Topological mirror symmetry (Q5944878)

Some years ago, \textit{A. Strominger}, \textit{S.-T. Yau} and \textit{E. Zaslow} [Nucl. Phys., B 479, No. 1--2, 243--259 (1996; Zbl 0896.14024)] proposed an approach to mirror symmetry involving special Lagrangian submanifolds of Calabi-Yau manifolds. A famous conjecture of theirs, the so-called SYZ conjecture, suggests -- in a properly modified form -- that mirror symmetry can be explained by the existence, in a mirror pair of Calabi-Yau manifolds, of special Lagrangian \(T^n\)-fibrations. More precisely, the SYZ conjecture predicts that Calabi-Yau manifolds that are what is called ``close to the large complex structure limit'' admit special Lagrangian torus fibrations which allow to construct, via a certain dualizing procedure, their mirror manifold. In the paper under review, the author continues his exploration of the SYZ conjecture. In a series of foregoing papers, whose contents have been masterly surveyed by himself, in the meantime, in his treatise [in: Calabi-Yau manifolds and related geometries (Berlin: Springer), Universitext. 69--159, 227--236 (2003; Zbl 1014.14019)], the author has focused on the geometry and topology of conjectural special Lagrangian fibrations. Now, in the present paper, he turns to the topology, monodromy, and dualization of a particular topological fibration on the simplest classical Calabi-Yau threefold, the celebrated quintic. Extending some more recent work by \textit{I. Zharkov} [Duke Math. J. 101, No. 2, 237--257 (2000; Zbl 0957.14031)], who first constructed topological torus fibrations on the quintic, and by \textit{W.-D.Ruan} [cf.: Lagrangian torus fibration of quintic Calabi-Yau hypersurfaces. I: Fermat quintic case, preprint, \texttt{http://arxiv.org/abs/math.DG/9904012}], who established concrete Lagrangian fibrations, the author gives affirmative answers to the following crucial questions: 1. Does there exist a class of torus fibrations for which duals can be explicitely constructed? 2. Do such fibrations exist on Calabi-Yau threefolds, e.g., on the quintic? 3. If they do, does a suitable dualizing procedure produce the mirror manifold? His positive answers to these questions prove that in at least one non-degenerate case, the SYZ conjecture explains the mystery of mirror symmetry on a topological level. The fundamental problem in this context is that, in order for the SYZ conjecture to be useful in constructing mirror manifolds, one has to produce the mirror manifold by dualizing and compactifying torus fibrations without any a priori knowledge of the wanted mirror manifold. Inspired by W.-D. Ruan's monodromy calculations, the author solves this problem by building a topological \(T^3\)-fibration over the Calabi-Yau threefold at hand, and analyzes it thoroughly by means of his results established in his earlier papers. Another basic tool of his fine analysis is \textit{C. T. C. Wall's} classification of 6-manifolds [Invent. Math. 1, 335--374 (1966; Zbl 0149.20601)], together with D. Joyce's local models of special Lagrangian torus fibrations [cf.: \textit{D. Joyce}, Singularities of special Lagrangian fibrations and the SYZ conjecture, preprint, \texttt{http://arXiv.org/abs/math.DG/0011179}]. One of the author's main theorems (Theorem 0.2) states the following: The quintic threefold \(X\subset\mathbb{P}^4\) has a ``well-behaved'' (in some sense) \(T^3\)-fibration with semi-stable fibres \(f: X\to B\). Also, there is a ``dual'' (in some sense) fibration \(\check f:\check X\to B\) such that \(\check X\) is diffeomorphic to a specific non-singular minimal model of the mirror quintic. Along the proof of this theorem, the author also discovers a geometric explanation for the phenomenon of topology change in mirror symmetry, which was first observed by \textit{P. S. Aspinwall}, \textit{B. R. Greene} and \textit{D. R. Morrison} [Nucl. Phys., B 416, No. 2, 414--480 (1994; Zbl 0899.32006)]. Altogether, this is a highly substantial and impressive paper which reveals many aspects of the topology of torus fibrations, in general, and which represents another milestone in the study of mirror symmetry in view of the SYZ conjecture.