Computing cohomology spaces of left-invariant involutive structures on $\mathrm{SU}(2)$: examples

Gabriel Ara\'ujo

REVIEW 4 major objections 3 minor 2 references

Computing cohomology spaces of left-invariant involutive structures on $\mathrm{SU}(2)$: examples

T0 review · 4 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read On $SU(2)$, the corank-1 left-invariant cohomology spaces are $H^{1,1}=0$ and $H^{0,1}\simeq\mathbb{C}$.

desk verdict The corank-1 cohomology theorem is false: the sign error in §5.2 is load-bearing, and even the printed maps make d'(1,0) non-injective at ℓ=1. read the letter →

arxiv 1908.10207 v1 pith:CRTGLXWY submitted 2019-08-25 math.DG

Gabriel Araújo This is my paper · ORCID

classification math.DG MSC 35R0335A0158J10

keywords left-invariantoperatorssolvabilitydifferentialcomplexeslocallyintegrablestructuresSU(2)smoothcohomologyCRstructurePeter-Weyldecomposition

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 studies left-invariant involutive structures—complex subbundles of the complexified tangent bundle closed under Lie bracket—on $SU(2)$, the simplest non-commutative compact Lie group. It establishes that the single complex vector field $\partial_-$ spanning the standard CR structure has closed range in the smooth topology, and that its associated first-order operator $d'$ does too. For the corank-1 structure spanned by $\partial_-$ and $\partial_0$, it derives explicit formulas for $d'$ on each irreducible representation block and concludes that the smooth cohomology spaces satisfy $H^{1,1}_V(SU(2);C^\infty)=0$ and $\dim H^{0,1}_V(SU(2);C^\infty)=1$. The concrete payoff is that an abstract cohomology theory becomes computable in a non-abelian example through representation matrix coefficients.

What carries the argument

The load-bearing machinery is the Peter-Weyl decomposition of $C^\infty(SU(2))$ into isotypic components $M_{T_\ell}$ indexed by the irreducible representations $T_\ell$, together with the standard formulas for the action of the invariant vector fields: $\partial_+ t^\ell_{mn}=-\sqrt{(\ell-n)(\ell+n+1)}\,t^\ell_{m,n+1}$, $\partial_- t^\ell_{mn}=-\sqrt{(\ell+n)(\ell-n+1)}\,t^\ell_{m,n-1}$, and $\partial_0 t^\ell_{mn}=n\,t^\ell_{mn}$. These formulas turn every differential $d'$ in the complex into a family of finite-dimensional linear maps on the blocks $M_{T_\ell}$, so kernel/range computations reduce to rank-nullity arguments; a theorem from the companion paper converts the resulting closed-range and exactness statements into statements about smooth cohomology on the whole group.

What would settle it

On the $\ell=1$ isotypic component, apply the paper's displayed formula $d'_{(1,0)}u=(-\partial_-u,-\partial_0u-u)$ to the matrix coefficient $t^1_{m,-1}$; the second component acts as $-(n+1)=0$, so the kernel is nontrivial. A direct rank-nullity calculation on this four-dimensional block therefore decides whether the exactness argument behind $H^{1,1}_V=0$ survives, and it fails as written.

Watch

Extended reading notes

Core claim

The paper's claim is that on $SU(2)$ the left-invariant CR vector field $\partial_-$ is globally almost hypoelliptic, equivalently $\partial_-:C^\infty(SU(2))\to C^\infty(SU(2))$ has closed range, and that for $v=\mathrm{span}\{\partial_-,\partial_0\}$ the smooth cohomology is $H^{1,1}_V(SU(2);C^\infty)=0$ with $\dim H^{0,1}_V(SU(2);C^\infty)=1$. The route is to decompose smooth functions by the Peter-Weyl theorem into finite-dimensional isotypic components $M_{T_\ell}$ spanned by the matrix coefficients $t^\ell_{mn}$, write down how $\partial_\pm$ and $\partial_0$ act on these coefficients, and reduce the differential complex on each block to a finite-dimensional linear algebra problem. Exactness on each block, combined with the closed-range property, is then lifted back to the smooth Fréchet cohomology.

Load-bearing premise

The corank-1 theorems rest on the sign of the commutator coefficient $b_{2,3}$ in Section 5.2: the displayed $d'$ maps take $b_{2,3}=-1$, i.e. $[\partial_0,\partial_+]=-\partial_+$, whereas the paper's bracket table gives $[\partial_0,\partial_+]=\partial_+$, so the proof of Theorem 5.1 depends on that sign convention and the exactness argument on the $\ell=1$ block is the fragile point.

Editorial extensions

If this is right

The closed-range result for $\partial_-$ implies that $\partial_-$ is globally almost hypoelliptic on $SU(2)$ in the quantitative sense of the earlier theory, so the equation $\partial_-u=f$ has solutions whose regularity is controlled by that of $f$.
The blockwise computation shows that smooth cohomology of left-invariant structures on compact Lie groups can be read off from finite-dimensional representation theory, without constructing explicit parametrices.
For the corank-1 structure $v=\mathrm{span}\{\partial_-,\partial_0\}$, the single surviving cohomology class $H^{0,1}_V\cong\mathbb{C}$ is carried by the trivial representation block, so the nonzero class is the natural 'constant' class visible already at the level of $\ell=0$.
Because the same vector-field action formulas exist for all compact Lie groups, the paper's examples provide a template for computing cohomology of other invariant structures, provided the analogous closed-range estimates hold.

Reading between the lines

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

The same representation-block method should extend to other compact Lie groups, where the highest-weight classification plays the role of the half-integer labels $\ell$; the main new work would be proving the closed-range lower bounds that the current paper gets from the elementary inequality $(\ell+x)(\ell-x+1)\ge 2\ell$.
One can test the method on the Heisenberg group or on tori with non-standard invariant structures, where the Peter-Weyl blocks are replaced by Fourier modes and the analogous $d'$ maps are constant-coefficient.
The closed-range estimate for $\partial_-$ suggests a general principle: a left-invariant vector field that lowers the representation index by a bounded amount and has no zero eigenvalue on orthogonal complements will be globally hypoelliptic; this could be formulated as a representation-theoretic criterion.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Short version: the computation for ∂− is fine and useful, but the corank-1 section is wrong in a way that kills the main theorem. I checked by hand, and the stress-test concern lands.

What is actually new and good: the closed-range proof for ∂−, the explicit kernel computation for d'(1,0), and Corollary 4.2 are concrete and, modulo the inequality issue, credible. The representation-theoretic setup is clean, and the abstract is honest that this is an application of [1].

Soft spots, in order of severity:

1. Section 5.2, b2,3. The paper's own bracket relations give [∂0, ∂+] = +∂+, so b2,3 should be +1. The printed −1 propagates into d'(1,0) and d'(1,1). With the correct sign, d'(1,1)(u1,u2) = ∂0u1 − ∂−u2, and the constant form (1,0) is d'-closed. It cannot be in the image of d'(1,0): if d'(1,0)f = (1,0), then ∂−f = −1, impossible because ∫ ∂−f = 0 on SU(2). So H^{1,1}_V ≠ 0, contradicting Theorem 5.1. This is not a typo; the claimed result is false.

2. Even using the printed maps, the proof of Theorem 5.1 asserts d'(1,0) is injective because n+1 is never zero for half-integers. For ℓ = 1, n = −1 occurs, and ∂−t_{m,−1} = 0, so d'(1,0) has a nontrivial kernel. The rank-nullity exactness argument collapses.

3. Proposition 4.1: the inequality √(2ℓ) ≥ C(1+ℓ(ℓ+1))^{1/3} is false for large ℓ (left ~ℓ^{1/2}, right ~ℓ^{2/3}). This is repairable with a smaller exponent, but as written the closed-range proof is incomplete.

4. Minor: in Theorem 5.2 the chosen φ = −(n+1)^{-1}t fails at n = −1; surjectivity can be fixed by another choice, so this is minor.

Verdict: the ∂− example may survive repair, but the corank-1 result, the main new computation, is false. This needs major revision and a different claim, not copyediting. If this is heading toward submission, I would not send it to a referee as is; I would tell the author to redo the corank-1 computation. The ∂− part alone might be worth a short note after the inequality is fixed, but the paper as a whole should not be relied on.

Referee Report

4 major / 3 minor

Summary. The manuscript applies the author's earlier abstract theory [1] to two left-invariant involutive structures on SU(2). For the corank-2 structure generated by ∂−, it proves (Prop. 4.1) that ∂− has closed range in C∞ and (Cor. 4.2) that the associated d′(1,0) has closed range. For the corank-1 structure v=span{∂−,∂0}, it computes the smooth cohomology, claiming H^{1,1}_V(SU(2);C∞)=0 (Thm. 5.1) and dim H^{0,1}_V(SU(2);C∞)=1 (Thm. 5.2). The computations are carried out by selecting complements and writing the differentials on matrix coefficients of the irreducible representations of SU(2).

Significance. The paper is a useful concrete exercise in applying the abstract framework of [1] to explicit left-invariant involutive structures on a compact Lie group. Its representation-theoretic setup is transparent, the coefficients are computed directly, and there are no hidden free parameters or data-fitting steps. The closed-range result for ∂− and the structure of the proof are plausible, and the paper is honest that the results are examples rather than deep new theory. However, the corank-1 computation is invalid as stated because of a sign error in §5.2, and the proof of Corollary 4.2 contains a non-sequitur. As a result, the advertised cohomology computation and one of the closed-range corollaries are not established.

major comments (4)

[§5.2, constants for v=span{∂−,∂0}] The coefficient b2,3 is set to −1, contradicting the bracket relations stated in §2.1. Since [∂+,∂0]=−∂+, antisymmetry gives [∂0,∂+]=∂+, and with L2=∂0 and M=∂+ one must have b2,3=1. Every displayed map in §5.2 and in the proof of Theorem 5.1 depends on this value: the correct maps are d′(1,0)u=(−∂−u,−∂0u+u) and d′(1,1)(u1,u2)=∂0u1−∂−u2, not the printed (−∂−u,−∂0u−u) and ∂0u1−∂−u2+2u1.
[§5.2 / Theorem 5.1] With the corrected sign, Theorem 5.1 is false. On the trivial representation MT0, d′(1,0)c=(0,c), so its image is {(0,c)}. The pair (1,0) lies in ker d′(1,1) because ∂0(1)=0 and ∂−(0)=0. If (1,0)=d′(1,0)f for some f∈C∞(SU(2)), then ∂−f=−1; integrating against Haar measure gives 0=−1, since ∫∂−f dμ=0 for every f. Hence H^{1,1}_V(SU(2);C∞) contains a nonzero class, contradicting the theorem. The spurious +2u1 term in the printed d′(1,1) is exactly what kills this class.
[§4.2, proof of Proposition 4.1] The displayed inequality √(2ℓ) ≥ C(1+ℓ(ℓ+1))^{1/3} is false for large ℓ: the left side grows like ℓ^{1/2} and the right side like ℓ^{2/3}. Therefore the cited condition from [1, eqn. (6.1)] with s=1/3 is not verified, and the closed-range conclusion does not follow from the argument as written. The claim may be repairable with a smaller exponent (for example 1/4), but the estimate must be corrected.
[§4.2, proof of Corollary 4.2] The inference “since ∂− has closed range, there exists u2 such that u2,ν→u2” is invalid. Closed range only gives a lift u2 with ∂−u2 equal to the limit of ∂−u2,ν; it does not make the original sequence converge. The proof therefore does not establish closed range of d′(1,0). A correct proof would need a uniform estimate on a complement or an explicit Peter-Weyl argument.

minor comments (3)

[Theorem 5.1 statement] The statement is missing a closing parenthesis: it should read H^{1,1}_V(SU(2);C∞(SU(2))).
[§5.2 / Theorem 5.2] In the proof of surjectivity of d′(0,1), the choice φ=−(n+1)^{-1}tℓ_{mn} is undefined when n=−1 (for example, ℓ=1). The surjectivity is still true because the ∂− term from the second component can produce t_{m,−1}, but the proof should be adjusted.
[References] Reference [1] is cited as “Ann. Glob. Anal. Geom., 2019” without volume or article number; please supply full bibliographic data if the paper has been published.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the examples are independent applications of the author's prior general theory, not reductions of the conclusions to their inputs.

full rationale

The paper explicitly frames itself as applications of the theory in [1] and, on inspection, the derivation chain is not circular. Proposition 4.1 verifies a concrete symbol inequality on the matrix-coefficient spaces and then invokes [1, Theorem 6.4(1) and 6.7(1)] as a general closed-range criterion; Corollary 4.2 reduces closed range of the two-component operator to Proposition 4.1. In Section 5, the cohomology of the corank-1 structure is computed by deriving the d' maps from the bracket coefficients a_j, b_{j,k} via the general formulas of Section 5.1, then checking injectivity, surjectivity, and exactness on each finite-dimensional representation space, and finally applying [1, Theorem 7.2] to identify smooth cohomology with the direct sum of finite-dimensional cohomologies. That cited theorem is a general structural result and does not by itself assert H^{1,1}=0 or dim H^{0,1}=1, so the specific cohomology values come from the paper's own computations, not from the citation. There are no fitted parameters presented as predictions, no quantity is defined in terms of the result it is used to prove, and no known result is merely renamed. The manuscript contains apparent mathematical defects—for example, the asserted b_{2,3}=-1 in Section 5.2 contradicts the bracket relation [∂+,∂0]=-∂+ recorded in Section 2.1, and the inequality sqrt(2ℓ) ≥ C(1+ℓ(ℓ+1))^{1/3} is false for large ℓ—but these are correctness issues, not circularity. Because the central claims are not equivalent by construction to any input or self-citation, the circularity score is 0.

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

No free parameters are fitted to data and no new physical or geometric entities are introduced. The paper relies on standard Peter-Weyl theory, on explicit matrix-coefficient formulas from [2], and on a heavy black-box use of the author's prior theorems in [1]. The fragile point is the sign computation in the corank-1 example, which is an ad hoc computational assumption that is currently wrong as stated.

assumptions (4)

standard math Peter-Weyl decomposition and orthonormality of matrix coefficients for SU(2), including the formula (2.4) expanding smooth functions in matrix coefficients.
Used throughout to reduce the analysis to finite-dimensional irreducible components; the coefficients are borrowed from [2, Chapter 11].
standard math Classification of irreducible unitary representations of SU(2): every irreducible representation is equivalent to T_ℓ for some ℓ ∈ 1/2 Z+, and these exhaust the unitary dual.
Section 2.2; needed for the decomposition of functions and for index ranges in the computation of kernels.
domain assumption Theorems [1, Thm 6.4(1), Thm 6.7(1), Thm 7.2]: a subelliptic estimate with the stated loss implies almost C∞ global hypoellipticity, which is equivalent to closed range; and closed range together with per-irreducible exactness computes smooth cohomology.
This is the load-bearing black box. The paper invokes these theorems without reproving them, and if their hypotheses are not met, the closed-range and cohomology conclusions do not follow.
ad hoc to paper The chosen complement M = ∂+ and the dual-basis computation in Section 5.1 correctly encode the operator d′ for v = span{∂−, ∂0}.
Section 5.2; this is where the sign error enters. The stated constant b2,3 = −1 is inconsistent with the paper's own commutator relations, so the displayed d′ maps computed from this axiom are unreliable.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Computing cohomology spaces of left-invariant involutive structures on $\mathrm{SU}(2)$: examples." pith.science (2026). https://pith.science/paper/CRTGLXWY

@misc{pith2026190810207,
  author       = {Pith},
  title        = {Pith review of: Computing cohomology spaces of left-invariant involutive structures on $\mathrmSU(2)$: examples},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CRTGLXWY}},
  note         = {Machine review of arXiv:1908.10207}
}

abstract

In these notes we study left-invariant involutive structures on $\mathrm{SU}(2)$, the most na\"ive non-commutative compact Lie group. We determine closedness of the range (in the smooth topology) of a single complex vector field spanning the standard CR structure of $\mathrm{SU}(2)$ and also compute the smooth cohomology spaces of a corank $1$ structure. In our approach, it is fundamental to understand concretely the irreducible representations of the ambient Lie group and how left-invariant vector fields operate on their matrix coefficients (which we borrow from the book of Ruzhansky and Turunen (2010)). Our purpose is solely to provide some easy applications of the theory developed in a previous paper (2019), as the results shown here can probably be obtained by more direct methods.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages

[1]

Ara´ ujo

G. Ara´ ujo. Global regularity and solvability of left-invariant diﬀerential systems on compact Lie groups. Ann. Glob. Anal. Geom., 2019

work page 2019
[2]

Ruzhansky and V

M. Ruzhansky and V. Turunen. Pseudo-diﬀerential operators and symmetries , volume 2 of Pseudo-Diﬀerential Opera- tors. Theory and Applications . Birkh¨ auser Verlag, Basel, 2010. Background analysis and advanced topics. University of S ˜ao Paulo, ICMC-USP, S ˜ao Carlos, SP, Brazil E-mail address: gccsa@icmc.usp.br

work page 2010

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Ara´ ujo

G. Ara´ ujo. Global regularity and solvability of left-invariant diﬀerential systems on compact Lie groups. Ann. Glob. Anal. Geom., 2019

work page 2019

[2] [2]

Ruzhansky and V

M. Ruzhansky and V. Turunen. Pseudo-diﬀerential operators and symmetries , volume 2 of Pseudo-Diﬀerential Opera- tors. Theory and Applications . Birkh¨ auser Verlag, Basel, 2010. Background analysis and advanced topics. University of S ˜ao Paulo, ICMC-USP, S ˜ao Carlos, SP, Brazil E-mail address: gccsa@icmc.usp.br

work page 2010