Finite-term recurrences in a generalized Bochner--Krall family

B. Shapiro; D. Barrios Rolan\'ia; L.M. Anguas; M. Tater

Finite-term recurrences in a generalized Bochner--Krall family

T0 review · 0 major / 5 minor · reviewed 2026-08-27 · deepseek-v4-flash

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

Pith's one-line read The monic eigenpolynomials of $T=z^j\partial_z^j+z^m\partial_z^\ell$ satisfy a fixed-term recurrence exactly when $j=1$ and $\ell-m$ divides $\ell$, with all coefficients given in closed form.

desk verdict Correct, complete classification of a natural three-parameter Bochner–Krall family; the proof is checkable and the result is new, though the scope is deliberately narrow. read the letter →

arxiv 2608.21802 v1 pith:CB6VFLNZ submitted 2026-08-22 math-ph math.CAmath.MP

L.M. Anguas , D. Barrios Rolanía , B. Shapiro , M. Tater This is my paper · ORCID

classification math-phmath.CAmath.MP MSC 42C0534A3033C4547E05

keywords generalizedBochner–Krallproblemexactlysolvableoperatorpolynomialeigenfunctionfinite-termrecurrenced-orthogonalitydifferenceorderresidue-classdecompositionChu–Vandermondesummation

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

The paper studies the differential operators $T=z^j\partial_z^j+z^m\partial_z^\ell$ with $0\le m<\ell$ and $1\le j<\ell$, acting on polynomials, and asks when their unique monic eigenpolynomials satisfy a recurrence with a fixed number of terms. The answer is a complete classification: such a recurrence exists exactly when $j=1$ and $k=\ell-m$ divides $\ell$. In that case all recurrence coefficients are given in closed form, and the associated difference operator has order $\ell$. This confirms, for this family, the order equality predicted by a general conjecture in the higher-order Bochner–Krall problem, and it places each admissible operator inside the conjectured Type (2) family. A concrete new case, $T=z\partial_z+z^2\partial_z^4$, shows that symmetry alone does not force a one-lower-term recurrence.

What carries the argument

The load-bearing structure is the decomposition of the polynomial ring into residue-class subspaces $V_r=\operatorname{span}\{z^{r+ik}:i\ge0\}$, $0\le r<k$, on which $T$ acts bidiagonally: $Te_i=\Lambda_i e_i+M_i e_{i-1}$ with $\Lambda_i=(r+ik)_j$ and $M_i=(r+ik)_\ell$. On each $V_r$, the coordinate functionals $\psi_R$ of the eigenpolynomial basis satisfy an explicit closed form (Lemma 3.2), turning recurrence coefficients into finite sums. For $j=1$, Lemma 5.1 evaluates the relevant functionals, and a terminating Gauss hypergeometric series summed by Chu–Vandermonde collapses the finite sum to the product formula (23); the factor $\prod_{a=0}^{\tau-1}(\ell-ak)$ then forces $k\mid\ell$ for finiteness. For $j\ge2$, the same bidiagonal model produces a growth contradiction: one side of an identity tends to $-\infty$ while the other stays bounded, because the right-hand side converges when $j\ge2$.

What would settle it

Take $T=z^2\partial_z^2+z\partial_z^3$ (so $j=2$, $\ell=3$, $m=1$, $k=2$) and compute the normalized monic eigenpolynomials from the coefficient recurrence (10)–(12). The theorem predicts that no fixed $N$ bounds all lower diagonals; finding any fixed $N$ for which all coefficients $\alpha_{n,R}$ with $n-R\ge N$ vanish would contradict Theorem 1.1.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is Theorem 1.1: for $0\le m<\ell$ and $1\le j<\ell$, the normalized monic eigenpolynomials $P_n$ of $T=z^j\partial_z^j+z^m\partial_z^\ell$ have finite lower bandwidth if and only if $j=1$ and $k=\ell-m$ divides $\ell$. Under these conditions the recurrence takes the form $$zP_n=P_{n+1}+\sum_{q=1}^{\$\sigma$}\$gamma_n^{{(q)}}$P_{n+1-qk},\qquad \$\sigma$=\ell/k,$$ with $$\$gamma_n^{{(q)}}$=\frac{(-1)^q}{q!k^q}\left(\prod_{a=0}^{q-1}(\ell-ak)\right)\left(\prod_{c=0}^{q-1}(n-ck)_{\ell-1}\right),$$ and the optimal lower bandwidth is $N=\ell$, independently of $m$. The proof also shows that every admissible operator can be factorized as $T=z\partial_z+q'(G)G$, with $G=p(D)\partial_z$, $D=z\partial_z$, $p(t)=\prod_{r=1}^{\sigma-1}(t-(rk-1))$, and $q(t)=t^k/k$, so each admissible operator is of Type (2) in Conjecture 1.10 of [5].

Load-bearing premise

The classification depends on the recurrence having a fixed number of terms $N$ that does not grow with $n$; if the bandwidth were allowed to depend on $n$, the proof that $j\ge2$ is impossible would no longer go through.

Editorial extensions

If this is right

For every admissible operator with $j=1$ and $k\mid\ell$, the monic eigenpolynomials obey the explicit $(\sigma+1)$-term recurrence, so the coefficients can be read off directly from $\ell$, $k$, and $n$.
The optimal lower bandwidth is always $N=\ell$, independent of $m$; in the extreme cases $m=0$ and $m=\ell-1$ this gives, respectively, a sparse $\ell$-step recurrence and a full $(\ell+1)$-term recurrence.
For composite $\ell$, intermediate values of $m$ give genuinely new cases, the smallest being $T=z\partial_z+z^2\partial_z^4$, whose recurrence has two lower diagonals rather than the single one that symmetry alone might suggest.
Every admissible operator admits the factorization $T=z\partial_z+q'(G)G$ with $G=p(D)\partial_z$, placing it inside the Type (2) family and giving an instance where differential and difference orders both equal $\ell$.
If $\ell$ is prime, the only admissible choices are $m=0$ and $m=\ell-1$, so the classification is degenerate for prime differential order.

Reading between the lines

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

The paper's proof leaves open the possibility that allowing the number of recurrence terms to grow with $n$ could admit new cases for $j\ge2$; the necessity proof relies on one fixed bandwidth $N$, so relaxing uniformity deserves separate study.
The explicit coefficients in formula (6) are products of falling factorials in $n$, which makes them natural candidates for known $d$-orthogonal moment sequences; checking this for small $\ell$ would connect the classification to the existing theory of $d$-orthogonal polynomials.
The factorization recipe suggests a concrete probe of sharpness: replace the fixed polynomial $p(t)=\prod_{r=1}^{\sigma-1}(t-(rk-1))$ by another polynomial of the same degree and test whether the resulting $T=z\partial_z+q'(G)G$ still has a fixed-term recurrence.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Request a human review

A listed scientist reviews the paper for a fee and the review publishes here regardless of verdict. See the reviewers or get listed.

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

Two things you should know. First, the main theorem is real: for T = z^j ∂^j + z^m ∂^ℓ the normalized monic eigenpolynomials have a fixed-term recurrence iff j=1 and k=ℓ−m divides ℓ, with explicit coefficients and optimal bandwidth ℓ. Second, the proof is honest and self-contained; the hard parts are the bidiagonal model on residue classes and an asymptotic contradiction for j≥2. I went through the steps with the stress-test in hand and did not find a hidden gap.

What is genuinely new: the iff criterion, the closed form (6), and the factorization realizing every admissible operator as Type (2) in Horozov–Shapiro–Tater. The paper confirms two conjectures from [5] for this family, but the conjectures are not used as input; the derivation of the recurrence coefficients from the operator action is direct. The paper also usefully warns against a tempting misuse of Douak–Maroni symmetry: symmetry alone only gives the congruence of Proposition 2.2, and the proofs control all lower diagonals simultaneously.

Soft spots are minor. The significance is moderate: this is one three-parameter slice of a large classification program, and the reader's score of 6 feels about right. The necessity proof in Theorem 4.1 uses the definition of fixed-term recurrence with a uniform N; that is the right definition, but the argument does not address n-dependent bandwidth that grows slowly. That is a choice of problem, not a flaw. The polynomial-extension step in Theorem 5.2 is a little compressed: it states that both sides are polynomials and therefore it suffices to prove the identity for large n. The justification is sound, but a referee might ask for one sentence on why α_{n,n+1−τk} is polynomial in n for all n with R≥0; the answer is in the displayed summands. The Chu–Vandermonde condition w>τ−1 is explicitly verified, and the spot-check against Example 2.3 matches. The AI-assisted calculation disclosure is transparent and does not affect the reasoning.

This is a paper for specialists in orthogonal polynomials and special functions, not a wide-audience piece. For that audience it is a solid, publishable contribution. I would send it to a serious referee and expect acceptance after a light round of small clarifications. I would cite it if I worked on the Bochner–Krall problem.

Referee Report

0 major / 5 minor

Summary. The paper classifies, within the three-parameter family T=z^j∂_z^j+z^m∂_z^ℓ (0≤m<ℓ, 1≤j<ℓ), exactly which normalized monic eigenpolynomial sequences satisfy a fixed-term recurrence in the sense of definition (4). The main result (Theorem 1.1) states that such a recurrence exists if and only if j=1 and k=ℓ−m divides ℓ; in that case the recurrence is zP_n=P_{n+1}+∑_{q=1}^{σ}γ_n^{(q)}P_{n+1−qk} with the explicit coefficients (6), and the optimal lower bandwidth is N=ℓ. The proof is self-contained: Lemma 2.1 gives explicit coefficient formulas, Section 3 reduces the problem to bidiagonal models on residue-class subspaces, Theorem 4.1 excludes j≥2 by an asymptotic contradiction, and Theorem 5.2 derives the coefficients via a terminating Chu–Vandermonde evaluation. Proposition 5.6 then identifies every admissible operator with the Type (2) family of a conjecture of Horozov–Shapiro–Tater and verifies the predicted equality of differential and difference orders.

Significance. If correct, the result is a sharp and non-obvious classification within the generalized Bochner–Krall problem: it confirms Conjectures 1.9 and 1.10 of Horozov–Shapiro–Tater for this family, provides closed-form recurrence coefficients, and shows that the optimal bandwidth is always ℓ, independent of m. The proof is checkable and largely self-contained, with explicit formulas in Lemma 2.1, a rigorous asymptotic contradiction in Theorem 4.1, and a detailed terminating hypergeometric evaluation in Theorem 5.2; Example 2.3 provides a concrete verification of formula (6). The paper also usefully cautions that (k−1)-symmetry alone does not force the Douak–Maroni shape of the recurrence. The family studied is natural and the structural factorization in Proposition 5.6 is elegant.

minor comments (5)

[§2, proof of Lemma 2.1(2)] The phrase 'with s=n-k' presumes n≥k; for n<k the vanishing follows from part (1) because no nonnegative exponent below n is congruent to n modulo k, so the argument should be rephrased to cover both cases.
[§5, proof of Theorem 5.2] The claim that each summand of (24) is a polynomial in n is compressed: the functional ψ depends on R=n+1−τk, and the authors should explicitly cite (22) with R=n+1−τk and i=τ−t to justify the polynomial dependence.
[Eq. (20)] The notation for the transformed constants in (20) is visually confusing; using a tilde over c_R would improve readability.
[Remark 4.2] The conclusion 'one reads off that N≥ℓ' would be clearer if the degrees of the two polynomials on each side of the identity were stated explicitly.
[Acknowledgements] The disclosure of GPT-assisted checking is transparent, but specifying which calculations were checked with the tool would aid reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the classification and explicit recurrence coefficients are derived from the operator action and standard identities, with self-citations used only as context.

full rationale

The paper's derivation chain is self-contained. Lemma 2.1 obtains the eigenpolynomial coefficients directly from the eigenvalue equation (10) and the monomial action (7), without assuming any recurrence. Section 3's bidiagonal model and the coordinate-functional formulas (Lemma 3.2) follow from triangularity and the T-action on monomials (18), again without recurrence input. The necessity proof (Theorem 4.1) assumes only the definition of finite lower bandwidth (4); the uniformity of N is the definition itself, not a hidden gap. The contradiction between the asymptotic growth of the left side of (20) and the boundedness of the right side uses only j ≥ 2 and the polynomial growth of Λ_q, so no target property is smuggled in. The sufficiency proof (Theorem 5.2) computes α_{n,R} from the expansion (24) using the already-proved coefficient formulas (21) and (22), then applies a terminating Gauss-series evaluation (Chu–Vandermonde); no recurrence coefficient is fitted or assumed. The references to Conjectures 1.9 and 1.10 of Horozov–Shapiro–Tater are post-hoc contextual identifications, not premises: Proposition 5.6 explicitly proves the Type (2) factorization from the definition of the operator, and the order equality is read off from the derived formula (5). Citations to [1] and [3] are contextual or cautionary and are not load-bearing. The acknowledgement about GPT-assisted checking is transparent and does not enter the proof. Hence there is no step in which a prediction reduces, by construction, by fitting, or by self-citation, to its own inputs.

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

The proof is self-contained. No free parameters are fitted: the constants in formula (6) follow from the operator action and standard identities. No new entities are postulated. The only external inputs are standard mathematical facts and the cited hypergeometric summation identity.

assumptions (4)

standard math Falling factorial identities and the decomposition (x)_r = (x)_s (x-s)_{r-s}
Used repeatedly in Lemma 2.1 and Theorem 5.2 to manipulate the coefficient formulas.
standard math Chu-Vandermonde summation for terminating Gauss hypergeometric series
Used in Theorem 5.2 to evaluate the finite sum in (26); cited to [6].
standard math Polynomial identity principle: equality on infinitely many integers implies equality as polynomials
Used in Theorem 5.2 to extend formula (23) from all sufficiently large n to all n.
standard math Linear algebra of direct sums: C[z] = ⊕_{r=0}^{k-1} V_r and coordinate functionals are well-defined
Used in Proposition 2.2 and Lemma 3.2 to diagonalize the operator on residue-class subspaces.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Finite-term recurrences in a generalized Bochner--Krall family." pith.science (2026). https://pith.science/paper/CB6VFLNZ

@misc{pith2026260821802,
  author       = {Pith},
  title        = {Pith review of: Finite-term recurrences in a generalized Bochner--Krall family},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CB6VFLNZ}},
  note         = {Machine review of arXiv:2608.21802}
}

abstract

We classify the differential operators $T=z^j\partial_z^j+z^m\partial_z^\ell$, where $0\le m<\ell$ and $1\le j<\ell$, whose monic eigenpolynomials satisfy a finite-term recurrence relation. Writing $k=\ell-m$, such a recurrence exists if and only if $j=1$ and $k\mid\ell$. In that case we determine all recurrence coefficients in closed form and prove that the associated difference operator has order $\ell$. We also give an explicit factorization showing that every admissible operator is of Type~(2) in Conjecture~1.10 of Horozov--Shapiro--Tater; the equality of the differential and difference orders is the conclusion predicted by their Conjecture~1.9.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

6 extracted references · 6 canonical work pages

[5]

Horozov, B

E. Horozov, B. Shapiro, and M. Tater,In search of a higher Bochner theorem, J. Approx. Theory305(2025), Paper No. 106114, 23 pp

work page 2025
[1]

L. M. Anguas and D. Barrios Rolanía,New tools for the study of Bochner differential oper- ators, Rend. Circ. Mat. Palermo (2)75(2026), Article 60. doi:10.1007/s12215-026-01379-w

work page doi:10.1007/s12215-026-01379-w 2026
[2]

Bergkvist,On asymptotics of polynomial eigenfunctions for exactly solvable differential operators, J

T. Bergkvist,On asymptotics of polynomial eigenfunctions for exactly solvable differential operators, J. Approx. Theory149(2007), 151–187

work page 2007
[3]

Douak and P

K. Douak and P. Maroni,Les polynômes orthogonaux classiques de dimension deux, Analysis 12(1992), 71–107

work page 1992
[4]

Horozov,d-orthogonal analogs of classical orthogonal polynomials, SIGMA14(2018), Paper No

E. Horozov,d-orthogonal analogs of classical orthogonal polynomials, SIGMA14(2018), Paper No. 063, 27 pp

work page 2018
[6]

L. J. Slater,Generalized Hypergeometric Functions, Cambridge University Press, Cambridge, 1966. 12 L. M. ANGUAS, D. BARRIOS ROLANÍA, B. SHAPIRO, AND M. TATER Saint Louis University, Madrid Campus, A venida del V alle 34, 28003 Madrid, Spain Email address:luismiguel.anguas@slu.edu ETSI Industriales, Universidad Politécnica de Madrid, C/José Gutiérrez Abasc...

work page 1966

Pith tools

Reviewed August 27, 2026 · model on record in the stance chip above.

[1] [5]

Horozov, B

E. Horozov, B. Shapiro, and M. Tater,In search of a higher Bochner theorem, J. Approx. Theory305(2025), Paper No. 106114, 23 pp

work page 2025

[2] [1]

L. M. Anguas and D. Barrios Rolanía,New tools for the study of Bochner differential oper- ators, Rend. Circ. Mat. Palermo (2)75(2026), Article 60. doi:10.1007/s12215-026-01379-w

work page doi:10.1007/s12215-026-01379-w 2026

[3] [2]

Bergkvist,On asymptotics of polynomial eigenfunctions for exactly solvable differential operators, J

T. Bergkvist,On asymptotics of polynomial eigenfunctions for exactly solvable differential operators, J. Approx. Theory149(2007), 151–187

work page 2007

[4] [3]

Douak and P

K. Douak and P. Maroni,Les polynômes orthogonaux classiques de dimension deux, Analysis 12(1992), 71–107

work page 1992

[5] [4]

Horozov,d-orthogonal analogs of classical orthogonal polynomials, SIGMA14(2018), Paper No

E. Horozov,d-orthogonal analogs of classical orthogonal polynomials, SIGMA14(2018), Paper No. 063, 27 pp

work page 2018

[6] [6]

L. J. Slater,Generalized Hypergeometric Functions, Cambridge University Press, Cambridge, 1966. 12 L. M. ANGUAS, D. BARRIOS ROLANÍA, B. SHAPIRO, AND M. TATER Saint Louis University, Madrid Campus, A venida del V alle 34, 28003 Madrid, Spain Email address:luismiguel.anguas@slu.edu ETSI Industriales, Universidad Politécnica de Madrid, C/José Gutiérrez Abasc...

work page 1966