Strong linkage for function fields of surfaces (Q2139850): Difference between revisions

The article discusses a generalisation of the notion of a global field from class field theory, which they call a strongly linked field. Precisely, a field \(F\) is strongly linked in degree \(m\) if for any finite collection of central simple \(F\)-algebras, whose order in the Brauer group divides \(m\), there is a cyclic extension \(E\) of \(F\) of degree dividing \(m\) such that every algebra \(A\) in the collection is split by \(E\), i.e. \(A \otimes E\) is a matrix ring over \(E\). This property is well-known to be satisfied by global fields. The aim of this article is to divine a class of fields that can be proved to be strongly linked, focusing on examples arising from geometry and number theory. The examples they focus on are when \(F\) is the field of fractions of certain excellent Henselian local domains with algebraically closed residue fields, as well as the function field of an algebraic curve over such a domain, or over a pseudo-algebraically closed field, and the paper is dedicated to proving that these examples are all strongly linked. The approach the author takes is to use \(K\)-theory, adapting techniques from a previous work by Milnor which proved strong linkage for various examples of fields arising as second \(K\)-groups of global fields. The article recalls the basics of Milnor \(K\)-theory in section 2, and it goes on in section 3 to compare these \(K\)-groups with Brauer groups, and use this to outline the proof strategy the article adopts; to prove that the K-group \(K_2^{(m)}(F)\) is strongly linked, which using Corollary 3.5 is sufficient to deduce a property closely related to strong linkage for \(F\). In section 5, the authors go on to prove that if \(F\) is the field of functions of a curve over a pseudo-algebraically closed field, then the \(K\)-group will satisfy strong linkage. The key steps in the proof require further explorations into the theory of Brauer groups developed in section 4. However, to approach the other two examples of focus, a greater exploration of algebraic geometry is needed. The theory of ramification in regular local rings, and of curves and surfaces over Henselian rings is explored in sections 6 and 7, and this theory is applied in section 8 to show that when \(F\) is the field of functions on such a curve satisfying the appropriate conditions, then \(K_2^(m)(F)\) is strongly linked as required. The article then concludes in section 9 by speculating on the strong linkage of \(K_2^(m)(F)\) when \(F\) is a field of algebraic functions over a quasi-finite field. It concludes that this is true in certain limited cases, but unknown in general, suggesting a direction for future research. Overall, this article is very well-written and thorough, and the theory it is developing is very novel, and any academic interested in class field theory and algebraic geometry over rings of integral type should find the development of this theory very interesting, and this paper would certainly be worth reading for them.