The first derived limit and compactly \(F_\sigma\) sets (Q1298032)

For a~sequence \(a=\{a(n)\}\) of integers (a~member of~\(\mathbf N^{\mathbf N}\)) let \(A_a\) be the group of all functions from the set \(D_a=\{(m,n):n\leq a(m)\}\) in~\(\mathbf Z\) which are equal to~\(0\) on all but finitely many pairs \((m,n)\in D_a\), i.e., \(A_a= \bigoplus_{m=1}^\infty \bigoplus_{n=1}^{a(m)}\mathbf Z\). The set \(\mathbf N^{\mathbf N}\) is directed with the ordering defined by \(a\leq b\) if and only if \(a(n)\leq b(n)\) for all~\(n\). For \(a\leq b\) in \(\mathbf N^{\mathbf N}\) let \(\pi^b_a:A_b\to A_a\) be the natural projection. The author shows some connections of the first derived limit \(\varprojlim^{(1)}\mathcal A\) of the inverse system \(\mathcal A= \langle A_a,\pi^b_a,\mathbf N^{\mathbf N}\rangle\) with descriptive set theory. He proves: (1)~If \(\varprojlim^{(1)}\mathcal A\neq 0\) then there is a~subset \(X\) of \(\mathbf R\setminus\mathbf Q\) which is not analytic but its intersection with every compact subset of \(\mathbf R\setminus\mathbf Q\) is~\(F_\sigma\); and (2) PFA implies that there exists a \(\sigma\)-closed poset which adds no new subsets of \(\omega_1\) and which forces \(\varprojlim^{(1)}\mathcal A\neq 0\). Therefore Martin's axiom together with negation of CH does not imply that a subset of \(\mathbf R\setminus\mathbf Q\) is Borel if its intersection with every compact subset of \(\mathbf R\setminus\mathbf Q\) is Borel.