Lie algebras graded by finite root systems and intersection matrix algebras (Q1923255)

Recently, the dominance of the study of affine Kac-Moody algebras in Lie theory has been challenged by a growing interest in various classes of algebras graded by root systems. For example, \textit{R. V. Moody} and \textit{J. Pianzola} [Lie algebras with triangular decompositions (Wiley, New York) (1995; Zbl 0874.17026)] foreground the issue of the root system, and this approach underliesthe program of \textit{S. Berman} et al., as in [Extended affine Lie algebras and their root systems, Mem. Am. Math. Soc. 603 (1997; Zbl 0879.17012)]. The authors complete the classification up to central isogeny (i.e., isomorphic (perfect) universal central extensions) of Lie algebras over a field of characteristic zero graded by an irreducible reduced finite root system, \(\Delta\). Previously, \textit{S. Berman} and \textit{R. V. Moody} [Invent. Math. 108, 323-347 (1992; Zbl 0778.17018)] gave such a classification for simply-laced algebras of rank at least 2, while \textit{E. Neher} [Am. J. Math. 118, 439-491 (1996; Zbl 0857.17019)] gave a classification for all root systems except those of types \(E_8\), \(F_4\) and \(G_2\) from a Jordan algebra point of view. In this paper, the authors consider the non-simply laced cases and \(A_1\)-graded algebras. The classification essentially splits into two parts. The case of \(C_n\) (including \(C_1=A_1\)) is very similar to that of \(A_n\), and the recognition theorem the authors derive is presented as a consequence of Jacobson's Coordinatization theorem for Jordan algebras. For the remaining cases, the authors introduce a generalization of the Tits construction to obtain a suitable canonical representative of each central isogeny class. The techniques used derive ultimately from Seligman's rational methods, which can be seen as a pioneering work in this field. If \(\mathfrak g\) is a split semisimple Lie algebra of type \(\Delta\), then a \(\Delta\)-graded Lie algebra \(L\) is centrally isogenous with a canonical algebra \(\mathcal D\) of the general form \[ \mathcal D = ({\mathfrak g}\otimes A) \oplus (V \otimes B) \oplus C. \] The bulk of the paper is devoted to determining the precise choices for algebras \(A\), \(B\) and \(C\) and the space \(V\) (and the multiplication). This is an analysis which draws on the spectrum of associative, alternative and Jordan algebras. In the later parts of the paper, the authors consider superalgebras and intersection matrices and also note that, for sufficiently large \(p\), the results in characteristic \(p\) are identical to those in characteristic zero. The authors have produced a beautiful and timely paper. They have given an admirably clear and well-organized account of central classification results in an area of growing importance.