Bernhard Riemann 1861 revisited: existence of flat coordinates for an arbitrary bilinear form (Q6063908)

This important paper completely solves a classical problem of Riemannian geometry, going back to Riemann himself, who solved it in a special case, after having enunciated it in his \textit{Habilitationsvortrag} of 1854, in his 1861 submission for a prize question of the Paris Academy, \textit{Commentatio mathematica, qua respondere tentatur quaestioni ab Illma Academia Parisiensi propositae}. There Riemann considered what is now called a Riemannian metric, that is, a symmetric positive definite 2-form \(g = g_{ij}(x)\), asked and answered the question under what conditions there exists a coordinate system such that \(g\) is given by a constant matrix. The answer is that such coordinates exist locally if and only if what came to be known as the Riemann curvature tensor is identically zero. One notices that the assumption of positive definiteness is not essential for Riemann's proof, that it is enough for the symmetric form to be nondegenerate. The case in which the bilinear form is skew-symmetric was considered and solved, thereby laying the foundation for symplectic geometry, by \textit{G. Darboux} [Bull. Sci. Math. 6, 14--36, 49--68 (1882; JFM 14.0294.01)], where he proved that there exists a local coordinate system such that a nondegenerate differential 2-form \(\omega = \omega_{ij}(x)\) is given by a constant matrix, if and only if \(\omega\) is closed. The aim of the paper under review is to provide necessary and sufficient conditions that offer a complete answer to Riemann's and Darboux's question for an arbitrary bilinear form, i.e., a tensor field of type \((0, 2)\), which may have nontrivial symmetric and skew-symmetric parts that can be degenerate. Since the case in which the symmetric part is nondegenerate can be approached with the methods of Riemann, the results in the paper under review are new only in the case in which \(g\) is degenerate and \(\omega\) is arbitrary. The results are formulated such that ``the hypothesis on \(g\) and \(\omega\) can effectively be checked using only differentiation and algebraic manipulations, as was the case in the results of Riemann and Darboux''. Moreover, particular care has been taken to use a notational language and methods that ``were available to, and used by, Riemann, Darboux and other fathers of differential geometry. These methods include basic real analysis, basic linear algebra and the standard results on the existence and uniqueness of solutions of systems of ordinary differential equations.'' The proofs are first presented in a form that ``would be understood by Bernhard Riemann and mathematicians coming shortly after him, such as Sophus Lie, Gregorio Ricci-Curbastro, Gaston Darboux, Tullio Levi-Civita and Ferdinand Georg Frobenius'', i.e., under the assumption ``that all objects are sufficiently smooth'', to be later, whenever possible, given in minimal regularity. There are, however, three concepts used in the paper under review, which were not available in Riemann's or Darboux's time: Levi-Civita's parallel transport, the idea of the holonomy group, together with the Ambrose-Singer Theorem, and ideas from the theory of integrable Hamiltonian systems.