Almost-free groups in varieties (Q1183287)

Let \(\mathcal V\) be a non-trivial variety of groups. A group \(G\) of \(\mathcal V\) is said to be almost-free (in \(\mathcal V\)) if every subset of \(G\) of cardinality less than that of \(G\) is contained in a subgroup of \(G\) which is free in \(\mathcal V\). (See the paper for a discussion of other possible definitions of the concept.) The main results of the paper are summarized as follows (Theorem 18). At least one of the following possibilities holds for \(\mathcal V\). (a) Every almost-free group in \(\mathcal V\) is free. (b) For every infinite cardinal \(\kappa\) there is a non-free almost-free group in \(\mathcal V\) of cardinality \(\kappa^ +\) (the successor cardinal of \(\kappa\)). (c) For every uncountable cardinal \(\kappa\), if there is a non-free almost-free abelian group of cardinality \(\kappa\) then there is a non-free almost-free group in \(\mathcal V\) of cardinality \(\kappa\). Furthermore (a) holds if and only if \(\mathcal V\) is a variety of nilpotent groups of prime-power exponent. In particular, \(\mathcal V\) has non-free almost-free groups of cardinality \(\aleph_ n\) for every positive integer \(n\) if and only if \(\mathcal V\) is not a variety of nilpotent groups of prime-power exponent. The last result generalizes a result stated by \textit{A. Pope} [J. Algebra 108, 174-178 (1987; Zbl 0625.20022)]. But the author points out that there is an error in Pope's proof (Lemma 1 of his paper is incorrect).