Consistent interactions and cluster separability in Dirac's three forms of relativistic dynamics

In his 1949 classification of relativistic Hamiltonian dynamics into the instant, point and front forms, Dirac left open which interactions can be made consistent: writing interactions into the dynamical Poincare generators must preserve the ten bracket relations, whose quadratic consistency conditions he called the real difficulty, and an additional, then-unformulated condition requires interactions to become small between space-like separated systems. The later literature sharpened the question decisively. No-interaction theorems (1963-1990) show that no interaction is possible at all if particle positions are canonical and transform as space-time points; the Bakamjian-Thomas construction (1953) realizes interactions through the invariant mass operator but violates cluster separability - the precise modern form of Dirac's large-separation requirement - for three or more particles; and Sokolov's packing operators (1978) solve the cluster problem only formally. This lineage leaves a precise, nonempty open core attested by the current literature: an explicit, computable, exactly Poincare-invariant and cluster-separable Hamiltonian dynamics for three or more particles with spin and identical particles has never been constructed - packing operators are explicit only for three spinless distinguishable particles, and no dynamical few-body calculation has ever used the full Sokolov construction - while consistent electromagnetic current operators and the quantification of separability corrections for strongly bound systems remain open directions of the same program. The problem is open.

Background

A relativistic theory must give the same observable probabilities in every inertial frame. For quantum theory on a Hilbert space, Wigner (1939) and Bargmann (1954) showed that this holds if and only if states in different frames are related by a single-valued unitary representation of the Poincare group - the group of space-time translations and Lorentz transformations of flat Minkowski space-time. Such a representation is generated by ten self-adjoint quantities: the four energy-momentum components $P_\mu$ and the six independent components $M_{\mu\nu}$ of the angular-momentum tensor (three spatial rotations and three Lorentz boosts), satisfying the commutation relations of the Poincare algebra. In a classical Hamiltonian theory the same structure appears with Poisson brackets (the antisymmetric bilinear bracket of classical mechanics) in place of commutators; Dirac's 1949 paper, the source of the present problem, was written in this classical language with the transition to quantum theory explicitly deferred.

Dirac's central observation is that a Hamiltonian theory places the state of a system on a three-dimensional space-like surface, and the choice of surface determines which generators are kinematic - expressible without interaction - and which must act as Hamiltonians carrying the interaction. He distinguished three forms of dynamics: the instant form (surface $t=0$), in which spatial translations and rotations are kinematic while the Hamiltonian $P_0$ and the three boosts are dynamical; the point form (hyperboloid $t^2-\mathbf{x}^2=\kappa^2$), in which all Lorentz transformations are kinematic and all four components of $P_\mu$ are dynamical; and the front form (light front $x_0+x_3=0$), in which seven generators, including two boost combinations, are kinematic, only three Hamiltonians remain, and no square roots appear. Writing an interaction term $V$ into the dynamical generators is constrained linearly by covariance, but the brackets of two dynamical generators - for instance the bracket of $V$ with the interaction part $V_r$ of a boost in the instant form - impose quadratic consistency conditions on the interaction, which Dirac identified as the real difficulty; he gave no general method for solving them, leaving open which interactions admit consistent solutions in each form and whether all three forms can be realized for realistic atomic systems. The source paper adds two further requirements: interactions between physical systems separated by a large space-like distance should become small, a condition Dirac found difficult to formulate mathematically (strict localizability of variables possibly being too strong, a fundamental length possibly being needed); and, for fields, the point-form and front-form surfaces can fail to support all wave configurations - packets that never intersect the hyperboloid, or waves propagating exactly along the front - so that extra variables are needed, Fourier modes resolving this only for free fields.

The later literature gave these questions precise shape. First, the no-interaction theorem of Currie, Jordan and Sudarshan (1963), refined by Leutwyler (1965), states that a classical relativistic Hamiltonian particle dynamics whose position variables are canonical and transform as space-time points (the world-line condition) admits no interaction at all; Jaen, Molina and Iranzo (1986) extended the theorem to the front and point forms, and Li (1990) to any number of particles. Interacting dynamics therefore exists only when the positions are allowed to be non-canonical. Second, Bakamjian and Thomas (1953) exhibited the first quantum realization: placing the interaction in the invariant mass operator $M$ (the square root of the Poincare Casimir $P_\mu P^\mu$, whose eigenvalues are the rest masses), commuting with the kinematic generators, makes all ten generators close the algebra automatically. This works exactly for two particles; Sokolov (1978) and Coester and Polyzou (1982) generalized it, and Sokolov and Shatnii (1978) proved the physical equivalence of the three forms, converting the choice of form into a matter of convenience. Third, Dirac's large-separation requirement became cluster separability (also called macroscopic causality): when the interactions coupling two groups of particles are switched off, the unitary Poincare representation must become the tensor product of the subsystem representations. The Bakamjian-Thomas construction violates this for three or more particles; Sokolov (1978) gave a formal solution by packing operators - unitary transformations that redistribute interaction between the mass operator and the four-velocity operator for each clustering; Keister and Polyzou (1991) systematized the graded separability notions (for the generators or only for the S-matrix, i.e. the scattering operator, with different limiting senses); and Polyzou (2002) gave a general construction for any number of particles satisfying Poincare invariance, cluster properties and the spectral condition (mass spectrum bounded below), formally.

What remains unsettled, and what the modern literature itself names as the problem, is explicitness and realism. Packing operators have been computed explicitly only for three spinless, distinguishable particles (Reichelt, Schweiger and Klink 2017); no dynamical few-body calculation has ever used the full Sokolov construction (Polyzou and Keister 2011-2012); electromagnetic current operators consistent simultaneously with covariance, separability and charge normalization have been constructed by Lev (1995) for two and three particles, with existence claimed for any number of particles, but within the same formal packing-operator framework, and their unphysical spectator-momentum dependence in applications is still under investigation; and for four or more particles the Sokolov construction does not even preserve the S-matrix. On the field side, the front form grew into light-front quantum field theory, where Dirac's extra-variables issue survives as the zero-mode/vacuum-sector problem: field components independent of the coordinate along the front are not fixed by the canonical data and carry vacuum information that the on-shell light-front Hamiltonian cannot see (Mannheim, Lowdon and Brodsky 2021).

Problem Statement

The problem is the still-open core of Dirac's 1949 program of interacting relativistic Hamiltonian dynamics. As posed in the source: for each of the three forms of dynamics (instant, point, front), determine which interactions can be made consistent - the quadratic Poisson-bracket, later commutator, conditions on the interaction terms being the essential difficulty - subject to the additional requirement that interactions between systems separated by a large space-like distance become small, and realize the result for realistic quantum systems.

The intervening results summarized in Background settled parts of this: interactions exist in all three forms once the particle positions are non-canonical (the three forms being physically equivalent), the large-separation requirement is now formulated precisely as cluster separability, and formal solutions of the consistency-plus-separability conditions exist (Sokolov's packing operators; Polyzou's general N-particle construction). The explicit, computable dynamics that the source's 'realistic atomic systems' demands is still missing. What remains is an interacting, exactly Poincare-invariant Hamiltonian dynamics for a system of three or more particles, including particles with spin and identical particles, such that:

1. the ten generators satisfy the Poincare algebra exactly, i.e. the quadratic consistency conditions on the interaction terms hold; since the source poses the question for each of the three forms, the construction is realized in each of Dirac's forms or carried to the other forms explicitly, rather than assumed to transfer;

2. cluster separability holds: for every partition of the particles, setting the inter-cluster couplings to zero reduces the unitary Poincare representation to the tensor product of the subsystem representations, with the subsystem scattering operators (S-matrices) preserved;

3. the construction is explicit enough to compute - the interacting generators and the unitary transformations realizing cluster separability, whether Sokolov-type packing operators, operators in the sense of Polyzou (2002), or any other construction, must be given explicitly rather than merely asserted to exist, and must extend beyond the currently explicit case of three spinless, distinguishable particles to a construction valid for three or more particles with spin and identical particles.

The three conditions have the common objective of a usable relativistic interacting Hamiltonian particle dynamics: condition 1 is the consistency requirement Dirac emphasized, condition 2 is his large-separation requirement in its modern precise form, and condition 3 is what 'realistic systems' and the deferred quantum transition now demand. Consistent electromagnetic current operators (covariance, cluster separability and charge normalization) and the quantification of separability corrections, including for strongly bound systems, are literature-attested open directions and motivations of the program, additional to the three conditions above. It also remains open whether such a dynamics is provably impossible: whether an independently motivated axiom set exists, one that is not already settled by the known results summarized in Background and that admits nontrivial interacting models (and does not itself stipulate the absence of interaction or of a non-trivial S-matrix), within which such a dynamics is provably impossible and a single principle whose removal restores existence can be named (as local quantum field theory gives up fixed particle number). The source's scope also covers the application of the forms to the electromagnetic field, including the completeness issue for front-form and point-form variables (extra variables / zero modes) for interacting fields; this field-theoretic component, like the electromagnetic-current directions above, is part of the source's wider scope rather than a required part of the answer to the three conditions: light-front and instant-time quantization are known to be unitarily equivalent in the non-vacuum sector, while light-front vacuum diagrams are not describable by the on-shell Hamiltonian formalism (Mannheim, Lowdon and Brodsky 2021), so the zero-mode/vacuum sector remains open.

References

1. E. P. Wigner, 'On Unitary Representations of the Inhomogeneous Lorentz Group', Annals of Mathematics 40, 149 (1939). DOI: 10.2307/1968551. URL: https://doi.org/10.2307/1968551
2. V. Bargmann, 'On Unitary Ray Representations of Continuous Groups', Annals of Mathematics 59, 1 (1954). DOI: 10.2307/1969831. URL: https://doi.org/10.2307/1969831
3. P. A. M. Dirac, 'Forms of Relativistic Dynamics', Reviews of Modern Physics 21, 392-399 (1949). DOI: 10.1103/RevModPhys.21.392. URL: https://doi.org/10.1103/RevModPhys.21.392
4. B. Bakamjian and L. H. Thomas, 'Relativistic Particle Dynamics. II', Physical Review 92, 1300-1310 (1953). DOI: 10.1103/PhysRev.92.1300. URL: https://doi.org/10.1103/PhysRev.92.1300
5. D. G. Currie, T. F. Jordan and E. C. G. Sudarshan, 'Relativistic Invariance and Hamiltonian Theories of Interacting Particles', Reviews of Modern Physics 35, 350-375 (1963). DOI: 10.1103/RevModPhys.35.350. URL: https://doi.org/10.1103/RevModPhys.35.350
6. H. Leutwyler, 'A no-interaction theorem in classical relativistic Hamiltonian particle mechanics', Il Nuovo Cimento 37, 556-567 (1965). DOI: 10.1007/BF02749856. URL: https://doi.org/10.1007/BF02749856
7. S. N. Sokolov, 'Relativistic addition of direct interactions in the point form of dynamics', Theoretical and Mathematical Physics 36, 682-692 (1978). DOI: 10.1007/BF01036481. URL: https://doi.org/10.1007/BF01036481
8. S. N. Sokolov and A. N. Shatnii, 'Physical equivalence of the three forms of relativistic dynamics and addition of interactions in the front and instant forms', Theoretical and Mathematical Physics 37, 1029-1038 (1978). DOI: 10.1007/BF01018583. URL: https://doi.org/10.1007/BF01018583
9. H. Sazdjian, 'Relativistic and separable classical Hamiltonian particle dynamics', Annals of Physics 136, 136-189 (1981). DOI: 10.1016/0003-4916(81)90088-9. URL: https://doi.org/10.1016/0003-4916(81)90088-9
10. F. Coester and W. N. Polyzou, 'Relativistic quantum mechanics of particles with direct interactions', Physical Review D 26, 1348 (1982). DOI: 10.1103/PhysRevD.26.1348. URL: https://doi.org/10.1103/PhysRevD.26.1348
11. X. Jaen, A. Molina and V. Iranzo, 'Front form and point form formulation of predictive relativistic mechanics. Noninteraction theorems', Journal of Mathematical Physics 27, 512-518 (1986). DOI: 10.1063/1.527200. URL: https://doi.org/10.1063/1.527200
12. Fu-Bin Li, 'Proof of the noninteraction theorem for a system of N relativistic particles in direct interaction', Journal of Mathematical Physics 31, 1395-1399 (1990). DOI: 10.1063/1.528726. URL: https://doi.org/10.1063/1.528726
13. B. D. Keister and W. N. Polyzou, 'Relativistic Hamiltonian dynamics in nuclear and particle physics', in J. W. Negele and E. Vogt (eds.), Advances in Nuclear Physics, Vol. 20, pp. 225-479, Plenum Press, New York (1991). Book ISBN 978-1-4613-9910-0, book DOI: 10.1007/978-1-4613-9910-0. URL: https://link.springer.com/book/10.1007/978-1-4613-9910-0
14. F. Coester and W. N. Polyzou, 'Relativistic Quantum Dynamics of Many-Body Systems', arXiv:nucl-th/0102050 (2001). URL: https://arxiv.org/abs/nucl-th/0102050
15. W. N. Polyzou, 'Cluster properties in relativistic quantum mechanics of N-particle systems', Journal of Mathematical Physics 43, 6024-6063 (2002). DOI: 10.1063/1.1516627. arXiv:nucl-th/0201013. URL: https://arxiv.org/abs/nucl-th/0201013
16. B. Desplanques, 'Dirac's inspired point form and hadron form factors', Nuclear Physics A 755, 303-306 (2005). DOI: 10.1016/j.nuclphysa.2005.03.032. arXiv:nucl-th/0501051. URL: https://arxiv.org/abs/nucl-th/0501051
17. W. N. Polyzou and B. D. Keister, 'Constraints of cluster separability and covariance on current operators', Few-Body Systems 52, 427-430 (2012). DOI: 10.1007/s00601-011-0302-1. arXiv:1110.1367. URL: https://arxiv.org/abs/1110.1367
18. B. D. Keister and W. N. Polyzou, 'Model tests of cluster separability in relativistic quantum mechanics', Physical Review C 86, 014002 (2012). DOI: 10.1103/PhysRevC.86.014002. arXiv:1109.6575. URL: https://arxiv.org/abs/1109.6575
19. W. N. Polyzou and B. D. Keister, 'Cluster properties in Poincare invariant quantum mechanics', arXiv:1210.1636 [nucl-th] (2012). URL: https://arxiv.org/abs/1210.1636
20. N. Reichelt, W. Schweiger and W. H. Klink, 'Cluster Separability in Relativistic Few Body Problems', arXiv:1711.04555 [nucl-th] (2017). URL: https://arxiv.org/abs/1711.04555
21. W. Schweiger, N. Reichelt and W. H. Klink, 'The Problem of Cluster Separability in Relativistic Few-Body Systems', Springer Proceedings in Physics, pp. 801-805 (2020). DOI: 10.1007/978-3-030-32357-8_125. arXiv:1811.09405. URL: https://arxiv.org/abs/1811.09405
22. F. M. Ciaglia, F. Di Cosmo, A. Ibort and G. Marmo, 'Descriptions of Relativistic Dynamics with World Line Condition', Quantum Reports 1, 181-192 (2019). DOI: 10.3390/quantum1020016. arXiv:1910.06447. URL: https://arxiv.org/abs/1910.06447
23. P. D. Mannheim, P. Lowdon and S. J. Brodsky, 'Comparing light-front quantization with instant-time quantization', Physics Reports 891, 1-65 (2021). DOI: 10.1016/j.physrep.2020.09.001. arXiv:2005.00109. URL: https://arxiv.org/abs/2005.00109
24. F. M. Lev, 'Exact Construction of the Electromagnetic Current Operator for Relativistic Composite Systems', Annals of Physics 237, 355-419 (1995). DOI: 10.1006/aphy.1995.1013. arXiv:hep-ph/9403222. URL: https://arxiv.org/abs/hep-ph/9403222