Proper-Momentum Prescription in Theoretical Physics

• Proper-momentum prescription is a methodological pattern in theoretical physics, defining momentum via complete geometric, algebraic, and variational constraints.
• It eliminates spurious contributions and ensures consistency with commutation relations and Noether charges, as shown in constrained systems, string theories, and holographic setups.
• Applications include operator quantization, nonlocal QCD matching, and curved momentum-space kinematics, emphasizing its broad relevance in modern physics.

Searching arXiv for papers using or defining “proper momentum” / “proper-momentum prescription” across contexts. Search 1: exact phrase and close variants. The proper-momentum prescription is not a single universal formalism but a recurrent methodological pattern in theoretical physics: a momentum variable, momentum density, or momentum-space kernel is deemed “proper” when it is fixed by the full geometric, algebraic, or variational structure of the problem rather than by a naive canonical choice. In the literature, this designation appears in constrained quantization on curved hypersurfaces, effective string theories, holographic complexity, nonlocal-operator matching in QCD, and several adjacent gravitational and field-theoretic settings (Liu, 2013, Liu et al., 2011, Dass, 2010, Barbon et al., 2020, Caputa et al., 2024, Ji et al., 31 May 2025).

1. Scope of the term and recurring criteria

Across the cited works, the term is used for distinct objects rather than a single operator.

Context	Proper momentum	Defining criterion
Constrained quantum mechanics	$P=-i\hbar(\nabla_s+H\,n)$	Dirac-bracket / commutator consistency
Effective string theory	$P^\mu=T\int d\sigma\,\partial_\tau X^\mu$ up to improvements	Noether current with improvement terms integrating to zero
Holographic complexity	Infall momentum or proper radial momentum	Equality with complexity growth rate
QCD nonlocal mixing	DR-defined momentum-space kernel	Unambiguous Fourier transform of singular coordinate kernel

A common structural feature is the rejection of a merely formal canonical momentum whenever that quantity fails one of the following tests: it may violate the fundamental commutation relations, differ from the Noether charge by nontrivial improvement terms, fail to capture a geometric transport law, or remain ambiguous under coordinate-to-momentum transformation. This suggests that “proper” functions as a criterion of compatibility with the underlying dynamical framework rather than as a fixed mathematical definition.

A second recurring feature is that the prescription is often designed to eliminate spurious contributions. In effective string theory, higher-order corrections to the momentum density become pure improvements and therefore do not shift the integrated centre-of-mass momentum. In the QCD matching problem, dimensional regularization removes the ambiguity associated with the $1/z_{12}$ singularity. In constrained quantization, the proper momentum excludes normal components inconsistent with the constraint surface.

2. Geometric momentum for constrained quantum systems

In the most explicit use of the term, Liu’s strengthened canonical quantization scheme for a particle constrained to an $(N-1)$-dimensional smooth hypersurface $S\subset\mathbb R^N$ introduces two categories of fundamental commutation relations: the first for positions and momenta, and the second for the Hamiltonian with positions and momenta. The first category is

$$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$

$$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$

while the second demands that the Heisenberg equations reproduce the Dirac-bracket form of the classical constrained motion. With the Hamiltonian ansatz

$$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$

these requirements lead to the geometric, or proper, momentum

$P\equiv p=-\,i\hbar(\nabla_s+H\,n),$

where $\nabla_s$ is the surface gradient and $P^\mu=T\int d\sigma\,\partial_\tau X^\mu$0 is the mean-curvature convention used in that paper (Liu, 2013).

This operator is tangential in the sense that

$P^\mu=T\int d\sigma\,\partial_\tau X^\mu$1

and it satisfies the first-category commutators. For the sphere $P^\mu=T\int d\sigma\,\partial_\tau X^\mu$2, the prescription yields

$P^\mu=T\int d\sigma\,\partial_\tau X^\mu$3

together with the curvature potential

$P^\mu=T\int d\sigma\,\partial_\tau X^\mu$4

In the same setting, defining $P^\mu=T\int d\sigma\,\partial_\tau X^\mu$5 and $P^\mu=T\int d\sigma\,\partial_\tau X^\mu$6, one obtains the commutators

$P^\mu=T\int d\sigma\,\partial_\tau X^\mu$7

which realize the Lie algebra $P^\mu=T\int d\sigma\,\partial_\tau X^\mu$8 (Liu, 2013).

The two-dimensional sphere provides the sharpest uniqueness result. For a particle on $P^\mu=T\int d\sigma\,\partial_\tau X^\mu$9, Liu, Tang, and Xun show that the usual canonical momentum

$1/z_{12}$0

fails Dirac’s prescription because

$1/z_{12}$1

By contrast, the Cartesian components $1/z_{12}$2 of the geometric momentum satisfy

$1/z_{12}$3

as well as the Hamiltonian commutators

$1/z_{12}$4

Among five candidate forms in the literature, imposing these relations fixes uniquely the choice $1/z_{12}$5; the same choice is also singled out by self-adjointness on $1/z_{12}$6 (Liu et al., 2011).

The constrained-system usage therefore defines the proper-momentum prescription as the replacement of coordinate-canonical momenta by an ambient-space geometric operator determined by the full Dirac algebra.

3. Proper string momentum in effective string theories

In effective string theory, the issue is not operator ordering on a constrained manifold but the identification of the correct centre-of-mass momentum in the presence of higher-order terms in the action. Das, Matlock, and Rabin calculate the momentum by the Noether prescription for two conformally invariant effective string theories of the Polchinski–Strominger type: the Polyakov–Liouville theory and an extension of the original Polchinski–Strominger action (Dass, 2010).

For the free Polyakov action

$1/z_{12}$7

the translation Noether current has

$1/z_{12}$8

and the centre-of-mass momentum is

$1/z_{12}$9

For the Polyakov–Liouville action, with

$(N-1)$0

the correction to the $(N-1)$1-component of the momentum current, to order $(N-1)$2, is

$(N-1)$3

Because this is a pure $(N-1)$4-derivative, the integrated correction vanishes for periodic boundary conditions: $(N-1)$5

An analogous result holds for the extended Polchinski–Strominger action

$(N-1)$6

which is conformal only up to $(N-1)$7. Its momentum correction is again of improvement type,

$(N-1)$8

so that

$(N-1)$9

to the orders considered (Dass, 2010).

The physical significance is explicit in the paper: the oscillator part of the improved momentum drops out of the total $S\subset\mathbb R^N$0, the centre-of-mass kinematics remains identical to that of the free theory, and the mass-shell condition

$S\subset\mathbb R^N$1

is free from spurious $S\subset\mathbb R^N$2 or $S\subset\mathbb R^N$3 shifts in the background momentum. In this setting, the proper-momentum prescription is therefore the Noether momentum modulo genuine improvement currents.

4. Holographic complexity and bulk proper momentum

A different usage arises in holography, where momentum is defined so as to reproduce the rate of complexity growth. In the Complexity $S\subset\mathbb R^N$4 Volume proposal, one considers maximal spacelike hypersurfaces $S\subset\mathbb R^N$5 anchored on boundary Cauchy slices and defines

$S\subset\mathbb R^N$6

Using the Hamiltonian constraint on $S\subset\mathbb R^N$7, one introduces an infall vector field $S\subset\mathbb R^N$8 and the matter infall momentum

$S\subset\mathbb R^N$9

This gives

$$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$0

with the remainder $$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$1 vanishing when $$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$2 is a conformal Killing field. The generalized construction starts instead from the full Codazzi–Mainardi equation, introduces an antisymmetric–cyclic infall tensor $$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$3, and yields

$$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$4

where

$$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$5

is a purely gravitational Weyl-flux contribution. When $$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$6, one obtains the exact tensorial prescription

$$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$7

The construction is checked explicitly in a Ricci-flat pp-wave, where $$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$8, the remainder vanishes by construction, and $$[x_i,x_j]=0,\qquad [x_i,p_j]=i\hbar(\delta_{ij}-n_i n_j),$$9 exactly reproduces the rate of change of the extremal volume (Barbon et al., 2020).

An even more direct “proper momentum” relation appears in the AdS$$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$0/CFT$$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$1 study of spread complexity for locally excited states. There the CFT quantity is

$$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$2

and in three standard geometries its growth rate is matched to the canonical momentum conjugate to the proper radial distance $$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$3 of the dual bulk particle. For planar BTZ, global AdS$$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$4, and the Poincaré patch, the bulk computation gives $$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$5, $$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$6, and $$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$7, with the universal identity

$$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$8

after identifying the particle mass with the operator dimension, $$[p_i,p_j]= -\,i\hbar\,\{(n_i\partial_j n_k-n_j\partial_i n_k)p_k\}_{\mathrm{Hermitian}},$$9, and $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$0. The result is formulated in the heavy-operator, semiclassical regime, with the UV regulator $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$1 fixing the initial bulk radial position (Caputa et al., 2024).

Taken together, these holographic works use “proper momentum” in two closely related ways: as an infall quantity controlling the growth of extremal volume, and as the radial canonical momentum in the proper-distance coordinate that exactly tracks spread-complexity growth.

5. Momentum-space prescriptions in QCD and curved momentum geometry

In perturbative QCD, the phrase designates neither a particle operator nor a Noether charge, but an unambiguous coordinate-to-momentum transformation. The mixing of nonlocal gluon and quark bilinear operators contains a $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$2 singularity in coordinate space, so a naive Fourier transform to momentum space is ill defined. Ji et al. show that this ambiguity arises from the lack of a proper regularization prescription when the field separation approaches zero. The remedy is to keep the coordinate-space kernel in $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$3 dimensions, define the momentum-space kernel through the full $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$4-dimensional Fourier transform,

$$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$5

and only afterward expand in $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$6. This produces a unique momentum-space matching coefficient, reproduces the local $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$7 mixing pole, removes the ambiguity associated with integration constants or lower limits in integrated kernels, and matches direct momentum-space one-loop calculations. The same prescription is stated to be compatible with lattice-QCD extractions of quasi-distributions (Ji et al., 31 May 2025).

A related but distinct geometric usage appears in curved momentum-space kinematics. In the de Sitter case study of relative locality, Amelino-Camelia, Gubitosi, and Palmisano propose a new prescription that associates the affine connection of momentum space to the law of momentum composition by geodesic parallel transport rather than by the derivative-based rule common in the $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$8-Poincaré literature. For “proper de Sitter momentum space” in $$H=-\,\frac{\hbar^2}{2m}\nabla^2_{LB}+V_{\mathrm{ext}}(x)+V_g(x),$$9 dimensions, with de Sitter metric

$P\equiv p=-\,i\hbar(\nabla_s+H\,n),$0

and its Levi-Civita connection, the prescription yields a unique commutative but deformed composition law $P\equiv p=-\,i\hbar(\nabla_s+H\,n),$1, together with a deformed mass Casimir determined by the metric distance from the origin. The authors emphasize that this is inequivalent to the standard prescription and provides a DSR-relativistic momentum space with a commutative, non-associative law of composition (Amelino-Camelia et al., 2013).

These examples broaden the meaning of “proper-momentum prescription” from operator selection to the proper definition of momentum-space objects themselves: kernels, composition laws, and on-shell structures can all require a non-naive construction.

6. Adjacent extensions, limitations, and controversies

Several adjacent literatures apply the same “proper” criterion to momentum density or energy-momentum, even when the phrase is not used in exactly the same sense.

In general relativity, Shimizu proposes a gravitational energy-momentum tensor $P\equiv p=-\,i\hbar(\nabla_s+H\,n),$2 derived from Noether’s theorem in tetrad variables,

$P\equiv p=-\,i\hbar(\nabla_s+H\,n),$3

with $P\equiv p=-\,i\hbar(\nabla_s+H\,n),$4 transforming as a genuine tensor under general coordinate transformations because one index is a local Lorentz index. The construction includes a Gibbons–Hawking boundary contribution and reproduces explicit weak-wave and FLRW results, including the vanishing of the total gravitational-plus-matter energy density in the FLRW example (Shimizu, 2016). This is not a momentum-operator prescription in the constrained-quantization sense, but it shares the same aim: replacing coordinate-dependent pseudotensors by an object fixed by the full covariant structure.

Also in general relativity, Flanagan, Nichols, Stein, and Vines develop an operational prescription for measuring local linear and angular momenta $P\equiv p=-\,i\hbar(\nabla_s+H\,n),$5 from local curvature data and transporting them between points. The transport law

$P\equiv p=-\,i\hbar(\nabla_s+H\,n),$6

selects

$P\equiv p=-\,i\hbar(\nabla_s+H\,n),$7

as the unique choice yielding the desired asymptotic consistency in stationary, vacuum, asymptotically flat spacetimes. In that regime, the measured and transported momenta agree with the standard ADM/Bondi quantities to the stated order (Flanagan et al., 2016).

In classical electrodynamics, a controversy concerns which local momentum density is the “proper” one for structured light. The canonical tensor gives

$P\equiv p=-\,i\hbar(\nabla_s+H\,n),$8

whereas the Belinfante tensor gives

$P\equiv p=-\,i\hbar(\nabla_s+H\,n),$9

Although both integrate to the same total momentum and total angular momentum when boundary terms vanish, they differ point by point. The paper on superkicks concludes that local torques, recoil momenta, and even the presence or absence of tractor-beam regions depend on which density is adopted, and explicitly states that the debate is not resolved by principle within that work (Afanasev et al., 2022). This is a reminder that “proper” may remain context-dependent when local coupling mechanisms are under dispute.

A further extension appears in Einstein–Cartan theory, where torsion induces noncommutative momentum commutators,

$\nabla_s$0

reducing in a suitable frame to

$\nabla_s$1

Popławski then replaces momentum-space integration in loop diagrams by a sum over discrete momentum eigenvalues and derives a prescription that regularizes ultraviolet-divergent integrals while preserving gauge invariance in the vacuum-polarization example (Popławski, 2017). Here again, the proper-momentum idea means that the physically relevant momentum-space structure is not the naive continuum one.

The literature therefore supports two general conclusions. First, proper momentum is typically the momentum notion that survives all consistency tests imposed by geometry, constraints, symmetry, and regularization. Second, there is no single trans-contextual definition: the same phrase can denote a geometric differential operator, a Noether charge stripped of improvements, an infall quantity sourcing complexity growth, or a fully regulated momentum-space kernel. The unifying theme is methodological rather than ontological.