I am studying a theory with a Lagrangian density that I don't want to write because I would like to try to tackle the problem by myself. The fact is that it is a theory with two fields $\{ A_{\mu\nu},B_\mu \}$ where $A_{\mu\nu}$ is an anti-simmetric field and $B_\mu$ is a vector field.
I found that the action is invariant under a local transformation of the $A_{\mu\nu}$ field and then, to calculate the propagators, I introduce a gauge fixing term that breaks the symmetry which depends on field $A_{\mu\nu}$ and on a lagrange multiplier.
What I have found is that despite the breaking of the local symmetry the Hessian matrix to be inverted to calculate the propagators is still degenerate.
So my question is: is the Hessian matrix degenerate if and only if there is a local symmetry? Or is it possible that it is even if the action has no local symmetries?