Bulk–Boundary Master Equations and Anomaly Inflow

Anomaly inflow is a relative master-equation statement. A bulk variation can cancel a boundary cocycle only when both are defined on the same gauge data, their local cohomology classes transgress correctly, and the exponentiated bulk functional is globally well defined. Local cancellation of an infinitesimal variation does not settle level quantization, large gauge transformations, or the holonomy of an anomaly line.

Required background. Quantum master equation and anomaly obstructions supplies the degree-one obstruction class. BV–BFV structures, boundaries, and gluing supplies the relative classical equation. Boundary phase spaces, constraints, and the BFV charge supplies the boundary differential.

Helpful background. Anomaly polynomials and inflow gives the physical descent equations. Symmetry TFT explains the invertible bulk interpretation.

Relative classical and quantum equations

For exact BV–BFV data, the boundary term in

$$\iota_{Q_M}\omega_M =(-1)^d\delta S_M+\pi_M^*\alpha^\partial_{\partial M}$$

implies that the bulk action does not satisfy the closed-manifold master equation by itself. In a polarization with residual fields $\mathcal V_M$, perturbative quantization assigns a boundary operator $\Omega^\partial_{\partial M}$ and a state $\widehat\psi_M$ satisfying

$$\left(\hbar^2\Delta_{\mathcal V_M} +\Omega^\partial_{\partial M}\right)\widehat\psi_M=0.$$

The two summands commute and square to zero in an anomaly-free construction. A failure appears as a degree-one cocycle. Changing the gauge fixing or local counterterm shifts a representative by an exact term, but a nonzero cohomology class remains an obstruction. The quantum data and their controlled ambiguity are stated in Cattaneo, Mnev, and Reshetikhin 2018, §2.3, pp. 18–19.

This equation is relative in two senses. It couples a bulk BV Laplacian to a boundary BFV operator, and its state depends on a selected boundary polarization. A proposed cancellation must therefore match ghost number, locality, polarization, and global gauge action. Adding two terms whose ordinary variations cancel is insufficient if their quantum line bundles do not tensor to a trivial equivariant line.

Chern–Simons transgression

For a compact gauge group and invariant trace, the three-dimensional Chern–Simons action is

$$S_{\mathrm{CS}}[A] =\frac{k}{4\pi}\int_M \operatorname{tr}\left(A\wedge dA+\frac23A\wedge A\wedge A\right).$$

Its variation contains the boundary potential

$$\alpha^\partial_{\partial M} =\frac{k}{4\pi}\int_{\partial M}\operatorname{tr}(A\wedge\delta A).$$

Under an infinitesimal gauge transformation $\delta_\lambda A=d_A\lambda$, the on-shell variation reduces to a boundary descent term. In BV language the superconnection and its boundary BFV charge package the gauge parameter, curvature constraint, and higher ghost terms. The induced data and the fact that a gauge transformation changes the classical action by a boundary functional are explicit in Cattaneo, Mnev, and Reshetikhin 2014, §§7.1–7.2, pp. 45–50.

Choose a complex structure and polarization on the boundary. The resulting chiral boundary functional has a current anomaly whose local variation is opposite to the Chern–Simons descent term. The combined bulk–boundary state can therefore satisfy the relative master equation. This is the mathematical content behind the Chern–Simons/Wess–Zumino–Witten inflow statement; the physical descent calculation is developed at Anomaly Polynomials and Inflow.

The global condition

The exponentiated action, not merely its infinitesimal variation, must be gauge invariant. With standard trace normalization, $k$ must lie in the appropriate integral lattice so that extending a boundary gauge transformation into one higher dimension changes $S_{\mathrm{CS}}$ by an integer multiple of $2\pi$. For non-simply connected groups, spin refinements, or torsion backgrounds, the admissible lattice changes. The boundary anomaly line must have inverse equivariant holonomy to the bulk line on every loop in field space.

This distinguishes three statements:

• a local descent cocycle is cancelled;
• the perturbative relative QME is solved to a declared loop order;
• the exponentiated coupled theory is globally gauge invariant.

Each implication requires additional input. Vanishing of the local anomaly polynomial does not exclude a global sign anomaly, and a correctly quantized topological inflow term does not prove convergence or positivity of the boundary quantum theory.

Failure test and surviving claim

Take the Chern–Simons coefficient to be a nonintegral real number and couple the local boundary current so that infinitesimal variations cancel. A large gauge transformation of winding number one changes the exponentiated bulk action by a nontrivial phase. The local Ward identity survives, but the combined object does not descend to the gauge quotient. No choice of a local boundary counterterm can remove a nontrivial global holonomy.

A second failure occurs if a boundary counterterm cancels the variation but changes the chosen symplectic primitive incompatibly with the polarization. Then the supposed state lives in the wrong boundary space. Both examples show why anomaly inflow is a statement about relative cohomology and global quantization, not just signs in a variation.

Exercises

Why is a degree-one anomaly removable when it is $\Omega^\partial$-exact?

Solution

If $\mathcal A=\Omega^\partial B$, adding the local counterterm $-\hbar B$ shifts the quantum operator or state so that the anomaly representative cancels at that order. The argument assumes $B$ is an admissible local functional compatible with the polarization and symmetries. A nonlocal primitive is not an allowed renormalization.

What extra test distinguishes local from global inflow cancellation?

Solution

Transport the anomaly line around every noncontractible loop generated by large gauge transformations and background-field variations. The tensor product of bulk and boundary lines must have trivial holonomy and a compatible trivialization, not merely zero infinitesimal curvature.

References

• Cattaneo, Alberto S., Pavel Mnev, and Nicolai Reshetikhin. “Classical BV Theories on Manifolds with Boundary.” Communications in Mathematical Physics 332 (2014): 535–603. DOI; Open PDF.
• Cattaneo, Alberto S., Pavel Mnev, and Nicolai Reshetikhin. “Perturbative Quantum Gauge Theories on Manifolds with Boundary.” Communications in Mathematical Physics 357 (2018): 631–730. DOI; Open PDF.
• Witten, Edward. “Quantum Field Theory and the Jones Polynomial.” Communications in Mathematical Physics 121 (1989): 351–399. DOI.