The 1976 analysis of De Rújula, Georgi and Politzer showed that electroproduction local duality (the Bloom–Gilman correspondence between the nucleon resonance spectrum and the deep-inelastic scaling function) and precocious scaling follow from the operator-product-expansion moment predictions of QCD to the accuracy of the twist expansion, with higher-twist corrections of relative size nΛ2/Q2n\Lambda^2/Q^2 in the nn-th Nachtmann moment. Half a century later the empirical side is in excellent shape — Jefferson Lab data confirm duality resonance by resonance, including in separately measured longitudinal and transverse responses at Q2Q^2 of a few GeV2\mathrm{GeV}^2, and in inclusive cross sections and truncated moments up to Q2≈10 GeV2Q^2\approx 10~\mathrm{GeV}^2 — but the first-principles side is not: it is still not understood from QCD itself why the leading-twist description works at Q2∼1Q^2\sim 1–2 GeV22~\mathrm{GeV}^2 (why the relevant higher-twist matrix elements are small or cancel), the large-xx longitudinal strength that the original analysis underpredicted is still attributed to uncomputed dynamical higher twists, and no QCD derivation exists of the individual resonance transition strengths whose sum over the spectrum reconstructs the scaling function. The problem is to close this gap: derive local duality and precocious scaling from first-principles QCD — or rigorously delimit where they cannot hold — including the higher-twist mechanism, the longitudinal channel, and the fine-grained resonance structure that the moment expansion provably cannot reach.

Background

In inclusive electron–nucleon scattering only the outgoing electron is detected; when the photon virtuality Q2Q^2 and energy transfer ν\nu are large compared with hadronic scales, this is deep-inelastic scattering (DIS). The hadronic response is encoded in structure functions, chiefly F2(x,Q2)F_2(x,Q^2) with x=Q2/(2Mν)x=Q^2/(2M\nu) the Bjorken variable (MM the proton mass), together with the transverse and longitudinal virtual-photon absorption cross sections σT\sigma_T and σL\sigma_L, whose ratio R=σL/σTR=\sigma_L/\sigma_T measures the response to longitudinally polarized photons. In the parton model, structure functions measure parton distribution functions (PDFs), the momentum-fraction distributions of quarks and gluons. In quantum chromodynamics (QCD), the strong-interaction gauge theory, asymptotic freedom — the weakening of the coupling at short distances — makes this description quantitative at large Q2Q^2: the structure functions scale (become approximately Q2Q^2-independent functions of xx) up to logarithmic violations governed by the DGLAP (Dokshitzer–Gribov–Lipatov–Altarelli–Parisi) evolution equations for the parton distributions, plus power corrections in M/QM/Q. The kinematic part of the power corrections (target-mass effects) is largely removed by using the Nachtmann scaling variable ξ=2x/[1+1+4M2x2/Q2]\xi=2x/[1+\sqrt{1+4M^2x^2/Q^2}], as suggested by Georgi and Politzer (1976). The dynamical part is organized by the operator product expansion (OPE): the moments (integrals over ξ\xi with weight ξn−2\xi^{n-2}, up to known kinematic factors) of a structure function evolve with Q2Q^2 according to the anomalous dimensions (computable logarithmic quantum corrections to their scaling) of twist-two operators — twist is the operator dimension minus spin, and leading twist (twist 2) describes incoherent scattering from a single quark — while multi-parton correlations enter as higher twists, suppressed by powers of a scale M02/Q2M_0^2/Q^2 with M0M_0 expected to be of order Λ\Lambda, the fundamental QCD scale parameter.

At Q2Q^2 of a few GeV2^2 and below, the inclusive response at large xx is dominated by nucleon resonances: sharp enhancements at invariant mass WW up to about 2 GeV, each described by a transition form factor that falls rapidly with Q2Q^2. Bloom and Gilman (1971) observed that, plotted against the target-mass-corrected variable, the resonances move with increasing Q2Q^2 along — and average to — the same smooth scaling curve measured at much higher Q2Q^2: Bloom–Gilman duality. Equality of full moments (integrals over all xx) is called global duality; equality resonance by resonance, or over restricted intervals of WW, is called local duality. That approximate scaling already holds at Q2∼1Q^2\sim 1–2 GeV22~\mathrm{GeV}^2, far below naive expectations, is called precocious scaling.

De Rújula, Georgi and Politzer (1976) gave this the standard QCD account. Interpreting the resonance region through OPE moments of νW2\nu W_2 taken with the Nachtmann variable ξ\xi (including the elastic peak at ξ=ξp\xi=\xi_p), they showed that local duality is a consequence of the moment predictions: the perturbatively evolved nn-th moment carries a relative higher-twist uncertainty of order nΛ2/Q2n\Lambda^2/Q^2, so resonance averages at low Q2Q^2 should track the evolved scaling curve, the more accurately the lower the moment order. A fit to the data of the time gave Λ≈0.5\Lambda\approx 0.5 GeV — a small Λ\Lambda being the origin of precocious scaling — and a next-to-leading-twist scale M0≃400M_0\simeq 400 MeV extracted from resonance-region strength, and the scaling-region fit was shown to describe the resonance region. The same analysis flagged unresolved items: the explicit formulae covered only the non-singlet valence region (ξ≳0.3\xi\gtrsim 0.3; the region where only valence-quark flavor combinations contribute), because below that the SU(3)-flavor-singlet quark distribution (the flavor-symmetric sum of quarks and antiquarks, which mixes with the gluon distribution under evolution) and the gluon distribution enter and could not then be separated from the data; measurements of RR at large xx lay systematically above the predictions; and fine-grained information on individual resonances would require moment orders so high that all higher twists contribute at once — genuinely nonperturbative territory (exchanges of arbitrarily many soft gluons).

The subsequent half century settled the empirical side. Global QCD analyses now determine the singlet and gluon PDFs, so the low-ξ\xi limitation is obsolete. Jefferson Lab measurements confirmed duality quantitatively: valence-region duality in F2F_2 (Niculescu et al., 2000), duality in separately measured transverse and longitudinal structure functions in the resonance region (experiment E94-110, as analyzed in the 2005 review of Melnitchouk, Ent and Keppel), spin-structure-function duality, and most recently data from the CLAS12 spectrometer (a large-acceptance detector at Jefferson Lab) up to Q2≈10 GeV2Q^2\approx 10~\mathrm{GeV}^2 (Wunderlich et al., 2026), where truncated moments (moments integrated over a restricted invariant-mass interval rather than the full range) of the resonance-region F2F_2 agree with a global QCD fit including target-mass and higher-twist corrections. The large-xx RR remains above perturbative QCD even with target-mass corrections and is interpreted as dynamical higher twist. The theoretical side, however, stalled at the 1976 level of understanding. The OPE organizes but does not explain why the relevant higher twists are small or cancel at Q2∼1Q^2\sim 1–2 GeV22~\mathrm{GeV}^2; relativistic constituent-quark models (hadrons treated as bound states of massive effective quarks) exhibit duality through interference of resonance contributions (Isgur, Jeschonnek, Melnitchouk and Van Orden, 2001), while a model based on light-front holography (an approach combining light-front quantization with ideas from holographic, gauge/gravity, duality) fails to satisfy global and local duality sum rules at all (Sheckler and Miller, 2021); lattice QCD (QCD formulated numerically on a discrete Euclidean space-time grid) computes leading-twist matrix elements but not the resonance-region inclusive response. De Rújula himself, reviewing QCD's stubbornly unsolved problems in 2019, answers the question 'Is Bloom–Gilman duality a prediction of QCD?' with 'not (yet)': that would require a complete understanding of bound-state production.

Problem Statement

The problem is to establish, from first principles in QCD, whether local (Bloom–Gilman) duality and precocious scaling in the nucleon resonance region are consequences of the theory — and, if they are, to derive their quantitative content; if they are not in full generality, to delimit rigorously where they hold and where they must fail. This is the question one of the original authors posed in 2019: 'Is Bloom–Gilman duality a prediction of QCD?' — as opposed to a regularity that the operator-product expansion organizes but does not explain. Concretely, an answer should deliver, within one coherent framework grounded in QCD itself (not in phenomenological models or fits to the same data):

  1. The higher-twist mechanism at low Q2Q^2. Starting from QCD, account quantitatively for the empirical fact that resonance-averaged structure functions at Q2∼1Q^2\sim 1–2 GeV22~\mathrm{GeV}^2 match the perturbatively evolved leading-twist scaling functions: derive the magnitude and pattern of the higher-twist corrections (of relative size nΛ2/Q2n\Lambda^2/Q^2 in the nn-th Nachtmann moment), determine whether the relevant higher-twist matrix elements are small or cancel at these momentum transfers and, if so, why, and state whether a universal higher-twist scale M0M_0 of order Λ\Lambda emerges. This must include the longitudinal channel: the original analysis underpredicted the measured R=σL/σTR=\sigma_L/\sigma_T at large xx, and modern data still exceed perturbative QCD with target-mass corrections there.

  2. The fine-grained resonance structure. Extend the account to the level the moment expansion cannot reach: derive from QCD the individual resonance transition strengths and their Q2Q^2 dependence — the hadronic content whose sum over the spectrum reconstructs the scaling function — in the regime where resolving individual resonances would require moment orders so high that all higher twists contribute simultaneously.

  3. The dynamical mechanism and its limits. Identify the mechanism and derive from it where duality holds and where it fails — channel by channel, resonance by resonance — so that the observed violations (for example in spin observables or in the longitudinal channel) are predicted rather than fitted.

The common objective of the three requested results is a first-principles QCD explanation of the resonance–scaling (quark–hadron) correspondence; they are jointly the modern content of the limitations that the 1976 analysis had to leave open. Each deliverable is resolved by a first-principles QCD derivation of the requested result; where no such derivation exists, it is resolved instead by a no-go argument that establishes a stated obstruction, naming the class of obstruction and the standard of proof the argument must meet.

References

  1. A. De Rújula, H. Georgi and H.D. Politzer, 'An explanation of local duality and precocious scaling', Physics Letters B 64 (1976) 428–432, DOI: 10.1016/0370-2693(76)90113-1, https://doi.org/10.1016/0370-2693(76)90113-1
  2. A. De Rújula, H. Georgi and H.D. Politzer, 'Demythification of electroproduction local duality and precocious scaling', Annals of Physics 103 (1977) 315–353, DOI: 10.1016/S0003-4916(97)90003-8, https://doi.org/10.1016/S0003-4916(97)90003-8
  3. E.D. Bloom and F.J. Gilman, 'Scaling and the behavior of nucleon resonances in inelastic electron-nucleon scattering', Physical Review D 4 (1971) 2901–2916, DOI: 10.1103/PhysRevD.4.2901, https://doi.org/10.1103/PhysRevD.4.2901
  4. H. Georgi and H.D. Politzer, 'Precocious scaling, rescaling and xi scaling', Physical Review Letters 36 (1976) 1281, erratum: Physical Review Letters 37 (1976) 68 (erratum DOI: 10.1103/PhysRevLett.37.68.2), DOI: 10.1103/PhysRevLett.36.1281, https://doi.org/10.1103/PhysRevLett.36.1281
  5. W. Melnitchouk, R. Ent and C. Keppel, 'Quark–hadron duality in electron scattering', Physics Reports 406 (2005) 127–301, DOI: 10.1016/j.physrep.2004.10.004, arXiv:hep-ph/0501217, https://arxiv.org/abs/hep-ph/0501217
  6. A. De Rújula, 'QCD, from its inception to its stubbornly unsolved problems', International Journal of Modern Physics A 34 (2019) 1930015, DOI: 10.1142/S0217751X19300151, arXiv:1910.13891, https://arxiv.org/abs/1910.13891
  7. A.B. Sheckler and G.A. Miller, 'Mystery of Bloom-Gilman duality: a light-front holographic QCD perspective', Physical Review D 103 (2021) 096018, DOI: 10.1103/PhysRevD.103.096018, arXiv:2101.00100, https://arxiv.org/abs/2101.00100
  8. Y. Wunderlich, A. Bulgakov, K. Joo, T.-S.H. Lee and V.I. Mokeev, 'Quark–hadron duality in inclusive electron–proton scattering at high Q^2: structure functions and truncated moments from CLAS12', arXiv:2606.29690 (2026), https://arxiv.org/abs/2606.29690
  9. N. Isgur, S. Jeschonnek, W. Melnitchouk and J.W. Van Orden, 'Quark hadron duality in structure functions', Physical Review D 64 (2001) 054005, DOI: 10.1103/PhysRevD.64.054005, https://doi.org/10.1103/PhysRevD.64.054005
  10. I. Niculescu, C.S. Armstrong, J. Arrington, K.A. Assamagan, O.K. Baker, D.H. Beck, C.W. Bochna, R.D. Carlini, J. Cha, C. Cothran, D.B. Day, J.A. Dunne, D. Dutta, R. Ent, V.V. Frolov, H. Gao, D.F. Geesaman, P.L.J. Gueye, W. Hinton, R.J. Holt, H.E. Jackson, C.E. Keppel, D.M. Koltenuk, D.J. Mack, D.G. Meekins, M.A. Miller, J.H. Mitchell, R.M. Mohring, G. Niculescu, D. Potterveld, J.W. Price, J. Reinhold, R.E. Segel, P. Stoler, L. Tang, B.P. Terburg, D. Van Westrum, W.F. Vulcan, S.A. Wood, C. Yan and B. Zeidman, 'Evidence for valencelike quark-hadron duality', Physical Review Letters 85 (2000) 1182–1185, DOI: 10.1103/PhysRevLett.85.1182, https://doi.org/10.1103/PhysRevLett.85.1182
  11. A. Psaker, W. Melnitchouk, M.E. Christy and C. Keppel, 'Quark-hadron duality and truncated moments of nucleon structure functions', Physical Review C 78 (2008) 025206, DOI: 10.1103/PhysRevC.78.025206, https://doi.org/10.1103/PhysRevC.78.025206
  12. I. Niculescu, C. Keppel, S. Liuti and G. Niculescu, 'Extraction of higher twists from electron–proton inclusive data at large Bjorken x', Physical Review D 60 (1999) 094001, DOI: 10.1103/PhysRevD.60.094001, https://doi.org/10.1103/PhysRevD.60.094001
  13. C.E. Carlson and N.C. Mukhopadhyay, 'Quantum chromodynamics and Bloom-Gilman duality', Physical Review D 41 (1990) 2343, DOI: 10.1103/PhysRevD.41.2343, https://doi.org/10.1103/PhysRevD.41.2343