An \(R = T\) theorem for imaginary quadratic fields (Q623328): Difference between revisions

The authors prove the modularity of certain reducible Galois representations of an imaginary quadratic number field \(F\), assuming the uniqueness of the residual representation, via an \(R=T\) theorem. As a by-product to their proof, it follows that the Eisenstein ideal is principal and the Hecke algebra \(T\) is a complete intersection. The proof of the \(R=T\) theorem is based on the following argument: Let \(\Sigma\) be a finite set of places of \(F\) and \(F_\Sigma\) denote the maximal extension of \(F\) unramified outside \(\Sigma\). Let \(\rho_0: \text{Gal}(F_\Sigma/F)\rightarrow \text{GL}_2(\mathbb{F})\) be a continuous representation, where \(\mathbb{F}\) is a finite field. Let \(I_{re}\) be the ideal of reducibility of the universal deformation ring \(R\) and let \(J\) be the Eisenstein ideal of the Hecke algebra \(T\). Then there is a surjection \(R/I_{re}\twoheadrightarrow T/J\). In [Compos. Math. 145, No. 3, 603--632 (2009; Zbl 1247.11081)], the first named author obtained a lower bound on the size of \(T/J\) (say by \(\mathfrak{B}\) for the sake of exposition) explicitly in terms of a special value of a relevant Hecke \(L\)-function. On the other hand, the authors prove that if \(\#R/I_{re}>\mathfrak{B}\), then the main conjecture of Iwasawa theory (as proved by Rubin) implies that there are no upper-triangular \(\Sigma\)-minimal deformations of \(\rho_0\) to \(R/I_{re}\), which, together with the non-existence of non-trivial upper-triangular deformations (proved by the authors in an earlier work [J. Inst. Math. Jussieu 8, No. 4, 669--692 (2009; Zbl 1248.11040)]) imply that \(\#R/I_{re}\leq\mathfrak{B}\). This in turn shows that \(R/I_{re}\) is isomorphic to \(T/J\). Then a commutative algebra lemma the authors prove allow them to deduce that \(R=T\). The paper is very well written. The introduction gives a very nice summary of the details to follow and also of the prior results related to the main theorem of the paper.