Nambu's 1973 generalized Hamiltonian dynamics replaces the Poisson bracket and a single Hamiltonian by a totally antisymmetric ternary bracket, the Jacobian ∂(F,H,G)/∂(x,y,z)\partial(F,H,G)/\partial(x,y,z), and two Hamiltonians. The source paper left open whether this mechanics admits a bona fide quantization: an operator realization of the ternary bracket that is totally antisymmetric, acts as a derivation (Leibniz law) on the observable product, satisfies the consistency identities of the classical bracket (later codified as the fundamental identity), admits central Nambu-Heisenberg relations such as [X,Y,Z]=i[X,Y,Z]=i, and reproduces the classical Nambu bracket in the correspondence limit, without collapsing to ordinary Heisenberg quantum mechanics with a single effective Hamiltonian. Fifty years of subsequent work -- Nambu-Poisson geometry, Filippov n-Lie algebras, deformation and Zariski quantization, quantum Nambu brackets, 3-algebras in M2-brane physics -- has produced partial frameworks, obstruction results, and consistent reformulations that change the desiderata, but no accepted solution and no definitive no-go theorem. The associated structural questions Nambu flagged likewise remain unsettled in general: the weaker combinations of algebraic conditions he enumerated for the operator bracket, and nonassociative (Jordan/octonionic) realizations. The problem remains open.

Background

Classical Hamiltonian mechanics describes a system with an even-dimensional phase space of canonical coordinate pairs (qi,pi)(q_i,p_i): observables are functions on phase space, and time evolution is generated by a single Hamiltonian HH through the Poisson bracket, a bilinear, antisymmetric operation satisfying the Jacobi identity. The antisymmetry makes HH itself conserved, and the Jacobi identity guarantees the consistency of the generated dynamics; the associated flow preserves phase-space volume (Liouville theorem), which underlies equilibrium statistical mechanics. Canonical quantization promotes observables to operators and the Poisson bracket to the commutator divided by iℏi\hbar.

Nambu's 1973 proposal generalizes this picture to a three-dimensional phase space with a single canonical triplet r⃗=(x,y,z)\vec r=(x,y,z) and two Hamiltonians H(x,y,z)H(x,y,z), G(x,y,z)G(x,y,z):

dr⃗dt=∇⃗H×∇⃗G,dFdt=[F,H,G]≡∂(F,H,G)∂(x,y,z),\frac{d\vec r}{dt}=\vec\nabla H\times\vec\nabla G, \qquad \frac{dF}{dt}=[F,H,G]\equiv\frac{\partial(F,H,G)}{\partial(x,y,z)},

where the Nambu bracket [F,H,G][F,H,G] is the Jacobian determinant of the three functions. The bracket is totally antisymmetric, both HH and GG are conserved, and the velocity field is divergence-free, giving a Liouville theorem and equilibrium ensembles with weight exp⁡(−βH−γG)\exp(-\beta H-\gamma G). Canonical transformations are the unimodular (unit-Jacobian) maps, with linear ones forming SL(3,R)\mathrm{SL}(3,\mathbb{R}), and the torque-free Euler top is realized exactly by taking r⃗\vec r as the body-frame angular momentum, H=L2/2H=L^2/2 and GG the kinetic energy. The formalism extends to nn-dimensional phase spaces with n−1n-1 Hamiltonians and the nn-fold Jacobian, although in the source only the three-dimensional case is worked out in detail.

The quantization question is the paper's central acknowledged limitation. For an operator (quantum) version one wants a trilinear bracket on observables that (a) is totally antisymmetric, (b) obeys the derivation or Leibniz law in each argument with respect to operator multiplication, so that the bracket of a product is determined by the brackets of the factors, and (c) satisfies consistency identities guaranteeing that the generated time evolution is well-defined. One would also want a canonical commutation relation analogous to [X,P]=iℏ[X,P]=i\hbar, namely a Nambu-Heisenberg relation [X,Y,Z]=i[X,Y,Z]=i on a quantum triplet with the left side a central c-number (a scalar commuting with every operator) in an irreducible representation. Nambu tested the cyclic antisymmetrized product ABC+BCA+CAB−BAC−ACB−CBAABC+BCA+CAB-BAC-ACB-CBA and found a structural obstruction: imposing the derivation law forces [H,G]=0[H,G]=0, and requiring the relevant commutators to be invertible forces H=αG+βH=\alpha G+\beta with c-numbers α,β\alpha,\beta (scalar operators commuting with everything), so that the evolution i dF/dt=[F,H,G]i\,dF/dt=[F,H,G] reduces to an ordinary Heisenberg equation i dF/dt=[F,K]i\,dF/dt=[F,\mathcal K] with the effective Hamiltonian K=i[H,G]\mathcal K=i[H,G]. He enumerated weaker combinations of the algebraic conditions (several logical possibilities, some inconsistent or uninteresting) and suggested non-associative algebras as a way around the obstruction; Lie-algebra realizations of [X,Y,Z]=i[X,Y,Z]=i exist in SO(4)\mathrm{SO}(4), SO(3,1)\mathrm{SO}(3,1), SO(2,2)\mathrm{SO}(2,2), SO(3)\mathrm{SO}(3), SO(2,1)\mathrm{SO}(2,1) and contracted nonsemisimple algebras, with the cubic Casimir element (a polynomial in the Lie-algebra generators that commutes with all of them) playing the role of the central c-number; and the exceptional Jordan algebra (the algebra of 3×33\times3 Hermitian matrices over the octonions, the eight-dimensional non-associative normed division algebra; it is a commutative but non-associative algebra with a determinant-based triple product) was left as an unexplored possibility.

The mathematical framework was codified later. The consistency condition missing at first sight, the fundamental identity (FI), states that the adjoint action of n−1n-1 bracket entries acts as derivations of the nn-ary bracket:

{f1,…,fn−1,{g1,…,gn}}=∑i=1n{g1,…,{f1,…,fn−1,gi},…,gn},\{f_1,\dots,f_{n-1},\{g_1,\dots,g_n\}\} =\sum_{i=1}^{n}\{g_1,\dots,\{f_1,\dots,f_{n-1},g_i\},\dots,g_n\},

which generalizes the Jacobi identity and guarantees that the bracket of integrals of motion is again an integral of motion. Filippov introduced this identity in pure algebra in his theory of n-Lie algebras, and Takhtajan built the canonical formalism of Nambu mechanics on it, introducing Nambu-Poisson manifolds (manifolds whose function space carries such a bracket) and posing the quantization problem precisely. The classical Jacobian brackets on Rn\mathbb{R}^n satisfy the fundamental identity and the Leibniz rule.

The modern status of the obstruction is as follows. No example is known of a vector space with an nn-ary bracket, n>2n>2, satisfying simultaneously the fundamental identity and the Leibniz rule other than the classical Nambu-Poisson algebra itself. In the deformation-quantization approach (quantization by deforming the pointwise product of observables into a noncommutative star-product), Dito, Flato, Sternheimer and Takhtajan recorded several impossibilities -- Gelfand's theorem (every Abelian involutive Banach algebra is an algebra of continuous functions with its pointwise product) and the triviality of the second Harrison cohomology (the cohomology that classifies deformations of a commutative associative product) forbid nontrivial Abelian associative deformations of the polynomial algebra, and the naive bracket built from fully antisymmetrized operator products fails the fundamental identity -- and then constructed the Zariski quantization, a commutative, associative deformed product on a polynomial algebra of fields in which both the fundamental identity and the Leibniz rule survive; later literature characterizes this as a formal, field-like exception rather than an operator quantum mechanics. Curtright and Zachos showed that quantum Nambu brackets defined as fully antisymmetrized operator products can be used consistently within standard Hilbert-space quantum mechanics for superintegrable systems (systems with more conserved quantities than degrees of freedom), at the price of abandoning the naive Leibniz law -- they introduce derivators to measure its failure -- and of disputing the fundamental identity as the correct quantum consistency condition. Reduced or kinematical quantizations (fuzzy-sphere and matrix regularizations -- the fuzzy sphere being a finite-dimensional matrix algebra approximating the algebra of functions on a sphere -- mapping Nambu-Poisson structures to n-Lie algebras) solve only the kinematical part, and the Bagger-Lambert 3-algebras (ternary-bracket structures used in the gauge-theory description of parallel M2-branes, the two-dimensional membranes of M-theory) use Nambu-bracket-type structures classically. None of these realizes Nambu's original desiderata as stated; whether they can be realized, or are impossible, is the open problem below.

Problem Statement

The problem is the quantization of Nambu's generalized Hamiltonian dynamics, left open by the source paper and unresolved by the subsequent literature: construct a bona fide quantum theory -- or rigorously prove that none exists under a faithful formalization of Nambu's stated desiderata, namely conditions (a)-(f) below -- whose observable algebra carries a ternary bracket such that:

(a) the bracket is totally antisymmetric and multilinear;

(b) it acts as a derivation (Leibniz law) on the observable product in each argument;

(c) it satisfies the consistency identities of the classical Nambu bracket -- the fundamental identity, or a precisely identified replacement identity with a demonstration that the replacement still guarantees consistency of the generated dynamics;

(d) it admits central Nambu-Heisenberg commutation relations, e.g. [X,Y,Z]=iℏ[X,Y,Z]=i\hbar for a canonical triplet in an irreducible representation, with the bracket of the triplet a c-number;

(e) it reproduces the classical Nambu bracket ∂(F,H,G)/∂(x,y,z)\partial(F,H,G)/\partial(x,y,z) in the correspondence limit; and

(f) the resulting quantum dynamics is genuinely distinct from ordinary Heisenberg quantum mechanics: the framework must admit at least one nontrivial Hamiltonian pair for which the generated evolution is not reducible to a Heisenberg equation with a single effective Hamiltonian (such as K=i[H,G]\mathcal K=i[H,G]) and for which neither Hamiltonian is forced to be a c-number (a scalar operator).

The source's own acknowledged limitations frame the remaining structural questions: (i) which of the weaker sets of algebraic conditions enumerated in the source admit consistent, nontrivial operator realizations; and (ii) whether non-associative algebras -- in particular the exceptional Jordan algebra of 3×33\times3 Hermitian octonionic matrices -- realize such a bracket and yield a quantum dynamics genuinely different from standard quantum mechanics while producing a c-number central element like [X,Y,Z]=i[X,Y,Z]=i. Conditions (a)-(f) state the objective to be settled, and these two structural questions belong to the same program: completing, or definitively excluding, the quantum extension of Nambu's classical formalism.

References

  1. Y. Nambu, Generalized Hamiltonian dynamics, Physical Review D 7, 2405-2412 (1973), DOI: 10.1103/PhysRevD.7.2405, https://doi.org/10.1103/PhysRevD.7.2405 (source paper)
  2. V. T. Filippov, n-Lie algebras, Siberian Mathematical Journal 26, 879-891 (1985), DOI: 10.1007/BF00969110, https://doi.org/10.1007/BF00969110
  3. L. Takhtajan, On foundation of the generalized Nambu mechanics, Communications in Mathematical Physics 160, 295-315 (1994), DOI: 10.1007/BF02103278, arXiv:hep-th/9301111, https://arxiv.org/abs/hep-th/9301111
  4. R. Chatterjee, L. Takhtajan, Aspects of classical and quantum Nambu mechanics, Letters in Mathematical Physics 37, 475-482 (1996), DOI: 10.1007/BF00312678, arXiv:hep-th/9507125, https://arxiv.org/abs/hep-th/9507125
  5. G. Dito, M. Flato, D. Sternheimer, L. Takhtajan, Deformation quantization and Nambu Mechanics, Communications in Mathematical Physics 183, 1-22 (1997), DOI: 10.1007/BF02509794, arXiv:hep-th/9602016, https://arxiv.org/abs/hep-th/9602016
  6. J. Bagger, N. Lambert, Modeling multiple M2-branes, Physical Review D 75, 045020 (2007), DOI: 10.1103/PhysRevD.75.045020, https://doi.org/10.1103/PhysRevD.75.045020
  7. T. Curtright, C. Zachos, Classical and quantum Nambu mechanics, Physical Review D 68, 085001 (2003), DOI: 10.1103/PhysRevD.68.085001, arXiv:hep-th/0212267, https://arxiv.org/abs/hep-th/0212267
  8. J. DeBellis, C. Saemann, R. J. Szabo, Quantized Nambu-Poisson manifolds and n-Lie algebras, Journal of Mathematical Physics 51, 122303 (2010), DOI: 10.1063/1.3503773, arXiv:1001.3275, https://arxiv.org/abs/1001.3275
  9. T. Yoneya, Generalized Hamilton-Jacobi theory of Nambu mechanics, Progress of Theoretical and Experimental Physics 2017, 023A01 (2017), DOI: 10.1093/ptep/ptx008, arXiv:1612.08509, https://arxiv.org/abs/1612.08509
  10. T. Yoneya, Canonical Nambu mechanics: relevance to string/M-theory and approaches to quantization, Progress of Theoretical and Experimental Physics 2021, 12C101 (2021), DOI: 10.1093/ptep/ptab050, arXiv:2103.12918, https://arxiv.org/abs/2103.12918
  11. M. Ashwinkumar, L. Schmidt, M. Tan, Matrix quantization of classical Nambu brackets and super p-branes (published as: Matrix regularization of classical Nambu brackets and super p-branes), Journal of High Energy Physics 2021, 172 (2021), DOI: 10.1007/JHEP07(2021)172, arXiv:2103.06666, https://arxiv.org/abs/2103.06666
  12. Z. Yoshida, Nambu mechanics viewed as a Clebsch parameterized Poisson algebra -- toward canonicalization and quantization, Progress of Theoretical and Experimental Physics 2024, 03A103 (2024), DOI: 10.1093/ptep/ptac096, arXiv:2206.12519, https://arxiv.org/abs/2206.12519
  13. S. Katagiri, Quantization of Nambu brackets from operator formalism in classical mechanics, International Journal of Modern Physics A 38, 2350101 (2023), DOI: 10.1142/S0217751X23501014, arXiv:2209.09798, https://arxiv.org/abs/2209.09798
  14. Y. Kawamura, Generalized Heisenberg dynamics revisited, Progress of Theoretical and Experimental Physics 2025, 123A01 (2025), DOI: 10.1093/ptep/ptaf170, arXiv:2507.09848, https://arxiv.org/abs/2507.09848
  15. P. Hanlon, M. Wachs, On Lie k-algebras, Advances in Mathematics 113, 206-236 (1995), DOI: 10.1006/aima.1995.1038, https://doi.org/10.1006/aima.1995.1038
  16. J. A. de Azcarraga, A. M. Perelomov, J. C. Perez Bueno, The Schouten-Nijenhuis bracket, cohomology and generalized Poisson structures, Journal of Physics A: Mathematical and General 29, 7993-8009 (1996), DOI: 10.1088/0305-4470/29/24/023, arXiv:hep-th/9605067, https://arxiv.org/abs/hep-th/9605067