On the Hartshorne-Hirschowitz theorem

TL;DR

This paper presents a new degeneration method using $(2,s)$-cone configurations to prove the Hartshorne-Hirschowitz theorem for lines in $ ext{P}^3$, with potential extensions to higher dimensions.

Contribution

It introduces a novel degeneration approach based on $(2,s)$-cone configurations, simplifying the proof of the theorem in $ ext{P}^3$ and suggesting applicability to higher-dimensional cases.

Findings

$(2,s)$-cone configurations are degenerations of disjoint lines in $ ext{P}^3$

The method proves generic lines impose independent conditions on surfaces of degree $d$

Potential for extending the approach to higher-dimensional postulation problems

Abstract

The Hartshorne--Hirschowitz theorem says that a generic union of lines in $\mathbb{P}^n$, $(n\geq 3)$, has good postulation. The proof of Hartshorne and Hirschowitz in the initial case $\mathbb{P}^3$ is difficult and so long, which is handled by a method of specialization via a smooth quadric surface with the property of having two rulings of skew lines. We provide a proof in the case $\mathbb{P}^3$ based on a new degeneration of disjoint lines via a plane $H\cong\mathbb{P}^2$, which we call $(2,s)$-cone configuration, that is a schematic union of $s$ intersecting lines passing through a single point $P$ together with the trace of an $s$-multiple point supported at $P$ on the double plane $2H$. In the first part of this paper, we discuss our degeneration inductive approach. We prove that a $(2,s)$-cone configuration is a degeneration of $s$ disjoint lines in $\mathbb{P}^3$, or more generally in $\mathbb{P}^n$. In the second part of the paper, we use this degeneration in an effective method to show that a generic union of lines in $\mathbb{P}^3$ imposes independent conditions on the linear system $|\mathbb{O}_{\mathbb{P}^3}(d)|$ of surfaces of given degree $d$. The basic motivation behind our degeneration approach is that it looks more systematic that gives some hope of extensions to the analogous problem in higher dimensional spaces, that is the postulation problem for $m$-dimensional planes in $\mathbb{P}^{2m+1}$.