Soft propagators and the smooth $m\to 0$ limit of massive spin-2 (graviton) theories coupled to realistic interacting matter

The 1975 letter of Delbourgo and Salam introduced the Stueckelberg formalism for spin 2 — rewriting the massive spin-2 (graviton) field $h_{\mu\nu}$ as $f_{\mu\nu}+(\partial_\mu V_\nu+\partial_\nu V_\mu)/m$ with a vector Stueckelberg field, in the manner long known for spin 1 — and asked whether it yields a massive gravity with soft propagators, that is, propagators possessing a well-defined $m\to 0$ limit. Their answer was negative: the reformulated free theory is equivalent to the Fierz–Pauli massive spin-2 theory only if the source $\theta_{\mu\nu}$ satisfies $(n-2)\partial^\mu\partial^\nu\theta_{\mu\nu}=m^2\theta^\lambda{}_\lambda$ (with $n$ the spacetime dimension), a condition that realistic interacting matter fails once the total stress tensor — including the spin-2 field's own contribution — is used as the source. Later literature confirmed and sharpened this picture. The van Dam–Veltman–Zakharov discontinuity makes the linearized $m\to 0$ limit of local, Lorentz-invariant, Fierz–Pauli-type theories discontinuous; the Vainshtein mechanism restores agreement with general relativity only nonperturbatively and non-analytically, with strong coupling that grows without bound in the limit; nonlinear completions propagate the Boulware–Deser ghost until the dRGT construction removed it, at the price of a strong-coupling scale $\Lambda_3=(m^2M_\mathrm{Pl})^{1/3}$ that itself vanishes as $m\to 0$. Whether any framework — Lorentz-violating, nonlocal, non-Fierz–Pauli, or higher-dimensional — can realize a consistent massive spin-2 theory coupled to realistic interacting matter with genuinely soft propagators and a smooth reduction to massless general relativity remains open: no accepted construction exists, and no theorem rules one out.

Background

Massive gravity is the program of modifying general relativity (GR) so that the gravitational field — the spin-2 particle, or graviton, described by the symmetric tensor field $h_{\mu\nu}$ — acquires a mass $m$ while the theory remains consistent. The unique free (linear) theory of a massive spin-2 field is the Fierz–Pauli Lagrangian of 1939, which propagates exactly five polarizations with positive norm; alternative linear mass terms propagate a ghost, a negative-norm state signaling an inconsistency. This record concerns the structural question, first posed sharply by Delbourgo and Salam in 1975, of whether a massive spin-2 theory can have a smooth massless limit — a question about the interacting theory, not the free one.

The Stueckelberg formalism (introduced for spin 1 in 1938 and reviewed by Ruegg and Ruiz-Altaba) is a device that restores gauge invariance to a massive gauge theory: for a massive vector one introduces an additional scalar field so that the mass term becomes gauge invariant under a shifted symmetry, and the $m\to 0$ limit is then smooth because the massive vector propagator decomposes into a soft piece — one with a well-defined limit as $m\to 0$, free of terms that diverge like $k_\mu k_\nu/m^2$ — plus pieces controlled by the scalar. Delbourgo and Salam constructed the spin-2 analogue: the massive field is written as $h_{\mu\nu}=f_{\mu\nu}+(\partial_\mu V_\nu+\partial_\nu V_\mu)/m$ with a vector Stueckelberg field $V_\mu$, the linear gauge invariance of massless gravity is restored, and the free propagators of $f_{\mu\nu}$ and $V_\mu$ are both soft. However, the propagator of the physical combination $h_{\mu\nu}$ then agrees with the Fierz–Pauli propagator — the one that couples correctly to an external conserved current $\theta_{\mu\nu}$ — only up to an extra finite term that vanishes precisely when the current obeys $(n-2)\partial^\mu\partial^\nu\theta_{\mu\nu}=m^2\theta^\lambda{}_\lambda$, with $n$ the spacetime dimension. For certain free-field combinations this condition can hold, but in a realistic interacting theory the graviton couples universally to the total stress tensor, which includes the spin-2 field's own contribution and generically fails the condition: Delbourgo and Salam concluded that, for example, scalar matter with Lagrangian $\tfrac12(\partial\phi)^2+\lambda\phi^{2n/(n-2)}$ in $n$ dimensions, and QED in four dimensions, do not acquire soft propagators from the trick.

This obstruction is the source-paper version of the van Dam–Veltman–Zakharov (vDVZ) discontinuity, established in 1970: for any $m>0$ the linearized exchange amplitude between conserved sources in Fierz–Pauli gravity differs from that of massless GR by an $O(1)$ amount — the helicity-0 polarization contained in the massive graviton couples to the trace of the stress tensor, changing, for instance, the bending of light by 25% — so the linearized $m\to 0$ limit is not the massless theory. Vainshtein showed in 1972 that the discontinuity is an artifact of perturbation theory: the helicity-0 mode's own nonlinear interactions screen its coupling to matter within a Vainshtein radius that grows without bound as $m\to 0$, so the full classical predictions can recover those of GR — but the recovery is non-analytic in $m$, and the interactions responsible become strongly coupled at a scale that vanishes in the limit. (A theory is strongly coupled at a scale $\Lambda$ when interactions become as important as kinetic terms at energies of order $\Lambda$; $\Lambda\to 0$ means the perturbative expansion collapses as $m\to 0$.) In the same year Boulware and Deser showed that generic nonlinear completions of the Fierz–Pauli theory propagate a sixth, ghost-like polarization, now called the Boulware–Deser ghost. The modern Stueckelberg formulation of massive gravity — decomposing the metric into helicity-2, helicity-1 and helicity-0 fields, revived by Arkani-Hamed, Georgi and Schwartz in 2003 and completed by de Rham, Gabadadze and Tolley in 2011 into ghost-free (dRGT) massive gravity — is the direct descendant of Delbourgo and Salam's construction. It eliminates the Boulware–Deser ghost, but its strong-coupling scale $\Lambda_3=(m^2M_\mathrm{Pl})^{1/3}$ (with $M_\mathrm{Pl}$ the Planck mass) goes to zero as $m\to 0$, so the interacting theory's massless limit remains singular.

Within local, Lorentz-invariant, ghost-free massive gravity on an asymptotically flat background, the negative answer is thus settled in two complementary senses — discontinuity of the linearized limit, non-analyticity and strong coupling of the full limit. What has never been settled is the general existence question that the 1975 result poses: is there any consistent framework — breaking Lorentz invariance, locality, the Fierz–Pauli structure, or embedding the graviton in extra dimensions — in which a massive spin-2 field coupled to realistic interacting matter has soft propagators and reduces smoothly to massless GR as $m\to 0$? Works cited below do claim progress on individual sub-questions — a nonlocal massive gravity without the vDVZ discontinuity, non-Fierz–Pauli theories claimed to avoid it, and a higher-dimensional embedding claimed to remove both the vDVZ and the strong-coupling problems — but none is accepted as resolving the question posed here: each leaves the soft-propagator and smooth-limit requirement unestablished, carries an unresolved consistency question of its own, or is reported only in unrefereed form.

Problem Statement

Determine whether there exists a consistent relativistic theory of a massive spin-2 field (massive graviton) of mass $m$, coupled to realistic interacting matter — such as QED or interacting scalar matter — through the universally conserved total stress tensor, which necessarily includes the spin-2 field's own stress tensor, such that: (i) the propagators of the theory are soft, meaning each admits a well-defined $m\to 0$ limit with no pieces that diverge like $k_\mu k_\nu/m^2$ or otherwise blow up; and (ii) the theory's classical predictions and on-shell amplitudes reduce continuously to those of massless general relativity as $m\to 0$. A consistent theory meeting (i) and (ii) would be specified by an explicit Lagrangian or set of equations of motion, would propagate exactly five healthy degrees of freedom (in particular, no Boulware–Deser ghost or other ghost and no acausal mode), and would have a strong-coupling or cutoff scale that does not vanish as $m\to 0$. A negative resolution would be a no-go over a precisely stated class of theories — with explicit assumptions on locality, Lorentz invariance, unitarity, background, and the structure of the mass term and of the matter coupling — establishing that within that class the $m\to 0$ limit is necessarily discontinuous or the interacting theory necessarily becomes strongly coupled, the stated assumptions delimiting which of the known circumvention candidates (Lorentz-violating mass terms, nonlocal or infinite-derivative form factors, non-Fierz–Pauli mass terms, anti-de Sitter backgrounds, higher-dimensional embeddings) the class includes and which it excludes. The question ranges over every such framework — including the circumvention candidates — so a result confined to a subclass of theories bears only on that subclass, candidates outside it are not addressed, and a no-go restricted to a class that the Background already treats as settled adds nothing beyond that class. The Stueckelberg reformulation of Fierz–Pauli gravity does not by itself answer the question: it restores the gauge symmetry and gives soft constituent propagators, but equivalence to the Fierz–Pauli theory coupled to a realistic source requires the current condition $(n-2)\partial^\mu\partial^\nu\theta_{\mu\nu}=m^2\theta^\lambda{}_\lambda$ (with $n$ the spacetime dimension), which fails once the source is the total stress tensor, since the spin-2 field's own stress tensor contribution cannot maintain it.

References

1. R. Delbourgo and Abdus Salam. The Stueckelberg formalism for spin two. Lettere al Nuovo Cimento 12, 297-299 (1975). DOI: 10.1007/bf02790604. https://doi.org/10.1007/bf02790604 — The source paper.
2. M. Fierz and W. Pauli. On relativistic wave equations for particles of arbitrary spin in an electromagnetic field. Proceedings of the Royal Society of London A 173, 211-232 (1939). DOI: 10.1098/rspa.1939.0140. https://doi.org/10.1098/rspa.1939.0140
3. E. C. G. Stueckelberg. Die Wechselwirkungskräfte in der Elektrodynamik und in der Feldtheorie der Kernkräfte (Teil I). Helvetic Physica Acta 11, 225-244 (1938). No DOI. Bibliographic record: https://inspirehep.net/literature/45337 — The origin of the Stueckelberg formalism for spin 1; the first reference of the source letter.
4. H. van Dam and M. Veltman. Massive and mass-less Yang-Mills and gravitational fields. Nuclear Physics B 22, 397-411 (1970). DOI: 10.1016/0550-3213(70)90416-5. https://doi.org/10.1016/0550-3213(70)90416-5
5. V. I. Zakharov. Linearized gravitation theory and the graviton mass. Journal of Experimental and Theoretical Physics Letters 12, 312 (1970). No DOI. Bibliographic record: https://inspirehep.net/literature/63463
6. A. I. Vainshtein. To the problem of nonvanishing gravitation mass. Physics Letters B 39, 393-394 (1972). DOI: 10.1016/0370-2693(72)90147-5. https://doi.org/10.1016/0370-2693(72)90147-5
7. D. G. Boulware and S. Deser. Inconsistency of finite range gravitation. Physics Letters B 40, 227-229 (1972). DOI: 10.1016/0370-2693(72)90418-2. https://doi.org/10.1016/0370-2693(72)90418-2
8. N. Arkani-Hamed, H. Georgi, and M. D. Schwartz. Effective field theory for massive gravitons and gravity in theory space. Annals of Physics 305, 96-118 (2003). DOI: 10.1016/S0003-4916(03)00068-X. arXiv:hep-th/0210184. https://arxiv.org/abs/hep-th/0210184
9. H. Ruegg and M. Ruiz-Altaba. The Stueckelberg field. International Journal of Modern Physics A 19, 3265-3347 (2004). DOI: 10.1142/S0217751X04019755. arXiv:hep-th/0304245. https://arxiv.org/abs/hep-th/0304245
10. J. B. Pitts and W. C. Schieve. Universally coupled massive gravity. Theoretical and Mathematical Physics 151, 700-717 (2007). DOI: 10.1007/s11232-007-0055-7. arXiv:gr-qc/0503051. https://arxiv.org/abs/gr-qc/0503051
11. C. de Rham, G. Gabadadze, and A. J. Tolley. Resummation of massive gravity. Physical Review Letters 106, 231101 (2011). DOI: 10.1103/PhysRevLett.106.231101. arXiv:1011.1232. https://arxiv.org/abs/1011.1232
12. N. Kaloper, A. Padilla, and N. Tanahashi. Galileon hairs of Dyson spheres, Vainshtein's coiffure and hirsute bubbles. Journal of High Energy Physics 10, 148 (2011). DOI: 10.1007/JHEP10(2011)148. arXiv:1106.4827. https://arxiv.org/abs/1106.4827
13. M. Jaccard, M. Maggiore, and E. Mitsou. Nonlocal theory of massive gravity. Physical Review D 88, 044033 (2013). DOI: 10.1103/PhysRevD.88.044033. arXiv:1305.3034. https://arxiv.org/abs/1305.3034
14. E. Babichev and C. Deffayet. An introduction to the Vainshtein mechanism. Classical and Quantum Gravity 30, 184001 (2013). DOI: 10.1088/0264-9381/30/18/184001. arXiv:1304.7240. https://arxiv.org/abs/1304.7240
15. C. de Rham. Massive gravity. Living Reviews in Relativity 17, 7 (2014). DOI: 10.12942/lrr-2014-7. arXiv:1401.4173. https://arxiv.org/abs/1401.4173
16. Z. Kakushadze. No vDVZ discontinuity in non-Fierz-Pauli theories. University Journal of Physics and Applications 8(7), 344-349 (2014). arXiv:1402.6989. https://arxiv.org/abs/1402.6989
17. A. N. Ivanov, M. Pitschmann, and M. Wellenzohn. Effective low-energy gravitational potential for slow fermions coupled to linearised massive gravity. arXiv:1511.07650 (2015). https://arxiv.org/abs/1511.07650
18. N. Moynihan and J. Murugan. Comments on scattering in massive gravity, vDVZ and BCFW. Classical and Quantum Gravity 35, 155005 (2018). DOI: 10.1088/1361-6382/aacb73. arXiv:1711.03956. https://arxiv.org/abs/1711.03956
19. G. Gabadadze, D. Older, and D. Pirtskhalava. Resolving the van Dam-Veltman-Zakharov and strong coupling problems in massive gravity and bigravity. Physical Review D 100, 124017 (2019). DOI: 10.1103/PhysRevD.100.124017. arXiv:1907.13491 (preprint listed under the variant title 'Resolving the vDVZ and Strong Coupling Problems in Massive Gravity and Bigravity'). https://arxiv.org/abs/1907.13491
20. C. de Rham, J. Kożuszek, A. J. Tolley, and T. Wiseman. Dynamical formulation of ghost-free massive gravity. Physical Review D 108, 084052 (2023). DOI: 10.1103/PhysRevD.108.084052. arXiv:2302.04876. https://arxiv.org/abs/2302.04876
21. Th. A. Rijken and J. W. Wagenaar. Massive spin-2: field-equations, propagators, massless-limit, and perihelion precessions. arXiv:2512.02050 (2025). https://arxiv.org/abs/2512.02050