Rational Pontryagin classes and functor calculus

DOI10.4171/JEMS/629MaRDI QIDQ738799zbMATH OpenFDO

Publication date 16 August 2016

Published in Journal of the European Mathematical Society (JEMS) (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1102.0233

functor calculus Pontryagin classes smoothing theory

Homology of classifying spaces and characteristic classes in algebraic topology (55R40) Smoothing in differential topology (57R10) Characteristic classes and numbers in differential topology (57R20)

Abstract: It is known that in the integral cohomology of BSO(2m), the square of the Euler class is the same as the Pontryagin class in degree 4m. Given that the Pontryagin classes extend rationally to the cohomology of BSTOP(2m), it is reasonable to ask whether the same relation between the Euler class and the Pontryagin class in degree 4m is still valid in the rational cohomology of BSTOP(2m). In this paper we use smoothing theory and tools from homotopy theory to reformulate the hypothesis, and variants, in a differential topology setting and in a functor calculus setting.

Review text

If \(n\) is even, then \(e^2= p_{n/2}\) in \(H^{2n}(BSO(n); \mathbb{Q})\), where \(e\) is the Euler class and \(p_{n/2}\) the Pontryagin class. This paper studies this relation in the rational cohomology of \(BSTOP(n)\). The authors offer three hypotheses. Hypothesis A states that if \(n\) is even, then \(e^2= p_{n/2}\) in \(H^{2n}(BSTOP(n); \mathbb{Q})\). Hypothesis B involves the space \(R(n,2)\) of smooth maps from \(D^n\times D^2\) to \(D^2\) which agree with the projection onto \(D^2\) near the boundary. For \(f\in R(n,2)\), \(df\) can be viewed as a point in \(\Omega^{n+2}Y\), where \(Y\) is the space of surjective maps from \(\mathbb{R}^{n+2}\) to \(\mathbb{R}^2\) taking the boundary of \(D^n\times D^2\) to the base point. The result is a map \(\nabla:R(n,2)\to \Omega^{n+2}Y\), where \(\nabla(f):= df\). The map \(\nabla\) is an \(S^1\)-map and if \(Y\) is replaced by its path connected rationalization, the target of \(\nabla\) is a path connected \(K(\mathbb{Q},n-3)\). Hypothesis B states that the cohomology class \(|\nabla|\in H^{n-3}_{S^1}(R(n,2),*;\mathbb{Q})\) is zero for even \(n\geq 4\). Hypothesis C involves two functors on the category of finite-dimensional inner product spaces with morphisms injective linear maps respecting inner products. If \(Bo\) is the functor \(V\mapsto BO(V)\) and \(Bt\) is the functor \(V\mapsto BTOP(V)\), then the Hypothesis C states that the inclusion \(Bo\to Bt\) admits a rational left inverse up to weak equivalence. The paper offers the results that C implies A for all even \(n\geq 4\) and A for a specific \(n\) implies B for the same \(n\).

Recommendations

Cites work

Cited in

(6)

This page was built for publication: Rational Pontryagin classes and functor calculus

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q738799)