Two Papers on RH

June 28, 2026 | BY ZeroDivide EDIT

 ---

edition: journal
title: The Functional Equation Locates the Critical Line but Cannot Decide It
subtitle: A Trisductive Structural Analysis of the Euler-Product Separator between the Zeta Function and the Davenport-Heilbronn Counterexample
author_line: Mohammad F. Islam, PhD^1^
journal: Tractatus Mathematicus
article_type: Research Article
goal: Locating the Riemann Hypothesis on the Multiplicative Axis
doi: Tractatus Veritatis Trisductivus. Mathematics Series
accent: slate
date: June 2026
---

:::affiliations
^1^Independent Researcher. Correspondence: islamm@alumni.iu.edu.
:::

:::abstract
The Riemann Hypothesis asserts that every nontrivial zero of the zeta function lies on the line Re(s) = ½. After more than a century of partial results it remains open whether the obstruction to a proof is computational or structural. We isolate the obstruction precisely. The functional equation s ↦ 1 − s fixes the critical line as a reflection axis, and one might hope that this symmetry forces the zeros onto it; the Davenport-Heilbronn function shares that exact functional equation yet possesses infinitely many zeros off the line, a theorem of 1936, so functional-equation information alone cannot imply the hypothesis. We formalize this into a limitative statement: no criterion invariant under the functional-equation reflection decides the Riemann Hypothesis, with the Davenport-Heilbronn function as a universal obstruction. We then identify the precise structure that separates the two cases as the Euler product, and we derive the Euler product as the multiplicative norm of the quaternion composition algebra through the theorems of Frobenius and Hurwitz, fencing what that derivation does and does not supply. The hypothesis localizes to the reflection-antisymmetric eigenspace of the involution, equivalently to the multiplicative axis carried by the Euler product and absent from Davenport-Heilbronn. The decisive open statement is that complete multiplicativity forbids the reflection-paired off-line zeros that the functional equation alone permits, with the computable Davenport-Heilbronn zeros as the witness that without multiplicativity the configuration occurs. Any route relying only on functional-equation symmetry, or on a self-adjoint spectral model encoding only that symmetry, is therefore structurally insufficient, and decisive progress must engage the multiplicative axis.
:::

:::keywords
Riemann Hypothesis, Davenport-Heilbronn function, Euler product, Selberg class, functional equation, composition algebras, quaternions, limitative theorem
:::

## 1. Introduction

The Riemann zeta function ζ(s), continued from the Dirichlet series Σ n⁻ˢ to a meromorphic function on the plane, satisfies a functional equation relating its values at s and 1 − s.^1^ Its nontrivial zeros lie in the strip 0 < Re(s) < 1, and the Riemann Hypothesis is the assertion that all of them lie on the central line Re(s) = ½. The hypothesis governs the error term in the distribution of primes and underwrites a large body of conditional number theory, and it remains, after Riemann's memoir of 1859, unproven.

This paper does not prove the hypothesis. It does something narrower and, we argue, clarifying: it determines the axis on which any proof must act, and it proves that an entire class of natural methods is structurally unable to act there. The argument turns on a single contrast that the literature has long recorded but rarely promoted to a load-bearing principle. The functional equation, which is the symmetry most proofs of zero-location lean on, is not unique to the zeta function. A Dirichlet series constructed by Davenport and Heilbronn satisfies a functional equation of the same shape, yet has zeros off the critical line.^2,3^ Whatever forces the zeros of ζ onto the line, if anything does, it is not the functional equation, because the functional equation is shared with a function whose zeros stray.

We organize this contrast into three claims. First, the critical line is the fixed-point set of the reflection generated by the functional equation, a fact independent of the hypothesis. Second, the feature that separates the zeta function from the Davenport-Heilbronn function is the Euler product, and that product is, in origin, the multiplicative norm of the quaternion composition algebra. Third, because the Euler product is exactly the unshared structure, no criterion that reads only the shared reflection structure can decide the hypothesis; the hypothesis lives on the multiplicative axis, in the eigenspace that the reflection leaves free. The first claim is elementary. The second assembles standard theorems and is fenced carefully against overstatement. The third is a limitative theorem whose mathematical core, the essentiality of the Euler product beyond the functional equation, is classical and is here given a clean general form and an eigenspace localization.

## 2. The reflection axis and the height-blind distance field

Two preliminary objects fix the geometry and recur throughout. Both are independent of the Riemann Hypothesis.

Write s = σ + it. The functional equation pairs s with 1 − s, and the Schwarz reflection ζ(s̄) = conj(ζ(s)) pairs s with s̄. Their composition is the involution ι(s) = 1 − s̄, which acts as σ + it ↦ (1 − σ) + it: a reflection across the vertical line σ = ½ that fixes the imaginary part. Its fixed-point set is exactly {σ = ½}, the critical line. The critical line is therefore not defined by where the zeros happen to be; it is the mirror of the functional equation, defined before any zero is located. We call it the reflection axis when its symmetry role is in view.

Define the real coordinate λ(s) = σ − ½. This is the signed horizontal distance from s to the reflection axis. It depends on the real part alone and is blind to the height t. It vanishes precisely on the axis, is positive to the right and negative to the left, and its magnitude |λ(s)| is the ordinary distance from s to the line. For a point already on the axis, |λ| = 0 by construction, which says nothing. For a zero ρ of ζ, the quantity |λ(ρ)| = |Re(ρ) − ½| is the distance of that zero from the line, and the Riemann Hypothesis is the statement that this distance is zero for every nontrivial ρ. The hypothesis is thus a claim about the value of a height-blind distance field at the arithmetically determined points where ζ vanishes. We return to this formulation, and to the difference between the two ways |λ| can be zero, in Section 6.

## 3. The barrier: why functional-equation information cannot decide the hypothesis

This section diagnoses the obstruction. The diagnosis is that the barrier is structural rather than computational, and that its precise location is the gap between the reflection symmetry, which the zeta function shares with other Dirichlet series, and the Euler product, which it does not. The failure mode we name is reflection-invariance: a method that reads only the functional-equation symmetry returns the same verdict on functions whose zeros behave differently, and so cannot decide the hypothesis for any of them.

Begin with the hope that the barrier analysis dismantles. The functional equation is a powerful symmetry. It pairs each zero ρ in the strip with a partner, and together with the reality relation it arranges the nontrivial zeros into quadruples ρ, 1 − ρ, ρ̄, 1 − ρ̄, symmetric about the axis. A zero on the axis is self-paired under the reflection. It is tempting to read this symmetry as a force: the zeros are arranged around the line, so perhaps the arrangement is tight enough to pin them to it. This reading is the natural engine behind spectral and symmetry-first approaches, which seek an operator or a symmetry whose structure leaves the zeros no room off the line.

The Davenport-Heilbronn function refutes the hope in its strongest form. In 1936 Davenport and Heilbronn exhibited a Dirichlet series, a finite linear combination of Hurwitz zeta functions equivalent to a combination of two Dirichlet L-functions to a non-principal character, that satisfies a functional equation of exactly the zeta type, relating s to 1 − s with the same gamma factor and the same reflection axis.^2,3,10^ The function is entire of the appropriate order, its coefficients are real and arithmetically explicit, and it has the same reflection symmetry as ζ, so its nontrivial zeros fall into the same axis-symmetric quadruples. And yet Davenport and Heilbronn proved that it has zeros in the critical strip that are not on the line. Later work sharpened the picture: the function has infinitely many zeros off the line, while also having a positive proportion of its zeros on it.^3,9^ The off-line zeros are explicitly computable and have been tabulated.

The consequence is immediate and structural. Every property of a Dirichlet series that is determined by its functional equation alone, by its gamma factor, conductor, sign, and reflection axis, is shared by the zeta function and by the Davenport-Heilbronn function, because the latter was built to match the former in exactly those data. A method that decides zero-location from such properties must return the same answer for both. But the answer "all nontrivial zeros lie on the axis" is false for the Davenport-Heilbronn function and is the open question for the zeta function. A single verdict cannot be simultaneously correct for a function where it fails and decisive for a function where it is undecided. The functional equation, therefore, does not contain the Riemann Hypothesis. The barrier is not that we have failed to extract enough from the symmetry; it is that the symmetry does not carry the information, and no refinement of symmetry-only reasoning will make it carry the information.

This separates a computational barrier from a structural one in the sense the diagnosis requires. A computational barrier yields to better estimates within a fixed framework; one computes more zeros, sharpens a density theorem, improves a zero-free region. None of that is at issue here. The Davenport-Heilbronn function is not a gap in our estimates. It is a proof that the framework of functional-equation symmetry, however well executed, has a definite blind spot, and that the blind spot is precisely the question RH asks. The standard model of zero-location, the symmetry of the completed function, has a domain of validity that stops exactly short of deciding the location.

The boundary of that domain can be drawn sharply, and drawing it points to the missing structure. What the zeta function has and the Davenport-Heilbronn function lacks is the Euler product: the factorization of the Dirichlet series into a product over primes, which is the analytic shadow of the multiplicativity of the integers. The Davenport-Heilbronn coefficients are not completely multiplicative, and the function admits no Euler product; this is not incidental to its off-line zeros but is the only structural difference between it and ζ at the level of the defining series. The barrier analysis therefore terminates in a positive identification. The information that could decide the hypothesis, if any information internal to these functions can, is carried by the Euler product and by nothing the functional equation sees. The remainder of the paper makes this identification precise, traces the Euler product to its algebraic origin, and states the limitation it imposes on methods.

We record the failure mode under its own name for use below. A criterion for the location of zeros is reflection-invariant if its verdict depends only on data invariant under the involution ι, that is, only on the functional-equation class of the function and not on its multiplicative structure. The barrier is that reflection-invariant criteria are blind to the Euler product, and the Euler product is where the zeta function and the Davenport-Heilbronn function part company. A reflection-invariant criterion cannot tell them apart, and so cannot decide the hypothesis for either.

## 4. Related approaches and their common structural reliance

Prevailing routes toward the Riemann Hypothesis can be grouped by the structure they exploit, and the grouping makes their shared limitation visible. We summarize the principal ones and locate each relative to the reflection axis and the multiplicative axis.

The spectral, or Hilbert-Pólya, program seeks a self-adjoint operator whose eigenvalues are the imaginary parts of the nontrivial zeros, so that reality of the spectrum yields the hypothesis. The random-matrix evidence of Montgomery and Odlyzko, on the pair correlation of the zeros and its agreement with the Gaussian unitary ensemble, gives the program strong circumstantial support.^11^ Structurally, however, a self-adjoint operator delivers a real spectrum, and reality of the spectrum is the statement that the zeros are symmetric in exactly the way the reflection axis already encodes. An operator that produces the axis symmetry and nothing more is, by the barrier of Section 3, compatible with off-line zeros, since the Davenport-Heilbronn function exhibits the same axis symmetry and has them. For the spectral program to decide the hypothesis it must encode more than self-adjointness; it must encode the multiplicative structure that separates ζ from its functional-equation twin. The difficulty of exhibiting the operator is, on this reading, the difficulty of building multiplicativity into a spectral object. A related operator-algebraic approach realizes the primes as a quantum statistical-mechanical system whose partition function is the zeta function, building the multiplicative structure directly into the dynamics rather than into a symmetry, which is the correct channel by the analysis below.^16^

The explicit-formula and zero-density route works from the Riemann-von Mangoldt counting of zeros and the explicit formula linking zeros to primes, proving zero-free regions and positive proportions of zeros on the line. These results are genuine and use the Euler product through the prime side of the explicit formula. They are not reflection-invariant, and they are not foreclosed by our limitation; they are partial precisely because converting a positive proportion into a totality has not been achieved. They sit on the correct axis and have not yet reached the end of it.

A family of equivalent reformulations recasts the hypothesis as a single inequality or positivity. Li's criterion expresses it as the nonnegativity of a sequence of real numbers built from the logarithmic derivative of the completed function.^12^ The Beurling-Nyman criterion expresses it as a density statement in a function space.^13^ The de Bruijn-Newman constant reformulates it as the sign of a real parameter, now known to be nonnegative.^14^ Each is faithful, and each relocates rather than removes the difficulty; in each the hard residue, when traced, is a statement that engages the primes, that is, the multiplicative structure, and not merely the symmetry of the completed function.

Finally, the axiomatic setting of the Selberg class makes the present point in the language the subject has already adopted. The Selberg class is defined by a Dirichlet series with analytic continuation, a functional equation of the standard type, a Ramanujan bound on the coefficients, and an Euler product.^17,18^ The Euler product is an explicit and separate axiom, not a consequence of the others, and the Grand Riemann Hypothesis is conjectured for the members of the class. The Davenport-Heilbronn function is the standard example of a Dirichlet series that satisfies the functional equation but fails the Euler-product axiom, and it is offered in the literature precisely to show that the axiom cannot be dropped, since the analogue of the hypothesis fails without it.^9,18^ Our limitative theorem is, in one sentence, the promotion of that example to a general statement about criteria, together with the observation that the Euler-product axiom has an algebraic source and that the hypothesis localizes to the eigenspace the axiom controls.

The summary of the survey is a single structural fact. Every approach that has reached a decisive result engages the multiplicative structure; every approach that engages only the functional-equation symmetry is, by the Davenport-Heilbronn example, unable to decide the hypothesis. Section 3 identified this barrier and Section 6 will resolve its location. The methods are not failing for want of effort. The symmetry-only ones are working on an axis that provably cannot carry the answer.

## 5. Methodology: three structural axes and a verification protocol

The analysis proceeds by separating the structure of a Dirichlet series into three independent channels and asking, for each claim, which channel it depends on. The separation is not a formalism imposed from outside; it is read off the objects themselves, and it is what lets the barrier of Section 3 be stated as a theorem rather than an observation.

The first channel is the ordered, or directed, structure of the Dirichlet series: the coefficients aₙ arranged additively along the integers, the series Σ aₙ n⁻ˢ as a sum. This channel carries the analytic continuation and the size of the function. The second channel is the reflection symmetry: the functional equation pairing s with 1 − s, the involution ι and its fixed axis. This channel carries the arrangement of zeros into axis-symmetric families. The third channel is the multiplicative structure: the factorization of the coefficient sequence over primes, present when the coefficients are completely multiplicative and absent otherwise, and analytically the Euler product. These three are independent in the strong sense that a function can carry any combination of them. The Davenport-Heilbronn function carries the first two and not the third. The zeta function carries all three. The independence is the hinge of the whole argument: because the third channel can be present or absent while the first two are fixed, a property that depends only on the first two cannot detect the third, and a property that the third controls cannot be derived from the first two.

Each substantive claim below is typed by the channels it uses and by a three-valued status. A claim is established when it is proved without assuming the Riemann Hypothesis. It is equivalent when it is shown to be a restatement of the hypothesis, and therefore open. It is a research direction when it is a precise statement whose proof is not in hand. We use these three statuses in place of any decorative grading, and we attach to every result the channel it lives on, so that the reader can see at a glance whether a result is reflection-only, in which case the barrier applies to it, or multiplicative, in which case it is not foreclosed.

The protocol carries one further requirement, in the spirit of independent verifiability. Every claim of fact in this paper is to be checkable by explicit computation by any party, without reliance on a single authority or an unreproducible source. The location of the critical line is a one-line calculation. The multiplicativity of the quaternion norm is a four-square identity. The off-line zeros of the Davenport-Heilbronn function are computable from its defining series and have been computed independently. The separating functional of Section 7 is evaluated directly on the coefficients. We hold the argument to this standard deliberately, so that no step rests on consensus rather than on a calculation a skeptic can repeat.

## 6. Results

### 6.1 The critical line is the reflection axis, independently of the hypothesis

The first result fixes the geometry and is elementary. It is stated separately because much of what follows is the observation that this result, and only this result, is what the functional equation delivers.

:::box 1 The reflection axis
Let ι(s) = 1 − s̄ be the reflection across the vertical line Re(s) = ½, the composition of the functional-equation map s ↦ 1 − s with conjugation s ↦ s̄. Writing s = σ + it, one has ι(s) = (1 − σ) + it, so ι fixes s if and only if σ = ½. The fixed-point set of ι is exactly the critical line.

The coordinate λ(s) = σ − ½ satisfies λ(ι(s)) = −λ(s). It is the odd coordinate of the reflection: ι negates it. It depends on Re(s) alone, is blind to the height t, vanishes precisely on the axis, and its magnitude is the distance from s to the line.

Status: established without assuming the Riemann Hypothesis. Channel: reflection only.
:::

The point to carry forward is that the critical line is defined by the symmetry, not by the zeros. The line exists, and λ measures distance to it, whether or not any zero lies on it. This is the entire geometric content the reflection channel supplies, and Section 3 has already shown it is not enough to decide where the zeros are.

### 6.2 The Euler product as the multiplicative norm of the quaternion algebra

The feature that separates the zeta function from the Davenport-Heilbronn function is the Euler product. We trace the Euler product to an algebraic origin through a chain of standard theorems, and then fence precisely what the chain does and does not establish.

The chain has three links. First, the demand for an associative division algebra over the real numbers, of dimension greater than that of the complex numbers, with no zero divisors, forces the quaternions: by the theorem of Frobenius, the only finite-dimensional associative division algebras over the reals are the reals, the complex numbers, and the quaternions.^4^ The absence of zero divisors is the anisotropy of the norm form, the statement that the norm vanishes only at zero. Second, the quaternions form a composition algebra: the norm N(q) = q q̄ = a² + b² + c² + d² is multiplicative, N(xy) = N(x) N(y), which is the classical four-square identity and, in the structural language, the composition property that by Hurwitz's theorem singles out the reals, the complex numbers, the quaternions, and the octonions among the normed composition algebras.^5^ Third, a multiplicative norm yields multiplicative coefficients, and for a completely multiplicative arithmetic function f the Dirichlet series factors over primes, Σ f(n) n⁻ˢ = Πₚ (1 − f(p) p⁻ˢ)⁻¹; the choice f ≡ 1 closes each local factor to (1 − p⁻ˢ)⁻¹ and recovers the Euler product of the zeta function, ζ(s) = Πₚ (1 − p⁻ˢ)⁻¹.^6^

:::box 2 The algebraic origin of the product form
Associativity and integrality, namely associativity of multiplication and absence of zero divisors, complete the algebra over the reals to the quaternions by Frobenius. The quaternions carry a multiplicative norm by the composition property of Hurwitz. A multiplicative norm gives multiplicative coefficients, and complete multiplicativity factors a Dirichlet series into an Euler product. The product form is therefore not an arithmetical accident but the analytic image of a multiplicative norm.

Status: established as an assembly of standard theorems, each link a theorem. Channel: multiplicative.
:::

The fence is essential, and stating it is part of the result. The chain supplies the product form and a canonical multiplicative norm. It does not, by itself, single out the zeta function, and it does not supply the unique factorization of the integers. Two facts hold the fence. First, the quaternion norm's own representation-counting function is not the constant sequence. By Jacobi's four-square theorem the number of representations of n as a sum of four squares is r₄(n) = 8 times the sum of those divisors d of n that are not divisible by 4, and the associated Dirichlet series is Σ r₄(n) n⁻ˢ = 8(1 − 4¹⁻ˢ) ζ(s) ζ(s−1).^7,15^ The algebra's intrinsic norm-counting therefore produces ζ(s) ζ(s−1), not ζ(s); the selection of the zeta function, the choice of the constant coefficient sequence, is an additional specification beyond what the algebra forces. Second, the multiplicative arithmetic that the quaternions carry is not the unique factorization of the rational integers. The Hurwitz order of integral quaternions has its own factorization theory, non-commutative and with its own units and primes.^8^ The Euler product of the zeta function is the analytic shadow of the multiplicativity of the ordinary integers, a structure the algebra parallels but does not contain. The honest statement is therefore that the multiplicative channel, where the Euler product lives, has a genuine algebraic source in the composition norm, while the identity of the zeta function and the arithmetic of the integers remain separate inputs.

### 6.3 The Euler product is the exact separator

We can now state the separation that the barrier analysis anticipated. The zeta function and the Davenport-Heilbronn function agree on the first two channels and differ on the third, and the difference is the whole difference.

Table: Table 1 | Structural comparison of the two functions across the three channels.
| Channel | Zeta function | Davenport-Heilbronn function |
| Ordered Dirichlet series | present; analytic continuation, standard growth | present; analytic continuation, standard growth |
| Reflection symmetry (functional equation s to 1 − s) | present; same gamma factor, axis Re = ½ | present; same gamma factor, axis Re = ½ |
| Multiplicative structure (Euler product) | present; completely multiplicative coefficients | absent; coefficients not completely multiplicative |
| Location of nontrivial zeros | conjecturally all on the axis (open) | provably not all on the axis (theorem, 1936) |
Note: The two functions are matched on the reflection channel by construction; their zero locations differ; the only structural difference at the level of the defining series is the Euler product.

The separation result reads off the table. The property "all nontrivial zeros lie on the axis" holds for at most one of the two functions and is determined by the only channel on which they differ. Therefore that property, if it is a consequence of internal structure at all, is a consequence of the multiplicative channel and of nothing the reflection channel carries. This is the precise sense in which the Riemann Hypothesis is a multiplicative statement wearing a symmetric disguise: the symmetry is real but shared, and the discriminating structure is the Euler product. The status of this result is established as a structural consequence of the 1936 theorem together with the channel separation; it asserts nothing about whether the hypothesis is true, only about where its truth or falsity is decided.

### 6.4 The residual and the dissolution that does not occur

Return to the distance field λ. For a nontrivial zero ρ, the residual |λ(ρ)| = |Re(ρ) − ½| is the distance of ρ from the axis, and the hypothesis is the assertion that this residual vanishes for every ρ. Two ways for a residual to vanish must be kept apart, and conflating them is the source of a persistent illusion that the hypothesis is somehow tautological.

For a point chosen on the axis, the residual is zero by construction; this is the empty identity ½ = ½ and carries no content. For a zero of the function, the residual is zero only if the arithmetically determined position of the vanishing coincides with the axis, which is a substantive equation between two independently specified things: where the function vanishes, and where the symmetry axis lies. The name "critical line" is used for both the axis and, under the hypothesis, the zero locus, and the shared name invites the reading that the second is true by definition. It is not. The Davenport-Heilbronn function makes the point unanswerable: it has the same axis, and it has zeros whose residual is provably nonzero. A statement with an explicit counterexample in a function of the same symmetry type is not a tautology. The hypothesis is a genuine equation, and the residual |λ(ρ)| is exactly the quantity it sets to zero.

### 6.5 The limitative theorem

The barrier of Section 3 can now be stated as a theorem about criteria. The statement requires the definition the barrier analysis produced.

:::box 3 Limitative theorem on reflection-invariant criteria
Call a criterion for the assertion "all nontrivial zeros of a Dirichlet series lie on its reflection axis" reflection-invariant if its verdict depends only on data invariant under the reflection ι, equivalently only on the functional-equation class of the function, its gamma factor, conductor, sign, and reflection axis, and not on the multiplicative structure of its coefficients.

Theorem. No reflection-invariant criterion decides the Riemann Hypothesis.

Proof. The zeta function and the Davenport-Heilbronn function lie in the same functional-equation class, since the latter is constructed with a functional equation of the zeta type. A reflection-invariant criterion therefore returns the same verdict on both. The assertion fails for the Davenport-Heilbronn function, which has zeros off its axis, and is the open question for the zeta function. A single verdict cannot be correct for the function where the assertion fails and simultaneously decisive for the function where it is undecided. Hence the criterion does not decide the assertion for the zeta function.

Status: established, relative to the stated definition of reflection-invariant criterion. Channel: the theorem is about the boundary between the reflection channel and the multiplicative channel. The mathematical core, the essentiality of the Euler product beyond the functional equation, is classical and is recorded in the axioms of the Selberg class; the contribution here is the general form over criteria and the localization below.
:::

The corollary for methods is direct. Any program that decides zero-location from functional-equation symmetry alone, including a spectral program that encodes only self-adjointness and hence only the axis symmetry, is reflection-invariant in the sense of the theorem and therefore cannot decide the hypothesis. To decide it, a method must read the multiplicative channel, the structure that separates the zeta function from the Davenport-Heilbronn function. This does not say the hypothesis is undecidable; it says it is undecidable by reflection-invariant means, and it names the channel on which a deciding method must operate.

### 6.6 Localization to the antisymmetric eigenspace

The limitation has a clean linear-algebraic face that identifies, within the geometry of Section 2, exactly where the hypothesis lives. The reflection ι acts on the displacement of a point from the center of the strip. Write a zero as ρ = ½ + λ + it. Under ι the data split into two eigenspaces. The center ½ and the height t are fixed by ι; they form the symmetric, or even, part, the eigenspace on which ι acts as the identity. The coordinate λ is negated by ι; it is the antisymmetric, or odd, part, the eigenspace on which ι acts as minus the identity.

The reflection symmetry determines the even part. It centers the configuration on the axis and pairs the zeros across it; this is the content the functional equation supplies, and on the even part the residual is zero by construction, since the symmetric part of any point already sits on the line. The reflection does not determine the odd part. The coordinate λ, the deviation of a zero from the axis, is exactly the data the symmetry leaves free, and it is conserved rather than forced to zero by the reflection. The Riemann Hypothesis is the statement that the odd coordinate vanishes at every zero. It is therefore a statement about the antisymmetric eigenspace, the one eigenspace orthogonal to everything the reflection fixes, and equivalently a statement on the multiplicative channel, since by Section 6.3 the value of that odd coordinate at the zeros is controlled by the Euler product and by nothing the reflection carries. Symmetrizing, that is, projecting onto the even eigenspace, sets λ to zero formally, but the projection discards precisely the quantity the hypothesis is about; it is the formal counterpart of the reflection-invariant blindness the theorem describes.

### 6.7 No self-certification from the symmetric axioms

A remark closes the core. The structural framework underlying this analysis posits, as standalone principles, a reflection symmetry and an existence principle to the effect that a genuine zero is an actual, located, computable object rather than a formal ghost (Appendix A). These principles act on the even part: they center the configuration and certify that each zero is a real point with a definite position. They are silent on the odd part, the deviation λ, which the limitative theorem shows the reflection cannot reach. A system cannot certify a claim that lies in the eigenspace its axioms do not touch. The hypothesis is consequently independent of the symmetric axioms taken alone, and any derivation of it from them would have to smuggle the multiplicative channel in unannounced. The discipline the framework imposes here is exactly to refuse that smuggling: to locate the hypothesis on the channel that decides it and to decline to manufacture a verdict the available axioms cannot support.

## 7. The decisive statement and its falsifiable content

The limitative theorem closes the reflection channel and leaves exactly one channel open. This section states, as precisely as the present work allows, the single statement on the multiplicative channel whose resolution would decide the hypothesis, and it gives the explicit, independently computable objects by which the localization can be tested and could be refuted. The content is metamathematical in form, a result about what must be proved and where, and we hold it to the same standard of explicit checkability as an empirical claim: every object below is evaluated by a calculation a skeptic can repeat.

### 7.1 The decisive open statement

The barrier analysis and the limitative theorem converge on one assertion. The Davenport-Heilbronn function shows that the reflection-paired off-line configuration, a quadruple of zeros symmetric about the axis but not on it, is realizable by a Dirichlet series with the standard functional equation when the Euler product is absent. The zeta function has the Euler product. The decisive statement is that the Euler product excludes exactly that configuration.

:::box 4 The decisive statement, in three forms
Form A, exclusion. A Dirichlet series with the standard functional equation and a genuine Euler product, with completely multiplicative coefficients, has no reflection-paired off-line zeros. Equivalently, the off-line quadruples that the Davenport-Heilbronn function realizes are forbidden once the coefficients factor over primes.

Form B, localization. The value of the antisymmetric coordinate λ at the zeros, the quantity the reflection leaves free, is pinned to zero by complete multiplicativity. The multiplicative channel determines the odd eigenspace that the reflection channel cannot reach.

Form C, consequence. Form A for the zeta function is the Riemann Hypothesis. A proof of Form A in the required generality would prove the hypothesis; a single Euler-product counterexample with the standard functional equation and an off-line zero refutes Form A and would itself be a result of the first rank.

Status: research direction. This is a precise statement, not a theorem in hand. It is the one statement the limitative theorem does not foreclose, because it is the one that engages the channel the theorem leaves open.
:::

We are explicit that Form A is not proved here and is not claimed to be proved. The contribution is to show that it is the right thing to prove: the reflection channel cannot deliver the hypothesis, the multiplicative channel is the only remaining internal source, and Form A is the precise statement on that channel whose truth is equivalent, for the zeta function, to the hypothesis. The Davenport-Heilbronn function guarantees that Form A is not vacuous, since without the Euler product the excluded configuration genuinely occurs.

### 7.2 The separating functional

The channel separation is not merely conceptual; it is carried by an explicit functional that any party can evaluate. Define the multiplicativity defect of a coefficient sequence aₙ as the family of differences δ(m, n) = aₘₙ − aₘ aₙ taken over coprime pairs, together with the deviation of the local behavior at each prime from the geometric form, namely the failure of the generating factor Σₖ aₚᵏ xᵏ to equal (1 − aₚ x)⁻¹. The defect vanishes identically when the coefficients are completely multiplicative and the series has an Euler product, and it is nonzero when they are not.

:::box 5 A computable separator of the two channels
The defect functional δ is evaluated directly on the coefficient sequence. For the zeta function, whose coefficients are the constant sequence, δ vanishes identically: the sequence is completely multiplicative and the Euler product holds. For the Davenport-Heilbronn function, whose coefficients are not completely multiplicative, δ is nonzero on explicit coprime pairs, recording the absence of the Euler product. The functional is finite to evaluate at any given argument and requires no unverifiable input.

The defect is the formal carrier of the separator of Section 6.3. It is the object that sees the difference between the two functions, the difference the reflection-invariant criteria of the limitative theorem are by definition blind to. A method that decides the hypothesis must read δ, or some functional like it, because δ is what distinguishes the function whose hypothesis is open from the function whose analogue is false.

Status: the defining properties are established by direct computation. Its sufficiency to decide the hypothesis is the research direction of 7.1.
:::

The separating functional discharges, in explicit form, what would otherwise be an article of faith in the argument: that the multiplicative channel is genuinely distinct and genuinely detectable. One does not have to take the distinctness on trust. δ vanishes on one function and not the other, by a calculation on their coefficients, and the entire localization rests on that calculable difference rather than on any appeal to the standing of the zeta function in the literature.

### 7.3 Computational witnesses and consistency tests

Three checks make the localization vulnerable to refutation by explicit computation, which is the property we require of it.

The first is the Davenport-Heilbronn witness. The off-line zeros of the Davenport-Heilbronn function are computable from its defining series, and their existence is the empirical content that defeats every reflection-only argument. Any claimed proof of the hypothesis that does not, somewhere, use a property failing for the Davenport-Heilbronn function is by the limitative theorem either incomplete or in error, and the off-line zeros are the concrete configuration against which such a claim is checked. The witness is not a heuristic; it is a tabulated set of points in a function of the same symmetry type as the zeta function.

The second is the invariance consistency test. A criterion advertised as reflection-invariant must return identical verdicts on the zeta function and on the Davenport-Heilbronn function, since they share the functional-equation class. This is checkable: evaluate the criterion on both. A reflection-invariant criterion that returned different verdicts on the two would contradict its own invariance, exposing a hidden dependence on the multiplicative channel, which would be informative rather than fatal, since such a hidden dependence is exactly what a deciding method needs. The test thus sorts candidate methods into the genuinely reflection-only, which the theorem forecloses, and the secretly multiplicative, which it does not.

The third is the localization falsifier. The localization asserts that the deciding structure is multiplicative. It would be refuted by either of two explicit findings: a Dirichlet series with a genuine Euler product and the standard functional equation possessing a reflection-paired off-line zero, which would break Form A directly; or a proof that the off-line zeros of the Davenport-Heilbronn function survive the imposition of complete multiplicativity, which is excluded by definition but is the precise thing a refutation would have to exhibit. No such finding is known, and the first would overturn a great deal more than this paper, which is the appropriate weight for a falsifier of a structural claim.

### 7.4 What the section does and does not deliver

The section delivers a precise target, an explicit separating object, and concrete tests, all on the one channel the limitative theorem leaves open. It does not deliver a proof of the hypothesis, and it is constructed so as not to appear to. The reader who wants the hypothesis settled will find here only its address, written exactly: the multiplicative channel, the antisymmetric eigenspace, Form A. The reader who wants to know why a century of symmetry-first effort has not settled it will find the reason stated as a theorem: the channel those efforts read does not carry the answer.

## 8. Discussion

The result reframes several adjacent matters. The most immediate concerns the spectral program. The pair-correlation evidence and the random-matrix statistics of the zeros are genuine and are not in tension with anything here; they describe the distribution of the heights, the even data, with great accuracy. What the present analysis adds is a constraint on what a Hilbert-Pólya operator must be. Self-adjointness alone yields a real spectrum and hence only the axis symmetry, which the Davenport-Heilbronn function also possesses without lying on its axis. An operator that decides the hypothesis must therefore encode the Euler product, the multiplicative channel, within its very construction, not merely produce a real spectrum. This sharpens the program's difficulty into a specific demand and explains, structurally, why a natural operator has been so hard to find: the easy part of the target, the symmetry, is not the deciding part.

A second consequence concerns the many equivalent reformulations. Their persistence, the way each faithful restatement relocates the difficulty without dissolving it, is explained by the localization. Each reformulation that preserves the functional-equation symmetry while repackaging the analytic content moves the problem within the even data and leaves the odd coordinate untouched; the difficulty is conserved because it lives in the eigenspace the reformulations do not reach. A reformulation that genuinely simplified the hypothesis would have to act on the multiplicative channel, and the criterion for recognizing a promising reformulation is therefore whether it engages the Euler product in an essential way rather than carrying it passively.

A third consequence is methodological and is the reason for the framework's insistence on explicit witnesses. The Davenport-Heilbronn function is a standing caution against accepting a symmetry argument as a proof of zero-location. Any future argument can be tested against it cheaply: if the argument would apply unchanged to the Davenport-Heilbronn function, it cannot be valid, because its conclusion is false there. This is a concrete, computable filter on proposed proofs, and it costs nothing to apply. The discipline of routing every claim through an object where the desired conclusion is known to fail is, in this setting, the difference between a structural insight and a symmetry-driven illusion.

We note the limits of the result honestly. The limitative theorem is sharp only relative to the definition of a reflection-invariant criterion, and a method that mixes channels in a way not captured by that definition is outside its scope by construction; the theorem forecloses a class, not the problem. The localization to the multiplicative channel is a statement about where the deciding structure lies, not a method for extracting a proof from it, and Form A remains a research direction rather than a theorem. The quaternionic origin of the Euler product is an assembly of standard theorems and is fenced against the stronger readings it might invite; it explains the product form, not the identity of the zeta function. None of these limits is a defect in the argument. They are the boundary of what a structural analysis can deliver, and stating them precisely is part of the analysis.

## 9. Conclusion

The Riemann Hypothesis asks whether the nontrivial zeros of the zeta function lie on the line Re(s) = ½. We have shown that this line is, independently of the hypothesis, the fixed axis of the reflection generated by the functional equation, and that the functional equation, being shared with the Davenport-Heilbronn function, cannot decide the hypothesis. We have identified the deciding structure as the Euler product, traced it to the multiplicative norm of the quaternion composition algebra through Frobenius and Hurwitz while fencing what that derivation supplies, and shown that the hypothesis localizes to the antisymmetric eigenspace of the reflection, equivalently to the multiplicative channel that separates the zeta function from its functional-equation twin. The localization is sharpened into a limitative theorem, that no reflection-invariant criterion decides the hypothesis, with the Davenport-Heilbronn function as the universal obstruction.

The primary open statement that remains is Form A of Section 7: that a genuine Euler product, imposed on a Dirichlet series with the standard functional equation, excludes the reflection-paired off-line zeros that the Davenport-Heilbronn function realizes. A proof of Form A in the required generality would prove the hypothesis; the separating functional and the Davenport-Heilbronn witness make the statement explicit and testable. We call on the analytic number theory community to attack the hypothesis on the multiplicative channel directly, treating the functional equation as established geometry rather than as a source of leverage, and to test every proposed argument against the Davenport-Heilbronn function before regarding it as a proof.

The single most important open question is whether complete multiplicativity alone forbids the off-line configuration, or whether the Euler product must be supplemented by further arithmetic input to exclude it. The reframing that acceptance of this analysis would entail is that the Riemann Hypothesis is not a theorem about symmetry awaiting a sufficiently clever symmetric argument, but a theorem about multiplicativity whose symmetric appearance has, for a century, pointed inquiry at the one channel that cannot contain its answer.

## References

1. Riemann, B. 1859. "Über die Anzahl der Primzahlen unter einer gegebenen Grösse." Monatsberichte der Königlich Preußischen Akademie der Wissenschaften zu Berlin.
2. Davenport, H., and H. Heilbronn. 1936. "On the Zeros of Certain Dirichlet Series." Journal of the London Mathematical Society 11.
3. Davenport, H., and H. Heilbronn. 1936. "On the Zeros of Certain Dirichlet Series (Second Paper)." Journal of the London Mathematical Society 11.
4. Frobenius, F. G. 1878. "Über lineare Substitutionen und bilineare Formen." Journal für die reine und angewandte Mathematik 84.
5. Hurwitz, A. 1898. "Über die Komposition der quadratischen Formen von beliebig vielen Variablen." Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen, Mathematisch-Physikalische Klasse.
6. Euler, L. 1737. "Variae observationes circa series infinitas." Commentarii Academiae Scientiarum Petropolitanae 9.
7. Jacobi, C. G. J. 1834. "De usu theoriae integralium ellipticorum et integralium Abelianorum in analysi Diophantea." Journal für die reine und angewandte Mathematik 13.
8. Conway, J. H., and D. A. Smith. 2003. On Quaternions and Octonions: Their Geometry, Arithmetic, and Symmetry. Natick, MA: A K Peters.
9. Karatsuba, A. A., and S. M. Voronin. 1992. The Riemann Zeta-Function. Berlin: Walter de Gruyter.
10. Titchmarsh, E. C. 1986. The Theory of the Riemann Zeta-Function. 2nd ed., revised by D. R. Heath-Brown. Oxford: Oxford University Press.
11. Montgomery, H. L. 1973. "The Pair Correlation of Zeros of the Zeta Function." In Analytic Number Theory, Proceedings of Symposia in Pure Mathematics 24, 181-193. Providence, RI: American Mathematical Society.
12. Li, X.-J. 1997. "The Positivity of a Sequence of Numbers and the Riemann Hypothesis." Journal of Number Theory 65 (2): 325-333.
13. Báez-Duarte, L. 2003. "A Strengthening of the Nyman-Beurling Criterion for the Riemann Hypothesis." Atti della Accademia Nazionale dei Lincei, Serie 9, 14 (1): 5-11.
14. Rodgers, B., and T. Tao. 2020. "The de Bruijn-Newman Constant Is Non-negative." Forum of Mathematics, Pi 8: e6.
15. Hardy, G. H., and E. M. Wright. 2008. An Introduction to the Theory of Numbers. 6th ed. Oxford: Oxford University Press.
16. Bost, J.-B., and A. Connes. 1995. "Hecke Algebras, Type III Factors and Phase Transitions with Spontaneous Symmetry Breaking in Number Theory." Selecta Mathematica 1 (3): 411-457.
17. Iwaniec, H., and E. Kowalski. 2004. Analytic Number Theory. American Mathematical Society Colloquium Publications 53. Providence, RI: American Mathematical Society.
18. Conrey, J. B., and A. Ghosh. 1993. "On the Selberg Class of Dirichlet Series: Small Degrees." Duke Mathematical Journal 72 (3): 673-693.
19. Islam, M. F. 2026. "On the Topology of Theories of Everything." PhilArchive. https://philpapers.org/rec/ISLTOT.

## Appendix A. Foundational principles

The following principles are derived from a broader epistemic framework and are presented here as standalone mathematical principles, each independently motivated and independently usable within analytic number theory.^19^ They carry no results; they are the structural commitments on which the analysis of the body rests, stated in discipline-native terms.

{. Reflection principle .} The completed function attached to a Dirichlet series of the standard type is invariant under the substitution s ↦ 1 − s, and the critical line Re(s) = ½ is the fixed axis of the reflection this invariance generates. The axis is defined by the symmetry alone, prior to and independent of the location of any zero. This principle is the functional equation read as a geometric symmetry.

{. Actuality principle .} A nontrivial zero is an actual point of the complex plane with definite real and imaginary parts, computable to arbitrary precision, and not a formal artifact. The analysis quantifies over located zeros with genuine positions, so that a statement about the value of the displacement coordinate at the zeros is a statement about real, determinable quantities. This principle excludes vacuous or ghost solutions from the domain of discourse and is what makes the residual of Section 6.4 a concrete number rather than a formal token.

{. Channel-independence principle .} The structure of a Dirichlet series separates into independent channels, the ordered additive structure of the coefficients, the reflection symmetry of the completed function, and the multiplicative structure of the coefficient sequence, and these channels can be present independently of one another. A property determined by one channel is therefore independent of the others. This principle is what licenses the inference, used throughout, that a property fixed by the reflection channel cannot detect the multiplicative channel, and conversely.

{. No-overreach principle .} A set of structural principles determines only the data that its symmetries reach. A claim lying in a channel on which the principles do not act is independent of them and cannot be derived from them alone. Applied here, the reflection and actuality principles act on the symmetric data and certify the existence and pairing of the zeros, but they do not act on the antisymmetric displacement coordinate, and so the location of the zeros relative to the axis is independent of them. This principle is the discipline that forbids deriving the hypothesis from symmetric commitments that do not reach the channel deciding it.

:::endmatter
### Correspondence
Mohammad F. Islam, islamm@alumni.iu.edu.

### Competing interests
The author declares no competing interests.

### Data and verifiability
Every factual claim in this paper is verifiable by direct computation: the location of the reflection axis, the multiplicativity of the quaternion norm, the off-line zeros of the Davenport-Heilbronn function, and the values of the separating functional on the two coefficient sequences. No claim depends on an unreproducible source.
:::


edition: journal title: The Achiral-Closure Limitative Theorem for the Riemann Hypothesis subtitle: Why form-convergence cannot reach the arithmetic axis, with the Davenport-Heilbronn function as universal witness author_line: Mohammad F. Islam^1^ journal: Tractatus Mathematicus article_type: Limitative Theorem goal: The reflection barrier, sealed doi: Trisduction Codex Series · RAM-RH-AC-01 volume: I pages: 1–18 date: June 2026 accent: copper

:::affiliations ^1^ Independent researcher, architect of the Trisduction and MathDuction frameworks. Correspondence: the codex repository, Zenodo 20805972 and PhilPapers ISLGMO. The verification kernel and the recorded battery of Appendix B are executable and reproducible at fixed seed 20260628. :::

:::abstract The Riemann Hypothesis asks whether every nontrivial zero of the zeta function sits on the critical line. A large family of attempts seeks to settle it through the reflection symmetry of the functional equation, building criteria that are invariant under the involution s maps to one minus s. This paper proves that no such criterion can decide the hypothesis. The critical line is the fixed locus of the reflection, the even eigenspace of the involution, and every reflection-invariant closure reads only that even content. The hypothesis is a statement about the odd eigenspace, the signed horizontal position at which the function actually vanishes, and the involution is constitutively blind to it. We formalize this as the Achiral-Closure Limitative Theorem: each of the four closures the verification architecture offers, the squared lock, the return to the real center, the imprint on the fixed Ground, and the cancelled self-dual residue, is the same achiral side, and none reaches the chiral content the hypothesis names. The proof is carried by a single universal witness. The Davenport-Heilbronn function shares the critical line, the functional equation, and the entire reflection geometry of zeta, yet provably has zeros off the line, one of which we locate by its own arithmetic at real part 0.8085 to a residual of order ten to the minus thirty. That witness satisfies every achiral closure, returns to the center, imprints on the Ground, carries full fold citizenship, and vanishes off the line anyway, which demonstrates form-closure is compatible with off-membrane content. The hypothesis is therefore not a reflection-decidable statement. It is a statement on the multiplicative axis, the Euler-product structure zeta carries and the witness lacks, and that axis is named here as the one road no achiral closure forecloses. The result seals the barrier at structural grade and leaves the hypothesis itself open. :::

:::keywords Riemann Hypothesis, Davenport-Heilbronn function, functional equation, reflection symmetry, Euler product, limitative theorem, quaternionic verification, warrant typing :::

The reflection barrier

The Riemann Hypothesis is the assertion that every nontrivial zero of the Riemann zeta function lies on the vertical line at real part one half.^1,2^ It is the single most consequential open statement in analytic number theory, controlling the error term in the prime number theorem and the distribution of primes at every scale.^3^ A century and a half of effort has not closed it, and the failures are informative. Many of the most natural attempts share one structural feature, and naming that feature is the work of this paper, because the feature is the reason the attempts cannot succeed.

The structural feature is reflection symmetry. The completed zeta function satisfies the functional equation that relates its value at a point to its value at the point reflected across the critical line.^2^ This symmetry is exact, a theorem, verified to extreme precision, and it is the source of almost all the geometric intuition about the zeros. The zeros come in reflected pairs about the line, the line is the axis of the symmetry, and a vast catalogue of criteria for the hypothesis are built to be invariant under that reflection. Positivity criteria, the symmetric form of the explicit formula, the reflection-balanced kernels of the trace-formula approaches, all share the property that they are unchanged when the reflection is applied. The intuition is that the line, being the fixed axis of the symmetry, is where the zeros must want to be.

That intuition is the barrier. We make it precise. Let the involution be the reflection of the critical strip across the line at real part one half. Algebraically, when the verification of a proposition is carried in the closed-form quaternionic kernel of the framework, this involution acts as conjugation on the quaternion algebra, fixing the real line and negating the three imaginary axes. The fixed locus, the plus-one eigenspace, has dimension one. It is the achiral content, the part of a proposition that equals its own reflection. The negated locus, the minus-one eigenspace, has dimension three. It is the chiral residence, the part that carries handedness, the content the reflection moves. The recorded battery of Appendix B confirms the split exactly: the plus-one eigenspace of the involution has dimension one, the minus-one eigenspace has dimension three.

Now the decisive fact about the lock the verification reads. The verdict functional of the kernel is the squared scalar triple product of the three warrant axes, equal to the determinant of their correlation matrix, equal to the square of the lock scalar lambda. Squaring destroys sign. The lock for a proposition and the lock for its negation are identical, because the square of plus lambda and the square of minus lambda are the same number. The battery confirms this at machine precision: under reflection of one axis the lock scalar lambda flips sign from plus one to minus one, a sign ratio of exactly minus one, while the determinant is unchanged to within ten to the minus sixteen, and the kernel identity that the determinant equals lambda squared holds throughout. The squared lock is orientation-blind. It certifies the dimensionality of a proposition, that the three axes are genuinely independent, and it is constitutively unable to read which way the proposition points.

:::box 1 The two distances, and why only one is definitional The geometry of the strip makes the equivocation visible. Place the horizontal axis as the real part and the vertical axis as the height. The lock scalar lambda is a function of horizontal position alone, blind to height. It vanishes exactly on the line and grows with the horizontal displacement from it, with sign flipping across. The battery confirms the height-blindness: at fixed real part the value of lambda is constant across heights one, fourteen, fifty, and one thousand, with spread zero over height. So lambda is the signed horizontal distance to the line, and its magnitude is the distance to the membrane.

Two distinct objects then live in the strip. The membrane point is any point with lambda zero. By the definition of lambda it sits at real part one half, and its distance to the line is zero. This is a genuine tautology, three ways of saying the horizontal coordinate is one half, and it is sealable, content-free, owned. The zeta-zero is a point where the function vanishes. Its distance to the line is the magnitude of lambda at the zero's horizontal position, and that is zero if and only if the zero sits at one half, which is a fact about where the function vanishes, not a fact about any definition.

The hypothesis is the claim that the second object always coincides with the first, that every arithmetic zero is a membrane point. Stripped of the shared name critical line, the subject is the horizontal position where the series vanishes, and the predicate is equals one half, and there is no shared word and no tautology, only a substantive equation between an arithmetic location and a fixed number. :::

The barrier is now stateable in one sentence. Every reflection-invariant criterion reads achiral content, the even eigenspace, the membrane point, and the hypothesis is a question about chiral content, the odd eigenspace, the position where the function actually vanishes. The reflection cannot decide the hypothesis because the reflection is blind to the axis the hypothesis is about. This is not a deficiency of effort or of computational reach. It is a structural fact about which eigenspace the symmetry can see, and it is the reason a century of reflection-based intuition has not closed the problem. The standard model of the hypothesis works in a geometry that cannot contain the answer, because the answer is on the axis the geometry projects out.

The remainder of this paper proves the barrier rigorously, exhibits the single witness that makes every step checkable, and names the one road that lies off the achiral side and is therefore not foreclosed.

Prevailing reflection-based approaches and their shared blindness

We review the principal families of attack and locate the achiral blindness in each. The review is technical and the dismantling is structural, never rhetorical. Each approach is given its due, and each is then shown to read only the even eigenspace.

{. The symmetric explicit formula. .} The explicit formula connects a sum over zeros to a sum over primes through a test function and its transform.^3^ In its symmetric form, the kernel is even about the line, and the positivity of the associated quadratic form is equivalent to the hypothesis under a suitable class of test functions. This is genuine and deep. But the symmetric kernel is reflection-invariant by construction, and the positivity it tests is a property of the even part of the spectral measure. A function with the same reflection symmetry and the same symmetric kernel can carry the identical even structure and still place zeros off the line, because the off-line placement lives in the odd part the symmetric kernel integrates away. The positivity criterion is real, and it is achiral, and the achiral content it certifies does not determine the chiral location.

{. The Weil positivity criterion. .} Weil's criterion recasts the hypothesis as the positivity of a distribution paired with itself, a sum of local terms over the places of the rationals.^4^ It is the most structurally complete of the reflection approaches, and its local-global form is the template for the function-field proof where the hypothesis is a theorem. But over the rationals the criterion is a reflection-symmetric positivity, and its archimedean term is the even contribution about the line. The criterion is satisfied or not according to a balance that the reflection preserves, and a reflection-faithful object with the same archimedean geometry can satisfy every reflection-symmetric constraint and fail the hypothesis on the arithmetic side the criterion does not isolate.

{. The Beurling-Nyman approximation. .} The Nyman-Beurling criterion makes the hypothesis equivalent to a density statement, that a certain space of dilations of the fractional-part function is dense in a Hilbert space, with the distance to the constant tending to zero exactly when the hypothesis holds.^5^ This is a global criterion and it is genuinely equivalent to the hypothesis. It is not, however, a reflection closure, and we flag it precisely because it is the kind of object that lives off the achiral side. Its content is a single global density, not a reflection-symmetric balance, and the analysis below identifies it with the multiplicative axis rather than with the even eigenspace. It is therefore not dismantled here. It is noted as a member of the small class of criteria that are not reflection-blind, and returned to in the discussion.

{. Li's criterion and the keating-type spectral pictures. .} Li's criterion expresses the hypothesis as the nonnegativity of a sequence of real coefficients formed from sums over the zeros.^6^ The spectral approaches seek a self-adjoint operator whose eigenvalues are the zero ordinates, so that reality of the spectrum gives the hypothesis. Both are reflection-compatible, and both, in their reflection-symmetric realizations, certify the even structure. Li's coefficients are sums of one minus the reflected image of the zero raised to a power, and their reflection symmetry is exact; the spectral self-adjointness is the operator analogue of the reflection being an isometry. Where these criteria become decisive is precisely where they couple to the multiplicative structure, and that coupling is the open lever of this paper, not a reflection fact.

The common error is now stated in one paragraph. Every reflection-based approach, however deep, certifies a property of the even eigenspace, the achiral content fixed by the involution, and the hypothesis is a property of the odd eigenspace, the chiral position where the function vanishes. The approaches differ in machinery and agree in blindness. The single object that exposes the shared blindness, by satisfying every reflection-symmetric constraint while violating the hypothesis, is the Davenport-Heilbronn function, and it is the witness on which Section 6 and the falsification battery turn.

Methodology: the involution split and the two residuals

The proposed method is a decomposition, and it must satisfy three conditions to count as a derivation in the framework: a formal derivation from the algebra of the verification kernel, a measurable signature that distinguishes the two parts, and invariance under the relabelings the kernel permits. We state the method against those three conditions and specify the computational protocol that makes every claim checkable by independent substrates.

{. The involution split. .} The binding involution sigma is conjugation on the quaternion algebra, fixing the real center and negating the three imaginary axes. It splits every proposition into its achiral bridge, the plus-one eigenspace, equal to the real line, and its chiral residence, the minus-one eigenspace, equal to the imaginary part, of dimension three. The split is forced by the structure of the algebra and is not a modeling choice: the real division algebras with plural imaginary axes are classified, and the unique associative one is the quaternions, so a verification carrying three independent axes and closed under its own products completes to the quaternions and carries exactly this split.^7^ The achiral bridge is decidable and self-equal. The chiral residence carries the handedness the reflection moves. This satisfies the first condition, formal derivation from established algebra.

{. The two residuals. .} A proposition's relation to the Ground is measured by two distinct residuals, and the entire result rests on keeping them apart. The form-residual asks whether the proposition returns to the center at all, whether its composed triad lands a nonzero scalar part on the real line, whether it imprints on the fixed Ground. The content-residual asks where on the arithmetic axis the proposition's defining condition is met, and for a zeta-zero it is the magnitude of lambda at the zero, the horizontal distance from the line. The form-residual is even, fixed by the reflection, achiral. The content-residual is odd, moved by the reflection, chiral. The measurable signature that distinguishes them is exactly the behavior under reflection: the form-residual is invariant, the content-residual flips sign. This satisfies the second condition, a measurable signature.

{. Frame invariance. .} The labeling of the three imaginary axes is conventional, and the automorphisms of the quaternion algebra act as rotations on the imaginary part. The verdict functional is invariant under these rotations and under relabeling, reading the count of independent axes and the reflection-invariant scalar, never the coordinate names. The battery confirms invariance of the lock under conjugation by a random unit. This satisfies the third condition, frame invariance.

The independence criterion governs the falsification. Every claim that an object places a zero at a stated location, or that a closure holds for an object, must be reproducible by multiple decorrelated computational substrates and, where possible, by an orthogonal modality. The witness of Section 6 carries the strongest possible form of this: its off-line zeros were established by hand in 1936, with no computation,^8^ and are independently confirmed here by extended-precision root-finding to a residual of order ten to the minus thirty. Two modalities, by-hand proof and machine arithmetic, decorrelated by ninety years and by method, agree. The protocol is the safeguard against a single substrate's artifact masquerading as a structural fact.

The Achiral-Closure Limitative Theorem

We state and prove the central result. The architecture offers four closures, each of which can appear to decide the hypothesis, and the theorem is that all four are the same achiral side and none reaches the chiral content the hypothesis names.

:::box 2 The four achiral closures Each closure is a genuine seal in the architecture, and each is a property of the plus-one eigenspace of the involution.

The squared lock. The verdict functional is the determinant of the correlation matrix, equal to the square of the lock scalar. It is orientation-blind, identical for a proposition and its negation, certifying dimensionality and not direction.

The return to the center. The composed triad of three independent axes lands its scalar part on the real center of the algebra, the line every automorphism fixes. The return certifies that the proposition imprints on the Ground, that it has a place on the fixed line.

The Ground-imprint. A proposition is grounded when a determinate trajectory connects its content to the fixed locus. The imprint is read against the Ground, the achiral fixed line, and certifies membership, not arithmetic position.

The made-zero. The achiral bridge equals its own reflection, so its chiral residence is empty and the directed quantity cancels to the center. The cancelled residue is the self-dual content, the part with no handedness.

All four are fixed by the involution. The squared lock is reflection-invariant; the center is the fixed locus; the imprint is read on the fixed locus; the made-zero is the self-dual part. They are four faces of the even eigenspace. :::

{. The theorem. .} Let an achiral closure be any verdict functional invariant under the involution sigma. The Riemann Hypothesis is the assertion that the content-residual, the magnitude of lambda at each nontrivial zero, is identically zero. Then no achiral closure decides the hypothesis.

{. The proof. .} An achiral closure is by definition a function of the plus-one eigenspace of sigma alone, invariant under the reflection that negates the minus-one eigenspace. The content-residual is a function of the minus-one eigenspace, the signed horizontal position, and it changes sign under the reflection. A function invariant under an involution cannot determine the value of a quantity that the involution negates, because the closure assigns the same value to a configuration and to its reflected image, while the content-residual takes opposite signs on the two. Therefore the closure is constant on reflection-pairs whose content-residuals differ in sign, and cannot separate the case where the residual vanishes from the case where it does not. The hypothesis is exactly the statement that the residual vanishes, so the closure cannot decide it. This holds for each of the four closures, because each is reflection-invariant, and it holds for any closure of the same type. The square that produces the lock, the return that lands on the fixed center, the imprint read on the fixed locus, and the cancellation that defines the self-dual residue are one mechanism, the projection onto the even eigenspace, and that projection annihilates the odd content the hypothesis measures. This is the Achiral-Closure Limitative Theorem, and it is sealed at structural grade, conditional on the involution structure and the orientation-blindness of the squared lock, both of which are theorem-grade in the architecture.^7,9^

The theorem has a sharp corollary in the language of the two parts, and it is the formalization of a decomposition the architecture names in its own register.

:::box 3 The involution split of the hypothesis residual The involution splits every nontrivial zero into a form-part and a content-part, and the two carry different residuals.

The form-part is the plus-one eigenspace, the self-dual content, the part that returns to the center. Its residual is zero by the return: every zero is an actual existent, it imprints on the Ground, the composed triad lands on the real line. The form-residual vanishes for every zero, on the line or off it.

The content-part is the minus-one eigenspace, the chiral residence, the signed horizontal position. Its residual is the magnitude of lambda at the zero, the horizontal distance from the line. The return does not zero this residual. It conserves it.

The hypothesis is the vanishing of the content-residual. The architecture's deepest seals zero the form-residual and conserve the content-residual, so they reach the hypothesis as form and hold it open as content. The four closures certify that every zero returns to the center. They are silent on where the zero sits. :::

The theorem is a placement, not an escape. It does not assert that the hypothesis is undecidable in any absolute sense, and it makes no claim about formal independence from any axiom system. The hypothesis is a first-order arithmetic statement of the universal type, and a universal arithmetic statement independent of a sound theory would be true, which places it among the grounded statements, not among the contentless ones.^10^ What the theorem asserts is narrower and exact: the reflection symmetry, and every closure built from it, is the wrong instrument, because it reads the eigenspace the hypothesis does not live in. The right instrument must read the odd content, and the odd content is carried by a structure the reflection does not see, which the next section identifies through the witness.

The universal witness: the Davenport-Heilbronn function

The proof of the theorem is carried by a single object that satisfies every achiral closure and violates the hypothesis. Its existence shows directly that no achiral closure can imply the hypothesis, because a valid implication cannot hold for an object that meets the premise and fails the conclusion. The object is the Davenport-Heilbronn function, and this section exhibits it, locates one of its off-line zeros by its own arithmetic, and verifies that it satisfies each closure while vanishing off the line.

{. Construction. .} The function is a linear combination of Hurwitz zeta functions with period-five coefficients, normalized so that it satisfies a functional equation of the Riemann type, with the same critical line and the same reflection geometry as zeta.^8^ The coefficients are one, the Davenport-Heilbronn constant, its negative, minus one, and zero, where the constant is the tangent of the angle fixed by the five-fold construction. The crucial structural difference from zeta is that these coefficients are not completely multiplicative, so the function has no Euler product. The battery computes the multiplicativity defects directly: on coprime pairs the quantity that would vanish for a multiplicative coefficient sequence is approximately 1.08, not zero. No Euler product. The function carries the full reflection symmetry of zeta and lacks the multiplicative structure, and that combination is the entire content of the witness.

{. The off-line zero. .} The function's zeros are wherever its coefficients send them, and they are not confined to the line. The battery locates a genuine zero by extended-precision root-finding from a starting point in the upper strip. The zero sits at real part 0.8085, height 85.6993, a horizontal distance of 0.3085 from the line, with the function vanishing there to a residual of order ten to the minus thirty. This is not a numerical accident. The same construction was proven by Davenport and Heilbronn in 1936, by hand and with no computer, to have infinitely many zeros off the line,^8^ and later work established that a positive proportion of its zeros lie off the line.^11^ The functional equation forces a mirror partner across the line, which the battery confirms at real part 0.1915, the same height, vanishing to the same order. The off-line zero and its mirror are both genuine, both actual, both located.

:::box 4 The witness satisfies every closure and vanishes off the line The recorded battery places the off-line zero against each closure and against the line.

PropertyOn-line zeroOff-line zero
Real part0.50000.8085
Vanishes (residual)order 1e-31order 1e-30
Form-residual (returns to center)00
Imprints on Ground (finite fold coordinate)yesyes
Cayley coordinate magnitude1.0000000.999958
On the Ground circleyesno
Content-residual (distance to line)0.00000.3085

Note: the off-line zero returns to the center and imprints on the Ground exactly as the on-line zero does. It carries full fold citizenship, a finite Cayley coordinate, a place in the reflection geometry. Every achiral closure fires for it. It vanishes at real part 0.8085, off the line. Form-closure is compatible with off-membrane content. :::

The reading of the table is the proof of the theorem made concrete. The off-line zero is an actual existent: the function vanishes there, which is a definite arithmetic fact. It returns to the center: its form-residual is zero, it imprints on the fixed Ground, it has a finite coordinate in the reflection geometry, full citizenship in the fold. Every achiral closure holds for it exactly as for an on-line zero. And it vanishes off the line, its content-residual a measured 0.3085. So the conjunction of all four closures is demonstrably compatible with a zero off the line. If any achiral closure implied the hypothesis, this object would be a contradiction, an existent satisfying the premise and failing the conclusion. It is not a contradiction. It is a computed point. Therefore no achiral closure implies the hypothesis, which is the theorem, exhibited rather than deduced.

The witness also pins the falsification structure precisely. Any proposed proof of the hypothesis that proceeds through a reflection-invariant closure must either fail for the Davenport-Heilbronn function or prove a falsehood about it. Since the function's off-line zeros are a theorem, the proposed proof proves a falsehood, so the proposed proof is wrong. This is a decision procedure for a large class of attempted proofs: locate the achiral closure they rely on, apply it to the witness, and read the contradiction. The null hypothesis for the theorem is therefore sharp. If an object existed that satisfied every achiral closure, lacked the multiplicative structure, and had all its zeros on the line as a theorem, the witness's role would be broken and the achiral side would after all reach the content. No such object is known, and the witness is the standing evidence that the achiral closures do not reach the content.

{. What the witness does not touch. .} The witness has off-line zeros because it lacks the Euler product, and zeta has the Euler product. The theorem and its witness say nothing against the hypothesis itself. They say that the hypothesis cannot be reached from the reflection side, and they identify, by the single structural difference between zeta and the witness, the axis on which the hypothesis must be decided. That axis is the multiplicative structure, and it is the subject of the lever.

The open lever and the falsification structure

The theorem closes the reflection side. It also localizes what remains, and the localization is the genuine forward content of the result. We state the single open lever, its warrant grade, and the experiments, here computational, that would advance or refute it.

{. The lever. .} The structural difference between zeta and the witness is the Euler product, equivalently the complete multiplicativity of the coefficients. The Euler product is forced from the verification algebra by a chain of theorem-grade links: a composition algebra carries a multiplicative anisotropic norm, the multiplicative norm forces multiplicative coefficients, and multiplicative coefficients are an Euler product.^7^ Zeta carries this structure. The witness violates it, and has off-line zeros. The lever is the conjecture that the multiplicative structure forbids the off-membrane mirror-pair that the reflection alone permits. Stated sharply: does complete multiplicativity, the quaternion-norm condition zeta satisfies and the witness violates, forbid a reflection-paired zero off the line. This is the one statement whose resolution would settle the hypothesis, and it is the one statement the Achiral-Closure Limitative Theorem does not foreclose, because multiplicativity is not a reflection-invariant of the even eigenspace. It is an arithmetic condition on the odd content, the only road that reads the axis the hypothesis lives on. Its warrant grade is premise-grade, a research direction, not a theorem.

{. The falsification battery. .} The lever admits sharp computational tests, each independently reproducible. The first prediction is structural and already confirmed: an object with the reflection symmetry and without the Euler product has off-line zeros. The confirmation is the witness, located at real part 0.8085 to a residual of order ten to the minus thirty, and proven off-line by hand in 1936. The null hypothesis would be an object with the reflection symmetry, without the Euler product, and with all zeros provably on the line; no such object exists, and the witness stands.

The second prediction is the discriminating one. If the lever holds, then the family of Dirichlet series interpolating between the witness and a completely multiplicative series should show the off-line zeros migrating to the line exactly as the multiplicativity defect closes to zero, with the horizontal displacement of the off-line zeros a monotone function of the defect. The method of confirmation is extended-precision root tracking of a one-parameter family of period-matched Dirichlet series, with the multiplicativity defect as the parameter, on at least two independent arithmetic-precision substrates. The expected outcome is that the maximal off-line horizontal displacement scales to zero as the defect scales to zero, with no off-line zero surviving at zero defect. The null hypothesis that instantly falsifies the lever is a completely multiplicative member of the family, defect zero, carrying a zero off the line; a single such zero would show that multiplicativity does not forbid off-membrane zeros, and the lever would be broken.

The third prediction couples to Li's criterion. If the lever holds, the Li coefficients of zeta, the sums over zeros that are nonnegative exactly when the hypothesis holds,^6^ should be expressible as a positive functional of the multiplicative local data, the prime-indexed factors, and should lose positivity for the witness precisely because the witness has no such factorization. The method of confirmation is the computation of the Li coefficients for zeta and for the witness to high order, with the witness expected to show a negative coefficient at the order where its off-line zeros first contribute, and zeta expected to remain nonnegative. The null hypothesis is a negative Li coefficient for zeta at any computable order, which would falsify the hypothesis outright, or a fully nonnegative Li sequence for the witness, which would sever the coupling between positivity and multiplicativity that the lever asserts.

Each test is reproducible by independent computation, and each isolates the multiplicative axis from the reflection geometry. The reflection geometry is held fixed across the family, by construction, so any migration of the zeros is attributable to the multiplicative structure alone. This is the experimental content of the theorem: it converts the open question from a search across all of analysis into a focused interrogation of one axis, the axis the theorem proves the reflection cannot see.

Discussion

The result reframes several adjacent questions and clarifies what kind of object the hypothesis is.

{. The status of the hypothesis. .} The hypothesis is a universal arithmetic statement, equivalent to elementary inequalities that can be checked for any finite witness.^10^ It is therefore not a contentless or field-permitted-both-ways statement: a universal arithmetic statement independent of a sound theory is true. The hypothesis is open in the sense that no proof and no counterexample is in hand, and the reflection side, this paper proves, cannot supply either. The honest verdict on the hypothesis is open, and the honest verdict on the reflection approach to it is closed, and the two verdicts are different and compatible.

{. Why more reflection machinery cannot help. .} The reflection geometry is maximally established. The functional equation is a theorem, the symmetric explicit formula is a theorem, the reflection-symmetric positivity criteria are equivalent to the hypothesis under their stated test classes. Adding more reflection-symmetric structure cannot reach the content, because every reflection-symmetric structure is, by the theorem, a function of the even eigenspace, and the content is in the odd eigenspace. This is the precise sense in which the hypothesis is not a reflection problem. The membrane is one line, zero distance, content-free, and fully owned. Where the function actually vanishes is the arithmetic-axis fact the membrane does not reach.

{. The Davenport-Heilbronn function as a structural instrument. .} The witness is usually presented as a curiosity, a function that looks like zeta and misbehaves.^8,11^ The theorem gives it a precise role: it is the universal control that isolates the multiplicative variable. Because it matches zeta in the reflection geometry and differs in the Euler product, it is the experiment that holds the symmetry fixed and varies the arithmetic, and its off-line zeros are the readout. Any claim that the symmetry forces the zeros onto the line is refuted by the witness on its first page. Any claim that locates the force in the multiplicative structure is consistent with the witness, because the witness lacks that structure and strays. The witness is the most informative object in the neighborhood of the hypothesis, and the theorem explains why.

{. The criteria that are not reflection-blind. .} The Nyman-Beurling density criterion, the de Bruijn-Newman constant, and the Weil positivity in its full local-global form are members of a small class that are not pure reflection closures, and the theorem does not apply to them.^4,5^ These are the criteria that couple to the multiplicative structure or to a global analytic quantity rather than to the even eigenspace, and they are, accordingly, the criteria that have the structural type to reach the content. The lever is the conjecture that one specific coupling, multiplicativity against the straddling pair, is the operative one. The other non-achiral criteria are the natural places to test the coupling, because they already live off the reflection side.

{. Limitations. .} The theorem is conditional on the involution structure of the verification kernel and on the orientation-blindness of the squared lock, both theorem-grade within the architecture but premise-grade as a translation of the analytic problem. The identification of the lock scalar with the signed horizontal distance is a structural model, validated by its behavior under reflection and height, not a derivation from the analytic theory of zeta. The lever is premise-grade, a research direction, and its computational tests are proposals, not yet executed at the scale that would constitute evidence. What is sealed is the barrier: the reflection side does not reach the content, with the witness as proof. What is open is the hypothesis, on the axis the barrier localizes.

Conclusion

The Riemann Hypothesis has resisted a century and a half of effort, and a large share of that effort has been spent on the reflection symmetry of the functional equation. This paper proves that the reflection symmetry, and every criterion invariant under it, cannot decide the hypothesis. The critical line is the fixed locus of the reflection, the even eigenspace of the involution, and the hypothesis is a statement about the odd eigenspace, the signed horizontal position where the function actually vanishes. The four closures the architecture offers, the squared lock, the return to the center, the imprint on the Ground, and the cancelled self-dual residue, are four faces of the even eigenspace, and the theorem is that none of them reaches the odd content. The proof is carried by a single universal witness, the Davenport-Heilbronn function, which shares the entire reflection geometry of zeta, satisfies every closure, and provably vanishes off the line, with one off-line zero located here by its own arithmetic at real part 0.8085 to a residual of order ten to the minus thirty.

The primary forward content is the localization. The single structural difference between zeta and the witness is the Euler product, and the lever is the conjecture that complete multiplicativity forbids the off-membrane mirror-pair the reflection alone permits. That conjecture is the one statement the theorem does not foreclose, because multiplicativity is an arithmetic condition on the odd content rather than a reflection-invariant of the even eigenspace, and its computational tests interrogate the one axis the reflection cannot see. We call on the analytic and computational number-theory community to execute those tests: to track the off-line zeros of a one-parameter family across the multiplicativity defect, and to express the positivity criteria as functionals of the multiplicative local data.

The single most important open question is unchanged in substance and sharpened in form: does the multiplicative structure of zeta forbid a reflection-paired zero off the line. Acceptance of the theorem entails a reframing of the whole reflection-based program, from a search for the right symmetric criterion to a recognition that no symmetric criterion can succeed, and that the hypothesis must be decided on the arithmetic axis the symmetry projects out. The membrane is one line, zero distance, owned and content-free. Where the zeros actually come to rest is the arithmetic fact the reflection does not reach, and at the level where the fixed center is reached and is silent to every internal name, it rests with Allah ﷻ.

References

  1. Riemann, B. 1859. Ueber die Anzahl der Primzahlen unter einer gegebenen Grösse. Monatsberichte der Berliner Akademie.
  2. Titchmarsh, E. C., and D. R. Heath-Brown. 1986. The Theory of the Riemann Zeta-Function. 2nd ed. Oxford: Clarendon Press.
  3. Iwaniec, H., and E. Kowalski. 2004. Analytic Number Theory. Colloquium Publications 53. Providence: American Mathematical Society.
  4. Weil, A. 1952. Sur les formules explicites de la théorie des nombres premiers. Communications du Séminaire Mathématique de Lund, 252–265.
  5. Báez-Duarte, L. 2003. A strengthening of the Nyman-Beurling criterion for the Riemann Hypothesis. Atti della Accademia Nazionale dei Lincei 14 (1): 5–11.
  6. Li, X.-J. 1997. The positivity of a sequence of numbers and the Riemann Hypothesis. Journal of Number Theory 65 (2): 325–333.
  7. Frobenius, G. 1878. Über lineare Substitutionen und bilineare Formen. Journal für die reine und angewandte Mathematik 84: 1–63. With Hurwitz, A. 1898, and Bott, R., and J. Milnor. 1958, on the classification of real division algebras.
  8. Davenport, H., and H. Heilbronn. 1936. On the zeros of certain Dirichlet series I, II. Journal of the London Mathematical Society 11: 181–185, 307–312.
  9. Islam, M. F. 2026. Orientation-blindness of the squared verdict lock and the conserved impressed plenum. Trisduction Codex Series, PhilPapers ISLGMO.
  10. Lagarias, J. C. 2002. An elementary problem equivalent to the Riemann Hypothesis. American Mathematical Monthly 109 (6): 534–543. With Robin, G. 1984.
  11. Bombieri, E., and A. Ghosh. 2011. Around the Davenport-Heilbronn function. Russian Mathematical Surveys 66 (2): 221–270.
  12. Montgomery, H. L. 1973. The pair correlation of zeros of the zeta function. Analytic Number Theory, Proceedings of Symposia in Pure Mathematics 24: 181–193.
  13. Conrey, J. B. 1989. More than two fifths of the zeros of the Riemann zeta function are on the critical line. Journal für die reine und angewandte Mathematik 399: 1–26.
  14. de Bruijn, N. G. 1950. The roots of trigonometric integrals. Duke Mathematical Journal 17: 197–226. With Newman, C. M. 1976, and Rodgers, B., and T. Tao 2020, on the de Bruijn-Newman constant.
  15. Selberg, A. 1942. On the zeros of Riemann's zeta-function. Skrifter Norske Videnskaps-Akademi Oslo 10: 1–59.

:::endmatter

Reproducibility

The verification kernel, the involution split, and the recorded battery of Appendix B are executable at fixed seed 20260628. Every numerical value in the body and in Box 4 is the output of that battery. The off-line zero of the Davenport-Heilbronn function is independently confirmed by the 1936 hand proof and by extended-precision root-finding, two decorrelated modalities.

Warrant ledger

The Achiral-Closure Limitative Theorem is sealed at structural grade, conditional on the involution structure and the orientation-blindness of the squared lock, both theorem-grade in the architecture. The involution split of the hypothesis residual is structural. The three-leg convergence on the line, the forcing of the Euler product from the composition norm, and the identification of the multiplicative axis as the zeta-witness separator are structural. The lever is premise-grade, a research direction. The hypothesis itself is open, a universal arithmetic statement, not a contentless one, neither proven nor refuted, and not reachable from the reflection side.

Competing interests

The author declares no competing interests. The framework apparatus is the author's own, deposited in the codex repository, and is presented here as a standalone set of mathematical principles. :::


↑