Pith. sign in

REVIEW 4 major objections 4 minor 4 references

The crossmetric tensor and the geometrical meaning of the imaginary numbers

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper argues that quaternion imaginary units are not abstract symbols but ordered pairs of mirror planes, and that Hamilton's rules follow from matching those planes head-to-tail.

desk verdict The crossmetric tensor is a handy algebraic tool for crystallographic quaternion products, but the claimed geometric derivation of Hamilton's rules rests on a false equality and should be reframed as a special orthogonal-case interpretation. read the letter →

arxiv 2608.05113 v1 pith:QFGTANUO submitted 2026-08-05 cond-mat.mtrl-sci math.MG

Cyril Cayron This is my paper · ORCID
classification cond-mat.mtrl-scimath.MG MSC 11R5215A6620L05
keywords QuaternionsRotationsCrossmetrictensorImaginarynumbersCrystallographicOrientedplanesGroupoidcompositionMetric
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to give the imaginary units of quaternions a concrete geometric meaning rather than treating them as algebraic curiosities. It shows that the product of two crystallographic quaternions can be written as a single quadratic form using a $4\times4$ "crossmetric tensor" built from the metric and cross tensors of a lattice. The paper then claims that every unit quaternion is geometrically a whole family of ordered pairs of oriented planes: two mirrors that meet along the rotation axis at half the rotation angle. Composing quaternions is the head-to-tail rule for these plane pairs, and Hamilton's $\mathbf{i}, \mathbf{j}, \mathbf{k}$ are just the pairs of perpendicular faces of a cube. If true, the mystery of $\mathbf{i}^2=-1$ dissolves into the double reflection across two perpendicular planes.

What carries the argument

The load-bearing object is the crossmetric tensor $\boldsymbol{\mathcal{Q}}_c$, the $4\times4$ symbolic matrix whose entries are the symbols $\mathbf{s},\mathbf{a},\mathbf{b},\mathbf{c}$ and which rewrites $q_1 q_2$ as a quadratic form $q_1^t \boldsymbol{\mathcal{Q}}_c q_2$. It is assembled from the metric tensor $\boldsymbol{\mathcal{M}}$ and the cross-product matrices $\boldsymbol{\Omega}_1,\boldsymbol{\Omega}_2,\boldsymbol{\Omega}_3$ as $\boldsymbol{\mathcal{K}}=-\boldsymbol{\mathcal{M}}\,\mathbf{s} + \boldsymbol{\Omega}_1\mathbf{a}+\boldsymbol{\Omega}_2\mathbf{b}+\boldsymbol{\Omega}_3\mathbf{c}$, so it packages scalar and cross multiplication in one table for any lattice. The companion machinery is the pair-of-planes representation, which turns that table's entries into geometric statements about intersecting mirror planes.

What would settle it

Take two unit quaternions $q$ and $r$. Since each has infinitely many equivalent plane pairs, pick two different equivalent pairs for $q$ and two for $r$, enforce exact head-tail coincidence in each combination, and compare the resulting rotation axis and angle with the usual product $qr$ computed from the crossmetric tensor. If different choices of equivalent pairs give different products, or if any choice disagrees with $qr$, the geometric rule is not well-defined; a minimal test case is $q=\mathbf{i}$ and $r=\mathbf{j}$ with the paper's representatives $\mathbf{i}=(\boldsymbol{m}_y,-\boldsymbol{m}_z)$ and $\mathbf{j}=(\boldsymbol{m}_z,-\boldsymbol{m}_x)$, which forces a re-representation before the head-tail rule can be applied.

Watch

Extended reading notes

Core claim

The paper's central discovery is that a unit quaternion $q = \cos(\alpha/2)\,\mathbf{s} + \sin(\alpha/2)\,\tilde{\mathbf{u}}$ can be read as the whole family of ordered pairs of oriented planes $(\boldsymbol{m}_1,\boldsymbol{m}_2)$ that intersect along $\mathbf{u}$ with dihedral angle $\alpha/2$; the two planes are the mirrors whose double reflection is the rotation represented by $q$. Multiplication of quaternions is then the source-target rule $(\boldsymbol{m}_1,\boldsymbol{m}_2)(\boldsymbol{m}_2,\boldsymbol{m}_3)=(\boldsymbol{m}_1,\boldsymbol{m}_3)$. On this reading the elementary Cartesian quaternions are $\mathbf{i}=(\boldsymbol{m}_y,-\boldsymbol{m}_z)$, $\mathbf{j}=(\boldsymbol{m}_z,-\boldsymbol{m}_x)$, $\mathbf{k}=(\boldsymbol{m}_x,-\boldsymbol{m}_y)$ for the faces of a cube, and because those pairs are simultaneously elementary and complementary, the six symbols needed for a general lattice collapse to the three Cartesian units, reproducing all of Hamilton's rules at once.

Load-bearing premise

The argument rests on the unproved postulate that quaternion multiplication is exactly the head-to-tail composition of oriented plane pairs, with the specific representatives chosen for $\mathbf{i},\mathbf{j},\mathbf{k}$; if that identification or those sign conventions fail, the geometric explanation of Hamilton's rules does not go through.

Editorial extensions

If this is right

  • Hamilton's rules $\mathbf{i}^2=\mathbf{j}^2=\mathbf{k}^2=-1$ and $\mathbf{i}\mathbf{j}=\mathbf{k}$ become corollaries of composing cube-face plane pairs head-to-tail, not separate axioms.
  • For a general crystal lattice the composition table needs six elementary symbols ($\mathbf{a},\mathbf{b},\mathbf{c}$ and their complements), and the crossmetric tensor gives the explicit table for each of the six crystal families from the metric tensor alone.
  • The 2D imaginary unit $\mathbf{i}$ should be understood as a pair of perpendicular oriented lines whose associated rotation is $180^\circ$, which removes the apparent factor-of-two mismatch between complex numbers and quaternion half-angles.
  • The non-uniqueness of the plane pairs makes quaternion composition a groupoid law with two units, $\mathbf{s}$ and $-\mathbf{s}$, so $q$ and $-q$ are geometrically distinct objects representing the same rotation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the pair-of-planes reading is right, spinor sign conventions and the $720^\circ$ rotation property have a geometric origin: the ordered pair records a choice of mirror sequence, and reversing or re-pairing the mirrors changes the sign of the quaternion without changing the rotation.
  • A testable extension would be to use the crossmetric tensor to define Clifford-algebra products directly in non-orthogonal crystallographic bases; the paper notes this connection but does not develop it.
  • One could probe the claim experimentally by comparing the groupoid composition law for grain-boundary misorientations with quaternion multiplication in the same lattice; agreement would tie the pair-of-planes picture to physical crystallography, and disagreement would localize where the geometric rule needs refinement.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper generalizes Hamilton's 4×4 symbolic multiplication table for quaternions to crystallographic bases. For a product q1 q2 written as a quadratic form, the author defines a 'crossmetric tensor' built from the metric tensor and a cross-product matrix, tabulates it for the six crystal families, and proposes a geometrical interpretation: a unit quaternion is represented by pairs of oriented mirror planes intersecting along its vector part, and quaternion composition is the source-target (head-tail) composition rule. The paper then claims that Hamilton's i, j, k are obtained as pairs of perpendicular faces of the cube, and that this explains why the square and tri-product rules i²=j²=k²=ijk=-1 hold with only three imaginary units.

Significance. The algebraic rewriting of the crystallographic quaternion product into a symbolic quadratic form is a useful and largely correct bookkeeping device, and the explicit tables for the six crystal families may be convenient for crystallographic computations. The proposed geometrical interpretation is attractive: if correct, it would give a genuinely elementary picture of quaternion units and of the special role of the Cartesian case. The paper deserves credit for presenting the crossmetric construction, for tabulating concrete formulas, and for identifying the cube-face geometry of i, j, k. However, the central geometric derivation is not established: the key algebraic equality on which it rests is false for non-perpendicular planes, so the general claims about 'any unit crystallographic quaternion' and about the geometrical origin of Hamilton's rules are not supported as stated.

major comments (4)
  1. [4.2, Eq. (20)] The equality (m1,m2)=(m2,-m1), and the related equality (m2,m1)=-(m1,m2), are asserted for arbitrary oriented planes. For a pair (m1,m2) whose planes form an angle θ and whose axis is u=m1×m2, the paper's own dictionary gives q=cosθ+sinθ u. The pair (m2,-m1) has angle π-θ between its planes, so it represents q'=cos(π-θ)+sin(π-θ)u=-cosθ+sinθ u, which is neither q nor -q unless cosθ=0, i.e. unless the planes are perpendicular. Equation (20) is therefore not a valid algebraic rule for general plane pairs; it is valid only in the orthogonal half-turn case that happens to be exactly the case of the Cartesian i, j, k.
  2. [4.4, Eq. (23) and following] The derivation of i², ij and ji from the head-tail rule silently uses the invalid identity (20). With i=(my,-mz) and j=(mz,-mx), the target plane of i is -mz while the source plane of j is mz, so the representatives do not satisfy the middle-plane matching condition required by (22). The step that replaces the product by (my,mx) amounts to an unstated re-representation of one factor, and that re-representation relies on Eq. (20). The calculation is therefore not a rigorous derivation of Hamilton's rules from geometry; it works only because the cube faces involved are pairwise perpendicular.
  3. [4.7, first paragraph and Figure 4] The 'geometrical proof' of the components of the crossmetric tensor is circular. The tensor components in Appendix 1 were already fixed by the algebraic product defined in Eqs. (4)-(6), and the plane-pair representation together with the source-target rule was introduced so as to match that product. Recovering a stored component from a deliberately matched geometric arrangement confirms internal consistency but does not independently derive the tensor. In addition, the hexagonal example positions planes using the same unrestricted equality (20), so the geometric derivation inherits the non-perpendicular-plane error.
  4. [4.2, paragraph before Eq. (20)] The statement that there is 'an infinity of pairs of planes defining the same quaternion' is true only if equivalent pairs are related by a common rotation about the axis u, which preserves the angle between the planes. The larger equivalence expressed in Eq. (20), which changes the angle from θ to π-θ, is not an equivalence of the represented quaternion. The equivalence class must be defined by angle-preserving rotations, not by the transformations in Eq. (20), if the geometric representation is to be coherent for non-orthogonal pairs.
minor comments (4)
  1. [1.2] Twice in the discussion of complex numbers the text states 'a=cos(α) and b=cos(α)' (and later 'a=cos(α/2) and b=cos(α/2)'); the second coefficient should be sin(α) and sin(α/2), respectively.
  2. [3, paragraph after Eq. (17)] The sentence defining q2 uses the undefined symbol d ('x2 b + y2 c + z2 d'); this appears to be a typographical error and should be corrected.
  3. [Appendix 1, hexagonal table] The hexagonal entry for the crossmetric tensor contains expressions such as a²(√3 c + c s)/(2c), where the same letter c is used for the lattice parameter and for the symbolic quaternion axis c; the notation should be disambiguated.
  4. [Acknowledgements and Note] The historical and polemical remarks about Hamilton, Gibbs, and Heaviside, while colorful, are not directly relevant and could be shortened; the reader's attention should remain on the mathematical claims.

Circularity Check

1 steps flagged · score 6.0 of 10

The geometric 'meaning' of the elementary quaternions is reverse-engineered: Section 4.6 chooses the plane-pair representation to satisfy the algebraic square rule and then presents that rule as a geometric consequence.

  1. fitted input called prediction [Section 4.6, paragraph defining the elementary crystallographic quaternions (equations (24) and the sentence after them)]
    "Initially we thought that the elementary crystallographic quaternions could be defined as pairs of planes 𝒎𝒂 = (𝐛, 𝐜), 𝒎𝒃 = (𝐜, 𝐛), 𝒎𝒄 = (𝐚, 𝐛), but we realized that this definition would disagree with the composition rule (18). We thus came to conclude that the unit crystallographic quaternions 𝐚/𝑎, 𝐛/𝑏 and 𝐜/𝑐 are actually the pairs of floating planes 𝐚̃ = (𝒎𝒂−, 𝒎𝒂+) such that the angle between 𝒎𝒂− and 𝒎𝒂+ is +90°."

    Rule (18), a~^2 = b~^2 = c~^2 = -s, was already obtained from the crossmetric tensor, which is itself built from the quaternion product rules (5)-(6). The paper explicitly rejects one geometric representation because it disagrees with (18) and then adopts a representation chosen to satisfy (18). Consequently, the subsequent statement in the same section that (18) 'also results from the geometrical rules (20)' is not an independent derivation: the geometric definition of the elementary quaternions was fitted to reproduce the target algebraic rule, so the 'result' is enforced by construction. The same fitted representatives are then used in Section 4.7 to 'prove geometrically' components of the crossmetric tensor, compounding the circularity.

full rationale

The paper contains one genuine circular step that affects its central interpretive claim. In Section 4.6 the elementary crystallographic quaternions are not derived from the mirror-pair geometry; they are selected to reproduce the already-known algebraic square rule (18), which came from the crossmetric tensor built on the quaternion product. The sentence 'we realized that this definition would disagree with the composition rule (18)' shows the geometric representation was fitted to the algebra, and the later statement that (18) 'also results from the geometrical rules' is therefore a re-derivation of an input. This is a fitted-input-called-prediction pattern. That said, the broader geometric representation of an arbitrary unit quaternion as a pair of mirror planes is an independent and standard fact (two reflections compose to a rotation), and the head-tail composition rule is a genuine geometric law, not an ad hoc assumption. The crossmetric tensor itself is an algebraic repackaging, and the paper explicitly concedes its components were already published by Katrusiak and Le (2024), with the novelty lying in the interpretation. Thus the circularity is partial: the specific 'geometrical meaning' assigned to the elementary quaternions is reverse-engineered from the algebra, while the overall quaternion-to-mirror-pair correspondence retains independent content. Score 6 reflects that one or more of the claimed geometric 'results' reduce by construction, without the entire derivation being vacuous.

Assumptions & free parameters 0 free parameters · 4 assumptions · 2 invented entities

The paper rests on prior quaternion product rules (self-cited) and on the unproven pair-of-planes interpretation. The crossmetric tensor is a reformulation of known algebra. No free parameters are fitted; the lattice parameters are inputs. The invented entities are mathematical interpretations, not physical objects.

assumptions (4)
  • domain assumption The product of two crystallographic quaternions is given by equations (4)-(6) from the author's prior work (Cayron, 2026), including the cross matrix X = sqrt(det M) M^{-1}.
    Sections 1.1 and 2 rely on this prior result without proof. If this product rule is flawed, the crossmetric tensor is invalid.
  • ad hoc to paper A unit quaternion q = cos(alpha/2) + sin(alpha/2) u~ can be represented by an equivalence class of pairs of oriented planes (m1,m2) intersecting along u with angle alpha/2.
    Section 4.2 introduces this as a 'key point' without derivation. It is the central interpretive leap.
  • ad hoc to paper The composition of two quaternions is the source-target rule (m1,m2)(m2,m3) = (m1,m3).
    Section 4.3. This rule is asserted and used to derive Hamilton rules; it is not proven equivalent to quaternion multiplication.
  • ad hoc to paper The equivalence relations on plane pairs given in eq. (20), e.g., (m1,m2) = (m2,-m1), hold for oriented planes.
    Section 4.2. These equivalence relations are geometric postulates.
invented entities (2)
  • Pair of oriented planes (m1,m2) as a representation of a quaternion
    purpose: To provide a geometrical meaning for imaginary numbers and to derive Hamilton's rules.
    This is an interpretive model. It does not make falsifiable predictions beyond reproducing known quaternion algebra.
  • Complementary quaternions a', b', c'
    purpose: To complete Hamilton-like rules for non-Cartesian bases where elementary quaternions only satisfy square rules.
    Defined in Section 4.6 as plane pairs (mb,mc), etc. They are new algebraic objects but carry no independent physical evidence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The crossmetric tensor and the geometrical meaning of the imaginary numbers." pith.science (2026). https://pith.science/paper/QFGTANUO

@misc{pith2026260805113,
  author       = {Pith},
  title        = {Pith review of: The crossmetric tensor and the geometrical meaning of the imaginary numbers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QFGTANUO}},
  note         = {Machine review of arXiv:2608.05113}
}
read the original abstract

The product of two Cartesian quaternions can be written as a quadratic form based on a 4x4 matrix made of the symbols 1,i,j,k where i,j,k are imaginary numbers introduced by Hamilton. We generalized this matrix to non-Cartesian bases, and showed that the matrix is made of the metric and the cross tensors. We called it crossmetric tensor. Its symbols are s,a,b,c; they are the elementary crystallographic quaternions. We determined the crossmetric tensors for the six crystal families. We also showed that any unit crystallographic quaternion can be geometrically represented by an infinity of pairs of oriented planes intersecting along the vectorial component of the quaternion such that the angle between them is the semiangle of the rotation. The composition of quaternions follows the intuitive source-target rule. The elementary quaternions a,b,c are geometrically represented by pair of perpendicular and oriented planes intersecting along the axis a,b,c, respectively. Other complementary quaternions noted a',b',c' were also introduced. They are the pairs of planes ma, mb, mc. The elementary quaternions a,b,c follow Hamilton rules on the squares of imaginary numbers. The complementary quaternions follow Hamilton rules on the bi and tri-products. For Cartesian basis, the quaternions i,j,k appear as a specific case of crystallographic quaternions. They are formed by the pairs of perpendicular faces of the cube mx, my, mz. Since these planes intersect along the x, y and z axis, respectively, the elementary and complementary quaternions are equal, i.e. i=i',j=j',k=k', which explains why the square, bi and tri-product Hamilton rules are satisfied all together with only three quaternions i,j,k.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

4 extracted references · 4 canonical work pages

  1. [1]

    Cayron, C. (2006). Acta Crystallogr A Found Crystallogr 62, 21–40

    work page 2006

  2. [2]

    Cayron, C. (2026). https://doi.org/10.48550/ARXIV.2607.16899. Clifford (1871). Proceedings of the London Mathematical Society s1-4, 381–395

    work page Pith review arXiv doi:10.48550/arxiv.2607.16899 2026

  3. [3]

    Imaeda, K. (1995). Vol. Clifford Algebras and Spinor Structures, edited by R. Ablamowicz & P. Lounesto. pp. 265–280. Dordrecht: Springer Netherlands

    work page 1995

  4. [4]

    Katrusiak, A. & Le, H. Q. (2024). Symmetry 16, 1366

    work page 2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.