Disjointifiable lattice-ordered groups (Q998782)

Denote by \(\mathcal C(G)\) the lattice of convex \(\ell\)-subgroups of an \(\ell\)-group \(G\). Then, as is well known, \(\mathcal C(G)\) is a frame -- in fact, an algebraic frame in which the meet of any two compact elements is compact. In the event that \(\mathcal C(G)\) is normal (in the usual frame-theoretic sense), then \(G\) is called disjointifiable. In the article an injective frame homomorphism is said to be capping if its right adjoint pulls back two-element covers to covers. An \(\ell\)-subgroup \(H\) of \(G\) majorizes \(G\) if \(G =\langle H\rangle_c\), where the latter signifies the convex subgroup (of \(G\)) generated by \(H\). The mapping \(C\mapsto\langle C\rangle_c\), from \(\mathcal C(H)\) to \(\mathcal C(G)\), is a frame homomorphism if \(H\) majorizes \(G\). It is capping iff the trace of \(H\) on any two-element cover of \(\mathcal C(G)\) is a cover of \(\mathcal C(H)\); in which case \(H\) is said to be a capping of \(G\). A number of results lead to a proposition part of which states that if \(H\) is a capping of \(G\), then \(\mathcal C(G)\) is normal iff \(\mathcal C(H)\) is normal. There are numerous deep results in the paper, and, as is his claim, the author presents them in a very lucid style. A question he would have loved to answer is: ``When is a Hahn group disjointifiable?'' In an attempt to tackle this, he establishes a number of theorems, including a lovely necessary condition that \(A\vee B = G\) in \(\mathcal C(G)\) implies the existence of a cardinal summand \(P\) of \(G\) such that \(P\subseteq A\) and \(P^\bot\subseteq B\); where, in the usual frame-theoretic parlance, \(P^\bot\) denotes the pseudocomplement of \(P\) in \(\mathcal C(G)\).