Split Widened: Can a Court Bypass Rooker-Feldman to Dismiss on Merits?

Per Cawley v. Celeste (8th Cir. May 9, 2013)

Rooker-Feldman is one of the more esoteric doctrines of federal jurisdiction.  In brief, it prevents state court losers from using a federal case to appeal their defeat.  Importantly, in 2005's SABIC decision, the Supreme Court clarified that the doctrine is statutory, based on 28 U.S.C. 1257, not constitutional.

Nearly every time that Rooker-Feldman could apply, the federal defendant will also have a strong preclusion defense.  After all, for Rooker-Feldman to apply, there must already be a state court decision on the issue.

The circuits have split on whether district courts must adjudicate the Rooker-Feldman jurisdictional issue first, or may reach the preclusion issue on the merits without determining jurisdiction.  The CAs 3,6,and 7 all have published decisions saying that Rooker-Feldman must go first, and the CAs 9, 10, and 11 agree in unpublished decisions.

On the other hand, this CA8 opinion joins  published opinions from the CAs 1 and 7 (yes, the CA7 has published on both sides), and unpublished authority from the CAs 2, 3, and 10 (the latter two demonstrating another intra-circuit split) to hold that the court may reach the merits.

The CA8's reasoning is so clearly correct that I cannot improve upon it:

Steel Co. acknowledged that a federal court may reach a merits question before deciding a statutory standing question because the merits inquiry and the statutory standing inquiry often overlap, and it would be artificial to draw a distinction between the two.  That rationale may not support bypassing all questions of statutory jurisdiction.  But we think it does allow a federal court to decide a question of preclusion without first resolving a murky problem under Rooker-Feldman, because our inquiries under preclusion law and the Rooker-Feldman doctrine would similarly overlap.

(p. 6 (citations omitted).)

Indeed, the interplay between Rooker-Feldman and preclusion seems to fall perfectly within footnote 2 of Steel Co.  A court should, in the interests of judicial economy, be permitted to bypass a difficult Rooker-Feldman question if the preclusion result is much simpler.

On an aside--and part of my reason for selecting this case for comment--the topic of Rooker-Feldman allows me to pay tribute to my favorite legal publication, The Green Bag.  Right after SABIC and another Supreme Court case, the journal published an amusing obituary for the entire Rooker-Feldman doctrine.

Split Created: Does 2255(e)'s Savings Clause Apply To Past Misapplications of the Guidelines?

Per Brown v. Caraway (7th Cir. May 10, 2013)

Royce Brown always maintained his arson conviction was not a crime of violence.  Turns out, under Begay, he was right.  But Begay came much too late.  Brown was sentenced in 1996 as a career offender (adding several years to his sentence) based in part on his arson conviction.

Ordinarily, after Begay, Brown could have filed a 2255 motion to challenge his sentence.  But Brown had already filed--and lost--a 2255 motion in 2000.  And 2255(h) bars second or successive motions.

Brown nonetheless sought to profit from Begay and filed a 2241 petition for habeas corpus.  But 2255(e) bars consideration of habeas petitions, unless a 2255 motion  "is inadequate or ineffective to test the legality of his detention."  (This quoted text is often referred to as the "savings clause").

In this case, the CA7 creates a circuit split by holding that the savings clause applies, and permits a habeas petition.  The CA 11 (en banc) and CA5 had previously concluded--also in the context of 2241 petitions based on Begay--that the savings clause did not apply.

I am of two minds.  On one hand, equity favors allowing some relief where a sentence is manifestly in error.  On the other, considerations of finality must trump at some point.  I lean slightly to the CA11 and CA5's view that the  savings clause does not apply because 2255 procedures are in fact effective and adequate, but unavailable only because of a previous collateral attack.

I lean that way in part because--unlike the CA7--I do not think this conclusion leaves prisoners without a remedy.  Even if both a 2255 motion and a 2241 habeas petition are unavailable, a prisoner could file a petition for a writ of coram vobis (not nobis, as discussed below).

As this blog has discussed earlier, coram vobis requires:

1. a fundamental error in the prior proceedings
2. reasonableness in not having acted earlier (i.e., though habeas)
3. collateral consequences from the prior proceedings (standing)
4. interests of justice require granting the writ (no alternative remedy)

Here, application of the career enhancement is a fundamental error increasing the sentence.  Brown both preserved the issue by objecting in the initial sentencing and acted reasonably in raising it again soon after Begay was decided.  Brown's lengthened sentence provides standing.  And--if both 2241 and 2255 are not available--there is no alternative remedy.

Thus, I ultimately come out at the same point as the CA7, but would use a different procedural tool to get there.  I feel that this procedural tool does less damage to the "second or successive" bar.  But I am not a criminal lawyer, and there may be something I am missing.

On a total aside, I am surprised that the vast majority of U.S. courts call the writ "coram nobis" (before us) rather than "coram vobis" (before you).  The names do not imply a difference in procedure or remedies.  Rather, the distinction arose in English common law depending on whether a writ petition was filed before the King's Bench--where the King was supposed to preside, and so using the royal "we/us"--or Common Pleas, where the King did not preside, so the writ was only before "you" judges.  As America has no king, it would appear that "coram vobis" is more appropriate.

Split Noted: Effect of Failing to Object to a Magistrate's Report

Per Dupree v. Warden (11th Cir. May 7, 2013)

The CA11 requires a district judge to rule on all grounds raised in a habeas petition.  Here, the magistrate overlooked one claim.  The petitioner did not object, and the district court accepted the magistrate's report.

This CA11 panel does the right thing, and abides by panel precedent requiring a finding of reversible error.  (p. 8-9.)  In so doing, however, the panel notes that the CA11 is an outlier in a three-way circuit split regarding the effect of failing to object to a magistrate's report, and recommends an en banc to shift position.  (p. 14-19.)

Most Lenient:  The CA11 rule, which is the same as the CA8 and CA9, holds that failure to object has no effect on the de novo review of legal issues.  For factual issues, the CA11 and CA8 employ plain error review (see middle paragraph below), while the CA9 deems such objections waived (see strictest paragraph).

Middle:  The CA3 and CA5 always employ plain error review for both factual and legal issues.  They permit reversal if the appellant can show (1) plain (2) error that both (3) affects substantial rights and (4) undermines the fairness or integrity of judicial proceedings.

Strictest:  The CAs 2, 4, 6, 7, and 10 apply a firm waiver rule.  Failure to object waives appellate review of both factual and legal issues.  Such waiver may be waived by the court in the interests of justice, and does not apply if the magistrate does not warn of the effects of failing to object.

Here the panel recommends adopting the strictest approach.  I agree.  A number of considerations support this approach.

First, judicial economy, as the Supreme Court stated in Thomas v. Arn, 474 U.S. 140 (1985)--a decision affirming, but not requiring, the CA6's strict approach.

Second, if--as required--a magistrate warns of the effect of failing to object, then the failure to object is indeed the intentional relinquishment of a known right.  United States v. Olano, 507 U.S. 725 (1993).  And even if only done by counsel, counsel is an agent for the party, and the party is responsible for the failings of his counsel.  Link v. Wabash R.R. Co., 370 U.S. 626 (1962).  Such waiver argues against the middle-ground approach of the CA3 and CA5.

Third, imposing a waiver rule is not jurisdictional, so appellate courts still retain authority to correct manifest injustices.  Hormel v. Helvering, 312 U.S. 522 (1941).

Fourth, adopting a waiver rule would bring Civil Rule 72 in line with Criminal Rule 59, adopted in 2005.

In sum, all indications point to adopting the strictest rule.  And while the CA11 may go en banc to adopt that approach, the rulemaking committee should make the waiver rule explicit in civil rule 72, as it did in criminal rule 59.  That would spare the resources necessary for the CA8 and CA9 to also go en banc to adopt this appropriate rule.

Split Widened: Can The Venue of Post-Judgment Garnishment Proceedings Violate The Fair Debt Collection Practices Act?

Per Smith v. Solomon & Solomon, P.C. (1st Cir. Apr. 24, 2012)

This unanimous opinion (including Retired Justice Souter) widens a split concerning the venue provisions of the Fair Debt Collection Practices Act (FDCPA).  In brief, the FDCPA requires debt collectors to file any suit "against a[] consumer" where the consumer either (a) "resides" or (b) signed the document giving rise to the debt.  15 U.S.C. 1692i.  Failure to do so can result in civil liability.

The split at issue does not concern initial suits against the consumer to reduce the debt to judgment.  Instead, it concerns the venue for post-judgment enforcement proceedings.  The venue provisions of the FDCPA are still relevant because the FDCPA defines "debt" as "any obligation or alleged obligation of a consumer . . . whether or not such obligation has been reduced to judgment."

The CA1 nonetheless joins the CA11 in holding that, under the relevant state law (Mass. and Ga.), post-judgment garnishment proceedings are not governed by the FDCPA's venue provisions, because it is not a suit "against [the] consumer" (wage-receiver), but rather are against the employer/wage-payor.  (p. 4, 6.)  The CA9 had earlier reached the opposite conclusion, holding that under California law a garnishment action was against the consumer.  (p. 5, 8).

I think the CA1 gets this exactly right.  But I am concerned about the extent to which these decisions make the availability of garnishment or other enforcement remedies depend on the vagaries of state law.  Why should a consumer in Massachusetts be subject to garnishment if his employment is in a different county, but not a consumer in California (or Ohio)?

I think there is another way to reach the right result, one that would have resulted in the CA9 also holding in favor of the collector  As the CA9 opinion notes, Congress passed the FDCPA venue provisions to ensure that consumers would not have to defend against suits in far-away, inconvenient courts.  But a garnishment suit--whether against the consumer or employer--filed wherever the consumer actually works cannot be inconvenient, as the consumer already commutes there daily.

In other words, I would treat "resides" in the venue provision as including every jurisdiction where the consumer would be subject to general (not specific--thats why the signature provision exists) personal jurisdiction.  If a consumer has a house in County A, but works in County B, his continuous and systematic presence in county B would dictate that he "resides"  in both for FDCPA purposes, and County B's exercise of jurisdiction surely would not offend notions of fair play.

True, the noun residence is ordinarily understood as a dwelling.  But--especially with the long hours at firms--a good argument could be made that I dwell both at work and home.  In the words of an old CA2 case:

Domiciliaries are those who have a fixed, permanent and principal home and to which, whenever absent, they always intend to return. At the opposite end of the scale are transients, those persons who are just passing through a locality. In between these notions of permanence and transience are residents. Residency means an established abode, for personal or business reasons, permanent for a time. 

Split Widened: Does U.S.S.G. 2G2.2(b)(3)(F) require knowledge?

Per United States v. Robinson (7th Cir. Apr. 9, 2013)

Although I am far more of a textualist than he, I am a fan of Judge Posner.  But this is one of the worst opinions from him that I have read.  It is internally contradictory, overlooks the obvious implications of the authorities it relies on, and creates an unnecessary make-work remand.

The split at issue is whether distribution has to be knowing for purposes of 2G2.2(b)(3)(F).  The text of the relevant commentary defining "distribution" and "distribution to a minor" provide:

"Distribution" means any act, including possession with intent to distribute, production, transmission, advertisement, and transportation, related to the transfer of material involving the sexual exploitation of a minor. Accordingly, distribution includes posting material involving the sexual exploitation of a minor on a website for public viewing but does not include the mere solicitation of such material by a defendant. 

"Distribution to a minor" means the knowing distribution to an individual who is a minor at the time of the offense

The CA10 held that knowledge was not required because (1) the text of the commentary is silent with respect to mens rea, (2) the usual presumption of a mes rea requirement, applicable to criminal laws, does not apply to Guidelines, and (3) reading a mens rea requirement into the definition of distribution would render the word "knowing" in the definition of distribution to a minor superfluous.

Here, the CA7 joins the CA8 (whose opinion is quite opaque) to reach the opposite conclusion, that knowledge is required.  (p.3.)  It provides only one reason for this conclusion:  that strict liability is disfavored in the criminal context.  In addition, Judge Posner rejects the argument (apparently not made) that knowledge of a peer-to-peer network's capabilities would make distribution "knowing," on the grounds that the criminal presumption of knowledge of the law does not apply in the Guidelines context.  Finally, Judge Posner rejects the CA10's superfluity analysis by reading the commentary's use of "knowing" to apply to the fact of the recipient's minority, rather than just to distribution.

As intimated in the introduction, I am inclined to agree with the CA10.  The relevant commentary is silent, the canon against superfluity applies, and the presumption of a mens rea does not apply to the guidelines.  With respect to Judge Posner's attempt to rebut the canon against superfluity, knowing should be read as modifying its closest noun, distribution, and not minor.

Moreover, Judge Posner's analysis is self contradictory in two respects.  First, he relies on the criminal law presumption of a mens rea, though numerous circuits have concluded that that presumption does not apply to the Guidelines, while rejecting the criminal law presumption of knowledge of the laws precisely because of the Guidelines context (though numerous circuits have applied this presumption to the Guidelines).  Second, Judge Posner notes that the Sentencing Commission has taken note of the split between the CA8 and CA10, and has stated "the guideline could be amended to better distinguish between more and less culpable distribution conduct.”  (p.5 (emphasis added).)  But he fails to draw the obvious conclusion:  That in the Sentencing Commission's view, unknowing conduct is still distribution, albeit less culpable.

What makes this all the worse, however, is that the opinion is reviewing for plain error.  Certainly, as the analysis above shows, the knowledge requirement is not plain.  (Judge Posner's opinion only introduces the plain error framework after undertaking what appears to be a de novo review of the Guidelines).  Plus, for a reversal on plain error, the defendant has the burden of showing that the error affected his substantial rights.  Here, I do not know how the Defendant could possibly have met that burden, where the ultimate sentence imposed was at the low-end of the guidelines range without taking the 2G2.2(b)(e)(f) enhancement into account.  Judge Posner entirely overlooks the burden of proof, and does not require the Defendant to offer any evidence that the sentencing judge might have or would have imposed a lesser sentence under a supposedly "proper" guidelines calculation.  This ultimately results in a make-work remand, unnecessarily burdening the judicial system..

Split Widened: Does Bankruptcy Stay a Tax Court Appeal?

Per Schoppe v. Comm'r (10th Cir. Mar. 28, 2013)

Never has my textualism been more difficult to follow.  I am 100% sure that Congress meant to accomplish what its words seemingly (but not too surely) preclude.  I strongly recommend that any legislation/statutory construction professors use excerpts of some of the historical (not necessarily current) cases on both sides of this split.  For example, the Ninth Circuit has a good discussion about when not to apply the canon against superfluity.

Even better, the issue is relatively simple to explain.  A bankruptcy filing automatically stays most proceedings against the debtor.  Courts have interpreted this stay to apply to cases where the debtor is the appellant, provided that he was the defendant in the underlying proceedings.

The split concerns whether the stay applies to debtor appeals of tax court proceedings, where the debtor must initiate the tax court proceedings against the Commissioner, but will never actually receive any affirmative relief, but only a potential reduction/elimination of a deficiency.  Are such proceedings "against" the debtor, so that they can be stayed?

Here, the CA10 joins the CA11 in saying no, the stay does not apply.  The CA9 said yes.  As indicated above, I have a really hard time figuring out where I would come out.

On one hand, it is entirely clear that Congress intended the stay to apply.  Indeed, the statute explicitly covers proceedings in the tax court, see 11 USC 362(a)(8), and there is no logical reason why lower court but not appellate proceedings should be covered.

On the other hand, it also fairly clear that the relevant statutory text does not cover tax court appeals.   The stay covers:

 the commencement or continuation . . . of a judicial, administrative, or other action or proceeding against the debtor . . ., or to recover a claim against the debtor that arose before the commencement of the case under this title.

Id. 362(a)(1).  For three reasons, this does not apply.  First, the tax court proceeding is initiated by the debtor, against the commissioner.  Second, the tax court proceeding is not a continuation of the administrative process against the debtor, because the Supreme Court has said that tax court proceedings are judicial proceedings not subject to deferential review.  Third, tax court proceedings are not to recover a claim against the debtor, because the court cannot afford the Commissioner affirmative relief (that's for the administrative mechanisms), just uphold or reject the deficiency calculation.

In the end, I believe I would concur dubitante (background) in whatever path my hypothetical panel chose.  Certainly, my textualism points me toward thinking (like the CA10 here) that the stay doesn't apply.  At the same time, the CA9 opinion on the other side is quite strong, and purposivism suggests that the stay should apply.  I could textually justify applying the stay in two ways:

1. Treating a tax court proceeding as a declaratory judgment action, with the debtor as the declaratory judgment plaintiff seeking to avoid potential liability.  Then, hypothetically realigning the parties (as permitted by the Supreme Court) to treat the tax court proceeding as against the debtor.  See Pub. Serv. Comm'n of Utah v. Wycoff Co., Inc., 344 U.S. 237, 248 (1952)
2. Treating the tax court appeal as the "continuation of a proceeding before the United States Tax Court," even though the appellate proceedings are no longer before the Tax Court.

Split Noted: Is Application of a Lengthened Limitations Period Retroactive?

Per U.S. ex rel. Carter v. Halliburton Co. (4th Cir. Mar. 18, 2013)

All qui tam plaintiffs should take note of this opinion, if only because has the effect of freezing the False Claims limitations bar in October 2002 (and thus also lengthens any potential damages period).

In dissent, Judge Agee notes (29 n.4) a circuit split on whether application of a limitations period that is lengthened after the relevant conduct is retroactive (and hence impermissible).  To provide an example, on Date X, Defendant does something for which there is a 1 year limitations period.  At some point after Date X, the legislature changes the period to 2 years.  In year 2, plaintiff files suit.  Can Defendant raise a viable limitations defense?

Judge Agree agrees with the CA6 that the answer is no.  Giving effect to the new limitations period is not retroactive, because limitations regulates not the underlying conduct, but when the Plaintiff may file suit.  In the words of the CA6 opinion cited by Judge Agee, "the 1991 Act applies to Forest's conduct, the filing of the complaint, which occurred after the enactment of the statute. Therefore, application of the 90-day statute is prospective in this case."  Thus, in this view, the new limitations period applies because the key date for retroactivity is the filing of the complaint.

I disagree, and side with the CA9.  The key date is not when Plaintiff finally files suit, but rather the date of passage/enactment of the limitations period.  Under this analysis, whether the hypothetical defendant can raise a limitations defense depends on whether the legislature acted in year 1 (no), or in year 2 (yes).

If the legislature acted in year 1, all is fine.  Plaintiff still had a viable claim, we presume he was on notice of the statute lengthening the limitations period.  The longer period thus did not affect anybody's substantive rights:  Plaintiff still had a cause of action, Defendant still had potential liability.

But if the legislature acted in year 2, then applying the new statute of limitations is retroactive.  Plaintiff no longer had a viable cause of action, defendant had escaped liability.  Applying the new limitations period would thus create a new substantive right in plaintiff, and a new liability in Defendant, when both had previously expired at the end of year 1.

Thus, the key point for retroactivity analysis is (1) the status of Plaintiff's claim on (2) the date the statute was passed/took effect.

Ultimately, this analysis calls to mind Justice Scalia's concurrence in Landgraf itself.  Justice Scalia, joined by Justices Thomas and Kennedy, criticized the majority for focusing on the substance/procedural divide, rather than looking at what conduct the statute was meant to regulate.  As any Civ Pro student knows, categorizing limitations periods as substantive or procedural so flummoxed the Supreme Court in Guarantee Trust that Justice Frankfurter crafted the outcome-determinative test to hold that--as with other parts of substantive law--state limitations periods should be applied in federal courts in diversity.  This result alone demonstrates how limitations periods can affect substantive rights.  By following the concurring opinion, one is left with the conclusion that the CA9 was correct in holding:

We do not find the substantive/procedural dichotomy helpful in deciding this case.  Regardless of whether a statute is "substantive" or "procedural," it may not apply to cases pending at the time of enactment if the new statute would prejudice the rights of one of the parties. . . . [W]e hold that a newly enacted statute that lengthens the applicable statute of limitations may not be applied retroactively to revive a plaintiff's claim that was otherwise barred under the old statutory scheme because to do so would "alter the substantive rights" of a party and "increase a party's liability."

Split Noted: Can A Defendant Obtain Fees Based On A Rejected Rule 68 Offer of Judgment?

Per UMG Recordings, Inc v. Shelter Capital Partners LLC (9th Cir. Mar. 14, 2013)

I have always been surprised how few defendants use Rule 68.  To facilitate settlement, that rule
authorizes defendants to make an "offer of judgment" before trial.  Such offer must specify terms (money, injunction, apology, etc.) to be entered as a formal court judgment if accepted.

The rule itself encourages defendants to make such offers because they come at no cost (other than the offered/confessed judgment) to defendants.  To encourage Plaintiffs to accept such offers, or at least consider them, the rule provides that, "[i]f the judgment that the [plaintiff] finally obtains is not more favorable than the unaccepted offer, the [plaintiff] must pay the costs incurred after the offer was made."  Fed. R. Civ. P. 68(d).

In Marek v. Chesney, 473 U.S. 1, 9 (1985), the Supreme Court held that "costs" for purposes of Rule 68 included "all costs properly awardable under the relevant substantive statute or other authority. . . .
Thus, . . . where the underlying statute defines 'costs' to include attorney's fees, we are satisfied such fees are to be included as costs for purposes of Rule 68."

Given this authority, it would seem clear that where a plaintiff (1) brings a federal suit under a fee shifting statute that defines costs to include fees, (2) rejects an offer of judgment, and (3) ultimately receives a judgment less favorable than the offer, the defendant would be entitled to costs, including its post-offer fees.

But that is not the position of the majority of courts, as this CA9 opinion notes.  (p. 58 n.25.)  Indeed, it appears that only the CA11 adheres to this view, while the CAs DC, 1, 4, 5, 7, 8, and 9 all disagree.

I agree with the CA11.  The majority misreads Marek's use of "properly awardable."  For example, many majority-school opinions find fees not "properly awardable" because the underlying statute authorizes  fees only to the "prevailing party", and defendants did not prevail.

This mistakes the baseline for determining who "prevailed."  The proper yardstick is not the status quo ante litem, but rather the status quo post confessionem (my translation of offer).  For example, before litigation, the damages baseline for prevailing is $0.  But after an offer, the damages baseline for prevailing becomes the amount of the offer.  Thus, if a plaintiff achieves more than an offer, it has still prevailed.  But, if a plaintiff rejects an offer and ultimately receives less, the defendant has prevailed.

The structure of the rules mandates utilizing the status quo post confessionem yardstick.  For example, Rule 54(d) presumes that costs--other than fees--will be awarded to the prevailing party.  But nobody, not even the CA9 here, disputes that Rule 68 mandates that post-offer, non-fee costs be awarded to the offeror.  Thus, Rule 68(d) in essence renders the offeror the prevailing party for purposes of non-fee costs.

No majority-school opinion offers any convincing reason why Rule 68(d) should alter the meaning of "prevailing party" for non-fee costs under Rule 54(d), or at least suspend operation of Rule 54(d)'s "prevailing party" provision for non-fee costs, but not do the same for "prevailing party" statutes where fees are defined as an element of costs.  (One could try to tease out a theory based on the Rules Enabling Act, but Marek rejected such a theory, which was espoused in Justice Brennan's dissent.)

Moreover, as a policy matter, the majority-school creates a one-way ratchet that is inconsistent with the objectives of Rule 68.  The fee shifting statute at issue here allows both prevailing defendants and plaintiffs to recover their fees equally.  (Thus distinguishing the Supreme Court's gloss on the civil-rights attorneys' fees.)  Thus, there is no policy reason to be more protective of plaintiffs in this situation.  Yet, the CA9 permits plaintiffs to avoid paying defendants' attorneys fees even if they recover less than an offer, but still recover their own attorneys' fees if they recover more.  This undermines the incentives to settle, contrary to the express purpose of Rule 68.

This is an issue ripe for review, or at least rulemaking.  Given the Court's concern about floods of litigation, one would think that a stronger interpretation of Rule 68--i.e., the CA11--should win out.

Split Noted: Who Bears The Burden On Earmarking for 11 U.S.C. 547(b)

Per In re ESA Environmental Specialists, Inc. (4th Cir. Mar. 1, 2013)

Under 11 U.S.C. 547(b), "[a] trustee in bankruptcy may avoid any transfer of an interest of the debtor in property" to a creditor for an antecedent debt made within 90 days of the petition, while the debtor was insolvent, that enables the creditor to recover more than it would in bankruptcy.

Courts have created an earmarking "affirmative defense" to such avoidance actions.  The earmarking defense applies when new creditor makes a loan to the debtor specifically for (or earmarked for) the payment of a designated old creditor.  In brief, the defense recognizes that the new loan never becomes part of the debtor's estate.  Moreover, the position of other creditors are not hurt by such an earmarked loan--the new creditor is simply substituted for the old creditor, and will receive the same pro-rata share of the bankruptcy proceeds as all other creditors.

In this opinion, the CA4 notes a circuit split on who bears the burden of proving the earmarking "affirmative defense."  (p. 8-9, n.6.)  It is unquestioned that the trustee bears the general burden of proving the avoidability of a transfer.  But a majority of circuits (the CAs 3, 6, and 9) hold that, if the trustee proves the general avoidability of a transfer, the burden shifts to the defendant to prove that the funds were earmarked.  In contrast, the CA8 holds that the trustee has the burden of proving that the earmarking defense does not apply.

I agree with the CA8.  To be sure, the majority school is consistent with the general rule that the burden of proving affirmative defenses is on the defendant.  But I do not believe that the earmarking defense is actually an affirmative defense (the CAs 3 and 9 hold as much).  Rather, it is a negative defense--if proven, it shows that the transfer was not "of an interest of the debtor."  The debtor never acquired full--i.e., equitable and legal--title to the transferred funds.  Instead, to the extent the debtor receives earmarked funds, it acts more in the capacity of a trustee, receiving legal but not equitable title to the funds, with no independent control but instead subject to an obligation to transfer the funds to the old debtor.  Thus, the burden of disproving this negative defense rests with the party bearing the burden of proof--the trustee.

Split Noted: Does U.S.S.G. 3C1.2 Require Nexus?

Per United States v. Gray (10th Cir. Mar. 5, 2013)

This unpublished disposition notes the CA10's continued disagreement with opinions from the CAs 5 and 6 holding that the 2-level adjustment for reckless endangerment by flight in U.S.S.G. 3C1.2 requires some nexus to the offense of conviction.  (p.10-11.)  Here, as before, the CA10 determines it is unnecessary to resolve this issue because--whether or not required--nexus is present.  (p. 12.)  The full text of the Guideline is:

If the defendant recklessly created a substantial risk of death or serious bodily injury to another person in the course of fleeing from a law enforcement officer, increase by 2 levels.

On its face, this language does not require nexus.  But the CA5 relied on a different provision (U.S.S.G. 1B1.3(a)(1)) to conclude that nexus was still nonetheless required.   I had written a post sharply critical of the CA5's position before I realized that the Commission's website provided an inaccurate html version of U.S.S.G. 1B1.3.  More on the website issues anon.  Now, with the accurate version, I am inclined to agree with the CA5 for four reasons.

First, U.S.S.G. 1B1.3(a)(1) provides in relevant part that "adjustments in Chapter Three[] shall be determined on the basis of . . . all acts and omissions committed . . . by the defendant . . . that occurred during the commission of the offense of conviction . . .  or in the course of attempting the avoid detection or responsibility for that offense."  The emphasized language indicates that must be a link between the flight proscribed by U.S.S.G. 3C1.2 and the underlying offense.

Second, the introductory commentary to Chapter 3 of the guidelines, where U.S.S.G. 3C1.2 is located, provides that the "following adjustments are included in this Part because they may apply to a wide variety of offenses."  Thus, the commentary also suggests an offense-specific linkage.

Third, the offense-specific linkage is supported by the history of U.S.S.G. 3C1.2.  When the Guidelines were first propounded, there was no "flight" adjustment.  And when the flight adjustment of 3C1.2 was added in 1990, the Commission made explicit that it derived from 3C1.1, governing obstruction.  (This relation is also documented in the title of Part C of Chapter 3--"Obstruction and Related adjustments).  The guideline for obstruction--to which the flight adjustment must be related--clearly requires a linkage to the offense, calling for a two-level increase if  "the defendant willfully obstructed or impeded . . . the administration of justice with respect to . . . the instant offense of conviction."

Finally, another part of the history of 3C1.2 supports an offense-specific linkage.  One year after the flight adjustment was created, the Commission made clear that "during flight" was to be construed broadly, such that reckless endangerment of a police officer in flight was at least subject to a 2-level increase under U.S.S.G. 3C1.2, if a 3-level increase under U.S.S.G. 3A1.2(b) did not apply.  At the time, 3A1.2(b) also created an explicit offense-specific link, requiring assaultative conduct creating "during the course of the offense or immediate flight therefrom", i.e., "proximate in time to the commission of the offense."  I view the distinction between these two as whether the flight is assaultative conduct.  That said, this is the weakest argument, becuase it could be argued that the amendment was intended to encompass situations that were not linked, or proximate in time, to the offense of conviction.

The counter-argument, though not explicitly made by the CA10, is that U.S.S.G. 1B1.3(a)(1) is not exclusive.  U.S.S.G. 1.1B3(a)(4) permits Chapter Three adjustments to be determined on the basis of "any other information specified in the applicable guideline," and the commentary to this provision explicitly states that it is intended to include information related to the defendant's "state of mind."  Thus, U.S.S.G. 3C1.2 calls for consideration of "other information" relating to a defendant's "state of mind":  namely whether "the defendant recklessly created a substantial risk of death or serious bodily injury to another person in the course of fleeing from a law enforcement officer."   Moreover, this counter-argument is bolstered by the fact that nexus does not appear to be required for U.S.S.G. 3A1.2(c)(2), which no published opinion has construed (at least that I found).

Ultimately, although I agree with the CA5, there is valid room for dispute.  This dispute may not matter all that much, because  if nexus is required, a sentencing court can (and should, in my view) vary 2 levels up to take into account the defendant's behavior where nexus is lacking.

What does matter to practitioners, however, is that the Sentencing Commission's website be accurate.  And it is not.  The Commission provides materially different versions of U.S.S.G. 1B1.3(a)(1),  depending on whether one is viewing the Guidelines in pdf form, or html form.

Here is the PDF version, as quoted above:

(1)      (A) all acts and omissions committed, aided, abetted, counseled, commanded, induced,  
                procured, or willfully caused by the defendant; and

          (B) in the case of a jointly undertaken criminal activity (a criminal plan, scheme, endeavor,
                or enterprise undertaken by the defendant in concert with others, whether or not    
                charged as a conspiracy), all reasonably foreseeable acts and omissions of others in
                furtherance of the jointly undertaken criminal activity,

          that occurred during the commission of the offense of conviction, in preparation for that      
          offense, or in the course of attempting to avoid detection or responsibility for that offense;

In this version, the structure of the paragraphs makes clear that the last sentence, beginning "that occurred", modifies both (A) and (B).  In contrast, the HTML version provides:

(1)      (A) all acts and omissions committed, aided, abetted, counseled, commanded, induced,  
           procured, or willfully caused by the defendant; and

          (B)in the case of a jointly undertaken criminal activity (a criminal plan, scheme, endeavor, or
           enterprise undertaken by the defendant in concert with others, whether or not charged as a
           conspiracy), all reasonably foreseeable acts and omissions of others in furtherance of the
           jointly undertaken criminal activity, that occurred during the commission of the offense of
           conviction, in preparation for that offense, or in the course of attempting to avoid detection
           or responsibility for that offense;

In the HTML version, the structure of the paragraphs erroneously suggests that the clause beginning "that occurred" modifies only (B).  This inference is grounded in the location of the semi-colons and the last antecedent rule ("that" would modify "reasonably foreseeable acts and omissions").

The Sentencing Commission needs to change its coding.

Split Widened: Is Use of a Fake Social Security Card for Employment Morally Turpitudinous?

Per Marin-Rodriguez v. Holder (7th Cir. Mar. 6, 2013)

This Seventh Circuit opinion widens one split, while continuing another.

The first, already extant split concerns whether immigration courts (IJs or the BIA) may consider material outside the record of conviction when determining whether a crime constitutes a crime of moral turpitude.  This blog already discussed the split here.  In brief, the CA7 and the Attorney General have concluded that, because the immigration context does not raise Sixth Amendment concerns, immigration courts should not be limited to the categorical and modified-categorical approaches to past convictions.  (p. 6-7.)  The CAs 3, 4, 8, and 11 have rejected this position.  (Id. at n.3.)  For the reasons discussed in my prior post, I am persuaded by the minority of the CA7 and AG, although the majority also has strong arguments (primarily based on a 1951 Supreme Court precedent holding that the words "moral turpitude" had a positive and fixed meaning).  To say the least, this is an issue ripe for certiorari review.

But not in this case.  This case did not even reach outside the modified categorical approach, which revealed that the alien had used a fake social security card for employment.  The CA7 joins with the CAs 5, 6, and 8 in determining that such fraudulent actions necessarily involve moral turpitude, because the false social security card is used to deceive the employer, or anyone examining the employer's records.  (Fraud is a prototypical example of moral turpitude.)  The minority--only the CA 9--relies on an entirely different statute, 42 U.S.C. 408(e)), which grants amnesty for having previously used false social security cards to aliens granted citizenship.  The CA9 believes that the exemption demonstrates that the use of a false social security card is only malum prohibitum, and not malum in se.  I would conclude the opposite.  The amnesty demonstrates that the use of a false social security card is malum in se, but to be forgiven when the alien is given a new shot at becoming a law-abiding citizen.

In all, a correct--and well-written--opinion by Judge Manion, but certainly some issues that may warrant certiorari review if a better vehicle presents itself.

Split Noted: When Does Restoration Civil Rights Discount a Conviction for the ACCA?

Per United States v. Swims (7th Cir. Feb. 28, 2013)

18 U.S.C. 922(g)(1) renders the interstate shipping, transporting, or receiving of a firearm a crime for any person who has been convicted in any court of a crime punishable by imprisonment for a term exceeding one year.  In turn, 18 U.S.C. 920(a)(2) excludes certain crimes from the statutory term "a crime punishable by imprisonment for a term exceeding one year," and further defines that term as:

What constitutes a conviction of such a crime shall be determined in accordance with the law of the jurisdiction in which the proceedings were held.  Any conviction which has been expunged, or set aside or for which a person has been pardoned or has had civil rights restored shall not be considered a conviction for purposes of this chapter, unless such pardon, expungement, or restoration of civil rights expressly provides that the person may not ship, transport, possess, or receive firearms.

In this case, upon serving his state sentence for aggravated robbery, Swims received a letter informing him "of the restoration of your right to hold offices . . . . [and] the right to restoration of licenses granted to you."  The letter did not mention Illinois's standing prohibition on gun ownership by felons.  Thus, Swims contended that this letter meant that he "has had civil rights restored," such that the aggravated robbery adjudication did not count as a conviction.

The Seventh Circuit rejects this argument, reasoning that the letter has to include restoration of "the big three" civil rights--i.e., the right to vote, hold office, and serve on a jury--in order to trigger the second sentence of the statute.  The Seventh Circuit notes, however, a circuit split over the "anti-mousetrapping" doctrine, which concerns the "unless" clause of the second sentence.  (p.4)

The anti-mousetrapping doctrine says that any restoration of civil rights itself must notify a convict that he is still forbidden from possessing firearms for the "unless" clause to apply.  Five circuits (the CAs 5, 7, 9, 11 & DC) follow this rule, while four (the CAs 4, 6, 8, & 10) hold that if the restoration is silent, but state law still would prohibit firearm ownership, the "unless" clause nonetheless applies.

 I side with in the middle of this split.  To recap only the relevant statutory language:

Any conviction  for which a person has had civil rights restored shall not be considered a conviction unless such restoration of civil rights expressly provides . . . [the] firearms [restriction]. 

The plain text of the "unless" clause, i.e., the words "such" and "expressly" require reference to the document restoring civil rights.  If--as here--a convict receives a letter notifying him of the restoration of civil rights, and that letter does not also notify him of a restriction on gun ownership, the convict could reasonably believe that the clause does not apply to his conviction.  To the extent there is any ambiguity, both the rule of lenity and the doctrine of fair notice should require the same result.  If, however, a convict does not receive any notification of the restoration of rights, but the restoration happens automatically--and silently--by operation of state law, the same "document" restoring civil rights, i.e., the code book,  expressly provides the firearms restriction.

 The circuit split at issue here is caused by an antecedent issue, on which all circuits and the Supreme Court agree: "'restoration of civil rights' does not" "connote[] a case-by-case determination."  Caron v. United States, 524 U.S. 308, 313 (1998).  Despite the uniformity, I disagree for multiple reasons.

First, the terms of the unless clause (the exception to the exception), referring to "such . . . restoration of civil rights . . . expressly ," suggest that the restoration must take place by an individualized, affirmative action, rather than by silent operation of a blanket law.  To the extent relevant, legislative history confirms this suggestion.  When the relevant language was originally passed in 1981, the Senate Committee report described its effect:  "In the event that the official granting the pardon, restoration of rights or expungement of record does not desire it to restore the right to firearm ownership, this provision is rendered inapplicable where the order or pardon expressly provides that the person may not possess firearms."  S. Rep. 97-476 at 18.

Second, the statutory term "restoration of civil rights" is ambiguous (e.g., as to what rights matter), so I would resort to noscitur a sociis.  All of the other statutory actions with respect to a conviction--i.e., expungement, setting aside, or pardoning--require an individual determination.  (Even routine expungements, such as for juvenile convictions, require the court to take an individual action with respect to one person's record, and can be prevented by an individualized showing of the need for continued collateral consequences.)

Third, requiring an individualized determination is likely the only way to harmonize the first and second sentences of the statute.  Where a state restores certain civil rights by operation of law, such as Illinois did here, its law still treats the convict as a convict for all purposes, including the rights restored and the rights still denied.  Thus, the first sentence (providing that state law governs whether a state conviction qualifies) would mandate that the convict still be treated as convict.  In contrast, where a state restores civil rights by individual order, that order itself will determine for what purposes state law will treat the convict as a convict.  Thus, even the first sentence would--as the second sentence actually does--require reference to the individual order granting a restoration of rights.

Fourth, the entire second sentence of the statute is an exception to the first, thus triggering the canon that provisos be narrowly construed.  Requiring case-by-case determinations is certainly more narrow than crediting all operations of state law.

Finally, the overall purpose of the second sentence's exception appears to be the discounting of convictions where public officials have determined that individual circumstances (whether rehabilitation, or lack of original criminal intent) render unnecessary continued, generally-applicable deprivations of civil rights.  Thus, if a state has determined that an individual convict does not pose a threat to the public such that he may have his rights restored, the criminal code provides no reason to overturn that determination and still count the conviction.

Split Noted: Must An IJ Provide Notice Of The Need For Corroboration?

Per Pulatov v. Holder (6th Cir. Feb. 25, 2013)

In 2005, Congress passed the REAL ID Act.  The effect of the statute was largely to abrogate several lines of outlier, pro-immigrant precedent from a certain "liberal" circuit.  One of the abrogated lines of precedent held that an Immigration Judge could not require corroboration from a credible alien.

In lieu of the no-corroboration rule, the REAL ID Act provided, in a clause now codified as 8 U.S.C. 1158(b)(1)(B)(ii):

The testimony of the applicant may be sufficient to sustain the applicant's burden without corroboration, but only if the applicant satisfies the trier of fact that the applicant's testimony is credible, is persuasive, and refers to specific facts sufficient to demonstrate that the applicant is a refugee.  In determining whether the applicant has met the applicant's burden, the trier of fact may weigh the credible testimony along with other evidence of record.  Where the trier of fact determines that the applicant should provide evidence that corroborates otherwise credible testimony, such evidence must be provided unless the applicant does not have the evidence and cannot reasonably obtain the evidence.

(Emphases added).  In this unpublished disposition, the CA6 notes, but does not resolve a split between the CA3 and CA9, on the one hand, and the CA7, over whether an IJ must provide notice and an opportunity to respond after reaching the determination that corroboration is necessary.  (p. 7 n.5).

The CA7 concludes that the statute already provides notice of the need for corroboration, because testimony alone only "may be" sufficient, and administrative efficiency does not require a second opportunity to obtain it.

In contrast, the CAs 3 and 9 hold that the statutory language requires opportunity to respond.  Specifically, the Ninth Circuit relies on the  tense of "should provide" and "must be provided" to conclude that the "language focuses on conduct that follows the IJ's determination, not precedes it[;] the statute's future directed language means that the applicant must be informed of the corroboration that is required."  (Third emphasis added).

I come down in the middle of this split.  Specifically, I agree with the Seventh Circuit that the statute provides notice of the general need for corroboration.  And I disagree with the Ninth Circuit that the statute uses future directed language--all of the verbs highlighted are in the present tense.  Thus, I don't think additional advance notice of the need for corroboration is required.

But once an IJ concludes that corroboration is necessary, the statute does mandate--as the CAs 3 and 9 conclude, that the alien be provided an opportunity to respond.  Such a response could take two forms:

1. An explanation that the alien had previously sought the requested corroboration, and was unable to obtain it, in which case the IJ would have to evaluate the explanation and determine whether the requested evidence is reasonable obtainable.  
2. An explanation that the alien had not previously sought the requested corroboration, and believes he would be able to provide it, in which case the IJ would have to determine whether the alien acted reasonably.  

• If the applicant reasonably believed the requested corroboration related to a minor point in his story or would be cumulative, the applicant should be afforded a continuance to obtain the requested evidence.  
• If the applicant acted unreasonably in failing to seek the requested corroboration despite the statutory notice, because it related to a central and uncorroborated point in his story, then the IJ could deny asylum because the alien may simply be seeking to delay the inevitable.

Finally, my apologies for the lagtime between posts.  Nine briefs, two in the Supreme Court, in a month will do that to a man.

Split Noted: Is Control Necessary for USSG 3B1.1?

Per United States v. Samolja (7th Cir. Jan. 25, 2013)

I rarely blog about unpublished cases, but two recent ones have caught my eye.  For other reasons, I cannot blog about one, out of the Fourth Circuit.

Here, however, the Seventh Circuit notes in passing (p.5 n.1) an intra-circuit split regarding whether   a defendant must control others in order to be eligible for an enhancement under U.S.S.G. 3B1.1.  Researching the split revealed that most circuits do require some degree of control, whereas the the CA10, and certain cases of the CA7, do not.

The majority is right.  Here is the relevant statutory text and commentary:

__(a) If the defendant was an organizer or leader of a criminal activity that involved five or more participants or was otherwise extensive, increase by 4 levels.

___(b) If the defendant was a manager or supervisor (but not an organizer or leader) and the criminal activity involved five or more participants or was otherwise extensive, increase by 3 levels.

4. In distinguishing a leadership and organizational role from one of mere management or supervision, titles such as "kingpin" or "boss" are not controlling. Factors the court should consider include the exercise of decision making authority, the nature of participation in the commission of the offense, the recruitment of accomplices, the claimed right to a larger share of the fruits of the crime, the degree of participation in planning or organizing the offense, the nature and scope of the illegal activity, and the degree of control and authority exercised over others. There can, of course, be more than one person who qualifies as a leader or organizer of a criminal association or conspiracy. This adjustment does not apply to a defendant who merely suggests committing the offense.

By definition, managers, supervisors, and leaders all exercise control.  While the definition of organizer could be broader, in theory, the statutory scheme confirms that it is not.

First,Organizers are equivalent to leaders, who must exercise control.  Second, control is inherent to management and supervision, which are deemed to be lesser included forms of organization.  Third, most tellingly, the commentary requires some "degree of control" for any form of organizing, leading, managing, and supervising.  Finally, if there is any degree of ambiguity remaining after reading the text and the canons, the rule of lenity would also favor a control requirement (circuits seem to uniformly apply lenity to the guidelines, though there is some question whether this application is appropriate).

Split Widened: Is the Definition of Arbitration a Matter of State or Federal Common Law?

Per Bakoss v. Certain Underwriters at Lloyds of London (2d Cir. Jan 23, 2012)

The Federal Arbitration Act announced a national policy in favor of arbitration, and the Supreme Court--through cases like Concepcion and CompuCredit--has been active in ensuring that policy is fulfilled.

Nonetheless, several circuit splits persist.  This split concerns the very foundation of the FAA, the meaning of "arbitration" in section 2:

A written provision in a . . . a contract evidencing a transaction involving commerce to settle by arbitration a controversy thereafter arising out of such contract or transaction . . .  shall be valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract.

Is arbitration to be defined as a matter of federal common law, or a matter of state law?

The CA2 joins the CAs 1, 6, and 10 in applying federal common law.  (p.5.)  In contrast, the CAs 5 and 9 define arbitration with respect to state law.

The majority is clearly right.  The FAA was intended to announce a national policy.  Having that national policy limited by the vagaries of state definitions of arbitration makes no sense.

Despite a clear split, I am not sure this is a candidate for review.  Not only is the CA2 on the right side of the split (or, at the very least, the side of the split clearly favored by recent Supreme Court FAA jurisprudence), but the opposing decisions are also quite old and might be revisited by their respective circuits in light of the last 20 years of that Supreme Court jurisprudence.

On the note of federal common law, the "Erieblogging" series on Prawfsblog is quite informative for those who do not have Civ Pro nightmares.

Split Widened: Must The Statement, Or The Falsity, Be Material For 18 U.S.C. 922(a)(6)

Per United States v. Abramski (4th Cir. Jan. 22, 2013)

Former police officer Abramski was suspected of robbing a bank.  In the course of investigating that crime, for which the Government apparently found no evidence (p.5), FBI agents uncovered Abramski's purchase of a handgun for his uncle.  Abramski purchased the gun because he was able to obtain a favorable police-officer price, and he concealed the fact that his uncle, who was legally entitled to own guns, was the ultimate purchaser.

As a result of this discovery (and the lack of proof of bank robbery), Abramski was indicted for violating 18 U.S.C. 922(a)(6), making a false statement material to the lawfulness of a firearm sale.  In relevant part, the statute criminalizes:

[K]nowingly mak[ing] any false or fictitious oral or written statement . . . intended to deceive such [licensed] importer, manufacturer, dealer, or collector with respect to any fact material to the lawfulness of the sale or other disposition of such firearm . . . .

(Emphasis added).  Abramski challenged the indictment, contending that the concealment of his uncle was not material because the uncle is legally entitled to purchase firearms.  The district court denied Abramski's motion, and Abramski pleaded guilty while reserving his right to appeal.  The Fourth Circuit affirms, in the process widening a split between the CA5 and the CAs 6, 11, and now 4.  (p. 12-13.)

The split concerns whether "material" simply modifies the statutory term "fact," or also modifies "false or fictitious."  Put another way, the split concerns whether the question answered falsely must be material (CA4, 6, 11), or whether the falsity of the answer must itself be material (CA5). 

I agree with the majority school.  The text of the statute indicates that material modifies fact.  Thus, Abramski's concession that the identity of the purchaser is sometimes material--e.g., where the purchaser cannot legally buy guns--dooms his argument. 

More importantly, to be material, a fact does not necessarily have to affect the outcome of the decision (here, the decision to sell the gun).  Instead, a fact must simply be "capable of influencing" the decision.  Kungys v. United States, 485 U.S. 759, 770 (1988).  The identity of the purchaser is capable of influencing the decision to sell a gun, because some individuals cannot buy guns.  Therefore, identity is material.

Split Widened: Must Evidence Be Admissible To Constitute Brady Material?

Per Johnson v. Folino (3d Cir. Jan. 16, 2013)

The standard of review governing alleged Brady violations is fairly well settled.  A new trial (or habeas) will be granted only if (1) the state suppressed (2) material evidence (3) favorable to the accused.

The split at issue here concerns the second factor, i.e., how to determine if the suppressed evidence is material.  Suppressed evidence is material if there is a reasonable probability of a different result if the evidence been disclosed. 

The CA4 holds, as a matter of law, that inadmissible evidence can never be material.  Other circuits--the CAs 1,2,6,11 and now 3--hold that while inadmissibility is a factor weighing against materiality, it is not a per se bar.

To be honest, the existence of this split surprised me.  The majority school is simply correct.  While inadmissibility is a factor (and perhaps a strong one) indicating that the supressed evidence could not have affected the trial, it is not outcome-determinative.  For example, inadmissible evidence may later become admissible for purposes of impeachment, or may lead to the discovery of additional, and admissible, evidence by altering case-preparation strategies.  And even if none of these events come to pass, the inadmissible evidence may at least alter case-presentation strategy, such as by inviting the jury to consider a theory that is plausible even though not supported by any (admissible) evidence.  

Thus, the CA4 is simply wrong.  Even the case it relies on goes beyond noting inadmissibility to show that the defense attorney would not have prepared any differently had the suppressed evidence been disclosed.

More generally, my Brady concern focuses mostly on the third factor--favorable to the accused.  I am baffled by the issue of how a prosecutor is supposed to determine what is, and what is not, favorable.

Consider, for example, the sentencing of a drug dealer.  Most prosecutors would think a former co-conspirator's admission regarding drug weight to be prejudicial.  Yet, in one of my cases, we successfully used this evidence to mitigate our defendant's personal responsibility and obtain a variance.  Nonetheless, had the prosecutor in that case not followed an open-file policy, he almost certainly would not have viewed the material as being subject to a Brady disclosure obligation.

My general line of work civil tort defense.  I am consistently surprised by the ability of most plaintiffs lawyers to turn any fact, no matter how harmful on its face, into an apparent strength.  Criminal defense attorneys often display the same adaptability.  Yet a strict interpretation of favorability would limit such adaptation.  Thus, I fall on the side of favoring open-file prosecutions in the interests of justice.

Split Created: Addict's Risk of Relapse As Disability Under ERISA

Per Colby v. Union Sec. Ins. Co. (1st Cir. Jan. 17, 2013)

I hated ERISA cases as a clerk.  They were like petitions for review of agency action, requiring record-intensive factual review in an area where the governing law is generally settled--i.e., defer to "reasonable" interpretations of statute/plan language provided there is "some evidence" to support factual positions.  Except that in ERISA, the "agency" record consisted of hundreds of pages of undecipherable doctor's scrawl and review of that record required some medical knowledge.

But enough about me, and onto this relatively straightforward split.  Can an ERISA plan administrator apply a per se rule barring disability benefits for on-the-wagon addicts?  The CA4 said yes, relapse is a choice not a disability.   The CA1 now says no, the present risk of relapse can (but does not have to) amount to a disability.  (See p. 16, 19-20.)  To the extent it influences your opinion, plaintiffs in both cases were anesthesiologists who became addicted to the drugs they administered.

I agree with the CA1.  The CA4's reasoning just looks sparse to me.  In essence, the CA4 ruled that the pre-existing division in precedent on this point ipso facto rendered the administrator's interpretation of the plan reasonable.  But the CA4 performed no analysis of the plan's text--i.e., the definition of disability--itself.  In contrast, the CA1 performs just such a textual analysis and demonstrates how addiction and risk of relapse fall within the plain language of the plan.

That said, I am not on board for all of the CA1 opinion.  For example, the Court bolsters its textual analysis with a reference to the contra proferentem doctrine.  (p. 17.)  But there is a circuit split on whether contra proferentem can ever apply to ERISA plans.  (CA9 yes; CA8 no).  Even assuming the doctrine could apply, moreover, its application is quite troubling where (as here) the plan administrator is explicitly granted discretion to resolve ambiguities.  Unlike certain canons of construction (such as expressio unius or noscitur a sociis), contra proferentem--like the rule of lenity--is a policy-based rule that only comes into play when the statute (or contract) is determined to be ambiguous.  But ambiguity should trigger the administrator's discretion not policy-based canons.  (This is the CA10's approach to contra proferentem in the ERISA context).

The existence of multiple circuit splits make this case an interesting candidate for review, especially where the Court semi-regularly grants cert to clarify ERISA jurisprudence.  If the Court does take cert, I hope it will deal with the contra proferentem issue in the broader context of Chevron deference.  The circuits are currently split on whether to apply traditional canons of statutory construction at Chevron step one, in determining whether the statute plainly addresses an issue.  My view is that some, but not all, of the canons should be applied.  More specifically, the text-based canons should be applied in determining whether Congress has spoken clearly, but not the policy-based canons, which operate more as a thumb on the scale than as an interpretative tool.

Since I opened with a side note, I might as well close with one.  It appears that the First Circuit must batch cases for each panel, and that this panel received the "circuit split" batch.  At least three of my previous posts have related to decisions also arising out of this same sitting.  As before, my best wishes go out to Judge Boudin and his family.

Split Created: What is a Second or Successive Habeas Petition

Per Suggs v. United States (7th Cir. Jan. 17, 2013)

AEDPA bars second or successive habeas petitions.  Here, the prisoner's first petition resulted in resentencing.  Now the prisoner challenges his conviction in a second petition.  Relying on past circuit precedent, a divided panel of the CA7 holds the petition barred, because the conviction issue could have been raised in the first petition.  (p. 8).

The CA2 and CA9 disagree.  (p. 11.)  The correct result is a very close call but I tend to agree with Judge Sykes' dissent and the other circuits:  The petition is not second or successive because the intervening resentencing creates a new judgment, and habeas petitions are directed at judgments.  At bottom, though, I question the panel procedure of the CA7.

The split revolves around the interpretation of a recent Supreme Court decision, Magwood v. Patterson, 130 S. Ct. 2788 (2010).  In Magwood, the Supreme Court held that a petition directed at the new sentence, following resentencing after an initial petition, could not be second or successive:  "[B]oth § 2254(b)'s text and the relief it provides indicate that the phrase 'second or successive' must be interpreted with respect to the judgment challenged." 

Magwood, however, specifically reserved the question here, and did so with a footnote apparently favorably citing past circuit practice:

The State objects that our reading of § 2244(b) would allow a petitioner who obtains a conditional writ as to his sentence to file a subsequent application challenging not only his resulting, new sentence, but also his original, undisturbed conviction. . . .  This case gives us no occasion to address that question, because Magwood has not attempted to challenge his underlying conviction.[fn]

fn.:Several Courts of Appeals have held that a petitioner who succeeds on a first habeas application and is resentenced may challenge only the "portion of a judgment that arose as a result of a previous successful action."

This reservation and citation would seem at first glance to leave past precedent intact.  Nonetheless, the statutory interpretation of Magwood--that second or successive refers to judgments, not claims--fatally undermines past precedent and demands that any petition following a new sentence not be barred because there is a new judgment.

I do not fault the Seventh Circuit majority for following past precedent because it was not explicitly overruled.  This is a very close issue.  I do fault the panel, however, for not pre-circulating their opinion.  As this blog has previously noted, Seventh Circuit Local Rule 40(e) demands precirculation where, as here, an opinion creates a conflict among circuits.  Precirculation would have also been required had the opinion adopted the position of other circuits and overruled past precedent.

In sum, regardless of the position adopted, this case should have been considered for en banc review prior to publication.  And the en banc court may have felt more liberty to depart from past precedent to follow the logic of Magwood.

Split Noted: Is Cumulative Error Doctrine Clearly Established For AEDPA? Plus AEDPA Commentary

Per Lott v. Trammell (10th Cir. Jan. 24)

I will not claim to have read all of this 100+ page effort.  Given the length and the subject--AEDPA--I assume without basis that it capital case.

Ah, AEDPA.  The other 1996 statute that certain liberal jurists love to hate.  Nowadays, to grant habeas relief to a state prisoner, a federal court must find that the state court's decision was either (1) contrary to, or (2) an unreasonable application of, (3) clearly established Federal law.

The split noted here concerns the third element; whether the cumulative error doctrine is clearly established.  Clearly established Federal law is limited to the holdings of SCOTUS.   In finding cumulative error analysis clearly established, the CA9 cites three different SCOTUS cases.  On the opposite side of the split is, bizarrely, the CA6.  I say bizarrely because the CA6 has actually been spanked more times on AEDPA recently than the CA9.

 Even more bizarrely, at least for me, I believe the CA9 is right.  The holding of  Chambers v. Mississippi, 410 U.S. 284 (1973), is on point.  Chambers's "claim, the substance of which we accept in this opinion, rests on the cumulative effect of those rulings in frustrating his efforts to develop an exculpatory defense."  Id. at 290 n.3.

But I actually want to discuss AEDPA's second element in light of some recent civil tort work I've been doing.  For a state court opinion to be "an unreasonable application of" Supreme Court law, it must be such that no reasonble jurist could possibly agree.  This standard is functionally the same as the "no reasonable litigant" standard established for objective baselessness by PRE, 506 U.S. 49, 55-61 (1993).

But there is one difficulty.  Any number of cases could be cited for the proposition that, if you get one judge (even in dissent) to agree with your position, that position cannot be objectively baseless as a matter of law.  Thus, for purposes of Noerr immunity, courts are willing to assume that all jurists act reasonably.

In contrast, any number of cases could be cited where habeas was granted based on a state appellate court's unreasonable application of Supreme Court law.  No longer is one lone voice enough to demonstrate reason.  Now federal courts are willing to overrule even a majority of their state brethren.

Obviously, this disparity can be explained by having a thumb on the scale of constitutional rights.  The any judge standard of PRE's progeny protects the First Amendment right to petition, while lack of any similar standard protects convicts right to due process.

But I think the more likely explanation is outcome-oriented jurisprudence.  In a private dispute, there generally are not any larger principles or constitutional visions at stake.  Whereas in habeas, the entire focus is on the larger principles of criminal justice.  Thus, even though the standards call for the same amout of deference, certain jurists are simply less willing to defer in the latter situation than the former.

Split Noted: Does Fed. R. Crim. P. 32(i)(1)(A) require an on-the record inquiry?

Per United States v. DeLeon (1st Cir. Jan. 11, 2013)

Federal Rule of Criminal Procedure 32(i)(1)(A) provides:  "At sentencing, the court must verify that the defendant and the defendant's attorney have read and discussed the presentence report and any addendum to the report."  (Emphasis added)

The split noted (p. 13-14, n.7) in this CA1 opinion is whether the verification required by Rule 32(i)(1)(A) need be explicit and on the record.  The CA1,3,5, and 6 all hold that this specific inquiry need not be explicit, so long as it is clear from the record that the defendant and the attorney have discussed the report.  The CA7 requires an explicit inquiry.

I agree with the CA7.  After all, the definition of verify is to "prove the truth of, as by evidence or testimony."  (Emphasis added.) 

That said, I dont think the split matters at all.  First, if--as per majority school--it is apparent from the record that the defendant and the attorney have discussed the report, the failure to perform an explicit inquiry is necessarily harmless and should be disregarded.  Fed. R. Crim. P. 52(a).  Second, a district court could fail to make an explicit inquiry only if there is no objection.  Thus, appellate review of this issue will always be for plain error under Fed. R. Crim. P. 52(b).

Nonetheless, I think the better rule would be to instruct district courts that it is error (albeit virtually irreversible error) not to conduct an on-the-record inquiry into whether the defendant and attorney have reviewed and discussed the PSR.  Such an instruction would allow district courts to improve their sentencing checklists, and cohere more closely with the terms of the rule.

Split Widened: Yardstick by which to evaluate Governmental Interest in Medicating Criminal Defendant

Per United States v. Gutierrez (5th Cir. Jan. 11, 2013)

Sell v. United States, 539 U.S. 166 (2003), sets the due process standard for forcibly medicating a criminal defendant in order to achieve competency.  In brief, the government must show "[1] the treatment is medically appropriate, [2] is substantially unlikely to have side effects that may undermine the fairness of the trial, and, [3] taking account of less intrusive alternatives, is necessary significantly to further [4] important governmental trial-related interests."  Id. at 179.

Sell provided additional instructions concerning each of its four factors.  The split at issue in this case (p. 13) concerns the fourth Sell factor, namely how to determine if the government has important trial-related interests.  Sell listed a number of subfactors bearing on this inquiry, including:

• whether the crime is serious,
• the availability of civil commitment, and
• the potential length of confinement after medication and trial.

Id. at 180.

Most circuits, including now the CA5, look to statutory maxima authorized for the indicted crimes to determine both whether the crime is serious and the potential length of confinement.  (See p. 13, citing decisions from the CA2, CA4, and CA10.)  In contrast, the CA9 looks to the probable guidelines range rather than the statutory maxima.

The CA5 provides a number of strong reasons for its view, including that the guidelines range cannot be determined without a pre-sentence investigation into the offender's history and characteristics, and even it could be, the guidelines range remains only advisory. 

Nonetheless, I come down somewhere in the middle of the split.  A court should not consider only the guidelines range, for the reasons stated by the CA5, but neither should a court consider only the statutory maxima, which are so rarely imposed.  Instead, both yardsticks can be used together.  A trial judge can use his experience--as under Twiqbal's plausibility standard--to determine what weight to give each yardstick.

At bottom, this split is likely caused by the use of a muddled, multi-factor due process test, and the inherent subjectivity of words like "important."  The legal realist in me assumes that with such fuzzy tests, trial judges likely first arrive at what they view to be the appropriate result, and only then express their reasoning in a manner permitted by precedent.  Recognizing this, I believe it better to permit trial judges to consider, and express, all relevant factors.  And, for the reasons explained by courts on both sides of this split, the guidelines and the maxima are both relevant considerations.

Split Noted: Individual Public Employer Liability Under The FMLA

Per Diaz v. Mich. Dep't of Corr. (6th Cir. Jan. 7, 2013)

The Family Medical Leave Act (FMLA) conains both family-care and self-care provisions authorizing suits against employers--including public agencies--for monetary damages and injunctive relief.  The Supreme Court has held that money damages are obtainable in family-care suits against public employers, because those provisions were passed pursuant to Congress's authority to enforce the Fourteenth Amendment and therefore abrogate sovereign immunity.  Nev. Dep't of Human Res. v. Hibbs, 538 U.S. 721, 725 (2003).  Similar money damages suits based on the self-care provisions, however, are not permitted because the self-care provisions do not enforce the Fourteenth Amendment.  Coleman v. Court of Appeals of Md., 132 S. Ct. 1327, 1332 (2012).

Plaintiffs in this case attempt to work around Coleman, bringing 1983 "and laws" suits to enforce the self-care provisions.  And they are partially successful. 

In a well-reasoned opinion, the CA6 holds that the damages claims fail for multiple reasons (more on that anon).  Nonetheless, the court reverses the dismissal of the equitable claim for reinstatement by analogy to Ex Parte Young.

Among the reasons for dismissal of the damages claims were that the claims were pleaded against individual, rather than corporate, public employers.  The CA6 notes an entrenched split between the CA6 and CA11, on one hand, and CA8 and CA5 on the other, concerning whether individual public employer liability is permissible.  (p.8.)  Having reviewed the relevant cited cases, I applaud both the CA6 and the CA5 for their indepth analysis of the issue.  IMO, The CA5 has the best of the argument.

Here is the relevant statutory language defining employer:

(4) Employer


     (A) In general

     The term “employer”—

(i) means any person engaged in commerce or in any industry or activity affecting commerce who employs 50 or more employees for each working day during each of 20 or more calendar workweeks in the current or preceding calendar year;

(ii) includes—

(I) any person who acts, directly or indirectly, in the interest of an employer to any of the employees of such employer; and

(II) any successor in interest of an employer;

(iii) includes any “public agency”, as defined in section 203 (x) of this title; and

(iv) includes the Government Accountability Office and the Library of Congress.

     (B) Public agency

For purposes of subparagraph (A)(iii), a public agency shall be considered to be a person engaged in commerce or in an industry or activity affecting commerce.

29 U.S.C. 2611 (emphases added).  Given this text, it is clear that 4(A)(ii)(I) creates individual liability for supervisory employees, while 4(A)(iii) creates public employer liability.

The recursive nature of the text, i.e., the use of the term employer in the definition of employer, shows that public supervisory employees may be liable.  Witness the substitution of 4(A)(iii) for employer in 4(A)(ii)(I):

(I) any person who acts, directly or indirectly, in the interest of a[] public agency to any of the employees of such public agency; and

Further, as the CA5 convincingly demonstrates, the CA6's two arguments against individual liability ultimately fail.  (A third argument is omitted here for space.)

First, the CA6 relies on statutory structure, noting the separation between distinctly enumerated provisions (ii) and (iii).  This argument fails because the statute indicates that romanettes i-iv are all linked by use of the serial semi-colon: "(i) . . .; (ii) . . .; (iii) . . . ; and (iv)."  Moreover, it also overlooks the recursion of the term employer.

Second, the CA6 relies on what it discerns to be superfluity.  Specifically, it submits that recursion analysis, when combined with the difference between "means" in (i) and includes in (iii) renders the definition of public agency superfluous because (iii) is already included in (i).  Not so.  By including "public agency" in (iii), the statute expands the meaning of employer solely with respect to public agencies and relieves a plaintiff of the need to prove that the agency  "e,ploys 50 or more people . . . . ."

In all, good efforts by both courts, but I agree with the CA5.  Two parting notes.  One argument in favor of the CA6 is the definition of public agency fails to require that the agency be, well, public.  This omission is likely what prompted the superfluity point.  Nonetheless, I think this omission can be remedied by the very term being defined.

Second, in light of the immunity discussion above, you might wonder why the strict construction canon does not come into play.  Once again, the CA5 has a ready answer.  As a general rule the Eleventh Amendment does not bar suits against officers in their individual capacities. Hudson v. City of New Orleans, 174 F.3d 677, 687 n. 7 (5th Cir. 1999).

Split Noted: Is U.S.S.G. 3E1.1(b) Mandatory or Permissive?

Per United States v. Castro (3d Cir. Jan. 8, 2013)

U.S.S.G. 3E1.1 reduces the offense level of a criminal who accepts responsibility for the crime.  It has two parts.  First, a two-level adjustment is appropriate "[i]f the defendant clearly demonstrates acceptance of responsibility."  Id. 3E1.1(a).  Second, an additional one-level decrease is called for if (1) the first section is satisfied, (2) the offense level prior to the first section is >15, and (3) the government formally requests.  In full, the second part provides:

(b) If the defendant qualifies for a decrease under subsection (a), the offense level determined prior to the operation of subsection (a) is level 16 or greater, and upon motion of the government stating that the defendant has assisted authorities in the investigation or prosecution of his own misconduct by timely notifying authorities of his intention to enter a plea of guilty, thereby permitting the government to avoid preparing for trial and permitting the government and the court to allocate their resources efficiently, decrease the offense level by 1 additional level

Id. 3E1.1(b) (emphases altered). 

The Third Circuit notes a split between the CA5 and the CA7 concerning whether this second provision remains discretionary with the court, or is mandatory upon satisfaction of the prerequisites, respectively.  (p. 29.)  Reviewing the cited opinions, the split appears to be lopsided in favor of discretion, in large part based upon one of the application notes:

Because the Government is in the best position to determine whether the defendant has assisted authorities in a manner that avoids preparing for trial, an adjustment under subsection (b) may only be granted upon a formal motion by the Government at the time of sentencing.

Id. cmt. n.6 (emphasis added).

Once again, I find myself in the minority, viewing the guideline as mandatory.  The plain text of the Guideline itself is in the imperative, follows the familiar "if . . . then" framework, and does not admit of discretion.  While the comment does sound a note of discretion with the use of may, it also notes that the Government, not the Court, is in the best position to exercise the discretion.  Further, guidelines trump comments in cases of inconsistency.  United States v. Stinson, 508 U.S. 36, 38 (1993). 

Finally, a mandatory reading does not eliminate judicial discretion for two reasons.  First, the judge determines whether section (a) applies, and thus whether one of section (b)'s prerequisites is satisfied.  Second, if the judge still disagrees with the adjustment, he can simply impose a variant sentence under the 3553(a) factors and tie the variance to what the offender's guidelines range would have been without the additional one-level downward adjustment.

UPDATE:  additional coverage from
Split Circuits: Third Circuit Declines to Weigh in on Split Re Whether Adjustment under Guideline 3E1.1(b) Is Mandatory under Certain Circumstances