Remarks on elementary duality (Q688808): Difference between revisions

Central to many areas of work in non-commutative algebra is the problem of connecting the `left' structure (ideal, module,\dots) of a ring to its `right' structure. Elementary duality, introduced by the author [in J. Lond. Math. Soc., II. Ser. 38, No. 3, 403-409 (1988; Zbl 0647.16019)] is one way of dealing with this problem, and recent developments of the idea by I. Herzog (forthcoming papers are cited in the article) have hinted at a potentially great power and scope for the method. The motivation for introducing elementary duality was a question from representation theory: ``Is every right pure-semi-simple ring of finite representation type?'', and the author's original paper made some contributions to work on this problem. But the paper under consideration, a clear and understandable survey of the topic, leaves one with the distinct impression (or hope!) that the potential applications will be much more wide-ranging. The concept of elementary duality comes from mathematical logic (model theory); and some of the basic ideas of the model theory of modules are reviewed. But elementary duality can also be discussed in terms of functor categories, and in terms of natural topologies on the space of pure-injective indecomposables; and the author carefully explores all of these connections in a way that makes the ideas easily accessible to the reader, whether algebraist or logician. Not many proofs are given, but much is explained, with a wide-ranging list of references. Much thought is given to future directions for research, and many will find inspiration in this carefully considered essay.