A semigroup-theoretic linkage theory for relative ideals: principal and canonical links

I. Ojeda · 2026

Read A semigroup-theoretic linkage theory for relative ideals: principal and canonical links by I. Ojeda (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-14478.txt.

Introduction

Ojeda builds a liaison analogue for relative ideals of a numerical semigroup, in semigroup terms. Two linkage notions run side by side: one from translates of the semigroup, one from translates of the canonical ideal. The paper sets out both notions for relative ideals. No application is claimed; the linkage theory itself is the point.

Catalog record

CREATORI. Ojeda
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE23461
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.14478

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A semigroup-theoretic linkage theory for relative ideals: principal and canonical links Thanks: The author is partially supported by project PID2022-138906NB-C21, funded by MCIN/AEI/10.13039/501100011033 and the European Union NextGenerationEU/PRTR, by grant GR24068, funded by the Junta de Extremadura and co-funded by the European Regional Development Fund (ERDF). I. Ojeda Abstract . We develop a semigroup-theoretic analogue of liaison for relative ideals of a numerical semigroup. Two parallel linkage notions are proposed: a theory based on translates of the semigroup and a theory based on translates of the canonical ideal. 2020 Mathematics Subject Classification 20M14, 13A30, 13H10 1. Introduction Classical liaison theory relates unmixed ideals in a polynomial ring, or more generally in a Cohen–Macaulay ambient ring, via residual constructions of the form I:J , typically with respect to complete intersection or Gorenstein linking ideals; see the expository notes of Migliore and Nagel [ 4 ] for background. In that setting, liaison classes often exhibit a rich and subtle structure, and even liaison plays a central role. In the context of numerical semigroups, however, it is natural to work intrinsically at the level of the semigroup of values, where the objects involved are combinatorial and monomial by nature. The purpose of this paper is to introduce a semigroup-theoretic framework inspired by liaison theory, formulated entirely within the poset of relative ideals of a numerical semigroup S , and to analyze the extent to which liaison-type phenomena persist in this much more rigid setting. Our approach is based on the semigroup-theoretic colon operation H-K\;=\;\{\,z\in\mathbb{Z}\mid z+K\subseteq H\,\}, which plays the role of an ideal quotient and is standard in the theory of numerical semigroups and commutative monoids [ 5 ] . Given a family \mathcal{L} of relative ideals of S , closed under translation, this operation allows us to define direct linkage and liaison classes for relative ideals of S . We focus on two natural choices of linking families. The first one consists of translates of the principal relative ideal S , leading to what we call principal linkage. The second one consists of translates of the canonical relative ideal K(S) , yielding canonical linkage. Both choices are motivated by classical constructions: principal linkage mirrors, at a formal level, linkage by complete intersections, while canonical linkage reflects the distinguished role of the canonical ideal in the theory of numerical semigroups, where it governs symmetry and almost symmetry phenomena [ 1 , 5 ] . From a broader perspective, the constructions considered here are naturally related to classical duality phenomena. The colon operation on relative ideals is formally analogous to the duality of fractional ideals, while the prominent role of the canonical ideal reflects familiar features of Gorenstein and almost Gorenstein numerical semigroups. More abstractly, the residuation H-K fits into the general framework of ordered and residuated commutative monoids. A key outcome of our analysis is that, once these two linkage notions are fixed, the semigroup-theoretic framework turns out to be highly rigid. Both principal and canonical linkage are governed by involutive duality operators, and as a consequence the corresponding liaison classes collapse in a systematic way: in each of the two settings they consist of at most two elements modulo translation and can be completely classified. In particular, even liaison is trivial in both the principal and the canonical theories. Therefore, much of semigroup liaison can be understood in terms of involutive duality operators. 2. Preliminaries Throughout, \mathbb{N}=\{0,1,2,\dots\} and \mathbb{Z} denote the nonnegative integers and the integers, respectively. A numerical semigroup is a submonoid S\subseteq(\mathbb{N},+) with finite complement in \mathbb{N} , and we write \operatorname{F}(S)=\max(\mathbb{N}\setminus S) for the Frobenius number of S . If S=\mathbb{N} , we adopt the convention \operatorname{F}(\mathbb{N})=-1 . Definition 1 . A relative ideal of S is a subset I\subseteq\mathbb{Z} such that (1) I+S\subseteq I , and (2) there exists a\in S such that a+I\subseteq S . If, in addition, I\subseteq S , then I is called an ideal of S . If I\subseteq S , then I is additively closed; moreover, I is a submonoid of (S,+) if and only if 0\in I , and in this case S is an overmonoid of I . Relative ideals are the natural semigroup-theoretic counterparts of fractional ideals in the semigroup-ring viewpoint; see, for instance, [ 5 , Chapter 1, Exercise 2.13] and [ 2 , Chapter 2, §§2.1–2.2] . For A,B\subseteq\mathbb{Z} we set A+B=\{a+b:\ a\in A,\ b\in B\} and, for t\in\mathbb{Z} , A+t=\{a+t:\ a\in A\} . Let H,K\subseteq\mathbb{Z} be relative ideals of S . We define the semigroup-theoretic colon (or difference) by H-K\;:=\;\{\,z\in\mathbb{Z}\mid z+K\subseteq H\,\} . This is the natural analogue of residuation (colon) in commutative monoids [ 2 , Chapter 2, Definition and Theorem 2.1.7] . The following elementary facts are standard; we record them (with brief proofs) for completeness and to keep the exposition self-contained. Lemma 2 . Let H,K,L be relative ideals of S . (1) H-K is a relative ideal of S . (2) If K\subseteq L , then H-L\subseteq H-K . (3) For every t\in\mathbb{Z} we have (H+t)-(K+t)=H-K . (4) K\subseteq H-(H-K) . Proof. (2) and (3) are immediate from the definition. For (4), if x\in K and z\in H-K , then z+K\subseteq H implies z+x\in H , hence x\in H-(H-K) . For (1), let z\in H-K and s\in S . Then (z+s)+K=s+(z+K)\subseteq s+H\subseteq H , so z+s\in H-K . Moreover, choose a\in S such that a+H\subseteq S . If z\in H-K , then z+K\subseteq H , and therefore (a+z)+K\subseteq a+H\subseteq S. Since K is a relative ideal, there exists b\in S with b+K\subseteq S . It follows that (a+z)+b\in S , proving that H-K is a relative ideal of S . ∎ Given a relative ideal L , we call L-I the residual of I with respect to L . Definition 3 . Let L and I be relative ideals of S . The residual of I with respect to L is \mathrm{Res}_{L}(I)\;:=\;L-I. Relative ideals are naturally considered up to translation. We say that two relative ideals I,J of S are isomorphic if there exists t\in\mathbb{Z} such that I=J+t . Definition 4 . Let I be a relative ideal of S . Since S\subseteq\mathbb{N} has finite complement and I is bounded below and stable under addition by S , the set I admits a minimum element. We define the normalization of I by I^{\mathrm{nor}}:=I-\min(I). We say that I is normalized if \min(I)=0 . Proposition 5 . Assume S=\mathbb{N} . Then: (1) Every relative ideal I of S is of the form I=m+\mathbb{N} for a unique m\in\mathbb{Z} . (2) If I=m+\mathbb{N} and J=n+\mathbb{N} , then I-J=(m-n)+\mathbb{N}. In particular, every normalization equals \mathbb{N} . Proof. (1) Let I be a relative ideal of \mathbb{N} . The condition I+\mathbb{N}\subseteq I implies that whenever x\in I then x+t\in I for all t\geq 0 . Since I has a minimum element m=\min(I) , it follows that I=m+\mathbb{N} . Uniqueness of m is clear. (2) Let I=m+\mathbb{N} and J=n+\mathbb{N} . By definition, z\in I-J if and only if z+J\subseteq I , i.e. z+n+\mathbb{N}\subseteq m+\mathbb{N} , which is equivalent to z+n\geq m . Hence I-J=(m-n)+\mathbb{N} . Finally, by Definition 4 , if I=m+\mathbb{N} then I^{\mathrm{nor}}=I-\min(I)=\mathbb{N} . ∎ Note that I^{\mathrm{nor}}\subseteq\mathbb{N} always holds, but I^{\mathrm{nor}} need not be contained in S in

Read the complete text