by Matthew Keller
Some more recent habeas rulings from the Courts of Appeals. In this post, I focus on the Sixth Circuit, which over the past month has issued a series of generally helpful procedural rulings:
Gillis v. United States (6th Cir. Sept. 9, 2013)
Gillis offers good advice on one possible method for avoiding filing deadlines for 2255 motions and, more generally, is a good reminder of the value of reading rules of procedure. Often there are little nuggests in there that may get you out of an occasional fix.
The petitioner was convicted in 2007 of dealing crack near a school in violation of 21 U.S.C. 841 and 860(a). The court initially sentenced him to 262 months, but in 2009 it lowered the sentence to 191 months following a remand to reconsider application of the Guidelines' career-offender provisions. Gillis asked his attorney to appeal his new sentence, but counsel failed to file a timely notice. In May, 2011, Gillis filed a pro se § 2255 motion claiming ineffective assistance. The district court dismissed the motion as barred by the one-year deadline of § 2255(f). Gillis waited 201 days to appeal that denial, and the Government moved to dismiss the appeal as untimely.
The Sixth Circuit held that the appeal was timely. The district court never entered a separate judgment after denying the § 2255 motion, and therefore Gillis effectively had 210 days to appeal. The 210 days adds up as follows: Federal Rule of Civil Procedure 58(a) requires a separate judgment for all decisions or orders, except for five exceptions. A § 2255 motion is not one of those exceptions. Therefore, the district court’s judgment was not considered “entered” until 150 days after it denied the § 2255 motion. See Fed. R. App. P. 4(a)(7)(A)(ii) (defining “judgment” as entered 150 days after entry in the civil docket in the absence of a separate judgment required under Civil Rule 58(a)). Then, you add to that 150 the standard 60 days to appeal provided for in Federal Rule of Appellate Procedure 4(a)(1)(B) (and Rule 11(b) of the Federal Habeas Rules). Because Gillis filed his appeal within the 210-day window, his appeal was timely.
For those of us practicing primarily in the Second Circuit, there is a wrinkle here. The Sixth Circuit noted that the Second is the only Circuit that has held, in a 1993 case, that , Civil Rule 58(a) does not apply to § 2255 motions because of the “quasi-criminal” nature of such motions. See Williams v. United States, 984 F.2d 28, 29–31 (2d Cir. 1993). However, there is a good argument to be made that Williams was wrongly decided. Federal Habeas Rule 11(b) is clear that Fed. R. App. P. 4(a) governs appellate filing deadlines. That rule, in turn, defines the entry of a judgment with reference to Fed. R. Civ. P. 58(a). There is thus no basis in the governing Rules for the Second Circuit’s “quasi-criminal” gloss. The other circuits to consider the issue have recognize this: the Third, Fourth, Tenth, Eleventh, and D.C. Circuits all agree with the Sixth Circuit’s 210-day rule.
Unfortunately for Mr. Gillis, AEDPA’s filing deadlines are not as forgiving as the rules of Federal and Apppellate Procedure. Here the latter rules gave Gillis an extra six months to appeal the denial of his § 2255 motion. But the former, as the Court noted, “are not so lenient with regard to the timeliness of his § 2255 motion in the first place.” Gillis had 14 days from his resentencing in December 2009 to file a direct appeal. (Fed. R. App. P. 4(b)(1)). His lawyer did not do so, and his conviction became final (and the one-year AEDPA clock began to run) on December 26, 2009. That clock ran out on December 26, 2010, but Gillis did not file his pro se § 2255 motion until May 20, 2011.
Gillis tried to argue that the one-year AEDPA clock should not have begun until the Sixth Circuit dismissed his untimely direct appeal in January 2011. The Sixth Circuit (correctly) found that this was not the way things work: “[i]f the one-year AEDPA statute of limitations could be extended by filing a late notice of appeal and getting that late appeal dismissed,” the Court said, “there would not be much left to the statute of limitations.”
Ajan v. United States (6th Cir. Oct. 3, 2013):
In another ruling on 2255 procedure, the Sixth Circuit joined the Fourth, Eleventh and D.C. Circuits in holding that a COA is not required to appeal the relief granted after a successful § 2255 motion.
Ajan was originally sentenced to 646 months for various drug, kidnapping and firearm offenses. However, his § 2255 motion was granted, and his sentence vacated, after the parties agreed that one of the § 924(c) firearm convictions was not an offense under the charged statute. Without conducting new sentencing proceedings, the district court ordered entry of an amended judgment reflecting a new sentence of 346 months (the total term for the remaining offenses). Ajan appealed without obtaining a COA.
The Sixth Circuit held that a defendant need not obtain a COA to appeal the remedy following a successful § 2255 motion, because the granting of the motion results in a “new judgment” and therefore a new right of direct appeal. The court held that this follows from the Supreme Court’s 2010 decision in Magwood v. Patterson. Magwood held that an appeal from a successful § 2254 petition was not a second or successive petition because a “new judgment” is entered when the § 2254 is granted. The Circuit held that this rule applies with equal force to § 2255:
“Ajan’s successful § 2255 petition led to a new judgment — the Amended Judgment — which did not exist at the time his § 2255 petition was brought. It is the Amended Judgment that Ajan appeals. Because Ajan seeks to challenge the relief granted — i.e., whether the relief was ‘appropriate’ under § 2255, whether the new sentence was in conformity with the Constitution or Sentencing Guidelines, etc. — he is appealing a new criminal sentence and therefore need not obtain a COA”
Jefferson v. United States (6th Cir. Sept. 12, 2013)
Petitioners often seek equitable tolling of AEDPA’s statute of limitations. The Sixth Circuit also had some good news for petitioner's on that score as well, at least in the context of Brady claims.
Kenneth Jefferson was originally convicted in federal court in 1999 for his role in a gang selling crack out of Ypsilanti, Michigan throughout that drug’s boom years of the 1980s and 1990s. He was sentenced to twenty years. The Sixth Circuit affirmed his conviction on direct appeal, and the conviction became final ninety days later, on May 12, 2003. However, Jefferson did not file his § 2255 motion until August 2004, making it untimely.
Here’s where it gets tricky (apologies for the bumpy procedural history, but its necessary). Throughout the post-trial process, Jefferson and his codefendants asserted that the Government withheld the promises it had made to several cooperating witnesses regarding the benefits they would receive in exchange for their help in prosecuting Jefferson and others. They also asserted that these CWs lied at trial about the benefits they were promised, and that the Government knew about this perjury and sat by silently. These claims were given some support when, in March 2004, one of the CWs stated to the judge that he expected the AUSA to recommend a downward departure in his case. This set in motion an internal investigation by the local US Attorney’s office, which ultimately concluded that the AUSA had in fact met with some CWs without their counsel present, and made some, “at least tacit, promises of further sentencing reductions.” The investigation also concluded that the AUSA “had witnesses testify without revealing the additional understandings; moved orally at sentencing or in Rule 35 motions for downward departures; and had the sentencing records of these witnesses sealed.”
Jefferson then filed a series of additional motions (and supplements to pending motions), and continued to receive belated Giglio information from the Government. At some point he was given an attorney to help him with his claims. At first, the district court denied Jefferson’s § 2255 motion and the various supplements to it arguing for a new trial based on the Brady, Giglio and Napue violations highlighted by the US Attorney’s investigation. The Court rejected Jefferson’s attempts to toll the start date for his one-year AEDPA limitations period from May 2003 (when his conviction became final) to either September 2004 (when a decision was issued in an unrelated case detailing disclosure violations by the same AUSA), or to September 2005 (when the government revealed information from its internal investigation essentially confirming the Brady/Giglio violations in Jefferson’s case).*
* Jefferson argued pursuant to § 2255(f)(4) that his one-year clock ran from the date “on which the facts supporting the claim or claims presented could have been discovered through the exercise of due diligence.”
The court rejected the September 20004 date on the ground that “a decision in an unrelated terrorism case could not constitute the factual predicate for prosecutorial-misconduct claims in Jefferson’s case.” And it rejected the September 2005 date because Jefferson had filed his motions in March 2005, which severely undercut an argument that he was not on notice of his claim until five months after he filed it.
On its first trip up to the Circuit, the court reversed. It agreed that neither the 2004 nor the 2005 dates could serve as the factual predicate of his claims. But, the Circuit held, the district court “did not resolve the critical question of whether Jefferson’s additional claims were brought within one year of ‘the date on which the facts supporting the claim or claims presented could have been discovered through the exercise of due diligence.’” The Court therefore remanded for consideration of “the statute of limitations and the possible application of equitable tolling.”
The district court complied, gave Jefferson’s motion additional consideration, and again denied it. This time, the court essentially deployed the “kitchen sink” reasoning the habeas practitioners are so used to: the claims are untimely, they do not warrant equitable tolling and, in any event, they are meritless. Case closed.
On appeal last month, the Sixth Circuit reversed the district court’s procedural rulings, but ultimately agreed that the claims failed on the merits. It found that, given the “realities of the prison system,” Jefferson could not have discovered the basis for his Brady/Giglio claims prior to 2005. Importantly, the Court rested this holding on the nature of the Government’s disclosure obligations under Brady:
“We hold that § 2255(f)(4)’s due-diligence standard did not require Jefferson continuously to seek out evidence that the government had a constitutional duty to disclose (evidence that, despite specific requests by the defense for the information, the government represented did not exist). We reject the district court’s determination that because Jefferson suspected undisclosed promises as early as 2000, his failure to seek information from “the cooperating witnesses themselves or their acquaintances” and his failure to seek sentencing records—which the prosecutor had sealed—rendered his diligence insufficient. We do not fault Jefferson for failing to scavenge for evidence of undisclosed promises when he already repeatedly asked for disclosure and the evidence was unconstitutionally withheld by the government.”
On the merits, the Circuit held that Jefferson’s Brady claims failed on the materiality requirement, the bane of so many Brady/Giglio claims. Where the Government withholds impeachment material for cooperating witnesses, such claims almost always fail for the same reason, which goes something like this: “these witnesses were bad guys ratting out their friends to save their own asses. Both sides knew it, the Court knew it, and the jury sure as hell knew it. If the jury chose to believe them anyway, it probably would have still believed them if it knew they were promised $10 (or ten years off their sentence) instead of $5 (or 5 years off).”
As Ive said on this blog before, I disagree with the wide application of the assumption that juries would do the same thing if they had more or different information to all manner of different circumstances. Our system operates largely on the theory that jury deliberations are (and should be) a black box. The frequency with which judges retroactively predict what a jury would have done under different circumstances (and the confidence with which they do so) violates that precept. To some extent this is necessary to conserve resources – the courts would be swamped if we had to retry every case where the jury didn’t hear evidence it should have (or where, in the more frequent case, the jury hears something it shouldn’t have). But make no mistake: the Brady materiality standard operates primarily as a practical resource-management tool. It is not a well-considered reflection on the ability of judges to accurately second-guess juries.
Henderson v. Palmer (6th Cir. Sept. 12, 2013)
In Henderson, the Sixth Circuit applied the prison mailbox rule to reverse the district court's finding that petitioner had procedurally defaulted on claims to the Michigan state courts. Henderson was convicted of armed robbery and carjacking based in part on a lineup identification. Law enforcement officers claimed that Henderson's attorney was present at the lineup, but the attorney denied that he was there. Henderson claimed that his trial attorney was ineffective for failing to raise the issue at trial. However, state courts rejected his claims as untimely because the filing arrived one day late due to failings in the prison mail system.
The Sixth Circuit reversed the district court's rejection of the 2254 petition because the lower court failed to apply Circuit precedent holding that “[w]here a pro se prisoner attempts to deliver his petition for mailing in sufficient time for it to arrive timely in the normal course of events,” that circumstance “is sufficient to excuse a procedural default based upon a late filing.”
Hooper v. Ryan (7th Cir. Sept. 9, 2013)
For the last case, we move to the Seventh Circuit. In Hooper, the Seventh Circuit granted a §2254 petition, reversing the district court and effectively reversing a 32-year-old murder conviction on Batson grounds.
At trial, the court drew a venire of 63 members, seven of whom were black. Two of the seven were removed for cause, and the prosecutor exercised peremptory challenges against the remaining five. The Seventh Circuit held that the Illinois Supreme Court applied Batson unreasonably in four respects. It’s worth going through all four grounds, both because a state Supreme Court making four mistakes on any one issue is notable, and because some of the mistakes would be funny if they were not so troubling.
First, the Illinois Supreme Court unreasonably held that, under Batson, a trial court cannot “infer a prima facie case of discrimination from a racially disproportionate use of challenges.” The Seventh Circuit noted that Batson and subsequent rulings clearly allowed the use of peremptory strike statistics to support a prima facie case. Second, the Illinois court held that the five strikes against black venire members could not be discriminatory because the prosecutor also struck six white or Asian members. The Seventh Circuit held that, in addition to being legal error, the comparison of absolute strikes as opposed to their proportion relative to the racial makeup of the venire reflected “confusion about how to use numbers.”
We all know that lawyers are bad with numbers - otherwise we’d all be doctors, right? So perhaps that second error is somewhat understandable. It’s a bit harder to excuse the third error. The Circuit held that it was error for the Illinois Supreme Court to reject the Batson challenge on the ground that the defendant, all three victims, and all principal witnesses were black. Not only was the Illinois Supreme Court plain wrong on this (three principal witnesses were white), the Circuit also pointed out that Batson and its progeny protect the rights of potential jurors (and the public at large) as well as the rights of defendants. The Circuit called the the Supreme Court of Illinois’ idea that “using race in jury selection is tolerable as long as the defendant, victims, and witnesses all are of the same race” a “serious legal blunder.”
Finally, the Circuit held that the state court erred unreasonably in excusing the trial judge’s incorrect application of Batson’s three-step sequence of proof.
The Seventh Circuit’s decision may force Illinois to free Hooper, who has served over 32 years for the murders and is now 67 years old. The Court acknowledged that it may be impossible for the state to conduct a “fruitful” Batson hearing more than three decades after the trial (not to mention a retrial). But I think the Court correctly reasoned that it could not let such an unacceptable decision stand. It may have helped that the Illinois Supreme Court decision under review was 24 years old – the Circuit was not throwing stones at the current court. It may have also felt that the trial was also tainted by its judge, Thomas Maloney, who was later convicted of bribery and spent the last 15 years of his life in prison. Maloney was not your ordinary crooked judge: a jury found that he fixed three gang murder trials. It’s not legally relevant to the Batson claim, but Maloney’s presence in the case was just one more taint on the appearance of justice.