\(K\)-theory of virtually poly-surface groups (Q1860332): Difference between revisions

A discrete group \(\Gamma\) is called strongly poly-free if there exists a finite filtration of \(\Gamma\) by subgroups: \(1=\Gamma_0 \subset \Gamma_1 \subset \dots \subset\Gamma_n = \Gamma\) such that the following conditions are satisfied: (1) \(\Gamma_i\) is normal in \(\Gamma\) for all \(i\), (2) \(\Gamma_{i+1}/\Gamma_i\) is a finitely generated free group, (3) for each \(\gamma \in \Gamma\) and \(i\) there is a compact surface \(F\) and a diffeomorphism \(f \colon F \to F\) such that the induced homomorphism \(f_{\#}\) on \(\pi_1(F)\) is equal to \(c_{\gamma}\) in Out\((\pi_1(F))\), where \(c_{\gamma}\) is the action of \(\gamma\) on \(\Gamma_{i+1}/\Gamma_i\) by conjugation and \(\pi_1(F)\) is identified with \(\Gamma_{i+1}/\Gamma_i\) via a suitable isomorphism. \textit{F. T. Farrell} and \textit{S. K. Roushon} [Int. Math. Res. Not. 2000, 515--526 (2000; Zbl 0985.20024)] proved that the FIC (``Fibered isomorphism conjecture'', as described by \textit{F. T. Farrell} and \textit{L. E. Jones} [J. Am. Math. Soc. 6, 249--297 (1993; Zbl 0798.57018)]) is true for virtually strongly poly-surface groups, and in particular the Whitehead groups of virtually strongly poly-free groups vanish. The purpose of this paper is to generalize this result to a larger class of groups, which the author calls poly-surface groups; a discrete group \(\Gamma\) is called strongly poly-surface if there exists a finite filtration of \(\Gamma\) by subgroups: \(1 = \Gamma_0 \subset \Gamma_1 \subset \dots \subset \Gamma_n = \Gamma\) such that the following conditions are satisfied: (1) \(\Gamma_i\) is normal in \(\Gamma\) for each \(i\), (2) \(\Gamma_{i+1}/\Gamma_i\) is isomorphic to the fundamental group of a surface, (3) for each \(\gamma \in \Gamma\) and \(i\), there is a surface \(F\) such that \(\pi_1(F)\) is isomorphic to \(\Gamma_{i+1}/\Gamma_i\) and either (a) \(\pi_1(F)\) is finitely generated or (b) \(\pi_1(F)\) is infinitely generated and \(F\) has one end. Also there is a diffeomorphism \(f \colon F \to F\) such that the induced outer automorphism \(f_{\#}\) of \(\pi_1(F)\) is equal to \(c_{\gamma}\) in Out\((\pi_1(F))\), where \(c_{\gamma}\) is the outer automorphism of \(\Gamma_{i+1}/\Gamma_i \cong \pi_1(F)\) induced by the conjugation action on \(\Gamma\) by \(\gamma\). The proof of the FIC for the virtually strongly poly-surface case runs along similar lines to the virtually strongly poly-free case. The key new ingredient is the ``Main Lemma'', which deals with the case when \(\Gamma\) has a normal subgroup \(H\) which is the fundamental group of a surface and \(\Gamma/H\) is infinite cyclic. However the proof of this Lemma is incorrect, as the author has admitted. Several unpublished (at the time of writing this review) papers are on his WEB pages at \url{http://www.math.tifr.res.in/~roushon} which describe various corrections required for this paper. However the result of this paper on the FIC for virtually strongly poly-surface groups (or any version of this result) should not be used until there is independent evidence of its validity.