Interpretation functors which are full on pure-injective modules with applications to R R -torsion-free modules over R R -orders.
Lorna Gregory · 2024
Read Interpretation functors which are full on pure-injective modules with applications to R R -torsion-free modules over R R -orders. by Lorna Gregory (2024) 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-2412-13396.txt.
Introduction
Gregory proves a claim about maps between groups of modules in pure algebra in twenty twenty-four. Modules are broad forms of the number like systems used across algebra. The maps she studies carry one well behaved class of modules to another. She asks when a map that is full on a small class stays full on a pure class.
She shows it does, under a set of mild rules for the classes in play. She then uses the result on a map first built by two earlier scholars. That lets her chart part of a key diagram for a tame class of orders. She also coins a new kind of module to play a guide role for lattices.
Catalog record
| CREATOR | Lorna Gregory |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 73 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-13396
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Interpretation functors which are full on pure-injective modules with applications to R R -torsion-free modules over R R -orders. Thanks: The author was supported by PRIN 2017-Mathematical Logic: models, sets, computability. Lorna Gregory Abstract . Let R , S R,S be rings, 𝒳 ⊆ mod \mathcal{X}\subseteq\text{mod} - R R a covariantly finite subcategory, 𝒞 \mathcal{C} the smallest definable subcategory of Mod - R R containing 𝒳 \mathcal{X} and 𝒟 \mathcal{D} a definable subcategory of Mod - S S . We show that if I : 𝒞 → 𝒟 I:\mathcal{C}\rightarrow\mathcal{D} is an interpretation functor such that I 𝒳 ⊆ mod I\mathcal{X}\subseteq\text{mod} - S S and whose restriction to 𝒳 \mathcal{X} is full then I I is full on pure-injective modules. We apply this theorem to an extension of a functor introduced by Ringel and Roggenkamp which, in particular, allows us to describe the torsion-free part of the Ziegler spectra of tame Bäckström orders. We also introduce the notion of a pseudogeneric module over an order which is intended to play the same role for lattices over orders as generic modules do for finite-dimensional modules over finite-dimensional algebras. 2020 Mathematics Subject Classification 03C60, 16G30 (primary), 03C98, 16H20 1. Introduction Interpretation functors are a uniform additive version of the model theoretic notion of interpretation for (definable subcategories of) module categories. Algebraically, they are additive functors between definable subcategories of module categories which commute with direct limits and products. They are key tools for studying module categories from the perspective of model theory and for the algebraic study of purity. When an interpretation functor I:\mathcal{C}\rightarrow\mathcal{D} is full on pure-injective modules the connection between \mathcal{C} and \mathcal{D} is much closer than for other interpretation functors (more precisely, the connection between \mathcal{C} “modulo” \ker I and I\mathcal{C} ). Model theoretically being full on pure-injective modules corresponds, [ Pre97 , 3.17] , to the interpretation given by I “preserving all induced structure”. The main theorem of this paper (see 2.4 ) is the following. Here, \text{mod}\text{-}R denotes the category of finitely presented (right) R -modules. Theorem . Let R,S be rings, \mathcal{X}\subseteq\text{mod}\text{-}R be covariantly finite, \mathcal{C} the smallest definable subcategory containing \mathcal{X} and \mathcal{D} a definable subcategory of \textnormal{Mod}\text{-}S . If I:\mathcal{C}\rightarrow\mathcal{D} is an interpretation functor and I is full on \mathcal{X} and sends modules in \text{mod}\text{-}R to modules in \text{mod}\text{-}S then I is full on pure-injectives. In general, interpretation functors do not send finitely presented modules to finitely presented modules. However, there are naturally occurring situations when they do. For instance, this is the case for k -linear interpretation functors between categories of modules over finite-dimensional k -algebras (see 2.7 for a generalisation of this observation). Moreover, many functors used in the literature to understand one category of finitely presented modules in terms of another extend to interpretation functors between categories of non-finitely presented modules. Thus, in these situations, our theorem can be applied. One such example is Butler’s functor, [ But74 ] , between a large subcategory of lattices over the \widehat{\mathbb{Z}_{2}} -order \widehat{\mathbb{Z}_{2}}C_{2}\times C_{2} to the category of finite-dimensional \mathbb{Z}/2\mathbb{Z} -vector spaces with 4 distinguished subspaces fulfills the hypotheses of our theorem [1] [1] [1] One needs to check that the category of “ b -reduced” lattices is a covariantly finite subcategory of mod - \widehat{\mathbb{Z}_{2}}C_{2}\times C_{2} . Using clever ad hoc methods, in [ PT11 ] , Puninski and Toffalori showed that its extension to an interpretation functor is full on pure-injective \widehat{\mathbb{Z}_{2}} -torsion free \widehat{\mathbb{Z}_{2}}C_{2}\times C_{2} -modules. The second half of this article, section 3 , will give another such application using a functor defined by Ringel and Roggenkamp in [ RR79 ] . Let R be a complete discrete valuation domain with maximal ideal generated by \pi and field of fractions Q . Let \Lambda be an R -order with A:=Q\Lambda a separable Q -algebra. Let \Gamma be a hereditary R -order such that \Lambda\subseteq\Gamma\subseteq A and let I\subseteq\textnormal{rad}(\Lambda) be an ideal of \Gamma such that \pi^{n}\in I for some n\in\mathbb{N} . Let D:=\left(\begin{smallmatrix}\Lambda/I&\Gamma/I\\ 0&\Gamma/I\\ \end{smallmatrix}\right). Ringel and Roggenkamp defined a full (but not faithful) additive functor from the category of \Lambda -lattices to the category of finitely generated D -modules. This functor extends to an interpretation functor \mathbb{F}:\text{Tf}_{\Lambda}\rightarrow\textnormal{Mod}\text{-}D where \text{Tf}_{\Lambda} denotes the category of R -torsion-free \Lambda -modules. Since categories of \Lambda -lattices are covariantly finite and the smallest definable subcategory containing the \Lambda -lattices is \text{Tf}_{\Lambda} , Theorem 2.4 implies that \mathbb{F} is full on pure-injectives (see 3.4 ). The kernel of \mathbb{F} consists exactly of the R -divisible R -torsion-free \Lambda -module i.e. the restrictions of Q\Lambda -modules. A useful property of interpretation functors which are full on pure-injective modules is that they induce homeomorphisms between parts of Ziegler spectra. The Ziegler spectrum of a ring S , denoted \textnormal{Zg}_{S} , is the set of isomorphism types of indecomposable pure-injective S -modules with closed sets those of the form \mathcal{D}\cap\textnormal{Zg}_{S} where \mathcal{D} is a definable subcategory. For \mathcal{C} a definable subcategory of Mod - S , we write \textnormal{Zg}(\mathcal{C}) for the subspace \textnormal{Zg}_{S}\cap\mathcal{C} . Prest showed in [ Pre97 , 3.19] that if I:\mathcal{C}\rightarrow\mathcal{D} is an interpretation functor which is full on pure-injective modules then the map N\mapsto IN induces a homeomorphism from the open subset \textnormal{Zg}(\mathcal{C})\backslash\ker I and its image in \textnormal{Zg}(\mathcal{D}) , which is closed. The ( R -)torsion-free part \textnormal{Zg}_{\Lambda}^{tf}:=\textnormal{Zg}(\text{Tf}_{\Lambda}) of the Ziegler spectrum of \Lambda was studied in [ MPT02 ] , [ PT11 ] , [ GLT20 ] and [ Gre23 ] . Under our hypotheses on \Lambda , every R -torsion-free \Lambda -module decomposes as D\oplus N where D is R -divisible and N is R -reduced i.e. \cap_{i\in\mathbb{N}}N\pi^{i}=\{0\} . Thus, 3.6 , the functor of Ringel and Roggenkamp induces a homeomorphism from the open subset of R -reduced modules in \textnormal{Zg}_{\Lambda}^{tf} to a closed subset of \textnormal{Zg}_{D} . Note that since Q\Lambda is semi-simple there are only finitely many points in \textnormal{Zg}_{\Lambda}^{tf} which are not R -reduced. In subsection 3.2 we review the general features of \textnormal{Zg}_{\Lambda}^{tf} and explain how to use \mathbb{F} to completely understand the topology on \textnormal{Zg}_{\Lambda}^{tf} in terms of \textnormal{Zg}_{D} . Generic modules, i.e. indecomposable non-finitely presented modules which are finite length as modules over their endomorphism rings, play an important role in the representation theory of Artin algebras. However, if an R -torsion-free \Lambda -module is finite length over its endomorphism ring then it is R -divisible. So, although the indecomposable divisible modules are important they do not play the same role for lattices over orders as generic modules play for finitely presented modules over Artin algebras. We propose the notion of pseudogeneric R -torsion-free \Lambda -modules as a replacement for generic modules over Artin algebras. In subsection 3.3 , after proving some routine results about pseudogenerics in analogy with generic modules, we show that Ringel and Roggenkamp’s functor \mathbb{F} sends pseudogeneric modules to generic modules. Moreover, we show that, under a variety of conditions, \mathbb{F} gives a bijection between the generic modules in \mathcal{D} and the pseudogeneric \Lambda -modules. As a consequence, we show that tame Bäckström orders of infinite representation type have finitely many pseudogeneric R -torsion-free \Lambda -modules. When \Lambda is a tame Bäckström order of infinite lattice type, D is a