The true prosoluble completion of a group: examples and open problems. (Q2642673)

This carefully structured report begins with a general introduction to topological completions of groups with respect to directed sets of normal subgroups. In particular, this leads to the notion of the `true prosoluble completion' of a group \(\Gamma\) which can be realised as the inverse limit of the projective system of all -- not necessarily finite -- soluble quotients of \(\Gamma\). (For instance, the true prosoluble completion of the infinite cyclic group \(\mathbb{Z}\) is again \(\mathbb{Z}\) equipped with the discrete topology, whereas the more frequently encountered prosoluble completion of \(\mathbb{Z}\), based upon all finite soluble images of \(\mathbb{Z}\), coincides with the familiar profinite completion \(\widehat\mathbb{Z}\).) Similarly, the `true pronilpotent completion' of a group is the inverse limit of the projective system of all nilpotent quotients of the group, and should not be mixed up with the more frequently encountered pronilpotent completion. The authors provide several examples, and they state various observations and open questions surrounding the profinite, pro-\(p\), prosoluble, true prosoluble and true pronilpotent completions of groups. The main focus of the report is on analogues of the following fundamental problem suggested by work of \textit{A.~Grothendieck} [Manuscr. Math. 2, 375-396 (1970; Zbl 0239.20065)] on linear representations and profinite completions of discrete groups: given a homomorphism \(\psi\colon\Gamma\to\Delta\) between residually finite groups which induces an isomorphism \(\widehat\psi\colon\widehat\Gamma\to\widehat\Delta\) between the respective profinite completions, what extra information is required to deduce that \(\psi\) itself is an isomorphism between the original groups? Addressing Grothendieck's more precise original questions, \textit{M. R. Bridson} and \textit{F. J. Grunewald} [Ann. Math. (2) 160, No. 1, 359-373 (2004; Zbl 1083.20023)] have shown that the extra information that \(\Gamma\) and \(\Delta\) are finitely presented is not sufficient. In the paper under review the authors point out how residually nilpotent groups which are parafree, first discovered by \textit{G.~Baumslag} [Bull. Am. Math. Soc. 73, 621-622 (1967; Zbl 0153.35001)], yield negative examples concerning the analogue of Grothendieck's problem for true pronilpotent completions. They also explain a possible approach to produce negative examples concerning the analogue of Grothendieck's problem for true prosoluble completions. This approach is based on the (as yet unproven) hypothesis that there exists a finitely generated subgroup \(\Delta\) of the pro-\(2\) completion of the first Grigorchuk group \(\mathcal G\) which contains \(\mathcal G\) properly and has the `congruence extension property'. In an appendix one finds a brief description of the first Grigorchuk group and an indication of how candidates for such a group \(\Delta\) can be constructed. The report is divided into seven sections and one appendix. The individual section headings are: 1. Introduction, 2. Completion with respect to a directed set of normal subgroups, 3. Universal property, 4. Examples of directed sets of normal subgroups, 5. True prosoluble completions, 6. Examples, 7. On the true prosoluble and the true pronilpotent analogues of Grothendieck's problem, 8. Appendix: Construction of elements in the closure of Grigorchuk group. The report ends with a substantial list of references.