Flag varieties. An interplay of geometry, combinatorics, and representation theory (Q5901524)

Let \(K\) be a field (often assumed in the book to be algebraically closed) and \(V\) be an \(n\)-dimensional \(K\)-vector space. A \textit{flag} in \(V\) is a chain of subspaces: \(\{0\} = V_0 \subset V_1 \subset V_2 \subset \cdots \subset V_n = V\) where the dimension of \(V_i\) is \(i\). The set of all flags in \(V\) is called the \textit{flag variety}. Consider the general linear group \(GL_n(K)\). The flag variety can be identified with the quotient of \(GL_n(K)\) by the Borel subgroup of upper triangular matrices. Equivalently, one can take the special linear group \(SL_n(K)\) modulo upper triangular matrices. For an arbitrary semisimple algebraic group \(G\) and Borel subgroup \(B\), the quotient \(G/B\) may be considered as a \textit{generalized flag variety}. The (classical) flag variety can also be considered as a subvariety of the \textit{Grassmannian variety} \(G_{d,n}\) where \(1 \leq d \leq n - 1\). The Grassmannian variety \(G_{d,n}\) is by definition the set of all \(d\)-dimensional subspaces of \(V\). Using the Plücker map, the variety \(G_{d,n}\) can be identified with a subvariety of the projective space on the \(d\)th exterior power of \(V\); in fact, as the zero set of the so-called \textit{Plücker relations}. The Grassmannians can also be placed into a general algebraic group context since \(G_{d,n}\) can be identified with the quotient of \(GL_n(K)\) by a parabolic subgroup. For either the flag variety or a Grassmannian variety, an important collection of varieties are the Schubert subvarieties. For the flag variety, considered as \(G/B\), given an element \(w\) of the Weyl group, the Schubert variety associated to \(w\) is the Zariski closure of the \(B\)-orbit of the coset \(wB\) inside \(G/B\). For the Grassmannian \(G_{d,n}\), let \(I_{d,n}\) denote the set of \(d\)-tuples \((i_1, i_2, \dots, i_d)\) such that \(1 \leq i_1 < i_2 < \cdots < i_d \leq n\). Classically and more combinatorially, a Schubert variety can be associated to each such tuple by using a fixed basis for \(V\). These varieties can also be identified using \(B\)-orbits (for \(B\) being the upper triangular matrices inside \(GL_n(K)\)). Flag varieties, Grassmannian varieties, and Schubert varieties have played a key role in the development of algebraic geometry and related subjects (such as algebraic groups). The goal of this book is to provide an introduction to these objects, presented (as suggested by the title) from the perspective that flag varieties involve the interplay of algebraic geometry, algebraic groups, combinatorics, and representation theory. This book stems from a series of lectures given by the first author at the Institute for Advanced Study in 2007 and is aimed at an introductory level. The authors are quite thorough in presenting the necessary background material to develop the subject. In principle, with a solid background in graduate algebra, a student or researcher could learn the basic ideas from this book, although some prior knowledge of algebraic geometry and algebraic groups would be beneficial. The discussion of flag varieties is focused on the classical case (working with the general or special linear group) in order to bring the reader more quickly to the key ideas. Sufficient references are provided to direct the interested reader to the more general case and further developments. To help the reader gain understanding of the ideas, a number of examples are given as well as proofs of most results. References are provided for the results (typically more significant) which are stated without proof. If used as a textbook, all but the first chapter contains a few exercises, which could be supplemented with problems from standard texts. We now briefly describe the contents of the book, from which one can see that it could serve as a concise reference for a number of topics. The first chapter provides a brief overview of key ideas from commutative algebra (e.g., Noetherian, localization, radicals, Krull dimension, regular, and Cohen-Macauley) and algebraic geometry (e.g., affine and projective varieties, schemes, sheaves of modules, vector and line bundles, the Picard group and tangent spaces). Chapter Two presents a basic discussion of semisimple modules and rings, including the Artin-Molien-Wedderburn Structure Theorem. Brauer groups, central simple algebras, and group algebras (including Maschke's Theorem) are also introduced. Chapter Three presents some basic ideas in the representation theory of finite groups over fields of characteristic zero or prime to the order of the group, including characters, irreducible characters, tensor products, restriction and induction. Chapter Four then focuses on the representation theory of the finite symmetric group \(S_n\) over fields of characteristic zero or prime larger than \(n\). Young tableaux are introduced, which appear throughout the remainder of the book. The simple (or irreducible) \(S_n\)-modules correspond to partitions of \(n\), and a tableaux can be formed from a partition. Two constructions of the simple modules are presented: Frobenius-Young modules (arising as left ideals within the group algebra) and Specht modules (arising from \(S_n\) acting on a polynomial ring in \(n\) variables). The chapter ends with a discussion of the representation theory of the alternating group \(A_n\). In Chapter Five, the discussion continues with computations on the characters and dimensions of Young modules. The chapter contains some general discussion of symmetric polynomials and various bases for them. Beginning with Chapter Six, the focus shifts toward algebraic groups. The chapter begins with a discussion of endomorphism algebras and then discusses Schur-Weyl Duality: the isomorphism of the group algebra of \(GL_n(\mathbb{C})\) with an endomorphism algebra over \(S_d\) and vice versa. This duality allows one to obtain some of the simple \(GL_n(\mathbb{C})\), the so-called Schur modules, from Young modules. The characters of the Schur modules are discussed. Over \(SL_n(\mathbb{C})\), all of the finite dimensional simple modules arise as certain Schur modules. From these, one can then give a complete description of the finite dimensional simple \(GL_n(\mathbb{C})\)-modules. Before beginning a general discussion of algebraic groups, Chapters Seven and Eight discuss Lie algebras. Chapter Seven provides a brief introduction to Lie algebras and the structure of semisimple Lie algebras, including root systems. Precise details are given only for the special linear Lie algebra \(\mathfrak{sl}_n(\mathbb{C})\). Chapter Eight turns to the representation theory of semisimple Lie algebras (in characteristic zero). The universal enveloping algebra and weight theory are introduced, along with the construction of irreducible highest weight modules. The chapter ends with Weyl's character and dimension formulas, and some specific discussion of the \(\mathfrak{sl}_n(\mathbb{C})\) case. Chapters Nine through Eleven discuss algebraic groups. The basic concepts are presented in Chapter Nine, including the associated Lie algebra, tori, Borel subgroups, parabolic subgroups, Jordan decomposition, semisimple groups, reductive groups, and group actions. The flag variety is introduced along with its connection to \(G/B\) (as discussed above). Finally, the collection of Borel subgroups is studied and seen to be identifiable with the flag variety \(G/B\). In Chapter Ten, the basic structure theories of reductive and semisimple algebraic groups are presented, including a discussion of groups of adjoint and universal type for a given root system. Schubert varieties are then introduced along with the Bruhat Decomposition of \(G/B\) into a union of \(B\)-orbits. In Chapter Eleven, the representation theory of semisimple algebraic groups is discussed, including weight theory and the correspondence between simple modules and dominant weights, and the simple modules are constructed (algebraically and geometrically) in characteristic zero. Weyl modules are introduced as duals to sections of a line bundle on \(G/B\), and their characters and dimensions are given (analogous to the Lie algebra case) by Weyl's character formula. In characteristic zero, the Weyl modules are simple, but that is generally not true in prime characteristic. Related to the algebraic group discussion, the book contains an Appendix on Chevalley groups. In Chapter 12, the reader arrives in a sense at the goal of the book. Grassmannian varieties and their Schubert subvarieties are introduced, including a thorough discussion of the Plücker map, coordinates, and relations. A Plücker coordinate \(p_{\tau}\) for \(G_{d,n}\) corresponds to a \(d\)-tuple \(\tau \in I_{d,n}\). The set \(I_{d_n}\) of \(d\)-tuples is partially ordered by \((i_1,i_2,\dots,i_d) \leq (j_1,j_2,\dots, j_d)\) if and only if \(i_s \leq j_s\) for each \(s\). A monomial \(p_{\tau_1}p_{\tau_2}\dots p_{\tau_m}\) in Plücker coordinates is said to be \textit{standard} if \(\tau_1 \geq \tau_2 \geq \cdots \geq \tau_m\). For a Schubert variety \(X(\omega)\) associated to a tuple \(\omega\), such a monomial is standard \textit{on \(X(\omega)\)} if in addition \(\omega \geq \tau_1\). The notion of \textit{Standard Monomial Theory} is developed, with the key result being that the monomials of degree \(m\) standard on \(X(\omega)\) give a \(K\)-basis for the degree \(m\) homogeneous coordinates of \(X(\omega)\). Further discussion is also given on unions and intersections of Schubert varieties. Next, the well-known sheaf cohomology vanishing theorem is presented. Specifically, for a Schubert variety \(X\), the higher cohomology of powers of \(\mathcal{O}_X(1)\) vanishes, and the zeroth cohomology (global sections) has a basis given by standard monomials. The remainder of the chapter presents an analogous development of standard monomial theory (and the vanishing theorem) for the flag variety. Here, one works with tableaux and must consider \textit{standard tableaux} (which extends the classical notion of standard). In Chapter Thirteen, computations of the singular locus (the non-smooth points) of Schubert varieties for \(SL_n(K)/B\) are presented. Information on the singular locus can be obtained from combinatorial information in the Weyl group. Finally, in Chapter Fourteen, two key applications are presented. The first application involves classical invariant theory. With \(V\) as above, let \(X\) be the sum of \(m\)-copies of \(V\) along with \(q\)-copies of the dual module \(V^*\) for \(m, q > n\). Let \(GL_n(K)\) act diagonally on \(X\). Of interest is the subring of \(GL_n(K)\)-fixed points of \(K[X]\). This can be identified with the coordinate algebra of a determinantal variety. The determinantal variety can be identified with an open subset of a Schubert variety, which allows one to use the standard monomial theory for Schubert varieties to identify generators and relations for the ring of invariants. The second application is the degeneration of Schubert varieties (for the Grassmannian or flag variety) to toric varieties. The toric varieties are associated to distributive lattices within \(I_{d,n}\) (or a union of such).