Groups with all centralizers subnormal of defect at most two. (Q351410)

A group \(G\) is called \(k\)-Engel if the normed commutator \([x,{_ky}]=1\) for all \(x,y\in G\). By a result of \textit{W. Kappe} [Ill. J. Math. 5, 187-197 (1961; Zbl 0098.25001)], a group \(G\) is \(2\)-Engel if and only if the centralizer \(C_G(x)\trianglelefteq G\) of all elements \(x\in G\). The authors of the paper under review introduce the class \(\mathfrak C_n\) of groups consisting of all groups \(G\) in which every centralizer of elements is subnormal in \(G\) of defect at most \(n\). So the class of \(2\)-Engel groups coincides with \(\mathfrak C_1\). It is observed by the authors that every \(\mathfrak C_n\)-group is locally nilpotent and \((n+1)\)-Engel. The main result of the paper under review is the following: Theorem 1.1. Let \(G\) be a group without involutions. Then \(G\) is a \(\mathfrak C_2\)-group if and only if it is a \(3\)-Engel group. The authors show that the class \(\mathfrak C_2\) strictly contains the variety of all nilpotent groups of nilpotency class at most \(3\), and is strictly contained in the variety of all \(3\)-Engel groups. The authors, by observing that \(\mathfrak C_2\)-groups are \(3\)-Engel, use the detailed description of \(3\)-Engel groups obtained by \textit{N. D. Gupta} and \textit{M. F. Newman} [Bull. Aust. Math. Soc. 40, No. 2, 213-228 (1989; Zbl 0675.20034)]. The authors have mentioned that their arguments for commutator calculus have been guided by computer calculations with GAP [The GAP Group, GAP -- Groups, Algorithms and Programming, Version 4.5.6 (2012), \url{http://www.gap-system.org}]. However the authors' commutator calculations do not depend on these computer calculations as they are completely hand proved.