From the Eastern District of Michigan comes a decision by Judge Denise Hood finding that Padilla cannot be applied retroactively. The case is United States v. Shafeek, Criminal Case No. 05-81129, Civil Case No. No. 10-12670 (E.D. Mich. Sept. 22, 2010). Opinion courtesy of Google Scholar here.
The wrinkle, however, is that she reaches this conclusion despite having concluded in the same decision that the Supreme Court’s decision in Padilla did NOT announce a “new rule” for retroactivity purposes. From the Shafeek decision:
Given the Supreme Court’s opinion in Padilla, it appears that the rule announced is not a “new rule” regarding a defense counsel’s duty to, at the minium, advise a client of a risk of adverse immigration consequences. The Padilla decision turned on the fact that the defense counsel could have easily determined from reading the removal statute that defendant’s deportation “was presumptively mandatory” and that his counsel’s advice to the contrary was incorrect. Id. Because the Padilla opinion may not be considered a “new rule,” Shafeek cannot show that the Padilla opinion should be applied retroactively.
Huh? It has been my understanding that under the Supreme Court’s seminal decision in Teague v. Lane on retroactivity in cases pending on collateral review, only “new rules” are barred from retroactive application unless they fall within one of two exceptions. The upshot, of course, is that rules that are not considered “new” under the Teague analysis should be applied retroactively. Such was the conclusion by the court in Chaidez which concluded, correctly in my opinion, that because Padilla was not a “new rule” it could be applied retroactively. See United States v. Chaidez (“Accordingly, the court holds that Padilla did not announce a new rule for Teague purposes and affirms its earlier opinion that no retroactivity problem is raised by petitioner’s claim”).
Am I missing something here?
Invisible Man 
Pingback: More Padilla Retroactivity | Invisible Man
What you’re missing is that Shafeek sought to file an out-of-time 2255, which he could have done if Padilla created a new right made retroactively applicable.
Mark,
Thank you for the clarification. We have a similar exception in Georgia which, if established, would allow for the filing of out-of-time habeas petitions. OCGA 9-14-42(c)(3). The difficulty I have in all this is that if Padilla qualifies for “old rule” retroactive application, which appears increasingly likely, then it necessarily follows that Padilla would not qualify for “new rule” retroactive application as was the case in Shafeek. Yet, it seems to me that both tests stand on the same footing — that is, the determination of retroactive applicability. And forcing litigants to establish “new rule” retroactivity is a near impossible task thanks to Teague v. Lane. This all seems fundamentally unfair and is akin to providing a right without a remedy.
The problem is that there really are two types of retroactivity, and two tests. So cases in which the first type of retroactivity is in question (“is Padilla an old rule, or a watershed rule, that applies to cases already final?”) and cases in which the second type of retroactivity is in question (“is Padilla a new rule made retroactive by the Supreme Court, such that it can be the basis of a successive writ?”) get mixed up together, to confusing effect.
For my money, Padilla does not create a new rule; as a consequence, anyone could argue its rule in his first writ. Mr. Shafeek loses now because he could have done so.
By the way, read section IV of Teague: the Supreme Court isn’t going to apply a new rule retroactively only to the person whose case is before it now.