A note on complete gauge-fixing and the constraint algebra

Ganga Singh Manchanda · 2026

Read A note on complete gauge-fixing and the constraint algebra by Ganga Singh Manchanda (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-16990.txt.

Introduction

Manchanda proves a factorization for the joint matrix built from gauge fixings and first-class constraints, via a Schur complement. Since the second-class part is invertible by definition, that part splits off from gauge fixing. In the algebra case, the result matches the Hamiltonian admissibility criterion with the Lagrangian completeness criterion.

Catalog record

CREATORGanga Singh Manchanda
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE14740
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.16990

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A note on complete gauge-fixing and the constraint algebra Ganga Singh Manchanda Abstract The admissibility of a gauge-fixing is governed by the invertibility of Δ = { σ a , γ b } \Delta=\{\sigma^{a},\gamma_{b}\} where σ a \sigma^{a} are gauge-fixing conditions and γ b \gamma_{b} are independent first-class constraints. We prove, via the Schur complement, that the determinant of the combined constraint matrix ℳ = { Φ A , Φ B } \mathcal{M}=\{\Phi_{A},\Phi_{B}\} built from all constraints and gauge-fixing conditions factorises as det ℳ ≈ ± ( det Δ ) 2 det C \det\mathcal{M}\approx\pm(\det\Delta)^{2}\det C , where C C is the second-class constraint matrix, providing an alternative criterion for admissibility. Since det C ≠ 0 \det C\neq 0 by definition, the second-class sector decouples entirely from the gauge-fixing sector. In the algebraic case, this factorisation identifies the Hamiltonian admissibility criterion of Henneaux and Teitelboim Henneaux with the Lagrangian completeness criterion of Motohashi, Suyama, and Takahashi Motohashi . We identify a metric ansatz as gauge-fixing at the action level and analyse completeness in the context of spherically symmetric spacetime. The factorisation ensures that completeness is robust to the second-class sector that arises in modified theories of gravity. I Introduction Every gauge theory of physical interest is represented by a singular Lagrangian. This poses a problem because the Legendre transform fails to account for non-invertible conjugate momenta and thus neglects relations of the form, \gamma_{a}(q,p)=0 , which cannot be inferred from a Legendre-transformed Hamiltonian but nevertheless express real physics. To pass between the Lagrangian and Hamiltonian forms of these theories requires the Dirac–Bergmann algorithm which identifies these constraints, adds them to the Hamiltonian by hand, classifies them, and ensures self-consistency Dirac ; Henneaux . For each independent first-class constraint, the corresponding Lagrange multiplier carries a gauge degree of freedom of the theory. The process of removing this freedom, gauge-fixing, is well understood at the Hamiltonian level. For F independent first-class constraints we impose F conditions, \sigma^{a}\approx 0 , which satisfy \det\Delta\equiv\det\{\sigma^{a},\gamma_{b}\}\neq 0 . In systems with S additional second-class constraints, \chi_{\alpha}\approx 0 , practitioners often deal with the combined constraint set, \Phi_{A}=(\gamma_{a},\chi_{\alpha},\sigma^{a}) , which should be entirely second-class once gauge freedom is eliminated (i.e. \det\mathcal{M}\equiv\det\{\Phi_{A},\Phi_{B}\}\neq 0 ). Naively, the admissibility criteria, \det\Delta\neq 0 and \det\mathcal{M}\neq 0 , seem to be equivalent. This much is clear when no second-class constraints are present ( S=0 ), and the combined constraint matrix is the 2F\times 2F block built from \gamma_{a} and \sigma^{a} , with determinant (\det\Delta)^{2} . When second-class constraints are present, the combined constraint matrix is (2F+S)\times(2F+S) , and one might worry that cross-couplings between gauge conditions and second-class constraints, R=\{\chi_{\alpha},\sigma^{b}\} , could spoil invertibility even when \det\Delta\neq 0 , or conversely rescue it when \det\Delta=0 . We show that neither occurs. The determinant factorises exactly as: \det\mathcal{M}\approx\pm(\det\Delta)^{2}\det C, (1) where C=\{\chi_{\alpha},\chi_{\beta}\} . Since \det C\neq 0 by definition, \mathcal{M} is invertible if and only if \Delta is invertible. The second-class sector decouples entirely from the gauge-fixing sector. Proof of ( 1 ) is simple via the Schur complement, and the result ensures that either \det\mathcal{M} or \det\Delta can verify the admissibility of a proposed gauge-fixing, with usage depending on context. In theories with complicated second-class sectors, this can provide a practical simplification. The decoupling is in essence a consequence of the Dirac bracket construction and can be understood through that lens. Making the decoupling explicit at the level of a determinant identity has two advantages: it provides a self-contained criterion that does not require constructing the Dirac bracket, and it connects directly to the Lagrangian completeness criterion of Motohashi, Suyama, and Takahashi, which characterises when gauge functions are determined uniquely by gauge conditions Motohashi . The plan of this paper is as follows. In Section II we establish some formal details of the constraint algebra and prove the relation between \det\Delta and \det\mathcal{M} . Section III demonstrates the connection to the Lagrangian completeness criterion, and discusses consequences. Finally, in Section IV we look at applications to gravity, diagnosing the completeness of a common spherically symmetric metric ansatz and noting the implications of our factorisation in modified theories of gravity where second-class constraints are generically present. II Proof of factorisation II.1 Preliminaries Consider a constrained Hamiltonian system with F first-class constraints, \gamma_{a}\approx 0 , and S second-class constraints, \chi_{\alpha}\approx 0 . Before moving forward, we first set out some definitions and conditions which we will rely on. We adopt the stronger definition of first-class as those constraints whose Poisson bracket with every constraint in the theory vanishes weakly Henneaux . This can always be achieved by redefining \gamma_{a}\to\gamma_{a}+k_{a}{}^{\alpha}\chi_{\alpha} with suitably chosen k_{a}{}^{\alpha} , such that \Delta^{a}{}_{b}\to\Delta^{a}{}_{b}-k_{b}{}^{\alpha}R_{\alpha}{}^{a} . While \det\mathcal{M} and \det C are intrinsic to the constraint surface, \det\Delta is not; its value changes with the redefinition. The vanishing of \det\Delta is, however, invariant, so that the conclusion \det\Delta\neq 0\Leftrightarrow\det\mathcal{M}\neq 0 holds regardless of the basis choice (under the strong definition of first-class). All equalities are weak, holding only on the constraint surface. In particular, the Schur complement manipulations require \det\Delta\neq 0 to hold on the constraint surface; if \det\Delta vanishes at isolated points of this surface, the factorisation and the admissibility conclusion hold only away from those points. At isolated points where \det\Delta=0 , the gauge-fixing conditions fail, corresponding to a Gribov-type obstruction (gauge slices are tangent to orbits) Gribov . Finally, we note that \det C\neq 0 on the surface \{\gamma_{a}=0,\chi_{\alpha}=0\} implies \det C\neq 0 on the surface \{\gamma_{a}=0,\chi_{\alpha}=0,\sigma^{a}=0\} , since \det C is a smooth non-vanishing function on the larger surface. The combined constraint set is then entirely second-class on the full constraint surface when \det\mathcal{M}\neq 0 , which the factorisation relates to \det\Delta\neq 0 . With these conventions in place, we turn to the proof. II.2 \det\Delta\neq 0\Longrightarrow\det\mathcal{M}\neq 0 We permute the combined constraints as \tilde{\Phi}_{A}=(\gamma_{a},\sigma^{a},\chi_{\alpha}) such that the permuted \det\tilde{\mathcal{M}} differs from \det\mathcal{M} only up to a sign. On the constraint surface the combined constraint matrix evaluates to: \tilde{\mathcal{M}}\approx\begin{pmatrix}0&-\Delta^{T}&0\\ \Delta&Q&-R^{T}\\ 0&R&C\end{pmatrix}, (2) where \Delta=\{\sigma^{a},\gamma_{b}\} , Q=\{\sigma^{a},\sigma^{b}\} , R=\{\chi_{\alpha},\sigma^{b}\} , and C=\{\chi_{\alpha},\chi_{\beta}\} . The zeroes in ( 2 ) follow from the stronger definition of first-class constraints. We label the upper-left 2F\times 2F block: \mathcal{A}\equiv\begin{pmatrix}0&-\Delta^{T}\\ \Delta&Q\end{pmatrix}. (3) When \det\Delta\neq 0 , this block has \det\mathcal{A}=(\det\Delta)^{2} , and inverse: \mathcal{A}^{-1}=\begin{pmatrix}\Delta^{-1}Q\Delta^{-T}&\Delta^{-1}\\ -\Delta^{-T}&0\end{pmatrix}. (4) The Schur complement of \mathcal{A} in \tilde{\mathcal{M}} Crabtree : \tilde{\mathcal{M}}/\mathcal{A}=C-\begin{pmatrix}0&R\end{pmatrix}\mathcal{A}^{-1}\begin{pmatrix}0\\ -R^{T}\end{pmatrix}=C, (5) demonstrates that the cross-couplings, R , drop out entirely such that: \det\tilde{\mathcal{M}}\approx\det\mathcal{A}\det C=(\det\Delta)^{2}\det C. (6) Since \det C\neq 0 by definition, \det\mathcal{\tilde{M}}\neq 0 and \det\mathcal{M}\neq 0 . II.3 \det\Delta=0\Longrightarrow\det\mathcal{M}=0 If \det\Delta=0 , there exists some non-zero v^{a} with \Delta^{b}{}_{a}v^{a}=0 . We can construct V^{A}=(v^{a},0^{a},0_{\alpha}) indexed over (\gamma_{a},\sigma^{a},\chi_{\alpha}) , and column-by-column, we see: \displaystyle\gamma\text{ column:}\quad \displaystyle v^{a}\{\gamma_{a},\gamma_{b}\}=v^{a}U^{c}{}_{ab}\gamma_{c}\approx 0, \displaystyle\sigma\text{ column:}\quad \displaystyle-v^{a}(\Delta^{T})^{b}{}_{a}=-\Delta^{b}{}_{a}v^{a}=0, \displaystyle\chi\text{ column:}\quad \displaystyle v^{a}\{\gamma_{a},\chi_{\beta}\}\approx 0, such that V^{A}\tilde{\mathcal{M}}_{AB}\approx 0 . \tilde{\mathcal{M}} is therefore singular on the constraint surface, implying \det\mathcal{M}\approx 0 . III Connection to Lagrangian completeness In the Lagrangian framework, gauge-fixing at the action level (“pre-fixing”) is safe if it is complete in the sense of Motohashi : the conditions must satisfy \delta_{\varepsilon}\sigma^{a}=0 , in order to uniquely determine the gauge parameters \varepsilon^{a} . In general, \delta_{\varepsilon}\sigma^{a}=M^{(0)a}{}_{b}\varepsilon^{b}+M^{(1)a}{}_{b}\dot{\varepsilon}^{b}+\cdots , and completeness requires M^{(0)a}{}_{b} to have trivial kernel. The connection to the Hamiltonian admissibility criterion is direct in

Read the complete text