Abelian extensions and solvable loops. (Q2258459)

After the development of commutator theory for congruence modular varieties by \textit{R. Freese} and \textit{R. McKenzie} [Commutator theory for congruence modular varieties. Lond. Math. Soc. Lect. Note Ser. 125. Cambridge: Cambridge University Press (1987; Zbl 0636.08001)], the authors of the article under review founded the commutator theory for loops [in J. Algebra 399, 290-322 (2014; Zbl 1319.20056)] based on the fact that the variety of loops is congruence modular. The main result of their earlier research is stated in this new article as Theorem 2.1. This present article is a continuation of the earlier work of the same authors mentioned above. They investigate two notions deduced from the Freese-McKenzie commutator theory; central and abelian subloops. Their main result, Theorem 4.1, is a syntactic and semantic characterization of abelian subloops, similar to the well-known characterization of central subloops which they review in Theorem 4.2. Their semantic characterization of abelian normal subloops conduces to the notion of abelian extensions which is a generalization of central extensions. These abelian extensions of loops are shown to be precisely the affine quasidirect products in the works of some other authors, and therefore concluded that a loop is Boolean complete if and only if it is not congruence solvable. This generalizes the connection between solvability and Boolean completeness from groups and the authors conceive that this will lead to new applications of commutator theory in computational complexity. Finally, they discuss the relationships between solvability and nilpotence in loops and their associated multiplication groups and inner mapping groups in form of open problems.