The original interest in sandwiched singularities comes from a natural question posed by J. Nash in the early sixties to H. Hironaka: “Does a finite succession of Nash transformations or normalized Nash transformations resolve the singularities of a reduced algebraic variety?” In 1975, A. Nobile proved that, in characteristic zero, a Nash transformation is an isomorphism only in case the original variety is already non-singular. It turns out, in particular, that curve singularities are resolved by a succession of Nash transformations. Rebasoo proved in his Ph. D. thesis that Nash transformations also resolve certain kinds of quasi-homogeneous hypersurface singularities in (C)3. In 1982, G. Gonzalez-Sprinberg proved that normalized Nash transformations resolve rational double points and cyclic quotients singularities of surfaces. Then, H. Hironaka proved that after a finite succession of normalized Nash transformations one obtains a surface “X” which birationally dominates a non-singular surface. By definition, the singularities of “X” are sandwiched singularities. Some years later, M. Spivakovsky proves that sandwiched singularities are resolved by normalized Nash transformations, thus giving a positive answer to the original question posed by Nash for the case of surfaces over C.
Since then, a constant interest in sandwiched singularities has been shown, and they have been deeply studied from the point of view of deformation theory by de Jong and van Straten, and also by Stolen and Mohring. Sandwiched singularities have been also studied as a nice testing ground for the Nash and the wedge Problem by Lejeune-Jalabert and Reguera, where the main idea is to extend combinatorial arguments for toric surface singularities to sandwiched ones.
Sandwiched singularities are the singularities obtained by blowing-up a complete ideal in the local ring of a regular point on a surface. They are rational surface singularities (roughly speaking, isolated singularities whose resolution has no effect on the arithmetic genus of the surface) and among them are included all cyclic quotients and minimal surface singularities. Sandwiched singularities are Cohen-Macaulay, but are not complete intersections and in general, there are no simple equations for them. The purpose of this memoir is to study sandwiched singularities through their relationship to the infinitely near base points of the complete ideals blownup to obtain them.
Now, we briefly summarize the main contents of each one of the chapters. Chapter I is of preliminary nature and gives references to the literature for proofs. Concepts and well-known facts about infinitely near points, weighted clusters, complete ideals and rational and sandwiched surface singularities are reviewed and some consequences that are needed in the memoir are derived. In Chapter II we establish the main link between the study of sandwiched singularities and the theory of Enriques diagrams of weighted clusters and we derive some results on sandwiched singularities by using the unloading procedure.
Chapter III deals essentially with the principality of divisors going through a sandwiched singularity. It is well known that Wei divisors going through a singularity (X, Q) are not Cartier divisors in general.
In Chapter IV we use the results of Chapter III to explore the connection between the ideal sheaves on “X” with finite cosupport contained in the exceptional locus and the complete m(o)-primary ideals in R. Chapter V is devoted to derive consequences related to the Nash conjecture of arcs for sandwiched singularities.
In Appendix A, we provide the listings of three programs in language C implementing some of the algorithms proposed. These programs have been used to compute some of the examples presented throughout the memoir.
Part of the results of this thesis has been published or will be published in:
• J. Fernandez-Sanchez, On sandwiched singularities and complete ideals,
J. Pure Appl. Algebra 185 (2003), no. 1-3, 165-175. [19]
• J. Fernandez-Sanchez, Nash families of smooth arcs on a sandwiched
singularity, To appear in Math. Proc. Cambridge. Philos. Soc. [18]
• J. Fernandez-Sanchez, Equivalence of the Nash conjecture for primitive
and sandwiched singularities, To appear in Proc. Amer. Math.
Soc. [17]