The number of models for theories of everywhere finitely defined polygonometries (Q1963402)

In the last decade the work of the author was directed to Shelah's problem of constructing theories without linear orders that have only finitely many models. In recent papers he has constructed hypothetical examples of such theories, the theories of polygonometries. In the article under review, he computes the number of models of these theories. The author uses the notions introduced in his articles [Sib. Math. J. 36, No. 2, 368-378 (1995), translation from Sib. Mat. Zh. 36, No. 2, 419-431 (1995; Zbl 0863.51010); Sib. Math. J. 37, No. 3, 614-617 (1996), translation from Sib. Mat. Zh. 37, No. 3, 700-703 (1996; Zbl 0874.03044); and Sib. Math. J. 38, No. 4, 801-807 (1997), translation from Sib. Mat. Zh. 38, No. 4, 925-931 (1997)]. Let \({\mathcal P}=\langle P,L,\in\rangle\) be a pseudoplane, where \(P\) is the set of its points, \(L\) is the set of its lines, and \(\in\) is an incidence relation between these sets. Polygonometries \(\text{pm}(G_1, G_2, {\mathcal P})\) are defined on \(\mathcal P\) by pairs of groups \(G_1\), \(G_2\). One of these groups, \(G_1\), acts on the points of \(\mathcal P\) and the second group, \(G_2\), acts on the lines of \(\mathcal P\). For every polygonometry \(\text{pm}\) the notion of an \(n\)-gon can be defined in an obvious way. Each \(n\)-gon can be defined by vertices and sides. A polygonometry \(\text{pm}\) is called everywhere finitely defined if, for every \(n\), there are only finitely many \(n\)-gons without repeated vertices and sides. The diameter \(d(\text{pm})\) of a polygonometry is the maximal number of sides of polygonal lines of the pseudoplane. Let \(T\) be the theory of an infinite model of an everywhere finitely defined polygonometry \(\text{pm}\) and let \(I(T,\lambda)\) be the number of pairwise nonisomorphic models for \(T\) of cardinality \(\lambda\). Theorem 4. Let \(T\) be the theory of an everywhere finitely defined polygonometry \(\text{pm}=\text{pm}(G_1,G_2,{\mathcal P})\). Then (i) \(I(T,\lambda)=1\) for every \(\lambda\geq\omega\) if \(G_1\) and \(G_2\) are finite groups and \(d(\text{pm})<\infty\); (ii) \(I(T,\omega)=\omega\) and \(I(T,\lambda)=1\) for every \(\lambda>\omega\) if \(G_1\) and \(G_2\) are finite groups and \(d(\text{pm})=\infty\); (iii) \(I(T,\omega_{\alpha})= (\max(|G_1|,|G_2|, |\alpha|))^{\omega}\) if \(\omega\leq \max(|G_1|, |G_2|)\leq \omega_{\alpha}\).