Metric symmetry by design in general relativity

Viktor T. Toth · 2024

Read Metric symmetry by design in general relativity by Viktor T. Toth (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-10607.txt.

Introduction

Viktor Toth, writing alone, revisits a silent assumption in the derivation of Einstein's equations, in a 2024 paper. Textbooks derive the field equations from the standard action while quietly assuming the metric, spacetime's measuring ruler, is symmetric. Drop the assumption and different theories result, including Einstein's own failed unified field theory.

Toth instead enforces symmetry openly, adding a Lagrange multiplier term to the action. Einstein's equations come back, but the multiplier survives as an extra antisymmetric piece with no constraint on it. The paper briefly explores what that leftover could mean: a cosmological angular momentum, a current of spin woven into gravity, or the known asymmetry of the usual bookkeeping for gravitational energy. It also touches on quantum gravity attempts. The note is short and speculative, flagging a term most derivations throw away without comment.

Catalog record

CREATORViktor T. Toth
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE16 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-10607

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Metric symmetry by design in general relativity Viktor T. Toth Abstract The usual derivation of Einstein’s field equations from the Einstein–Hilbert action is performed by silently assuming the metric tensor’s symmetric character. If this symmetry is not assumed, the result is a new theory, such as Einstein’s attempted Unified Field Theory or Moffat’s Nonsymmetric Gravitational Theory. Explicitly enforcing the constraint by means of a Lagrange-multiplier term restores Einstein’s field equations, but the multiplier appears as an additional, unconstrained antisymmetric term. We briefly discuss the possible significance of this term with respect to a nonvanishing cosmological angular momentum, a sourced spin current, the nonsymmetric nature of the Einstein pseudotensor characterizing the energy-momentum of the gravitational field, and possible implications on attempts to obtain a quantum theory of gravity. It is widely known that the field equations of gravitation, that is, Einstein’s field equations, can be derived using the variational principle from the Einstein–Hilbert Lagrangian [ 1 , 2 ] and the corresponding action, \displaystyle{\cal S}_{\rm grav}=\int d^{4}x{\cal L}_{\rm grav}=\int d^{4}x\sqrt{-g}\frac{1}{2\kappa}(R-2\Lambda), (1) where g is the determinant of the metric g_{\mu\nu} , R is the Ricci-tensor, and \Lambda is the cosmological constant, with \kappa=8\pi G . The derivation is conceptually straightforward, though there are technical subtleties, arising, in particular, due to the fact that the Ricci scalar R depends on both first and second derivatives of the metric tensor. Generally, the presence of such higher derivatives is bad news, as they can lead to an unstable theory with a Hamiltonian unbounded from below (Ostrogradsky instability [ 3 ] ). General relativity is an exception, as derivatives of the metric tensor beyond the second derivative cancel out in the resulting field equations. (This can also be inferred from the fact that general relativity can be re-expressed in terms of the connection coefficients in the Palatini formalism [ 4 ] , and in this case, only first derivatives are present in the Lagrangian.) The Euler-Lagrange equations that correspond to {\cal S}_{\rm grav} are given, as usual [ 5 ] , by functional differentiation with respect to the metric tensor and its derivatives: \displaystyle\frac{1}{\sqrt{-g}} \displaystyle\left[\frac{\partial{\cal L}_{\rm grav}}{\partial g_{\mu\nu}}-\frac{\partial}{\partial x^{\kappa}}\frac{\partial{\cal L}_{\rm grav}}{\partial g_{\mu\nu,\kappa}}+\frac{\partial^{2}}{\partial x^{\kappa}\partial x^{\lambda}}\frac{\partial{\cal L}_{\rm grav}}{\partial g_{\mu\nu,\kappa\lambda}}\right]+\text{[matter terms]}=0. (2) The matter terms in the general relativistic action are usually given in the form of the matter stress-energy tensor, defined as T^{\mu\nu}=-2(-g)^{-1/2}\delta{\cal S}_{\rm matter}/\delta g_{\mu\nu} , where {\cal S}_{\rm matter} is the action integral representing the matter content. The gravitational part, in turn, after a somewhat tedious but straightforward derivation, yields the well-known Einstein tensor, ultimately leading to Einstein’s field equations for gravitation: \displaystyle R^{\mu\nu}-\tfrac{1}{2}Rg^{\mu\nu}+\Lambda g^{\mu\nu}=8\pi GT^{\mu\nu}. (3) There is, however, a delicate issue here that remains largely unexplored in the general relativity literature, starting with Hilbert’s publication all the way back in 1915, the first to invoke the action principle (albeit without a cosmological constant) to derive the field equations of gravitation [ 2 ] . We ‘‘know’’ that the metric tensor is symmetric 1 1 1 In Hilbert’s own words: “ die zehn zuerst von Einstein eingefülhrten Gravitation-potentiale g_{\mu\nu} ( \mu,\nu=1,2,3,4 ) mit symmetrischen Tensorcharakter ”, i.e., the ten gravitational potentials first introduced by Einstein, g_{\mu\nu} ( \mu,\nu=1,2,3,4 ), with symmetric tensor character. Similarly, Rosenfeld [ 6 ] explored the properties, including the symmetry properties, of the energy-momentum tensor, but the symmetry of the metric tensor was simply assumed. : that is to say, g_{\mu\nu}=g_{\nu\mu} , which appears to follow from the definition of the invariant line element, ds^{2}=g_{\mu\nu}dx^{\mu}dx^{\nu}=g_{(\mu\nu)}dx^{\mu}dx^{\nu} , since g_{[\mu\nu]}dx^{\mu}dx^{\nu}\equiv 0 identically 2 2 2 Note that we follow the usual convention for symmetrizing indices, g_{(\mu\nu)}=\tfrac{1}{2}(g_{\mu\nu}+g_{\nu\mu}) and g_{[\mu\nu]}=\tfrac{1}{2}(g_{\mu\nu}-g_{\nu\mu}) . . However, just because we assume that g_{\mu\nu} is symmetric does not make it so. The dynamical theory of gravitation does not automatically imply its geometric interpretation 3 3 3 Indeed, Einstein himself argued that geometrization is not essential: “ Sie haben vollständig recht. Es ist verkehrt zu glauben, dass die ‘Geometrisierung’ etwas Wesentliches bedeutet. Es ist nur eine Art Eselsbrücke zur Auffindung numerischer Gesetze. Ob man mit einer Theorie ’geometrische’ Vorstellungen verbindet, ist [unleserlich] Privatsache. ”, or “You are perfectly right. It is wrong to think that the ‘geometrization’ has significant meaning. It is only a kind of a clue helping us find numerical laws. Whether you connect a ’geometric’ view to a theory is entirely a private matter.” – Einstein’s letter to Rosenbach, April 8, 1926, as published (with translation) by Lehmkuhl [ 7 ] . . And if g_{\mu\nu} is “only” the tensor field mediating the gravitational interaction, not the metric of spacetime, why would it be symmetrical? Specifically, when we apply the variational principle to a functional that depends on g_{\mu\nu} and its first and second derivatives, the sixteen components of g_{\mu\nu} vary independently unless they are constrained. Variation of the Einstein-Hilbert term with respect to the unconstrained metric tensor yields additional contributions to the field equations, leading, e.g., to Einstein’s attempt to develop a unified field theory [ 8 ] or to Moffat’s nonsymmetric generalization of Einstein’s theory of gravitation [ 9 ] . On the other hand, the metric tensor can be explicitly constrained to be symmetrical by way of introducing a Lagrange-multiplier into the Lagrangian formulation of the theory. The use of Lagrange-multiplier terms in general relativity to enforce constraints, of course, has been explored in the literature over the more than 100 years since Einstein and Hilbert published their respective papers. In particular, Lanczos [ 10 ] used Lagrange-multipliers to recast Einstein’s theory using a third-order field tensor that is antisymmetric in two of its indices. Later, Kichenassamy [ 11 ] , relying in part on the work of Lanczos, systematically explored the use of Lagrange-multipliers, establishing a consistent relationship between varying the metric components vs. varying the connection coefficients. Yet surprisingly, none of these authors appear to have addressed this more basic matter of the assumed symmetry of the metric tensor. This attitude is perhaps justifiable if the connection coefficients and the terms constructed from them, notably the Riemann curvature tensor and the associated Ricci-tensor and Ricci-scalar, are formed in such a manner that the metric tensor is effectively symmetrized, though that still allows for possible nonsymmetric contributions through the integration measure \sqrt{-g} that is an essential part of the generally covariant action. A more direct alternative, however, is to make the assumption that g_{[\mu\nu]}=0 explicit by way of introducing an appropriate Lagrange-multiplier term in the Einstein-Hilbert Lagrangian, namely a tensor-valued term \lambda_{\mu\nu} : \displaystyle{\cal S}_{\rm grav}=\frac{1}{2\kappa}\int d^{4}x\sqrt{-g}(R-2\Lambda+\lambda^{\mu\nu}g_{[\mu\nu]}). (4) Let us explore the consequences of this proposal. Variation with respect to the Lagrange-multiplier’s nondynamical degrees of freedom trivially yields the expected constraint, \displaystyle g_{[\mu\nu]}=0, (5) now explicitly restricting the metric to be symmetrical. Enforcing this constraint after computing the variation of the Einstein–Hilbert term, we find that any additional contributions vanish, and we are left with the usual Einstein field tensor. (This can also be verified through tedious but straightforward computer algebra calculations.) That still leaves open an intriguing question: What is the variation of the newly introduced Lagrange-multiplier term with respect to g_{\mu\nu} , and how does that contribute to the theory’s field equations? The answer can be obtained by calculating the following functional derivative: \displaystyle\frac{\delta}{\delta g_{\alpha\beta}}(\sqrt{-g}\lambda^{\mu\nu}g_{[\mu\nu]}) \displaystyle{}=\left[\frac{\delta}{\delta g_{\alpha\beta}}(\sqrt{-g}\lambda^{\mu\nu})\right]g_{[\mu\nu]}+\sqrt{-g}\lambda^{\mu\nu}\frac{\delta}{\delta g_{\alpha\beta}}g_{[\mu\nu]}. (6) The first

Read the complete text