Morse theory and Euler characteristic of sections of spherical varieties (Q1889941): Difference between revisions

The author generalizes the formula for the Euler characteristic of a hypersurface in the torus \((\mathbb{C}^*)^d\), due to \textit{D. N. Bernshtejn}, \textit{A. G. Kushnirenko} and \textit{A. G. Khovanskij} [Usp. Math. Nauk 31, No. 3(189), 201--202 (1976; Zbl 0354.14001)], [\textit{A. G. Kushnirenko}, Funct. Anal. Appl. 10, 233--235 (1976; Zbl 0341.32001)], [\textit{A. G. Khovanskij}, Funct. Anal. Appl. 12, No. 1, 38-46 (1978; Zbl 0406.14035)], to actions with spherical orbits, by using a variant of classical Morse theory. The author proves a formula for the Euler characteristic of hyperplane sections of generic orbits of a particular type of reductive group actions, namely actions with spherical orbits. (Let a complex connected reductive group \(G\) act linearly on a complex vector space. The action is called an action with spherical orbits if any orbit of the action is spherical, which means that a Borel subgroup of \(G\) has a dense orbit.) The main result of the paper is the following theorem. Theorem. (Corollary 2.12) Let \(G\) be a complex connected reductive group acting linearly on \(\mathbb C^n\) such that every generic orbit \(X= Gx\) is closed and spherical. Then there exists a Zariski dense open set \(U\subset(\mathbb C^n)^*\times \mathbb C\) such that if \((f, a)\in U\), we have \[ x(f^{-1}(a) \cap X)= \chi(X)+ (-1)^{\dim(X)+1}\deg(X). \] This theorem directely generalizes the theorem of Bernshtejn, Khusnirenko and Khovanskij [loc. cit.] on the characteristic of hypersurfaces in \((\mathbb{C}^*)^d\). (The author indicates also a result of \textit{M. Brion} [Duke Math. J. 58, No. 2, 397--424 (1989; Zbl 0701.14052)] for a combinatorial formula for the degree of a spherical orbit.) The main tool of the paper is a variant of Morse theory, which the author uses to relate the Euler characteristic of a hyperplane section and the number of critical points of the linear functional defining the hyperplane section. More precisely the author proves: Theorem 1.4. Let \(M\) be a (closed) complex algebraic submanifold of \(\mathbb C^n\) of complex dimension \(m\), and \(f\) a generic complex linear functional on \(\mathbb C^n\). Let \(c\) be a regular value for \(f\mid M\). Then \[ \chi(f^{-1}(c)\cap M)= \chi(M)+(- 1)^{m+1} \] (the number of critical points of \(f\mid M\)). The main result is illustrated by two interesting examples: (1) The Euler characteristic of a generic hyperplane section of \(\text{SL}(n,\mathbb{C})\) embedded in the usual way in the vector space of \(n\times n\) matrices is \((-1)^n.n\) (Cor. 3.3). (2) The Euler characteristic of a generic hyperplane section of a generic orbit of the standard representation of the complex group \(E_6\) on \(\mathbb{C}^{27}\) is equal to \(-3\) (Cor. 3.5).