Tangent cones to Schubert varieties in types \(A_{n}\), \(B_{n}\) and \(C_{n}\) (Q306542): Difference between revisions

Let \(G\) be a complex semisimple simply-connected algebraic group and let \(T\) be a maximal torus of \(G\), \(B\) a Borel subgroup of \(G\) containing \(T\) and \(W\) the Weyl group of \(G\). Let \(F = G/B\) be the full flag variety and let \(X_w\subset F\) be the Schubert variety, for any \(w\in W\). In the paper under review, the authors prove that if every irreducible component of \(G\) is of type \(B_n\) or \(C_n\), and \(w_1, w_2\) are two distinct involutions in \(W\), then the tangent cones at the base point \(eB\in X_{w_i}\) to the corresponding Schubert subvarieties \(X_{w_1} , X_{w_2}\) in \(F\) do not coincide as subschemes of the tangent space \(T_{eB}(F)\). Similarly, they also show that if every irreducible component of \(G\) is of type \(A_n\) or \(C_n\), then the reduced tangent cones to \(X_{w_1} , X_{w_2}\) do not coincide as subvarieties of \(T_{eB}(F)\). Their first result is proved by using (what they call) Kostant-Kumar polynomials (so was the following result by Eliseev and Ignatyev). Their second result is proved by using a connection between the tangent cones of \(X_w\) and the geometry of coadjoint orbits of \(B\). In a previous work [J. Math. Sci., New York 199, No. 3, 289--301 (2014); translation from Zap. Nauchn. Semin. POMI 414, 82--105 (2013; Zbl 1312.14116)], \textit{D. Yu. Eliseev} and \textit{M. V. Ignatyev} had proved that \(X_{w_1} , X_{w_2}\) (for distinct involutions \(w_1, w_2\)) do not coincide as subschemes of the tangent space \(T_{eB}(F)\) when the irreducible components of \(G\) are of type \(A_n\), \(F_4\) and \(G_2\) only (partially confirming an earlier conjecture by Panov in 2011 for any \(G\)).