Gravitational energy is well defined

Abstract

The energy of gravitating systems has been an issue since Einstein proposed general relativity: considered to be ill defined, having no proper local density. Energy–momentum is now regarded as quasi-local (associated with a closed 2-surface). We consider the pseudotensor and quasi-local proposals in the Lagrangian–Noether–Hamiltonian formulations. There are two ambiguities: (i) many expressions, (ii) each depends on some nondynamical structure, e.g. a reference frame. The Hamiltonian approach gives a handle on both problems. Our remarkable discovery is that with a 4D isometric Minkowski reference, a large class of expressions — those that agree with the Einstein pseudotensor’s Freud superpotential to linear order — give a common quasi-local energy value. With a best-matched reference on the boundary, this value is the nonnegative Wang–Yau mass.