The fundamental group of compact Kähler threefolds (Q2285037)

A fundamental question in Kähler geometry asked if the fundamental group \(\pi_1(X)\) of every compact Kähler manifold \(X\) is projective, namely there is there exists a projective manifold \(Y\) such that \(\pi_1(X)=\pi_1(Y)\). This question will be trivially answered if the famous Kodaira problem has a positive answer: any compact Kähler manifold is deformation equivalent to a projective manifold. However, \textit{C. Voisin}'s counterexamples [Invent. Math. 157, No. 2, 329--343 (2004; Zbl 1065.32010)] show that there exist compact Kähler manifolds of dimension at least four that do not deform to projective ones. In the paper under review, the authors completely solves this question for Kähler threefolds: Let \(X\) be a smooth compact Kähler threefold. Then \(\pi_1(X)\) is projective. The proof of the result comes in several step. if \(X\) is covered by rational curves, then the tools from maximal rationally connected fibrations by \textit{F. Campana} [Ann. Sci. Éc. Norm. Supér. (4) 25, No. 5, 539--545 (1992; Zbl 0783.14022)] and \textit{J. Kollár} et al. [J. Differ. Geom. 36, No. 3, 765--779 (1992; Zbl 0759.14032)] will reduce this problem to lower dimensional cases. If \(X\) is not covered by rational curves, then the authors make a case distinction based on the algebraic dimension \(a(X)\). The \(a(X)=0\) has been solved by \textit{F. Campana} and \textit{B. Claudon} [Proc. Edinb. Math. Soc., II. Ser. 57, No. 1, 55--78 (2014; Zbl 1337.32029)] and \textit{P. Graf} [Math. Ann. 371, No. 1--2, 487--516 (2018; Zbl 1405.32022)]. The main contribution of this paper is on \(a(X)=1\) or \(a(X)=2\). The most difficult case is when \(a(X)=2\), where the resolution of the algebraic reduction defines an elliptic fibration. The authors use the theory of elliptic fibrations developed by \textit{N. Nakayama} [Adv. Stud. Pure Math. 35, 185--295 (2002; Zbl 1059.14015)] to show the existence of a smooth bimeromorphic model \(X'\) of \(X\) which admits algebraic approximations, which thus solved the problem in this case. Let me mention that a more recent work [``Algebraic approximations of compact Kähler threefolds'', Preprint, \url{arXiv:1704.08109}] by \textit{H. Lin} solved the Kodaira problem for Kähler threefolds.