The ever mercurial Supreme Court has issued yet another opinion on the application of the modified categorical approach to the ACCA, this time tightening the reigns on district courts in a defense-friendly way.
Descamps asserts itself as the new go-to opinion for those wishing to understand the Shepard/Taylor approach, but also for those who wish to challenge its application to a defendant's criminal history. In a nutshell, the opinion limits the so-called "modified" categorical approach to those cases where a statute is "divisible" -- where, for example, the statute has an "(a)" section that meets the generic definition of, say, a burglary, but has a "(b)" section that does not. In those cases, a court may look to Shepard documents to determine what part of the statute the defendant was originally convicted of. Where, as here, the statute is simply broader in its definition of a crime than the "generic" definition but is not "divisible," the district court must simply apply the categorical approach and not look to outside Shepard documents.
Justice Kagan's opinion shows an impressive understanding of the issues behind this question, especially with respect to their oft-forgotten Sixth Amendment origins. She also evidently understands what goes on in courtrooms, noting that "during plea hearings, the defendant may not wish to irk the prosecutor or court by squabbling about superfluous factual allegations."
Justice Thomas jumps in with a concurrence suggesting once again that the real problem with all of this is not the ACCA, but is instead Almendarez-Torres, which he seems eager to reconsider.
All in all, an excellent defense win, and required reading for those working in the federal system.