Conjugation spaces

Abstract

There are classical examples of spaces X with an involution tau whose mod 2-comhomology ring resembles that of their fixed point set X^tau: there is a ring isomorphism kappa: H^2*(X) --> H^*(X^tau). Such examples include complex Grassmannians, toric manifolds, polygon spaces. In this paper, we show that the ring isomorphism kappa is part of an interesting structure in equivariant cohomology called an H^*-frame. An H^*-frame, if it exists, is natural and unique. A space with involution admitting an H^*-frame is called a conjugation space. Many examples of conjugation spaces are constructed, for instance by successive adjunctions of cells homeomorphic to a disk in C^k with the complex conjugation. A compact symplectic manifold, with an anti-symplectic involution compatible with a Hamiltonian action of a torus T, is a conjugation space, provided X^T is itself a conjugation space. This includes the co-adjoint orbits of any semi-simple compact Lie group, equipped with the Chevalley involution. We also study conjugate-equivariant complex vector bundles (`real bundles' in the sense of Atiyah) over a conjugation space and show that the isomorphism kappa maps the Chern classes onto the Stiefel-Whitney classes of the fixed bundle.

Published by Algebraic and Geometric Topology at http://www.maths.warwick.ac.uk/agt/AGTVol5/agt-5-39.abs.html

Cited by in corpus (14)

• Steenrod squares on conjugation spaces
• A variant of K-theory and topological T-duality for Real circle bundles
• The FKMM-invariant in low dimension
• Conjugation Spaces are Cohomologically Pure
• Involutions on S^6 with 3-dimensional fixed point set
• Cohomology of Quotients in Real Symplectic Geometry
• Conjugations on 6-manifolds with free integral cohomology
• Conjugation spaces and 4-manifolds
• Halving spaces and lower bounds in real enumerative geometry
• Realizing doubles: a conjugation zoo
• The equivariant cohomology for semidirect product actions
• Conjugation spaces and equivariant Chern classes
• Real loci of based loop groups
• Do manifolds have little symmetry?