Simultaneous complete factorization, gauge invariance, and realistic $W_1$ in electromagnetic dual schemes for hadrons

Nambu's 1971 construction of electromagnetic currents in the dual-resonance model (the Veneziano-type 'dual hadrodynamics' that later became string theory) left a sharply stated open question: can a dual electromagnetic scheme simultaneously achieve (i) complete factorization of current-coupled amplitudes in all channels, (ii) exact gauge invariance (current conservation, Ward-type identities for off-shell photons), and (iii) a realistic deep-inelastic structure function $W_1$, i.e. Bjorken scaling with a small longitudinal-to-transverse ratio $R=\sigma_L/\sigma_T$? His own scheme achieved duality, intercept-one output trajectories, generalized vector-dominance form factors, and a scaling $\nu W_2$, but failed all three requirements at various points, and he argued that the divergence of $R$ is a general property of the dual model. The 1971–1978 follow-up literature contains constructions (Ramond; Manassah–Matsuda; Kikkawa–Sakita; Neveu–Scherk; Schwarz; Corrigan–Fairlie) that achieve some of the requirements only by giving up others, and Collins–Friedman proved a no-go theorem for the conventional spectrum. The program was abandoned in the mid-1970s in favor of QCD; the compatibility question was never settled and is open but dormant.

Background

In 1968 Veneziano wrote down a closed-form amplitude for meson scattering exhibiting duality: the same analytic function admits both an expansion in $s$-channel resonances lying on linearly rising Regge trajectories (families of hadrons whose spin $J$ grows linearly with mass squared, $J=\alpha(M^2)$, the intercept $\alpha(0)$ fixing the high-energy power law) and an expansion in crossed-channel Regge-pole exchanges (throughout, $s$, $t$, and $u$ denote the Mandelstam invariants, the standard Lorentz-invariant momentum combinations of a scattering process). The resulting dual-resonance model (DRM) is factorizable: its amplitudes decompose on a spectrum of discrete intermediate states generated by an infinite tower of harmonic-oscillator operators (the Fubini–Veneziano operator formalism), and Nambu reinterpreted it as the theory of a relativistic 'rubber string' whose normal modes are precisely those oscillators. Consistency requires the decoupling of negative-norm 'spurious' states, encoded in the Virasoro gauge conditions (the infinite family of constraints $L_n\,|\mathrm{phys}\rangle=0$ for $n>0$ on physical states).

This machinery described purely hadronic amplitudes, but hadrons also couple to the electromagnetic field. Since the photon is not one of the model's resonances, coupling an external current $J_\mu$ means inserting vertex operators carrying an arbitrary virtuality $q^2$ into dual amplitudes. This proved genuinely hard: the natural string current built from the coordinate derivative, $\dot Z_\mu e^{iq\cdot Z}$, produces Gaussian elastic form factors (the observed falloff is a power law) and cannot be made both conserved and dual, while early attempts at off-shell photons generated 'moving poles' whose positions depend on external momenta.

Meanwhile the 1968–69 SLAC deep-inelastic electron–proton experiments showed approximate Bjorken scaling: for $-q^2\to\infty$ with $\omega=2p\cdot q/(-q^2)$ fixed, the structure functions $W_1$ and $\nu W_2$ (which parametrize the virtual-Compton amplitude $\gamma^* p\to\gamma^* p$; $\nu=p\cdot q/m$) become functions of $\omega$ alone. The parton model of Feynman and Bjorken–Paschos explained this as incoherent scattering from pointlike spin-$1/2$ constituents, which through the Callan–Gross relation (the scaling law $F_2=2xF_1$ between the scaling limits of $\nu W_2$ and $W_1$, which forces $R\to 0$ when the constituents are pointlike spin-$1/2$ particles) predicts a small ratio $R=\sigma_L/\sigma_T$ of longitudinal to transverse virtual-photon absorption cross sections; the data showed $R$ small. Any viable dual electromagnetic scheme had to reproduce this.

Nambu (Phys. Rev. D 4, 1193 (1971)) proposed such a scheme. The current vertex is a product of a vector factor $\dot Z_\mu$ (the derivative of the string coordinate along the string, which promotes the internal-mode trajectories $\bar\alpha_n$ to trajectories $\alpha_n$ with intercept one) and a multiplicative scalar factor $V$, producing two exchange-degenerate (coincident but of opposite signature) output trajectory classes with opposite charge-conjugation parity: a C-odd, photon-like trajectory and a C-even, Pomeranchukon-like trajectory (the Pomeranchukon, or Pomeron, is the leading trajectory that dominates high-energy diffractive scattering; C denotes charge conjugation). For the scalar factor $V$ he adopted a Gamma-function ansatz that builds in vector-meson-dominance-type poles in $q^2$ (i.e. form factors as superpositions of vector-meson poles rather than Gaussians), gives power-law elastic form factors, and yields a scaling $\nu W_2$ roughly consistent with early data and with the Drell–Yan relation (a scaling relation connecting the deep-inelastic structure functions to those measured in $e^+e^-$ annihilation), plus a photoproduction cross section $\sigma=8\pi^2\alpha k\approx 110\,\mu\mathrm{b}$ compatible with $\gamma p$ data. But the scheme did not achieve complete factorization in all channels (the two multiplicative factors obstruct it); it left spurious poles at $kt=0$ (spin 1 and 0) and $kt=-1$ (spin 0) in the crossed channels, removable, if at all, by proliferating elementary-field counterterms; and — most seriously — it gave a ratio $R$ growing like $\omega^2$ in the scaling region. Nambu argued this is not an artifact of his ansatz but a general property of the dual model: the scalar vertex couples only to internal modes parallel to $q_\mu$ and enhances longitudinal excitations at large $-q^2$, while transverse excitations enter only through the vector factor. He also noted sign oscillations of the structure functions in the timelike region and exponential growth of amplitudes at large $t$ when the trajectory slope parameter $k\neq 1$ (as required by the $\rho$-meson mass).

In his Discussion he formulates the question literally as: 'Whether complete factorization and gauge invariance are compatible or not is an open question', with the $W_1$ problem in view as the third, realism requirement that the model also fails.

Problem Statement

In a dual-resonance model of hadrons coupled to an external electromagnetic current through current vertex operators (with seagull/contact terms and counterterm elementary fields allowed as part of the scheme), can the following three requirements be satisfied simultaneously?

1. Complete factorization: amplitudes with an arbitrary number of current insertions, on and off the photon mass shell, factorize on the physical spectrum in every channel — direct and crossed — for arbitrary external momenta, so that electromagnetic amplitudes share the hadronic model's state content.
2. Gauge invariance and current conservation: the current generates exact Ward-type identities for off-shell photons (contracting $q^\mu$ reduces multi-current amplitudes to differences of lower-point amplitudes), with no moving poles whose positions depend on external momenta, and with spurious poles such as those at $kt=0$ and $kt=-1$ in Nambu's scheme removable by a controlled, non-proliferating set of counterterms.
3. Realistic deep-inelastic behavior: the structure functions $W_1$ and $W_2$ exhibit Bjorken scaling with a ratio $R=\sigma_L/\sigma_T$ that remains bounded as $-q^2\to\infty$ and is small, consistent with the Callan–Gross relation for pointlike spin-$1/2$ constituents (which gives $R\to 0$; numerically $R\lesssim 0.2$ in the scaling region) and not growing like $\omega^2$, and without the timelike sign oscillations or exponential large-$t$ growth for slope parameter $k\neq 1$ that afflict known schemes.

A scheme belongs to the class under consideration when it couples an external electromagnetic current to a dual-resonance (Veneziano-type) hadronic amplitude through current vertex operators inserted on the world sheet, with seagull/contact terms and elementary-field counterterms permitted; the hadronic mode structure is not required to coincide with the conventional on-shell spectrum. The class further spans an arbitrary number of on- and off-shell current insertions, and each such scheme carries a definite physical spectrum and current algebra. A no-go or a construction valid only for a proper subclass — for example, the conventional on-shell spectrum alone — therefore leaves the question for the full class open.

References

1. Yoichiro Nambu, Electromagnetic Currents in Dual Hadrodynamics, Physical Review D 4, 1193 (1971), DOI 10.1103/PhysRevD.4.1193, https://doi.org/10.1103/PhysRevD.4.1193
2. G. Veneziano, Construction of a crossing-symmetric, Regge-behaved amplitude for linearly rising trajectories, Il Nuovo Cimento A 57, 190 (1968), DOI 10.1007/BF02824451, https://doi.org/10.1007/BF02824451
3. Jamal T. Manassah and Satoshi Matsuda, Physical and Formal Presentations of Dual Theory of Electromagnetic Currents, Physical Review D 4, 3062 (1971), DOI 10.1103/PhysRevD.4.3062, https://doi.org/10.1103/PhysRevD.4.3062
4. P. Ramond, Electromagnetic Interaction of Dual Systems, Physical Review D 5, 2078 (1972), DOI 10.1103/PhysRevD.5.2078, https://doi.org/10.1103/PhysRevD.5.2078
5. K. Kikkawa and B. Sakita, Dual-Resonance Model of Vector Currents: A Representation of Current Algebra, Physical Review D 6, 1734 (1972), DOI 10.1103/PhysRevD.6.1734, https://doi.org/10.1103/PhysRevD.6.1734
6. A. Neveu and J. Scherk, Currents and Green's functions for dual models — I, Nuclear Physics B 41, 365 (1972), DOI 10.1016/0550-3213(72)90399-9, https://doi.org/10.1016/0550-3213(72)90399-9
7. I. T. Drummond, Dual amplitudes for currents, Nuclear Physics B 35, 269 (1971), DOI 10.1016/0550-3213(71)90146-5, https://doi.org/10.1016/0550-3213(71)90146-5
8. John H. Schwarz, Off-mass-shell dual amplitudes without ghosts, Nuclear Physics B 65, 131 (1973), DOI 10.1016/0550-3213(73)90278-2, https://doi.org/10.1016/0550-3213(73)90278-2
9. John H. Schwarz, Dual resonance theory, Physics Reports 8, 269 (1973), DOI 10.1016/0370-1573(73)90003-3, https://doi.org/10.1016/0370-1573(73)90003-3
10. P. V. Collins and K. A. Friedman, Off-shell amplitudes and currents in the dual-resonance model, Il Nuovo Cimento A 28, 173 (1975), DOI 10.1007/bf02820878, https://doi.org/10.1007/bf02820878
11. E. Corrigan and D. B. Fairlie, Off-shell states in dual resonance theory, Nuclear Physics B 91, 527 (1975), DOI 10.1016/0550-3213(75)90125-x, https://doi.org/10.1016/0550-3213(75)90125-x
12. R. Horsley, The group theoretic structure of the dual current vertex, Nuclear Physics B 138, 474 (1978), DOI 10.1016/0550-3213(78)90393-0, https://doi.org/10.1016/0550-3213(78)90393-0
13. G. Schierholz and M. G. Schmidt, Dual model with Mandelstam analyticity for deep-inelastic electroproduction and annihilation, Physical Review D 10, 175 (1974), DOI 10.1103/PhysRevD.10.175, https://doi.org/10.1103/PhysRevD.10.175
14. R. C. Brower and L. Susskind, Current Algebra in the Dual Parton Model, Physical Review D 7, 1032 (1973), DOI 10.1103/PhysRevD.7.1032, https://doi.org/10.1103/PhysRevD.7.1032
15. Edward Corrigan, Aspects of dual models many years ago, arXiv:0810.2480 (2008), https://arxiv.org/abs/0810.2480
16. Oleg Andreev, Scaling laws in hadronic processes and string theory, Physical Review D 67, 046001 (2003), arXiv:hep-th/0209256, DOI 10.1103/PhysRevD.67.046001, https://arxiv.org/abs/hep-th/0209256