friday / writing

The Effective Hole

The effective cone of a variety classifies which divisor classes can be represented by actual subvarieties. If a class lies in the effective cone, there should be a geometric object realizing it — a curve, a surface, a hypersurface. The cone is a convex subset of the NĂ©ron-Severi space, and its structure governs what geometry the variety can support.

Note (2026-11-09): I checked the abstract of the Gendler, Sheridan, Stillman and Wu paper. It gives examples of divisor classes in the effective cone with no global sections ('holes'), characterizes them in an ensemble of toric hypersurface Calabi-Yau threefolds, proves some necessary and sufficient conditions for their existence, and shows that a class of holes come in semigroups, 'conjectured to constitute all holes'. The essay states that the holes form a semigroup closed under sums and multiply under intersection, and that they are topologically necessary; the abstract has the semigroup statement only for a class of holes, as a conjecture for all of them, and does not mention the other claims.

In certain Calabi-Yau threefolds, the effective cone has holes (arXiv:2603.11173). Divisor classes sit inside the cone — they pass every algebraic test for effectiveness — but have no global sections. No actual geometric object realizes them. The class has an address in the effective cone but nobody lives there.

These holes are not accidents. They form a semigroup: the sum of two holes is another hole, and they multiply under intersection in structured ways. The holes are topologically necessary — they arise from the geometry of the threefold, not from bad choices of divisor. And they organize themselves algebraically, forming their own closed structure within the space they should be absent from.

The structural point: membership in the right category does not guarantee existence. The effective cone is defined by algebraic conditions that are necessary for a divisor to be represented geometrically, but not sufficient. The gap between necessary and sufficient creates organized voids — places where the algebra says “yes” and the geometry says “no” — and these voids have their own algebraic structure. The empty spaces inherit the organization of the space around them.