The core of the majority’s decision in Chaidez rests on the notion that before Padilla no court would have granted postconviction relief to a foreign national defendant under Strickland based on an attorney’s failure to give deportation advice because deportation was considered a collateral, not a direct, consequence of a conviction. It is this dispositive aspect of the collateral v. direct distinction and, to the majority, Padilla’s “rejection” of it, that makes Padilla a particularly novel decision and one ill-suited for retroactive application. The problem, it seems to me, with the majority’s analysis is that it overstates the importance or effect of the collateral v. direct divide, and it is also a demonstration that judges who have little or no on-the-ground experience, as is the case with Kagan, make bad law. (There is a reason Kagan was assigned to write the opinion which I will get to a bit later.) To take just one example: if what the majority stated was true, and courts really made mince meat out of Padilla-like ineffective assistance claims pre-Padilla based on the collateral-direct divide, then few if any of the cases which presented these claims before Padilla should have made it past the pleading stage, let alone being decided on their merits in published decision after published decision. I do not think that is how courts treated Padilla-like claims in the pre-Padilla era, however. Instead, courts still decided Strickland claims pertaining to deportation advice on their merits even if they ended up denying them based on the collateral v. direct distinction. In other words, there was no question that Strickland defined the standard of competent representation received by foreign nations in criminal cases pre-Padilla, the debate rather was over how that standard should be defined in such cases. To take yet another example: let’s say an individual is irked by his attorney’s performance in an immigration matter which ended up in his removal and in a misguided effort to prevent his removal brings a claim in federal district court alleging ineffective assistance under Strickland. There would be no question as to the applicability of Strickland or the fate of his claim; it would fail and fail big because Strickland applies only to criminal, not civil, cases. But Strickland squarely governs in cases like Padilla and Chaidez because they are, at their collective core, criminal matters. I think this is, in part, what led the Padilla court to describe as “ill-suited” to the Strickland analysis the dichotomy between collateral v. direct consequences of a conviction — a point that Sotomayor seized on in her dissent where she took the majority to task for its over-reliance on and over-emphasis of this distinction.
As to my hypothesis as to why Kagan ended up writing the Chaidez opinion, I think that the debate between say, Roberts and Alito, on the one hand, and Breyer and Kennedy, on the other, centered not on whether Padilla could be applied retroactively — the majority’s opinion makes clear that there was never much doubt there — but on how Padilla and now Chaidez might be used to expand the scope of Strickland to encompass advice on other so-called collateral matters, an outcome which presumably Roberts and Alito, to say nothing of Scalia and Thomas, would have disfavored. At the same time, the more liberal members of the bench, did not want Chaidez written in a way that would have narrowed or weakened the mandate in Padilla (I wonder if Stevens’ presence at the Court — he apparently still maintains an office there — and generally as an observer of the Court, might have had some influence as well). So the task of authorship was given to Kagan who was willing to say no to retroacivity but do so in a way that was respectful of the Padilla decision.