Contraction of nonsingular curves (Q1187740): Difference between revisions

Let \(Y\) be a smooth Riemann surface of genus \(g\) lying inside an analytic space \(X\). The author gives a cohomological criterion for contracting \(Y\) to a point, that is for constructing an analytic morphism \(f:X \to X'\) such that \(f(Y)=\{x'\}\) and \(f|_{X-Y}\) is an isomorphism onto \(X- \{x'\}\), establishing the following result: \(Y \subseteq X\) contracts to a point if and only if there is a line bundle \(L\) on a neighborhood of \(Y\) in \(X\) such that \(\deg(L |_ Y) \leq g-2\) and \(H^ 1(\hat X,\hat L)\) is finite-dimensional. He also establishes that this result gives restrictions on the second- order neighborhood of \(Y\) in \(X\). Furthermore he also proves that \(Y\) is formally seminegative in \(X\) if and only if there exists a formal analytic neighborhood \(V'\) of a point \(x'\) and an adic morphism of formal analytic space \(f:\hat X \to V'\). Finally the author gives a negative answer to the following question proposed by \textit{D. Morrison} [Classification of algebraic and analytic manifolds, Proc. Symp., Katata/Jap. 1982, Progr. Math. 39, 607-609 (1983)]: Is it true that either the curve contracts or an analytic subspace \(Z\) of \(X\) with \(\sup(Z)=Y\) moves in \(X\)?