Cohomological constraint on deformations of compact Kähler manifolds (Q1881098)

From MaRDI portal
Revision as of 05:02, 5 March 2024 by Import240304020342 (talk | contribs) (Set profile property.)
scientific article
Language Label Description Also known as
English
Cohomological constraint on deformations of compact Kähler manifolds
scientific article

    Statements

    Cohomological constraint on deformations of compact Kähler manifolds (English)
    0 references
    0 references
    4 October 2004
    0 references
    The paper under review presents an algebraic proof of the principle ``obstructions to deformations of compact Kähler manifolds annihilate ambient cohomology''. The proof is based on the relations between deformation problems and the differential graded Lie algebras associated to them [cf. \textit{M. Manetti}, Int. Math. Res. Not. 2002, No. 14, 719--756 (2002; Zbl 1063.58007)], and is done by using general methods of the theory of deformation functors of these algebras [see e.g., \textit{B. Fantechi} and \textit{M. Manetti}, J. Algebra 202, No. 2, 541--579 (1998; Zbl 0981.13009) and \textit{W. M. Goldmann} and \textit{J. J. Millson}, Publ. Math., Inst. Hautes Étud. Sci. 67, 43--96 (1988; Zbl 0678.53059)]. Let \(X\) be a compact Kähler manifold and \(K=KS_X\) the Kodaira-Spencer algebra of \(X\), i.e., the Dolbeault complex of the holomorphic tangent bundle of \(X\) with its natural structure of a differential graded Lie algebra. The vector space \(M=M_X=\)Hom\(^*_{\mathbb C}(H^*(X,\mathbb C),H^*(X,\mathbb C))\) of linear endomorphisms of the singular cohomology of \(X\) (with trivial differential and trivial Lie product) has a natural grading given by the Hodge decomposition. The author constructs a morphism \(\theta : H^*(X,T_X)=H^*(K)\rightarrow M\), mapping \(H^p(X,T_X)\) into \(M^{p-1}=:M[-1]^p\). He proves that every choice of a Kähler metric on \(X\) induces a canonical lifting of \(\theta\) to an \(L_{\infty}\)-morphism from \(K\) to \(M[-1]\) (Theorem A). The functor Def\(_K\) has an obstruction theory in the sense that for every small extension e\(:0\rightarrow J\rightarrow A\rightarrow B\rightarrow 0\) in the category of Artinian \(\mathbb C\)-algebras there is a map ob\(_e: \)Def\(_K\rightarrow H^2(K)\otimes J\) such that an element \(b\in\) Def\(_K(B)\) lifts to Def\(_K(A)\) if and only if ob\(_e(b)=0\). The obstruction space \(O_K\subset H^2(K)\) is defined as the vector space generated by the images of the maps \((\text{id}\otimes f)\circ \) ob\(_e\). The author constructs explicitly an \(L_{\infty}\)-morphism \(\mu: C(K)\rightarrow C(M[-1])\) between the associated graded coalgebras (Theorem 8), and from general results of the theory of deformation functors it follows that \(\mu\) induces a natural transformation \(\widetilde{\mu}:\) Def\(_K\rightarrow \)Def\(_{M[-1]}\) between the deformation functors of \(K\) and \(M[-1]\). Since the differential and the Lie product on \(M\) are trivial the obstruction space \(O_K\) is contained in the kernel of the induced map \(H^2(K)\rightarrow M[-1]^2\) (Proposition 13). These results and facts are applied to the deformations of \(X\), using the fact that the deformation functor Def\(_K\) of the Kodaira-Spencer algebra \(K\) is isomorphic to the functor Def\(_X\) of infinitesimal deformations of \(X\), see \textit{W.M. Goldman} and \textit{J. J. Millson} [Ill. J. Math. 34, No. 2, 337--367 (1990; Zbl 0707.32004)]. Finally, the above stated principle is made precise in Corollary 14: Let \(O\) be the kernel of \(\theta_2:H^2(X,T_X)\rightarrow \bigoplus_{r,s} \Hom_{\mathbb C}(H^r(\Omega^s_X),H^{r+2}\Omega^{s-1}_X))\). Then for every small extension \( 0\rightarrow J\rightarrow A\rightarrow B\rightarrow 0\) and every \(b\in\) Def\(_X(B)\) the associated obstruction belongs to \(O\otimes J\).
    0 references
    deformation functor
    0 references
    differential graded Lie algebra
    0 references
    obstruction space
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references