The Kazhdan-Lusztig cells in certain affine Weyl groups (Q1068941): Difference between revisions

Suppose (W,S) is a Coxeter group. In [Invent. Math. 53, 165-184 (1979; Zbl 0499.20035)], \textit{D. Kazhdan} and \textit{G. Lusztig} define certain polynomials \(P_{y,w}(X)\) in \({\mathbb{Z}}(X)\), indexed by pairs (y,w) of elements of W. Suppose G is a reductive group, with Weyl group W. The Kazhdan-Lusztig polynomials for W turn out to be closely related to the geometry and representation theory of G, over \({\mathbb{C}}\) or over a finite field. Let \(W_ a\) be the affine Weyl group for G. Then the Kazhdan- Lusztig polynomials for \(W_ a\) are probably related to the representation theory of G over a p-adic field, and (more subtly) to geometry and representation theory over \({\mathbb{C}}.\) The Kazhdan-Lusztig polynomials are extraordinarily complicated things, and one does not expect to understand them very explicitly in general. In many applications, however, one uses only various equivalence relations on W, defined in terms of the polynomials. The equivalence classes are the cells of the title, and these one can hope to understand fairly well. Some of this kind of information exists for Weyl groups. This book provides it for the affine Weyl group \({\mathcal A}_ n\) of SL(n). The first sixty pages provide expository background and motivation. It is reasonably good, but there are a few significant flaws. The fundamental definition of a Hecke algebra, on page 6, blurs the transition from a prime power q to an indeterminate X. The references for omitted proofs are sometimes inadequate or incorrect; those given for Lemma 2.2.2, for example, are of no value to those not already familiar with the result. Finally, the author repeats Joseph's incorrect definition of Goldie rank polynomials. The substance of the book is the elucidation of the various Kazhdan- Lusztig equivalence relations on \({\mathcal A}_ n\). This is beautifully done. One of the main points is a generalization of the Robinson- Schensted algorithm, which provides a simple description of the Kazhdan- Lusztig two-sided cells. The arguments are explained clearly and completely, with many examples and geometric pictures to help the reader. A complete index of notation facilitates browsing. The bulk of the technical definitions are made at the beginning, rather than piece by piece. The proofs are by no means merely messy; significant extensions of the techniques developed for Weyl groups are needed. This is a very nice piece of mathematics, presented with care. Those who wish to apply or extend it will be grateful to the author.