Pith. sign in

REVIEW 2 major objections 3 minor 2 cited by

On the Macdonald correspondence

T0 review · 2 major / 3 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Macdonald's correspondence is uniquely characterized by central characters and pair epsilon-factors.

desk verdict The advertised characterization theorem is unproven as written — the converse theorem is applied to a mixed pair without the required mixed-pair epsilon identity — but the survey parts and Theorem 1 are solid. read the letter →

arxiv 2501.02332 v1 pith:UAIF6MJD submitted 2025-01-04 math.RT

Anne-Marie Aubert This is my paper · ORCID
classification math.RT MSC 20C3322E5011S40
keywords Macdonaldcorrespondenceepsilon-factorsforpairscuspidalrepresentationsWeil-DelignefinitegenerallineargroupslocalLanglandsgammafactorscharacterizationtheorem
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 tries to establish that the Macdonald correspondence is the only bridge between inertial classes of tamely ramified Weil-Deligne representations and irreducible cuspidal representations of finite general linear groups that is compatible with the natural invariants. It reconstructs the correspondence through the classification of representations of finite groups of Lie type, defines epsilon-factors for pairs of cuspidal representations, and proves that these match the epsilon-constants on the Weil-Deligne side. The main theorem states that any collection of bijections preserving central characters and these paired epsilon-factors must be the Macdonald correspondence. If correct, this puts the finite-field correspondence on the same axiomatic footing as the p-adic characterization.

What carries the argument

The central object is the epsilon-factor for pairs of irreducible cuspidal representations, built from a normalized gamma factor attached to parabolic induction. Three supporting mechanisms carry the argument: the parametrization of cuspidal representations by regular characters of elliptic tori, the correspondence between Galois orbits of such characters and inertial classes of tamely ramified Weil-Deligne representations, and a finite-field converse theorem that converts equality of all pair gamma factors into equality of representations. The pair epsilon-factor identity is what lets the author compare any candidate bijection with the Macdonald correspondence and force them to match.

What would settle it

Compute the pair epsilon-factor with one representation taken from the Macdonald correspondence and the other from a candidate bijection, for a small rank where the two could disagree; any mismatch would show that the uniqueness proof as written does not go through.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2: if a family of bijections from inertial classes of n-dimensional tamely ramified irreducible Weil-Deligne representations to irreducible cuspidal representations of finite general linear groups preserves central characters and the epsilon-factor for every pair of such classes, then the family coincides with the Macdonald correspondence. The proof shows that the inverse Macdonald map has both properties, then uses a converse theorem for finite general linear groups to show that any competitor must agree with it on every representation. The paper also records a construction of the correspondence and proves in Theorem 1 that its epsilon-factors for pairs agree with the matching epsilon-constants of Weil-Deligne representations.

Load-bearing premise

The proof assumes that epsilon-factors for pairs remain unchanged when one member is chosen by the candidate bijection and the other by Macdonald's map, and this mixed equality is not derived in the text.

Editorial extensions

If this is right

  • If Theorem 2 is correct, the Macdonald correspondence is the unique matching of cuspidal representations to tamely ramified Weil-Deligne classes that preserves central characters and pair epsilon-factors.
  • The epsilon-factors for pairs defined from gamma factors agree with the epsilon-constants of the corresponding Weil-Deligne representations, giving a finite-field analogue of the local Langlands pair-factor compatibility.
  • The construction via the classification of finite groups of Lie type makes the Macdonald correspondence explicitly computable from regular characters and partitions.
  • The characterization implies that central characters together with all pair epsilon-factors form a complete set of invariants for this classification problem.
  • The compatibility with the depth-zero local Langlands correspondence locates the finite-field correspondence as the residue-field shadow of the p-adic correspondence.

Reading between the lines

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

  • If the mixed-pair equality used in the proof can be derived, the theorem would extend the p-adic uniqueness principle to finite fields: pair epsilon-factors alone determine the correspondence, without invoking the full local Langlands machinery.
  • The same strategy may characterize correspondences for other finite reductive groups whenever a compatible pair-factor theory and a converse theorem are available.
  • A direct computation of the mixed epsilon-factor for small q and small n would settle whether the only gap in the written proof is repairable.
  • The axiomatic formulation suggests viewing the Macdonald correspondence not as a list of matches but as the unique solution to a system of invariance equations.
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

2 major / 3 minor

Summary. The paper revisits Macdonald's 1980 bijection between irreducible representations of GL_n(k) and inertial equivalence classes of tame n-dimensional Weil-Deligne representations of W_F. It constructs the correspondence via Lusztig's classification of finite groups of Lie type, recalls its compatibility with the depth-zero local Langlands correspondence and Shintani descent, and defines epsilon factors for pairs of cuspidal representations using Soudry-Zelingher's Shahidi gamma factors. Theorem 1 states that these pair epsilon factors match Deligne epsilon factors under the Macdonald correspondence. The advertised new result is Theorem 2, which claims that any family of bijections T_n between the same two sets that preserves central characters and epsilon factors for pairs must equal the cuspidal Macdonald correspondence, with the proof resting on Nien's converse theorem through Corollary 5.0.3.

Significance. If valid, Theorem 2 would give a clean finite-field analogue of Henniart's characterization of the local Langlands correspondence, and the paper's construction of the Macdonald correspondence through Lusztig's classification is a useful perspective. Theorem 1 is a valuable matching statement for pair epsilon factors. However, the central characterization is not established by the written argument: the proof of Theorem 2 applies Corollary 5.0.3 to a mixed pair without proving the required mixed-pair epsilon identity. Since Theorem 2 is the main new claim of the paper, this is a load-bearing gap rather than a presentation issue.

major comments (2)
  1. [§5, proof of Theorem 2 (Eq. (5.0.5))] Corollary 5.0.3 requires, for the chosen ρ and every m ≤ n/2 and every irreducible m-dimensional ρ′, the equality ε(π × πρ′, ψ) = ε(ρ ⊗ ρ′, ψ). In the proof one takes π = τρ and needs exactly this mixed equality. The text establishes Eq. (5.0.5), which equates ε(πρ1 × πρ2, ψ) with ε(τρ1 × τρ2, ψ), plus the equality of central characters. These are same-correspondence comparisons: Theorem 1 gives ε(πρ × πρ′) = ε(ρ ⊗ ρ′), and property (2) gives ε(τρ × τρ′) = ε(ρ ⊗ ρ′). Neither controls the mixed quantity ε(τρ × πρ′, ψ), because T is not yet known to coincide with M. The sentence "Equations (5.0.4) and (5.0.5) allow us to apply Corollary 5.0.3" therefore conflates same-side identities with the mixed-pair identity demanded by the corollary. This is not a minor omission: the conclusion τρ = πρ is precisely the statement needed to convert the same-side equalities into the mixed equality, so the reasoning is circular at the load-bearing step.
  2. [§5, Corollary 5.0.3] The statement of Corollary 5.0.3 contains an internal inconsistency: it quantifies over an integer m with 1 ≤ m ≤ n/2 but then requires equality for every irreducible n-dimensional representation ρ′ of W_F, leaving m unused; presumably ρ′ should be m-dimensional. More substantively, the proof says the corollary follows from Nien's Converse Theorem by using Definition 3.0.6 and (3.0.5), but Nien's theorem requires equality of gamma factors against every irreducible generic representation τ of GL_m(k), whereas the corollary only supplies equalities for cuspidal representations of the form πρ′. A reduction from arbitrary generic τ to cuspidal data, presumably via multiplicativity of gamma factors, is not supplied. Since Corollary 5.0.3 is the bridge from the hypotheses of Theorem 2 to Nien's converse theorem, this step needs to be written out.
minor comments (3)
  1. [§2.5, Eq. (2.5.1)] The formula reads ψ(a1 + a1 + ⋯ + an−1); the second occurrence of a1 should almost certainly be a2.
  2. [§2.3.2, paragraph before Eq. (2.3.16)] There is a typo: "In particuler" should be "In particular".
  3. [§3, Definition 3.0.6] The notation ε(σ1 × σ2, ψ) is introduced for pairs of cuspidal representations, but in Theorem 2 the same notation is used in property (2) for all ρ1, ρ2 without explicitly restating the cuspidal restriction; clarifying this at the start of §5 would help the reader.

Circularity Check

1 steps flagged · score 3.0 of 10

Theorem 2's uniqueness proof applies Corollary 5.0.3 to the mixed pair (τρ, πρ′) without establishing the required mixed-pair ε-identity; the asserted reduction is the missing hypothesis in disguise.

  1. other [Section 5, proof of Theorem 2, equations (5.0.4)–(5.0.5) and application of Corollary 5.0.3]
    "Hence, we get (5.0.5) ε(πρ1 × πρ2, ψ) = ε(τρ1 × τρ2, ψ). Equations (5.0.4) and (5.0.5) allow us to apply Corollary 5.0.3. We obtain that τρ = πρ, for every integer n ≥ 1, and every ρ ∈ Irr(WDtF)IF,n."

    Corollary 5.0.3 requires, with π = τρ, the mixed-pair equality ε(τρ × πρ′, ψ) = ε(ρ ⊗ ρ′, ψ) for every ρ′ of dimension at most n/2. The paper has only established ε(πρ × πρ′, ψ) = ε(ρ ⊗ ρ′, ψ) by Theorem 1 and ε(τρ × τρ′, ψ) = ε(ρ ⊗ ρ′, ψ) by property (2) of T. Equation (5.0.5) therefore equates the two same-side products and gives no control over ε(τρ × πρ′, ψ). Since the only stated route to the missing mixed-pair identity would be the conclusion τρ = πρ, or an unstated induction on n, the application of the corollary assumes the very equality Theorem 2 is proving. The written proof does not supply the lemma needed to make the converse-theorem corollary applicable.

full rationale

No definitional circularity appears in the construction of the Macdonald correspondence or in the definition of ε-factors for pairs: ε(σ1 × σ2, ψ) is defined through Shahidi gamma factors from [SoZe], and Theorem 1 is a substantive matching statement citing external results [YZ] and [Z]. The paper is not driven by self-citations. The single load-bearing problem is in the proof of Theorem 2: the application of Corollary 5.0.3 conflates the established same-side ε-equalities with the required mixed-pair ε-equality. One can imagine repairing the argument by induction on n, using the already-proved equality for smaller-dimensional ρ′ to convert ε(τρ × πρ′) into ε(τρ × τρ′); but the text does not state such an induction, and as written the corollary's hypothesis is left unjustified. Because this gap is in the central characterization theorem, the circularity score is moderate; the claim itself has independent content and could plausibly be completed, so it is not a 6-10 'reduces by construction' case.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper does not fit any parameters to data and does not introduce new physical or mathematical entities. It relies entirely on established results in representation theory, including Lusztig's classification, Green's cuspidal parametrization, Nien's converse theorem, and the gamma-factor theories of Soudry-Zelingher, Ye-Zelingher, and Zelingher.

assumptions (6)
  • domain assumption Lusztig's classification of irreducible representations of finite groups of Lie type, including the bijection (2.3.21) and the Hecke algebra description (2.3.11).
    Invoked in §2.3.3 to construct the bijection P_n(X)^Γ ≅ Irr(GL_n(k)) that underlies the definition of the Macdonald correspondence.
  • domain assumption Green's parametrization of cuspidal representations of GL_n(k) by regular characters (Section 2.3.2, equation (2.3.14)).
    Used to attach a representation π[θ] to each Galois orbit of regular characters; this is the first step in the parametrization (2.3.16).
  • domain assumption Nien's converse theorem (Theorem 3.9 in [Nie]), stated as the Converse Theorem in Section 5.
    This is the engine behind Corollary 5.0.3 and hence Theorem 2; if Nien's theorem is not applicable in this form, the characterization collapses.
  • domain assumption Soudry-Zelingher theory of normalized Shahidi gamma factors for finite general linear groups (Definitions in §3, equation (3.0.4), and the relation (3.0.5)).
    The paper's epsilon-factors for pairs (Definition 3.0.6) are defined via these gamma factors, and Theorem 1 relies on the cited multiplicativity and normalization results.
  • domain assumption Macdonald's original construction of the correspondence and the matching of Godement-Jacquet epsilon-factors (2.5.9) from [Mac2].
    The paper identifies its construction with Macdonald's ([Mac2, §1]) and uses Macdonald's epsilon-factor preservation to connect to Deligne's constants.
  • domain assumption Silberger-Zink compatibility of the Macdonald correspondence with the depth-zero local Langlands correspondence (diagram (4.1.1)).
    Cited as a known property; used to support the interpretation of the correspondence, though not directly in the main proofs.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On the Macdonald correspondence." pith.science (2026). https://pith.science/paper/UAIF6MJD

@misc{pith2026250102332,
  author       = {Pith},
  title        = {Pith review of: On the Macdonald correspondence},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UAIF6MJD}},
  note         = {Machine review of arXiv:2501.02332}
}
abstract

In 1980 Ian G. Macdonald established an explicit bijection between the isomorphism classes of the irreducible representations of ${\mathrm{GL}}_n(k)$, where $k$ is a finite field, and inertia equivalence classes of admissible tamely ramified $n$-dimensional Weil-Deligne representations of $W_F$, where $F$ is a non-archimedean local field with residue field $k$ and $W_F$ the absolute Weil group of $F$. We describe a construction of the Macdonald correspondence based on the specialization to ${\mathrm{GL}}_n(k)$ of Lusztig's classification of irreducible representations of finite groups of Lie type, and review some properties of the correspondence. We define $\epsilon$-factors for pairs of irreducible cuspidal representations of finite general linear groups, and show that they match with the expected Deligne $\epsilon$-factors under the Macdonald correspondence. We use these $\epsilon$-factors for pairs to obtain a characterization of the Macdonald correspondence for the irreducible cuspidal representations

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. A Reduction over finite fields of the tame local Langlands correspondence for SLn

    math.RT 2025-01 accept novelty 7.0 of 10

    The paper constructs and proves the Vogan reduction map from irreducible SL_n(k)-representations to tame Langlands parameter inertia classes for SL_n(F), with fiber and compatibility theorems.

  2. Finite Langlands correspondence

    math.NT 2025-08 conditional novelty 6.0 of 10

    Every irreducible representation of a connected reductive group over a finite field is matched with a special Langlands parameter, with the fiber over each parameter given by irreducible representations of a finite co...

Reference graph

Works this paper leans on

21 extracted references · 21 canonical work pages · cited by 2 Pith papers

  1. [1]

    Deligne, ``Les constantes des \'equations fonctionnelles des functions L ", in Modular functions of one variable, II, Lecture Notes in Math

    P. Deligne, ``Les constantes des \'equations fonctionnelles des functions L ", in Modular functions of one variable, II, Lecture Notes in Math. 349, 501--597. Berlin, Heidelberg, New York: Springer 1972

    work page 1972

  2. [2]

    Deligne and G

    P. Deligne and G. Lusztig, ``Representations of reductive groups over finite fields", Ann. of Math. (2) 103:1 (1976), 103--161

    work page 1976

  3. [3]

    Geck and G

    M. Geck and G. Malle, The character theory of finite groups of Lie type. A guided tour, Cambridge studies in advanced mathematics 187, Cambridge University Press, Cambridge, 2020

    work page 2020

  4. [4]

    J. A. Green, ``The characters of the finite general linear groups", Trans. Amer. Math. Soc. 80 (1955), 402--447

    work page 1955

  5. [5]

    Harris and R

    M. Harris and R. Taylor, The Geometry and Cohomology of Some Simple Shimura Varieties, Annals of Mathematics Studies, vol. 151, Princeton University Press, Princeton, NJ, 2001. With an appendix by Vladimir G. Berkovich

    work page 2001

  6. [6]

    Henniart, ``Caract\'erisation de la correspondance de Langlands locale par les facteurs de paires", Invent

    G. Henniart, ``Caract\'erisation de la correspondance de Langlands locale par les facteurs de paires", Invent. Math. 113 (1993), 339--350

    work page 1993

  7. [7]

    , ``Une preuve simple des conjectures de Langlands pour (n) sur un corps p -adique, Invent. Math. 139:2 (2000), 439--455

    work page 2000

  8. [8]

    Jacquet, I

    H. Jacquet, I. I. Piatetski-Shapiro, and J. Shalika., ``Rankin–Selberg convolutions", Amer. J. Math. 105:2 (1983), 367--464

    work page 1983

Show all 21 references
  1. [9]

    Laumon, M

    G. Laumon, M. Rapoport, and U. Stuhler, `` -elliptic sheaves and the Langlands correspondence”, Invent. Math. 113:2 (1993), 217--338

    1993

  2. [10]

    Lusztig, Characters of Reductive Groups over a Finite Field, Annals of Mathematics Studies 107, Princeton University Press, Princeton, New Jersey, 1984

    G. Lusztig, Characters of Reductive Groups over a Finite Field, Annals of Mathematics Studies 107, Princeton University Press, Princeton, New Jersey, 1984

    1984

  3. [11]

    I. G. Macdonald, Symmetric functions and Hall polynomials, Oxford Mathematical Monographs, Clarendon Press, Oxford, 1979, viii +180 pp

    1979

  4. [12]

    , ``Zeta functions attached to finite general linear groups", Math. Ann. 249:1 (1980), 1--15

    1980

  5. [13]

    Nien, ``A proof of the finite field analogue of Jacquet's conjecture," Am

    C. Nien, ``A proof of the finite field analogue of Jacquet's conjecture," Am. J. Math. 136:3 (2014), 653--674

    2014

  6. [14]

    Roditty-Gershon, ``On gamma factors and Bessel functions for representations of general linear groups over finite fields", Master’s thesis, Tel Aviv University, 2010

    E.-A. Roditty-Gershon, ``On gamma factors and Bessel functions for representations of general linear groups over finite fields", Master’s thesis, Tel Aviv University, 2010

    2010

  7. [15]

    Scholze, ``The local Langlands correspondence for _n over p -adic fields”, Invent

    P. Scholze, ``The local Langlands correspondence for _n over p -adic fields”, Invent. Math. 192:3 (2013), 663--715

    2013

  8. [16]

    Silberger, and E.-W

    A. Silberger, and E.-W. Zink, ``The characters of the generalized Steinberg representations of finite general linear groups on the regular elliptic set”, Trans. Amer. Math. Soc. 352:7 (2000), 3339--3356

    2000

  9. [17]

    Algebra 319 (2008):10, 4147--4176

    , ``Explicit Shintani base change and the Macdonald correspondence for characters of _n(k) , J. Algebra 319 (2008):10, 4147--4176

    2008

  10. [18]

    Soudry and E

    D. Soudry and E. Zelingher, ``On gamma factors for representations of finite general linear groups", Essential Number Theory, 2023

    2023

  11. [19]

    T. A. Springer, ``The zeta function of a cuspidal representation of a finite group _n(k) ", in Lie groups and their representations, Proceedings of the Summer School of the Bolyai Janos Math. Soc., Budapest, 1971, Halsted Press (John Wiley & Sons) (1975), 645--648

    1975

  12. [20]

    Ye and E

    R. Ye and E. Zelingher, ``Epsilon factors of representations of finite general linear groups", J. Number Theory 221 (2021), 122--142

    2021

  13. [21]

    Zelingher, ``On values of the Bessel function for generic representations of finite general linear groups", Adv

    E. Zelingher, ``On values of the Bessel function for generic representations of finite general linear groups", Adv. Math. 434 (2023), art. id. 109314

    2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.