From Multirelations to Meet-Relations: A Relational Duality for Semilattices with Adjunctions
B. Gimenez, W. Zuluaga · 2026
Read From Multirelations to Meet-Relations: A Relational Duality for Semilattices with Adjunctions by B. Gimenez, W. Zuluaga (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-2605-21667.txt.
Introduction
Zuluaga and Gimenez build a duality for semilattices with adjunctions, systems that pair order with two linked maps. They introduce MoS-spaces, a category dual to modal semilattices. Then they define meet-relations and a category called RelSP, dual to the semilattices themselves. Under a normality rule, the older multirelational picture can be rebuilt from a meet-relation, and back again. The two relational categories turn out to be isomorphic.
Catalog record
| CREATOR | B. Gimenez, W. Zuluaga |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 87138 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.21667
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
From Multirelations to Meet-Relations: A Relational Duality for Semilattices with Adjunctions B. Gimenez, W. Zuluaga Abstract We develop a relational duality for semilattices with adjunctions (SLatas) based on binary meet-relations. First, we introduce the category of MoS-spaces and establish a dual equivalence with modal semilattices. Then, by means of A-relations, we define the category 𝖱𝖾𝗅𝖲𝖯 \mathsf{RelSP} and prove a dual equivalence between 𝖲𝖫𝖺𝗍𝖺 \mathsf{SLata} and 𝖱𝖾𝗅𝖲𝖯 \mathsf{RelSP} . To compare this framework with the multirelational semantics previously developed for SLatas, we introduce the notion of normal mS-space and show that, under this condition, the multirelational structure can be canonically recovered from a meet-relation, and conversely. As a consequence, we prove that the categories 𝖱𝖾𝗅𝖲𝖯 \mathsf{RelSP} and 𝖲𝖫𝖺𝗍𝖺𝖲𝗉 \mathsf{SLataSp} are isomorphic. 1 Introduction Duality theory provides a fundamental bridge between algebraic structures and their associated semantic models. By establishing contravariant correspondences between algebraic categories and categories of structured spaces, dualities allow algebraic phenomena to be studied through topological, order-theoretic, or relational methods. In the specific setting of semilattices, this perspective has proven particularly fruitful: the duality developed in [ 4 ] between semilattices and S-spaces provides a flexible framework for representing algebraic expansions, while subsequent work extended this approach to monotone semilattices and semilattices with adjunctions through multirelational structures [ 1 , 3 ] . Despite these developments, the use of multirelations—that is, subsets of X\times\mathcal{P}(X) instead of ordinary binary relations X\times X —comes at a conceptual and technical cost. While multirelations faithfully encode monotone operators and adjoint pairs at the dual level, they also move the semantics to a genuinely higher-order setting, obscuring the first-order character that is typical of classical relational semantics and making the associated dual structures more difficult to manipulate. This naturally leads to the following question: to what extent can the multirelational component in dualities for expanded semilattices be replaced by purely binary relational data without losing the underlying categorical information? The present paper answers this question affirmatively for the case of semilattices with adjunctions (SLatas), namely algebras \langle A,i,d\rangle where the unary operations i and d form an adjoint pair i\dashv d . Our main contribution is to show that, under a suitable normality condition on the associated dual spaces, the multirelational semantics introduced in [ 3 ] can be canonically reconstructed from binary meet-relations. More precisely, we define a category \mathsf{RelSP} of relational S-spaces and establish a dual equivalence between \mathsf{SLata} and \mathsf{RelSP} . Furthermore, we prove that \mathsf{RelSP} is not merely equivalent to the multirelational category \mathsf{SLataSp} introduced in [ 3 ] , but actually isomorphic to it. The categorical relationships developed throughout the paper are summarized in the following diagram: \textstyle{\mathsf{SLataSp}\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces} \scriptstyle{\mathbf{Q}} \textstyle{\mathsf{RelSP}\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces} \scriptstyle{\mathbf{P}} \textstyle{\mathsf{SLata}^{op}\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces} \scriptstyle{\mathbf{M}} The lower part of the diagram displays two dual representations of semilattices with adjunctions: on the left, the multirelational semantics developed in [ 3 ] , and on the right, the relational semantics based on meet-relations introduced in the present paper. The upper part shows that the corresponding categories of spaces are in fact isomorphic. Our strategy proceeds in two stages. First, we develop a relational semantics for modal semilattices, namely semilattices equipped with a meet-preserving operator. To this end, we introduce the category \mathsf{MoSsp} of MoS-spaces and establish its dual equivalence with the algebraic category \mathsf{MoSL} . This provides the fundamental relational machinery—meet-relations, compatible morphisms, and the associated functorial constructions—that will later be extended to the adjoint setting. In the second stage, we enrich this framework in order to represent adjoint pairs on semilattices. For this purpose, we introduce A-relations (adjoint-relations) and adjoint-preserving morphisms, obtaining the category \mathsf{RelSP} and the corresponding duality for SLatas. We then introduce the notion of normal mS -space and prove that every relevant multirelational structure can be canonically recovered from an associated meet-relation. This ultimately yields the main result of the paper: the categories \mathsf{SLataSp} and \mathsf{RelSP} are isomorphic. The paper is organized as follows. Section 2 reviews the necessary background concerning semilattices, S-spaces, monotone semilattices, and SLatas. In particular, we recall the topological duality for semilattices from [ 4 ] together with its multirelational extensions developed in [ 1 , 3 ] . Section 3 develops the relational duality for modal semilattices and extends it to semilattices with adjunctions via A-relations, culminating in the dual equivalence between \mathsf{SLata} and \mathsf{RelSP} . Section 4 contains the main technical contribution of the paper. There we introduce normal mS -spaces and prove the isomorphism between \mathsf{SLataSp} and \mathsf{RelSP} . Finally, we conclude with a discussion on the algebraic and semantic meaning of the constructions introduced throughout the paper. 2 Preliminaries The aim of this section is twofold. On the one hand, we fix the notation and terminology used throughout the paper. On the other hand, we review the dual frameworks underlying our approach, namely the topological duality for semilattices and its multirelational extensions to monotone semilattices and SLatas. We begin with some basic notation. Let f:A\to B be a function, and let U\subseteq A and V\subseteq B . We denote by f[U] and f^{-1}[V] the direct and inverse images of U and V under f , respectively. Let \langle X,\leq\rangle be a partially ordered set. For every x\in X , we write \uparrow x for the set \{y\in X:x\leq y\} . A subset U\subseteq X is called an upset if, whenever x\leq y and x\in U , then y\in U . The set of upsets of X is denoted by \mathrm{Up}(X) . The notion of downset is defined dually. A non-empty subset Z\subseteq X is called dually directed if for all x,y\in Z there exists z\in Z such that z\leq x and z\leq y . A meet-semilattice with greatest element , or simply a semilattice , is an algebra \mathbf{A}=\langle A,\wedge,1\rangle such that \wedge is idempotent, commutative, and associative, and a\wedge 1=a for all a\in A . The associated order is given by a\leq b if and only if a=a\wedge b . It is well known that \langle\mathrm{Up}(X),\cap,X\rangle is a semilattice. 2.1 Topological duality for semilattices Let \mathbf{A} be a semilattice. A subset F\subseteq A is a filter if it is an upset, 1\in F , and a,b\in F implies a\wedge b\in F . We denote the set of filters of \mathbf{A} by \mathrm{Fi}(\mathbf{A}) . A proper filter F is irreducible if, whenever F=F_{1}\cap F_{2} for filters F_{1},F_{2} , then F=F_{1} or F=F_{2} . We write \mathcal{X}(\mathbf{A}) for the set of irreducible filters of \mathbf{A} . A non-empty subset I\subseteq A is called an order-ideal of \mathbf{A} if it is a directed downset. That is, I is a downset such that every pair of elements a,b\in I has an upper bound in I . We denote by \mathrm{Id}(\mathbf{A}) the set of all order-ideals of \mathbf{A} . Consider the poset (\mathcal{X}(\mathbf{A}),\subseteq) and the map \beta_{\mathbf{A}}\colon A\to\mathrm{Up}(\mathcal{X}(\mathbf{A})) given by \beta_{\mathbf{A}}(a)=\{P\in\mathcal{X}(\mathbf{A}):a\in P\}. It was shown in [ 4 ] that \mathbf{A} is isomorphic to the subalgebra of \langle\mathrm{Up}(\mathcal{X}(\mathbf{A})),\cap,\mathcal{X}(\mathbf{A})\rangle with universe \beta_{\mathbf{A}}[A] . Throughout the paper we omit the subscript when no confusion arises. By a topological space we mean a pair \langle X,\mathcal{K}\rangle , where \mathcal{K}\subseteq\mathcal{P}(X) is a subbase such that X=\bigcup\mathcal{K} . We set S(X)=\{U^{c}:U\in\mathcal{K}\}. We denote by \mathcal{C}_{\mathcal{K}}(X) the closure system generated by S(X) , that is, \mathcal{C}_{\mathcal{K}}(X)=\left\{\bigcap\mathcal{A}:\mathcal{A}\subseteq S(X)\right\}. Given Y\subseteq X ,