All-orders large-spin asymptotics of the twist-two anomalous-dimension matrix in Abelian vector-gluon theory and the residual fixed-point loophole in the Callan-Gross no-go theorem for Bjorken scaling

In 1973 Callan and Gross asked when a renormalizable quantum field theory can reproduce exact Bjorken scaling in deep-inelastic scattering. Proving a large-spin conjecture of Parisi's -- that the anomalous dimensions of leading-twist operators of spin $j$ approach twice the anomalous dimension of the constituent field as $j\to\infty$ -- they showed that in most theories scaling forces the renormalization-group fixed point to lie at the origin of coupling space, i.e. the theory must be asymptotically free. Much of the program is settled: non-Abelian gauge theories supply the twist-two anomalous dimensions and logarithmic scaling violations, which are confirmed quantitatively in data, and modern lightcone-bootstrap theorems rigorously establish large-spin twist accumulation in unitary conformal field theories under a twist-gap assumption. Two residual threads survive. First, the vector-gluon case of the large-spin statement -- whether the lower eigenvalue of the twist-two anomalous-dimension matrix in an Abelian gauge theory with charged matter converges to $2\gamma_A$ (twice the anomalous dimension of the gauge field strength) to all orders, with the charged channels separating off logarithmically -- was verified by Callan and Gross only to order $g^{2}$ and has never been proven or refuted as stated; equivalently, twist accumulation has not been extended to gauge theories without a gap assumption, with operator mixing. Second, whether a nontrivial fixed point can render only a subset of twist-two operators canonical -- the Yukawa loophole in the no-go argument -- was never excluded with all-orders information.

Background

In deep-inelastic scattering (DIS) a lepton probes a hadron by exchanging a highly virtual photon of momentum $q$ ; the cross section is expressed through structure functions $F_i(x,q^{2})$ of the Bjorken variable $x=-q^{2}/(2p\cdot q)$ , which measures the fraction of the target's momentum carried by the struck constituent. Bjorken scaling is the observation that at large $q^{2}$ and fixed $x$ the structure functions become approximately independent of $q^{2}$ , suggesting pointlike constituents. Wilson's operator product expansion (OPE) -- the short-distance expansion of a product of local operators -- expresses the Mellin moments of the structure functions over $x$ as sums over matrix elements of local composite operators, classified by twist, defined as mass dimension minus Lorentz spin. The dominant terms at large $q^{2}$ come from the leading-twist or twist-two operators: fermion bilinears $\bar\psi\gamma(D)^{j}\psi$ and gauge-field bilinears $F D^{j} F$ , where $D$ is the covariant derivative and $j$ is the Lorentz spin; at each spin these form a finite basis whose size grows with $j$ (from expanding the covariant derivatives) and whose members mix under renormalization. Renormalization introduces a scale dependence: each operator acquires an anomalous dimension, the rate at which its renormalized matrix elements drift with the renormalization scale, governed by the anomalous-dimension matrix together with the beta function of the couplings, whose zeros are the renormalization-group fixed points -- the values of the couplings at which the couplings stop running. Exact Bjorken scaling of all moments requires every leading-twist anomalous dimension to vanish at the fixed point toward which the theory flows in the ultraviolet. Asymptotic freedom (Politzer; Gross-Wilczek, 1973) is the property of non-Abelian gauge theories whose beta function drives the coupling to zero at short distances, so the fixed point lies at the origin of coupling space. Callan and Gross (1973) asked the converse: which renormalizable theories could produce exact scaling at a nontrivial, interacting fixed point? Parisi had conjectured, and Callan-Gross proved for most theories, that the anomalous dimensions of the twist-two operators approach, as $j\to\infty$ , twice the anomalous dimension of the constituent field -- in modern language, twist accumulation at the double-twist value, i.e. the sum of the scaling dimensions of two constituents. Combined with the monotonicity of the anomalous dimensions in $j$ (a consequence of the positivity of structure-function moments) and the uniqueness of the spin-2 energy-momentum tensor, this forces all field anomalous dimensions to vanish at any scaling fixed point, which by Federbush-Johnson-type arguments (statements that a field whose current two-point function saturates the free-field bound behaves as a free field) makes the theory free at the fixed point: exact scaling requires the fixed point at the origin. Two cases were left outside the completed argument. The non-Abelian gauge theories were resolved within a year by the discovery of asymptotic freedom and by the explicit computation of the twist-two anomalous dimensions with full operator mixing (Gross-Wilczek II, 1974), which predicted logarithmic scaling violations later confirmed in DIS data. The Abelian vector-gluon theory -- the 1970s model in which the strong interaction is mediated by a neutral spin-one Abelian gauge field coupled to charged matter -- was, however, covered only conditionally: there the large-spin statement was verified to order $g^{2}$ and left as a conjecture -- the lower eigenvalue of the anomalous-dimension matrix should approach $2\gamma_A$ , twice the anomalous dimension of the gauge field strength (equal, in the Abelian case, to that of the vector potential, since the field strength is linear in it). This conjecture is the 1973 avatar of the modern lightcone bootstrap -- the study of operator spectra at large Lorentz spin by analytic continuation of crossing symmetry. Rigorous theorems now prove twist accumulation $\tau\to 2\Delta_{\phi}$ (twist $\tau$ = dimension minus spin; $\Delta_{\phi}$ the scaling dimension of the constituent) for operators of large spin in any unitary conformal field theory (CFT) whose operator product of two identical scalars has a twist gap, meaning that every operator appearing in that operator product has twist bounded below by some positive constant (Pal-Qiao-Rychkov 2023), and convexity results control the minimal twist as a function of spin (Komargodski-Zhiboedov 2013) -- with an explicit caveat that in gauge theories carrying flux the almost-conserved currents receive a universal $\log s$ correction to twist, the physics of the cusp anomalous dimension (conventionally denoted $\Gamma_{\rm cusp}$ ), the coefficient of the ultraviolet divergence of a Wilson line (a phase factor tracing a particle's path) with a corner, which governs the $\ln j$ growth of the anomalous dimensions of charged high-spin operators (computed to three loops in QCD by Grozin, Henn, Korchemsky and Marquard 2015). None of these theorems covers the literal 1973 statement: the all-orders asymptotics of the anomalous-dimension matrix, with its spin-growing multiplicity of mixing structures, in an Abelian gauge theory with charged matter, at a fixed point whose existence is itself not established.

Problem Statement

The problem, in the form left open by Callan and Gross (1973), is to complete the no-go argument for Bjorken scaling in the cases it did not cover. (1) Large-spin asymptotics in the vector-gluon theory. Consider a renormalizable four-dimensional theory of an Abelian gauge field $A_{\mu}$ coupled to charged matter, and its tower of leading-twist (twist-two) local operators of spin $j$ : fermion bilinears $\bar\psi\gamma(D)^{j}\psi$ , gauge-field bilinears $F D^{j} F$ , and all structures generated by expanding the covariant derivatives $D=\partial-igA$ , whose number grows with $j$ . At a renormalization-group fixed point of such a theory, let $\gamma^{(j)}$ be the anomalous-dimension matrix of this operator basis and $\lambda_{\min}^{(j)}$ its smallest eigenvalue. Prove or disprove, to all orders in perturbation theory (or nonperturbatively), that

$\lim_{j\to\infty}\lambda_{\min}^{(j)} = 2\gamma_A,$

where $2\gamma_A$ is twice the anomalous dimension of the gauge-field strength (equal in the Abelian case to that of the vector potential, since $F_{\mu\nu}$ is linear in $A_\mu$ ), with the gauge-neutral channels saturating at the sum of constituent anomalous dimensions while the charged channels separate off with corrections growing like $\ln j$ governed by the cusp anomalous dimension. Callan and Gross verified only the leading order $g^{2}$ . Equivalently in modern terms: extend the twist-accumulation and convexity theorems of the lightcone bootstrap to gauge theories with charged fields -- without a twist-gap assumption on the operator spectrum and with mixing across the spin-growing basis -- so that additivity of twists at large spin holds with the gauge-flux logarithmic corrections under control. (2) The partial-canonicity loophole. Determine whether there exists a nontrivial fixed point of a renormalizable four-dimensional theory (the Yukawa-plus-quartic-scalar class identified by Callan and Gross, or any other) at which some but not all leading-twist operators have canonical (free-field) anomalous dimensions, so that deep-inelastic moments scale exactly while the theory remains interacting; or prove that no such fixed point exists -- Callan and Gross noted that assessing this requires all-orders Yukawa beta-function knowledge, which is still unavailable.

References

C. G. Callan Jr. and David J. Gross, "Bjorken Scaling in Quantum Field Theory", Physical Review D 8, 4383-4394 (1973). doi:10.1103/PhysRevD.8.4383, https://doi.org/10.1103/PhysRevD.8.4383
H. David Politzer, "Reliable Perturbative Results for Strong Interactions?", Physical Review Letters 30, 1346-1349 (1973). doi:10.1103/PhysRevLett.30.1346, https://doi.org/10.1103/PhysRevLett.30.1346
David J. Gross and Frank Wilczek, "Ultraviolet Behavior of Non-Abelian Gauge Theories", Physical Review Letters 30, 1343-1346 (1973). doi:10.1103/PhysRevLett.30.1343, https://doi.org/10.1103/PhysRevLett.30.1343
David J. Gross and Frank Wilczek, "Asymptotically Free Gauge Theories. II", Physical Review D 9, 980-993 (1974). doi:10.1103/PhysRevD.9.980, https://doi.org/10.1103/PhysRevD.9.980
S. Ferrara and M. Testa, "Canonical dimensions in asymptotically conformal invariant theories", Physics Letters B 49, 95-98 (1974). doi:10.1016/0370-2693(74)90588-7, https://doi.org/10.1016/0370-2693(74)90588-7
Wu-Ki Tung, "Experimental Signature of Scaling Violation Implied by Field Theories", Physical Review Letters 35, 490-494 (1975). doi:10.1103/PhysRevLett.35.490, https://doi.org/10.1103/PhysRevLett.35.490
M. Glück and E. Reya, "Evidence Against Non-asymptotically-free Theories of Strong Interactions", Physics Letters B 69, 77-80 (1977). doi:10.1016/0370-2693(77)90137-X, https://doi.org/10.1016/0370-2693(77)90137-X
J. Geicke, "Are Asymptotically Non-Free Theories with Vanishing Anomalous Dimension Really Excluded?", Physics Letters B 86, 317-320 (1979). doi:10.1016/0370-2693(79)90847-5, https://doi.org/10.1016/0370-2693(79)90847-5
David J. Gross, "Twenty Five Years of Asymptotic Freedom", Nuclear Physics B Proceedings Supplement 74, 426-446 (1999). arXiv:hep-th/9809060, doi:10.1016/S0920-5632(99)00208-X, https://arxiv.org/abs/hep-th/9809060
S. E. Derkachov and A. N. Manashov, "Generic scaling relation in the scalar $\phi^{4}$ model", Journal of Physics A 29, 8011-8024 (1996). arXiv:hep-th/9604173, doi:10.1088/0305-4470/29/24/024, https://arxiv.org/abs/hep-th/9604173
Dongmin Gang, Jae-Sung Park and Satoshi Yamaguchi, "Operator with large spin and spinning D3-brane", Journal of High Energy Physics 11 (2009) 024. arXiv:0908.3938, doi:10.1088/1126-6708/2009/11/024, https://arxiv.org/abs/0908.3938
Juan Maldacena and Alexander Zhiboedov, "Constraining Conformal Field Theories with A Higher Spin Symmetry", Journal of Physics A 46, 214011 (2013). arXiv:1112.1016, doi:10.1088/1751-8113/46/21/214011, https://arxiv.org/abs/1112.1016
Zohar Komargodski and Alexander Zhiboedov, "Convexity and Liberation at Large Spin", Journal of High Energy Physics 11 (2013) 140. arXiv:1212.4103, doi:10.1007/JHEP11(2013)140, https://arxiv.org/abs/1212.4103
Charlotte Sleight and Massimo Taronna, "Anomalous Dimensions from Crossing Kernels", Journal of High Energy Physics 11 (2018) 089. arXiv:1807.05941, doi:10.1007/JHEP11(2018)089, https://arxiv.org/abs/1807.05941
Andrey Grozin, Johannes M. Henn, Gregory P. Korchemsky and Peter Marquard, "Three Loop Cusp Anomalous Dimension in QCD", Physical Review Letters 114, 062006 (2015). arXiv:1409.0023, doi:10.1103/PhysRevLett.114.062006, https://arxiv.org/abs/1409.0023
Sridip Pal, Jiaxin Qiao and Slava Rychkov, "Twist Accumulation in Conformal Field Theory: A Rigorous Approach to the Lightcone Bootstrap", Communications in Mathematical Physics 402, 2169-2214 (2023). arXiv:2212.04893, doi:10.1007/s00220-023-04767-w, https://arxiv.org/abs/2212.04893
Giorgio Parisi, "From Bjorken Scaling to Scaling Violations", Highlights in High-Energy Physics (HiHEP) 1, 17 (2025). arXiv:2506.03383, doi:10.53941/hihep.2025.100017, https://arxiv.org/abs/2506.03383