Comparing axiomatizations of free pseudospaces (Q1031828)

The paper under review studies the relationship between two similar constructions of pseudospaces, i.e., certain point-line-plane configurations, providing interesting examples of \(\omega\)-stable theories. The free pseudospace of \textit{A. Baudisch} and \textit{A. Pillay} [J. Symb. Log. 65, No.~1, 443--460 (2000; Zbl 0947.03050)] provides a non-CM-trivial theory (denoted by \(\Sigma\)) of rank \(\omega^2\) which does not interpret an infinite field; \textit{E. Hrushovski} and \textit{G. Srour} [``On stable non-equational theories'' (unpublished manuscript) (1989)] construct an \(\omega\)-stable non-equational theory (denoted by \(\Gamma\)), using similar ideas. CM-triviality has been introduced by \textit{E. Hrushovski} [Ann. Pure Appl. Logic 62, No.~2, 147--166 (1993; Zbl 0804.03020)]. It is a property of geometric complexity of stable theories which is incompatible with the existence of an infinite interpretable field and which is more general than one-basedness. The new strongly minimal set (a counter-example to Zilber's Trichotomy Conjecture) constructed by Hrushovski in [loc. cit.] is CM-trivial, and there are nilpotent CM-trivial groups which are not abelian-by-finite, by a result of \textit{A. Baudisch} [Trans. Am. Math. Soc. 348, No.~10, 3889--3940 (1996; Zbl 0863.03019)]. In the same way as the free pseudoplane of Lachlan (based on a certain point-line configuration) is the prototype of a stable structure which is not one-based, a pseudospace is the prototype of a non-CM-trivial stable theory. Equationality, a notion introduced and studied by G. Srour, provides another generalisation of stable one-based theories. A formula \(\phi(\overline{x},\overline{y})\) (where \(\overline{y}\) are the parameter variables) is called an \textit{equation} if an arbitrary intersection of instances \(\bigwedge_{i\in I}\phi(\overline{x},\overline{a}_i)\) is equivalent to an intersection \(\bigwedge_{i\in I_0}\phi(\overline{x},\overline{a}_i)\) for some finite \(I_0\subseteq I\). A theory is called \textit{equational} if every formula is equivalent to a Boolean combination of equations. One-based stable theories are equational, and equational theories are stable. In his paper, the author first considers a coloured version of Lachlan's pseudoplane which serves as a building block for the construction of a standard model of \(\Gamma\) from a standard model of \(\Sigma\). As a main result, it is shown that \(\Gamma\) is a reduct of \(\Sigma\), obtained by omiting the colouring. Moreover, a duality result for pseudospaces is established from which a simplified axiomatisation for \(\Sigma\) can be deduced. In the last section of the paper, a detailed exposition of the main result of Hrushovski and Srour [loc. cit.] is given, namely that \(\Gamma\) is an \(\omega\)-stable complete non-equational theory, the `coloured incidence relation' being a formula which is not equivalent to a Boolean combination of equations. It is noted that \(\Gamma\) does have equational forking, leaving as an open problem the existence of simple (or even stable) theories where forking is not controlled by equations.