The tensor density (1.10) includes the Einstein-Dirac pseudotensor
density  which is not symmetric.
Nevertheless, (55) can be rewritten in the form of (18) after introducing some "pseudotensor
of the gravitational field" t.
Einstein's proposed resolution of the idea is an energy-momentum pseudotensor
constructed from the metric components .
The tensor dual to vector [bar.a] is defined as [([[bar.a].sup.x]).sub.ik] = [[epsilon].sub.ijk][a.sub.j]([[epsilon].sub.ijk] is the antisymmetric Levi-Civita's pseudotensor
, which should be distinguished from permittivity [[epsilon].sub.ij])[81-83].
For example, for N = 2, Levi-Civita pseudotensor
[[epsilon].sup.IJ] is the antisymmetric metric, [[epsilon].sup.12] = +1 = -[[epsilon].sub.12], where [[epsilon].sup.IJ][[epsilon].sub.JK] = [[delta].sup.I.sub.K] and a field or variable transforms as [[phi].sup.I] = [[epsilon].sup.IJ][[phi].sub.J] and [[phi].sub.I] = [[epsilon].sub.IJ][[phi].sup.J].
isn't the best solution for elucidating the energy of a gravitational field, of course.
In addition, we introduce the pseudotensor
[F.sup.*[alpha][beta]] of the field dual to the field tensor
Hal Puthoff described the GR term to me as a "pseudotensor
, which can appear or disappear depending on how you treat mass".
In addition to the field tensor [F.sub.[alpha][beta]], we introduce the field pseudotensor
[F*.sup.[alpha][beta]] dual and in the usual way 
Gravitational fields bear an energy described by the energy-momentum pseudotensor