Variational, Symplectic, and Constraint Methods

The passage from an action to equations of motion depends on which variations are allowed. This chapter follows that dependence through boundary terms, fluctuations, Hamiltonian geometry, and constraints, explaining when reduction removes redundant directions and when boundaries change the physical problem.

Enter this chapter

Begin with Field Variations and Boundary Terms to separate bulk equations from boundary conditions. For Hamiltonian systems, continue through symplectic forms before tackling constrained reduction.

Topics

1. Field Variations and Boundary Terms — Derives the bulk and boundary parts of an action variation.
2. Second Variation, Hessians, and Jacobi Operators — Tests stationary configurations through the quadratic fluctuation operator.
3. Symplectic Forms, Hamiltonian Flows, and Poisson Brackets — Constructs Hamiltonian motion from a nondegenerate closed two-form.
4. Hamiltonian Group Actions and Moment Maps — Relates infinitesimal group actions to their Hamiltonian generators.
5. Constraints, Dirac Brackets, and Symplectic Reduction — Distinguishes second-class elimination from quotienting degenerate directions.
6. Presymplectic Systems and the Covariant Phase-Space Ambiguity Map — Explains nonuniqueness and boundary ambiguities when the two-form is degenerate.

How the ideas fit together

The first variation determines the equations and admissible boundary data; the second variation tests fluctuations within that same space of admissible variations. Dropping boundary terms too early can therefore change both the stationary problem and the fluctuation spectrum.

In Hamiltonian language, constraints can make the pulled-back two-form degenerate. A smooth quotient by its null directions can recover a symplectic space only when regularity and quotient hypotheses hold. Symplectic reduction is consequently a geometric construction with conditions, not merely deleting variables; see Cannas da Silva 2001, Lectures 23–24. Whether a null direction represents gauge redundancy also depends on the allowed transformations and boundary data.

Review the chapter

For $S[q]=\tfrac12\int_a^b\dot q^{\,2}\,\mathrm dt$, what changes if the endpoint variations are free instead of fixed?

Solution

Integration by parts gives $\delta S=-\int_a^b\ddot q\,\delta q\,\mathrm dt+[\dot q\,\delta q]_a^b$. Fixed endpoint variations leave only $\ddot q=0$. Independent free endpoint variations additionally require $\dot q(a)=\dot q(b)=0$, selecting a different stationary problem.

Where to go next

Continue to Free Spin-One Fields and Constraints to apply this distinction to Maxwell and Proca fields.

References

• Ana Cannas da Silva, Lectures on Symplectic Geometry, Lecture Notes in Mathematics 1764, Springer, 2001; author revision 2006. DOI. Open PDF.