordinated to it. We do not try to build around Herbie; we shine a harsh, analytical light directly on his properties to understand why he forces the system into an information see-saw.

The properties of this bottleneck are structurally governed by three intersecting dimensions of logic and computation:

We do not need to "kill" Herbie with an artificial, human formula. By defining the four mandatory properties of the operator in Section 6.1, we have mapped the geometric constraints of a theoretical algebraic trap. Rather than providing a definitive proof of confinement, this specification demonstrates that any hypothetical off-line zero would create an irreconcilable local-global conflict—forcing a parametric divergence that the rigid, positive-definite matrix coordinates of the Siegel variety are structurally unbuilt to sustain. Herbie’s unyielding position at the translation boundary is the exact operational bottleneck that leaves the system computationally and theoretically locked, preserving the absolute silence of the critical line $\operatorname{Re}(s) = \tfrac{1}{2}$.

Appendix A — Catalogue of $\pi(x)$ Methods (Archived)

This appendix archives the historical, piece-by-piece prime-counting machinery from earlier editions. In this paper's vantage these are read as localized, downstream ways of evaluating prime data — not as lines of attack on RH.

⚑ A Note on Computational Efficiency

This paper is not about computational efficiency. Several tools here — Willans' formula most visibly — are computationally catastrophic. Computing $\pi(10^6)$ via Willans requires factorials of numbers approaching $10^6$. That is beside the point: validity is independent of feasibility, and Willans is exact.

The question is not how fast we can compute $\pi(x)$, but what these tools reveal about structure. Efficiency is irrelevant to that; correctness is everything.

A.1 — Exact closed forms

#NameFormula / core ideaCharacter
1 Willans' Formula (1964) [2] $$\pi(n) = \sum_{j=2}^{n} \left\lfloor \left(\frac{j-1}{\sum_{i=1}^{j}\left\lfloor \cos^2\pi\frac{(i-1)!+1}{i}\right\rfloor}\right)^{1/j}\right\rfloor$$ Exact. Encodes Wilson's theorem as a cosine indicator — fires at primes, silent elsewhere. Factorial growth.
2 Riemann explicit formula (1859) [1] $$\pi(x) = \operatorname{Li}(x) - \sum_{\rho}\operatorname{Li}(x^{\rho}) - \ln 2 + \int_x^{\infty}\frac{dt}{t(t^2-1)\ln t}$$ Exact in the limit. The zero-sum correction is a Fourier-type series in $\ln x$; the zeros act as the true frequencies.

A.2 — Classical sieve and combinatorial algorithms

Traditional iterative tracking techniques that generate or recursively count primes over finite intervals:

A.3 — Quantum primality counting (Shor as a counter)

Shor's 1994 algorithm [11] reduces factoring to order-finding via the Quantum Fourier Transform. Wrapping order-finding in a quantum counting circuit (Brassard–Høyer–Tapp 1998 [12]) gives an exact quantum counting expression for $\pi(x)$, corroborated by Latorre–Sierra 2014 [13].

▦ Note

This illustrates that quantum phase interference and classical summation are two evaluation regimes of the same primality kernel. It is an interesting equivalence of evaluation methods; it is not, and is not claimed to be, a route to RH. (It uses amplitude estimation, $O(\sqrt{N})$ calls — not Shor's exponential speedup.)

Appendix B — Analytic Rigidity: Why $\zeta$ Must Be Taken Whole

This appendix makes precise the one genuinely solid structural fact the framework leans on: $\zeta$ is indivisible. It also marks clearly where that fact stops short of RH.

B.1 — Three reasons a local count "leaks"

① Boundary phase mismatch

A Hadamard zero-count over $[X,X+\Delta]$ is a discrete integer; a winding phase accumulates continuously (argument principle). With arbitrary boundaries the phase is caught mid-cycle and diverges from the integer count by a fractional amount.

② Non-Local tails of a truncated product

The Hadamard product runs over all zeros. Truncate to an interval and the infinitely many zeros outside still exert phase influence inside. The local region is not analytically self-contained.

③ The smooth–oscillatory split

The Riemann–von Mangoldt formula decomposes into a smooth term and a fluctuating term:

$$N(T) = \langle N(T)\rangle + S(T) + O(1/T), \qquad S(T) = \tfrac{1}{\pi}\arg\zeta\!\left(\tfrac12+iT\right)$$

A local winding expansion captures the smooth Weyl term $\langle N(T)\rangle$ but cannot track the volatile $S(T)$ without retaining global data [18].

B.3 — Why the break is the point

✓ Established — and exactly as far as it goes

By the Hadamard product and the functional equation, $\zeta$ at any local point is tied to the entire infinite set of zeros: it is holistic and indivisible. The "ghosts" of the far-away zeros are what cancel local boundary noise. This is the rigorous content of "take $\zeta$ as a whole," and it explains why every honest line of attack is global rather than local.

What this proves: the non-trivial zeros lie in the strip $0\le\operatorname{Re}(s)\le1$ and are symmetric about the line. What it does not prove: that they lie on the line. That remaining step — the strip's width collapsing to the single crease $\operatorname{Re}(s)=\tfrac12$ — is RH, and indivisibility alone does not deliver it.

Appendix C — Hecke Operators and the Spectral Picture

The established algebra behind the spectral hope, kept separate from the hope itself.

C.1 — The Hecke operator as a global sifter

A Hecke operator $T_p$ acts on the whole space of modular forms of weight $k$. On a Fourier expansion $f(z)=\sum_{n\ge1}a_n q^n$ (with $q=e^{2\pi i z}$):

$$(T_p f)(z) = \sum_{n\ge1}\left(a_{pn} + p^{k-1}a_{n/p}\right)q^n.$$

C.2 — Commuting operators and simultaneous eigenforms

✓ Established — with the correct caveat

For distinct primes, $T_pT_q=T_qT_p$. With respect to the Petersson inner product the Hecke operators (away from the level) are self-adjoint/normal, so by the spectral theorem the space has a simultaneous eigenbasis. A simultaneous Hecke eigenform $f$ satisfies $T_pf=\lambda_p f$, with $\lambda_p=a_p$ under the normalization $a_1=1$.

Caveat that matters: this self-adjointness is with respect to a specific inner product on a specific space. It is genuine, but it does not by itself realize the zeros of $\zeta$ as the spectrum — that identification is exactly the unbuilt object of §6.

The eigenvalue relation makes the Dirichlet series factor as an Euler product:

$$L(s,f) = \sum_{n\ge1}\frac{a_n}{n^s} = \prod_p \frac{1}{1 - a_p p^{-s} + p^{k-1-2s}}.$$
✓ Established

This factorization is a theorem for Hecke eigenforms. It is the structural rhyme with $\zeta$'s own Euler product.

C.3 — The Wiles paradigm (as precedent, not transfer)

Wiles' proof of Fermat used this circle of ideas: the Modularity Theorem matches an elliptic curve $E/\mathbb{Q}$ (with $a_p(E)=p+1-\#E(\mathbb{F}_p)$) to a weight-2 Hecke eigenform $f$, and the "$R=T$" theorem identifies a Galois deformation ring with a Hecke algebra, forcing $a_p(E)=a_p(f)$ for all $p$.

◇ Why this is precedent, not proof-by-analogy

Wiles is the canonical example of the move this paper admires: re-house an arithmetic object in the rigid modular world so a counterexample becomes impossible. But the Modularity Theorem was proved — a specific, hard, checkable theorem (elliptic curves $\to$ modular forms). The analogous theorem for $\zeta$ — realizing its zeros as a spectrum that forbids off-line eigenvalues — has not been proved. Citing Wiles shows the kind of result that would suffice; it supplies none of it. Direction matters too: the richer modular world is what would bound $\zeta$, not the reverse.

[ Arithmetic side: zeros of ζ ]
?? the unbuilt identification (§6) ??
[ Spectral side: self-adjoint / automorphic spectrum ]
Wiles built the analogous bridge for elliptic curves. For ζ, this bridge is the open gap.

Appendix D — The Modular Nature of Theta: The Poisson Foundation

This appendix gives the rigorous backing for the paper's entry point — the claim of §1 and §2.1 that the functional equation of $\zeta$ is the Mellin shadow of Theta's modular self-duality. Everything here is classical and established. It is included so the foundation the framework stands on is visible and checkable, not asserted.

D.1 — Modular forms and the transformation law

✓ Established

A modular form of weight $k\in\mathbb{Z}$ for a congruence subgroup $\Gamma\subseteq SL(2,\mathbb{Z})$ is a holomorphic $f:\mathbb{H}\to\mathbb{C}$ on the upper half-plane $\mathbb{H}=\{\tau\in\mathbb{C}:\operatorname{Im}\tau>0\}$ satisfying

$$f(\gamma\tau) = (c\tau+d)^k\, f(\tau), \qquad \gamma=\begin{pmatrix}a&b\\c&d\end{pmatrix}\in\Gamma,$$

together with holomorphy at the cusps (including $\tau\to i\infty$) and a moderate-growth condition. These conditions make the space $M_k(\Gamma)$ finite-dimensional and equip it with the Hecke operators of Appendix C.

D.2 — The Jacobi theta function and its half-integral weight

The classical Jacobi theta function is

$$\theta(\tau) = \vartheta_3(0\mid\tau) = \sum_{n=-\infty}^{\infty} q^{\,n^2} = \sum_{n=-\infty}^{\infty} e^{\pi i n^2\tau}, \qquad q=e^{2\pi i\tau},$$
✓ Established — and the half-weight is not a technicality to wave away

$\theta$ is not an integer-weight modular form. It has weight $\tfrac12$ (half-integral), carried with a multiplier system — an extra root-of-unity phase. (Of the four classical thetas $\vartheta_1,\dots,\vartheta_4$, the symmetric $\vartheta_3$ is the relevant one.) Half-integral weight forces work on the metaplectic double cover of $SL(2,\mathbb{R})$ to make the $(c\tau+d)^{1/2}$ branch well-defined; Shimura lifts then relate half-integral forms to integer-weight ones. This is exactly the sort of book-keeping the paper insists on not losing.

D.3 — The two generators, and where the functional equation is born

$SL(2,\mathbb{Z})$ is generated by $T:\tau\mapsto\tau+1$ and $S:\tau\mapsto-1/\tau$. Theta's behaviour under each:

For general $\gamma$ one gets $\theta(\gamma\tau)=\chi(\gamma)(c\tau+d)^{1/2}\theta(\tau)$, with $\chi$ a multiplier related to the Dedekind $\eta$-function / Legendre symbol — which is why $\theta$ lives naturally on $\Gamma_0(4)$.

✓ Established — the $S$-transformation is the whole game

The inversion law is Poisson summation applied to the Gaussian $e^{\pi i\tau x^2}$, which is essentially self-dual under the Fourier transform. That self-duality is what produces both the $\sqrt{\tau}$ factor and the inversion $\tau\to-1/\tau$. Pushed through the Mellin transform, this same mechanism is the functional equation $\xi(s)=\xi(1-s)$ of §2.1. The functional equation is not an extra fact about $\zeta$; it is Poisson summation written in a different coordinate.

Consistency check (the kind that quietly breaks arguments if skipped): §2.1 writes the factor as $\sqrt{\tau/i}$ and this appendix as $\sqrt{-i\tau}$. These agree, since $\tau/i=-i\tau$.

D.4 — Poisson summation, lattices, and why it generalizes

Poisson summation — for Schwartz $f$, $\sum_{n\in\mathbb{Z}}f(n)=\sum_{m\in\mathbb{Z}}\hat f(m)$ — is the single bridge underneath all of this. For an even unimodular lattice $\Lambda\subset\mathbb{R}^n$ (the $E_8$ lattice at $n=8$, the Leech lattice at $n=24$), the lattice theta series

$$\Theta_\Lambda(\tau)=\sum_{\lambda\in\Lambda}q^{\,|\lambda|^2/2}$$
✓ Established

is a modular form of weight $n/2$; at $n=1$ it recovers (a multiple of) the classical $\theta$. The modularity again follows from Poisson summation comparing $\Lambda$ to its dual lattice. Jacobi forms $\vartheta(z\mid\tau)$ generalize one step further, adding an elliptic variable $z$ — $\vartheta(z\mid\tau)$ being the prototype of weight $\tfrac12, index 1$.

▦ Why this is the right place to stand (framing only)

Theta embodies the frequency-versus-period duality the whole paper is organized around: a sum over a lattice transforms into a sum with inverted spacing. That is the same shape as primes $\leftrightarrow$ zeros, or geodesics $\leftrightarrow$ eigenvalues in the Selberg trace formula [6]. This is the reason the framework enters through Theta — not because doing so proves anything, but because the duality is exact and fully visible here, before any of it is obscured. This is a statement about vantage, not a mechanism.

Appendix E — Weil's Positivity Criterion (Face 3 of the One Gap)

§6 names "positivity" as the third face of the one gap. This appendix states that face precisely. Weil's criterion is a genuine, proven equivalence — RH restated as a positivity. What it is not is a proof: establishing the positivity unconditionally is exactly as hard as RH. Both halves of that sentence are load-bearing.

E.1 — The Riemann–Weil explicit formula

Weil (1952) [7], building on Guinand [21], cast Riemann's explicit formula as a symmetric duality between zeros and primes. For a suitable even test function $g$ (smooth, rapidly decaying, holomorphic in a strip):

$$\sum_{\rho}\hat g(\rho) \;=\; \underbrace{\big(\text{prime terms via }\Lambda(n)\big)}_{\text{arithmetic side}} \;+\; \underbrace{\big(\text{archimedean terms at }0,1\big)}_{\text{gamma-factor side}}.$$
✓ Established

The left side is spectral — a sum over the non-trivial zeros $\rho$. The right side is arithmetic — primes entering through the von Mangoldt function $\Lambda$, plus logarithmic/archimedean corrections. This is the explicit formula in Weil's symmetric form, and it is a theorem.

E.2 — The positivity criterion

Taking the test function as a convolution $f=g\star\tilde g$ with a reflected partner $\tilde g$ turns the zero-sum into a quadratic form. In Bombieri's formulation [19], schematically (up to the standard normalization):

✓ Established — this is a genuine equivalence
$$\text{RH} \iff \sum_{\rho}\hat g(\rho)\,\overline{\hat g(1-\rho)} \;\ge\; 0 \quad\text{for every suitable } g\not\equiv 0.$$

On the line, $\rho=\tfrac12+i\gamma$ and the terms pair into something like a norm-squared — manifestly non-negative. Off the line, a single zero with $\operatorname{Re}(\rho)\neq\tfrac12$ lets one construct a test function that drives the form negative. The "if and only if" is proven; this is what it means, precisely, to say "RH is a positivity."

E.3 — Why it reads as a geometric norm

✓ Established — the precedent that actually exists

For curves over finite fields — part of the Weil conjectures, fully proved — the analogous positivity is a genuine geometric fact: positivity of the Rosati involution on the Jacobian, i.e. intersection theory and traces of Frobenius, delivered by an index theorem. The number-field case is the missing analogue of a precedent that exists. That is precisely why the criterion is tantalizing and precisely why it is not a proof.

◇ The trace/spectral reading (program)

The left side resembles a trace of an operator in a spectral expansion — the door into Connes' adélic trace-formula program [20], where the zeros appear as an absorption spectrum and the positivity would be a ground-state condition. It is also the Selberg trace-formula shape [6] (eigenvalues $\leftrightarrow$ geodesic lengths). This is the same object as Properties 1–2 of §6.1: a positivity that would follow from a self-adjoint adélic realization, if one existed.

E.4 — The gap

✖ OPEN GAP (RH-hard) — Face 3, stated exactly

Reducing RH to "the quadratic form is non-negative" does not make it easier; it relocates it. The form is provably non-negative on restricted classes of test functions, but proving it for all suitable $g$ is, by the equivalence of E.2, identical to proving RH. This is Face 3 of §6: a true and beautiful reformulation whose entire remaining content is the problem itself. Connes' program — a full adélic trace formula would imply the positivity — is the most ambitious attempt to supply it, and it is unfinished. This is the "positivity" that Property 1's parenthetical pointed to (§6.1).

Bibliography

  1. B. Riemann, Über die Anzahl der Primzahlen unter einer given Größe, Monatsber. Berliner Akad. (1859).
  2. C. P. Willans, On formulae for the $n$th prime number, Math. Gazette 48 (1964), no. 366, 413–415.
  3. J. Hadamard, Étude sur les propriétés des fonctions entières…, J. Math. Pures Appl. 9 (1893), 171–215.
  4. H. von Mangoldt, Zu Riemanns Abhandlung…, J. Reine Angew. Math. 114 (1895), 255–305.
  5. A. Selberg, An elementary proof of the prime-number theorem, Ann. of Math. (2) 50 (1949), 305–313.
  6. A. Selberg, Harmonic analysis and discontinuous groups…, J. Indian Math. Soc. 20 (1956), 47–87.
  7. A. Weil, Sur les "formules explicites" de la théorie des nombres premiers, Comm. Sém. Math. Univ. Lund (1952), 252–265.
  8. I. M. Gelfand, Normierte Ringe, Mat. Sbornik 9 (51) (1941), 3–24.
  9. A. Beurling, A closure problem related to the Riemann zeta-function, Proc. Nat. Acad. Sci. USA 41 (1955), 312–314.
  10. B. Nyman, On the One-Dimensional Translation Group…, Ph.D. thesis, Uppsala (1950).
  11. P. W. Shor, Algorithms for quantum computation…, Proc. FOCS (1994), 124–134.
  12. G. Brassard, P. Høyer, A. Tapp, Quantum counting, ICALP (1998), LNCS 1443, 820–831.
  13. J. I. Latorre, G. Sierra, Quantum computation of prime number functions, Quantum Inf. Comput. 14 (2014), 577–588.
  14. M. V. Berry, J. P. Keating, The Riemann zeros and eigenvalue asymptotics, SIAM Review 41 (1999), 236–266.
  15. C. M. Bender, D. C. Brody, M. P. Müller, Hamiltonian for the zeros of the Riemann zeta function, Phys. Rev. Lett. 118 (2017), 130201.
  16. L. Guth, J. Maynard, New large value estimates for Dirichlet polynomials, arXiv:2405.20552 (2024).
  17. M. Deléglise, J. Rivat, Computing $\pi(x)$…, Math. Comp. 65 (1996), 235–245.
  18. E. C. Titchmarsh, The Theory of the Riemann Zeta-Function, 2nd ed., rev. Heath-Brown, Clarendon Press (1986).
  19. E. Bombieri, Remarks on Weil's quadratic functional in the theory of prime numbers, I, Atti Accad. Naz. Lincei (Rend. Lincei Mat. Appl.) 11 (2000), 183–233.
  20. A. Connes, Trace formula in noncommutative geometry and the zeros of the Riemann zeta function, Selecta Math. (N.S.) 5 (1999), 29–106.
  21. A. P. Guinand, A summation formula in the theory of prime numbers, Proc. London Math. Soc. (2) 50 (1948), 107–119.