Some Great Dicta about the Limits of the Almendarez-Torres Exception

United States v. Weisser, Docket No. 01-1588 (2d Cir. June 20, 2005) (Walker, Cardamone, and Owen, D.J.) (Op. by Walker): In this otherwise ho-hum opinion in which the primary appellate issue concerns the defendant's claim that he has been deprived of the right to appeal because much of the evidence presented against him at trial was destroyed during the September 11th attacks on the World Trade Center, the Circuit makes a very interesting statement about the limit of the Almendarez-Torres exception to the Apprendi-Blakely rule. The opinion is worth a quick perusal for that statement alone.

The essential facts are simple. Weisser lived in San Francisco and engaged in AOL chat sessions with someone claiming to be an 11-year-old boy in New York. The "boy" was of course an undercover agent fishing for customers (er, defendants). Much sexual conversation ensued, along with a planned liaison in a Big Apple hotel. To his unhappy surprise, Weisser found not a boy but a squad of federal agents in his hotel room.

Weisser went to trial and was convicted. At the pre-Blakely sentencing, the district court upwardly departed "horizontally", i.e., in the Criminal History Category, from III to VI. Finally, during the pendancy of Weisser's appeal, the 9/11 attacks occurred, which destroyed much of the Government's evidence against him at trial.

The primary claim on appeal was whether Weisser was deprived of his right to appeal by the destruction of the trial evidence. The Circuit ruled against him, concluding that (1) the appellant must demonstrate prejudice "before relief based on missing documents in the record can be granted," op. at 7; (2) prejudice "exists if the record is so deficient that it is impossible for the appellate court to determine if the district court has committed reversible error," op. at 9-10; and (3) Weisser had not adequately shown how his appeal was prejudiced by the destruction of the trial evidence.

Weisser also appealed various aspects of the sentence imposed, including the district court's three-Category horizontal departure from Category III to Category VI. In the course of evaluating the propriety of this departure, the Court offhandedly (and unnecessarily, given the Booker remedial ruling) states: "The district court's horizontal departure violated the Sixth Amendment because it was based on facts not found by the jury." Op. at 28 (emphasis added). This startling pronouncement -- after all, aren't facts underlying criminal history departures covered by the Almendarez-Torres exception to the Apprendi-Blakely rule? -- turns out to be somewhat less momentous than first appears. This is because the the factual bases for the horizontal departure in this case involved not the "fact" of a prior conviction or even the "nature" of a prior conviction, but three other kinds of facts concerning the defendant's criminal history. As a foonote explains, "the district court relied on (1) Weisser's repeated parole violations after his release from prison on a child molestation conviction; (2) his repeated failure to register as a sex offender, as required by law; and (3) his 'prior attempts to arrange sexual liaisons with other minors and his prior relationship with a teenage boy'" in departing to Category VI. Op. at 28, fn.10.

The 2nd and 3rd bases for the departure seem clearly outside the Almendarez-Torres exception. But the 1st basis seems quite close to the kind of facts that courts have routinely held to fall comfortably within that exception. To start, why should a defendant's "repeated parole violations" -- which of course are easily proved by court records -- be treated any differently than a defendant's "repeated convictions"? If the former cannot be relied upon to increase a statutory maximum unless admitted by the defendant or found by a jury (as this opinion states), neither should the latter. Even read more narrowly, moreover, the 1st basis is quite similar to the basis triggering the two criminal history points required under Section 4A1.1(d) of the Guidelines when the instant offense is committed while the defendant is on probation or parole for a prior offense. Finally, the 1st basis is also somewhat akin to a "fact" required for the application of the Armed Career Criminal Act -- the existence of three prior drug or violent felonies "committed on occasions different from one another". See 18 U.S.C. § 924(e). Of course, the Second Circuit previously held that such a fact falls squarely within the Almendarez-Torres exception. But if a court violates the Sixth Amendment when it increases a defendant's statutory maximum based on his repeated parole violations, why doesn't it also violate the Sixth Amendment when it does the same based on the fact that his 3 prior qualifying felonies were committed on different occasions?

In any event, if Shepard didn't hammer the point home clearly enough, Weisser should: Counsel should raise Sixth Amendment challenges to the use of any criminal history related fact to increase the statutory maximum. At worst, even if the Almendarez-Torres exception is not yet on its last legs, its scope is certainly an open question.

Batson’s Back, Baby!

Walker v. Girdrich, Docket No. 03-2645 (June 8, 2005) (Jacobs, Calabresi (C.J.J.) and Rakoff (D.J.)), (Op.by Jacobs).

Presaging the United States Supreme Court’s Miller-El decision by just five days, the Court of Appeals here rejected a state prosecutor’s reason for striking single black prospective juror, and granted the petitioner’s writ.

Facts

Tried in the New York Supreme Court, Kings County, on drug charges, Robert Walker lodged a
Batson objection after the prosecutor used twelve of thirteen peremptory challenges to remove African-Americans from the venire, alleging that the strikes were the product of purposeful discrimination. The trial judge concluded that there was no ‘pattern’ of discrimination, but invited the prosecutor to give her reason for striking the juror in question, Bernard Jones, who had not yet been discharged. The prosecutor indicated, inter alia, that Mr. Jones “gave one word answers” and was concerned about missing work, but noted that her “main ...problem[s]” with him were related to his race: “[T]his is an individual who was a Black man with no kids and no family.” She went out to point out that he had “ no experience whatsoever with police officers,” and that she perceived him as having “an attitude ... against a prosecutor.”

The trial judge accepted these comments as race-neutral, held that there was no “pattern” of purposeful discrimination, and denied the Batson objection. Interestingly, Mr. Jones, who apparently had been struck at a time when he would have been seated on the jury itself, ended up as an alternate, on consent, although he did not deliberate.

On appeal, Mr. Walker’s Batson claim got the typical perfunctory treatment from the New York State courts: a brief mention that it was “without merit” by the Appellate Division, and a denial of leave,without comment, by the New York Court of Appeals. On habeas review, Judge Weinstein denied the writ but granted a certificate of appealability.

The Court’s Holding

The Court of Appeals reversed and granted a conditional writ. After the usual preliminaries, including a concise recitation of the current law surrounding Batson claims, the Court had no trouble concluding that the A.D.A.’s striking of Mr. Jones “was not race neutral.” After all, she openly stated that one of her “main” concerns was Mr. Jones’ race, apparently aggravated by the fact that he had no family. The Court also astutely pointed out that some of her other reasons, such as Mr. Jones’ “one word answers” and his “attitude,” tended to “reinforce” rather than “dispel a race-based motive.” Lastly, the Court noted that the juror’s lack of experience with police would have favored keeping him on the panel, not removing him. Finding that the district court’s denial of the writ involved an unreasonable application of clearly established federal law, the Court reversed the denial of the writ.

The Real Significance

The Walker decision is significant for reasons other than its fairly obvious holding. The record in this case was notable for its lack of clarity. No record had been made of the composition of the jury pool up to the point of the Batson objection, which occurred during the fourth round of jury selection. Consequently, the Court was unable to determine whether the defendant might have been able to satisfy his preliminary burden of making out a prima facie case of purposeful discrimination. It got around this problem by noting that the striking of even a single veniremember for racial reasons violates Batson, and that the objection to the striking of Mr. Jones was sufficient.

But this case serves as a reminder of the importance of making an adequate record when registering aBatson objection, since it will be the rare case indeed where, as here, the prosecutor is stupid enough to give an overtly racial reason. Thus, the Court warns: “It is helpful in a Batson case to have a record as to the composition of the venire and the race and ethnicity of the jurors struck on either side. Moreover, the allocation of the burden for creating a record as to the prima facie case is unsettled; so it would be prudent for counsel to preserve a full opportunity for appeal by making a record for appeal.”

This case is significant for a second reason. Mr. Jones, the juror at issue, ended up on the jury, albeit as an alternate, but the Court still found a Batson violation.

Elderly Defendant Not Permitted to Withdraw Plea Despite Poor Advice from Attorney about Guidelines Exposure

United States v. Anthony Guidice, Docket No. 04-3729-cr (2d Cir. June 16, 2005) (Feinberg, Sack, Katzmann) (Op. by Feinberg): The Circuit sets a very high hurdle in this case for defendants wishing to withdraw their guilty pleas based on grossly erroneous advice from counsel about their sentencing exposure. Although the Court does not create any per se rules in this decision, it relies heavily on the fact that the defendant could not point to any "weaknesses" in the Government's case against him to conclude that, despite counsel's admittedly awful advice, there is no reasonably probability that defendant would have proceeded to trial if he had been given correct information about his sentencing exposure (despite his claims to the contrary). Unfortunately, many defendants will be unable to make such a showing.

The essential facts are simple. Guidice was charged in a 4-count indictment with extortion-related offenses. Shortly before trial, he pleaded guilty to a single count of conspiracy to commit extortion, pursuant to a plea agreement. Among other things, Guidice and the Government agreed that he faced a Guidelines range of 37 to 46 months. (This is of course a pre-Booker case; therefore, this was a "mandatory" range and not merely an advisory one).

Unfortunately, the Probation Office in preparing the PSR determined that Guidice qualified as a career offender under § 4B1.1, thus elevating his sentencing range to 151 to 188 months. When he learned of this, Guidice moved to withdraw his plea, claiming ineffective assistance of counsel. Guidice explained to the court that "I'm no angel, but twelve or more years [in prison] is a death sentence for me," and that, as a result, he would've taken his chances and gone to trial had he been told of the correct range by counsel. (Although the opinion does not disclose Guidice's precise age, he is apparently quite elderly and infirm -- so much so that the district court (Judge Cote, no less) departed downward on the basis of his age and poor health to an eventual sentence of 72 months.). The district court denied the motion, concluding that Guidice had not met the prejudice component of the Strickland because, among other things, he could not point to "any weaknesses in the government's case that would have influenced his decision to proceed to trial." Op. at 7.

The Circuit affirmed, largely adopting the district court's reasoning: Guidice could not meet his burden on the prejudice prong of Strickland because he did not demonstrate a "reasonable probability" that, but for the bad advice, he would have gone to trial. The Court dismissed Guidice's statements -- that given his age, he would've gone to trial if his sentencing exposure, even on a guilty plea, was 151 to 188 months -- as "self-serving". The Court also rejected Guidice's contention that the district court imposed an "unduly burdensome" task upon him -- i.e., the burden of showing weaknesses in the Government's case in order to demonstrate that he would have gone to trial but for his attorney's bad advice. The Circuit explained that the district court merely "suggested that evidence of weakness in the government's case might have been one means of proving that Guidice would have gone to trial had he known his true sentencing exposure." Op. at 17.

Criminal Forfeiture Not Governed by Requirements of Apprendi / Blakely Line of Cases

United States v. Fruchter, Docket No. 02-1422(L) (2d Cir. June 14, 2005) (Walker, Parker, and Wesley) (Op. by Walker): The Circuit concludes in this opinion that in determining the amount subject to forfeiture under the RICO criminal forfeiture statute, 18 U.S.C. § 1963, a judge need only rely upon the preponderance of the evidence standard. The Court specifically rejected appellant's argument that under Apprendi, Blakely, and Booker, the district court was required to determine the forfeiture amount under the beyond-a-reasonable-doubt standard. This was so because criminal forfeiture "is not a determinative scheme" (emphasis in original) -- like the Washington state regime invalidated in Blakely or the Sentencing Guidelines invalidated in Booker -- wherein a guilty verdict (or a guilty plea) "authorizes the imposition of a sentence within a specified [or determinate] range." Op. at 14. Rather, "criminal forfeiture is not subject to any statutory maximum," op. at 15, since the statute provides simply that the defendant shall forfeit "any property . . . derived from . . . any proceeds which the person obtained . . . from racketeering activity." Q.E.D.: "A judge cannot exceed his constitutional authority by imposing a punishment beyond the statutory maximum if there is no statutory maximum." Op. at 15.

Essentially, the reasoning is that because a guilty verdict (or a guilty plea) does not trigger a "previously specified range" in the forfeiture amount under § 1963, the Apprendi line of cases -- which "prohibit a judicial increase in punishment beyond a previously specified range" as authorized by a guilty verdict or plea -- is not applicable. See Op. at 15 ("Criminal forfeiture is, simply put, a different animal from determinate sentencing."). Rather, once the jury convicted Fruchter of RICO violations, the "maximum" forfeiture amount he became subject to was simply, as the statute provides, whatever proceeds the judge determined to be derived from his racketeering activities. The $20.7 million forfeiture order, in other words, was indeed "authorized" by the guilty verdict on the RICO counts. The Apprendi line of cases was therefore not implicated, and the preponderance standard is sufficient and the relevant findings need be made only by the judge. Indeed, the Court went on to uphold the district court's forfeiture order even though it included sums derived from counts on which the jury acquitted the defendant. Op. at 16-17.

A question naturally arises as to whether this case applies to restitution orders as well. Criminal forfeiture is not nearly as frequently encountered as restitution, and the governing statutes are of course distinct. One nonetheless wonders whether Fruchter means that restitution amounts also need not be found by the jury or on a beyond-a-reasonable-doubt standard -- a question that technically remains open in this Circuit . . . .

Drug Use Resulting in Revocation Can Be Premised, In Part, on Results Falling Below Cutoffs Established In Testing Companies' Contracts

United States v. Klimek, Docket No. 04-2459 (2d Cir. June 8, 2005) (Meskill, Newman, Cabranes) (Op. by Newman): In this case, briefed and argued by our own David Lewis, the Circuit ruled that, in the context of a supervised release revocation proceeding, a district court was not "per se precluded" from relying upon confirmatory drug test results that fall below the cutoff established by contracts between an outside testing company and the Adminstrative Officer of the United States Courts ("AO").

The facts are straigtforward. Mr. Klimek, after serving a prison sentence but while still on supervised release, used drugs. Initially, he admitted using LSD, pled guilty to violating the terms of his supervised release for that reason, and was sentenced to a term of home confinement. But even before home confinement could be set up, he tested positive again, this time for cocaine use. He denied using cocaine, so his urine sample was sent to an outside testing company for confirmation. The test, which used a gas chromatography/mass spectrometry technique (I couldn't describe it if I tried, but GC/MS is specifically approved in 18 U.S.C. section 3583(d)), detected cocaine metabolite at the level of 118 nanograms per milliliter. The contract between the testing company and the AO, however, required a cutoff of 150 nanograms per millileter in order to confirm a positive test. Mr. Klimek argued that a sub-cutoff test result precluded a finding of drug use.

The district court held a hearing at which a testing company representative testified. She testified that even a 118 nanograms per millileter finding could only be explained by cocaine use. She testified further that Mr. Klimek's urine had been diluted by his drinking copious amounts of water before submitting the sample. Once "normalized" to account for dilution, cocaine metabolite would have been present at over 400 nanograms per millileter. Based on this testimony and Mr. Klimek's admitted history of drug use, the district court concluded that he had, again, violated the terms of his supervised release.

On appeal, David Lewis argued that the district court was precluded from finding that Mr. Klimek ingested cocaine, given that the confirmatory test fell below the cutoff level specified in the contract between the AO and the testing company. His position was premised on two statutes, 18 U.S.C. sections 3583(d) and 3608, which govern drug testing for defendants who face imprisonment and deny the accuracy of initial tests (section 3583(d)), and require the Director of the AO to establish drug testing programs for Federal offenders on post-conviction release which include "standards" and "guidelines" deemed necessary to ensure accurate test results (section 3608).

The Panel, however, rejected defendant's argument. The Panel held that the district court's ruling was supported in this case based on the totality of the evidence -- specifically (1) the defendant's long and uncontroverted history of supervised release violations (why it is kosher to rely on this type of 404(b) propensity evidence is left unexplained), and (2) the fact that the confirmation test, once normalized for dilution, would have reflected the presence of cocaine metabolite far in excess of the cutoff amount.

The Panel declined to decide whether supervised release could be revoked solely on the basis of a confirmatory test result that falls below the cutoff level, so that argument should still be preserved.

New York Court of Appeals Upholds Persistent Felony Offender Law Despite Ring, Blakely, and Booker

People v. Rivera (June 9, 2005) (Op. by Rosenblatt): This Blog normally confines itself to Second Circuit decisions, but since we commented yesterday on the Second Circuit's narrow decision upholding New York's persistent felony offender statute on AEDPA review, see Brown v. Greiner, below, we thought it appropriate in the interest of completeness to discuss the New York Court of Appeals's decision, rendered today, upholding that law on direct review. (Our prediction of a contrary result was, alas, misguided). In Rivera, that Court ruled that N.Y. Penal Law § 70.10 did not violate either Apprendi, Ring, Blakely, or Booker. Of course, it did so only by interpreting the statue in a way that (1) flatly contradicts the plain language of the law itself, and (2) flouts years of case law and accepted practice in the New York courts.

Simply put, the Court upholds the statute by reading it to say that the two prior felony convictions "are both necessary and sufficient conditions for imposition of the authorized [enhanced] sentence for recidivism." Op. at 9; see op. at 6 ("[T]he prior felony convictions are the sole determinant of whether a defendant is subject to recidivist sentencing as a persistent felony offender.") (emphasis in original); op. at 14 ("[N]o additional fact-finding beyond the fact of two prior felony convictions is required under Penal Law § 70.10 or under Criminal Procedure Law 400.20.") (emphasis in original). And because a judge is permitted to find these "facts" under Almendarez-Torres, there is no Sixth Amendment problem with § 70.10. Q.E.D.

This is an incredibly strained reading of New York law. As Chief Judge Kaye points out in her dissent, "under the statute a defendant's classification as a persistent felon does not in and of itself subject the offender to enhanced punishment. . . . Rather, an enhanced sentence is available only for those who additionally are found to be of such history and character, and to have committed their criminal conduct under such circumstances, that extended incarceration and lifetime supervision will best serve the public interest." Dissent at 4. Indeed, under § 400.20, a court can impose the enhanced sentence only "based on fact findings made after a hearing at which the People have the burden of proof by a preponderance of the evidence." Id. at 6. Where an enhanced sentence was imposed without such additional fact findings, Judge Kaye notes, the Appellate Division "has repeatedly vacated persistent felony offender sentences." Id. at 7. Finally, Judge Kaye points out that § 70.10 -- until today -- has always stood "in stark contrast" with § 70.08, which "requires that all three-time violent felons be sentenced to an indeterminate life term on the basis of the prior convictions alone." Id. at 4-5.

(Kudos to Judge Kaye for one final, self-serving reason: She -- as our blog entry on Brown v. Greiner also points out -- notes that while § 70.10 (even correctly interpreted) was likely (or at least plausibly) constitutional in light of Apprendi alone, it was no longer valid in light of Ring. See dissent at 3 ("Thus, inasmuch as Walton was not overruled until Ring, the decision in Rosen neither was contrary to, nor involved an unreasonable application of, Apprendi at the time it was decided." (citing Brown v. Greiner)).

A Very Narrow Decision on New York's Persistent Felony Offender Law

Brown v. Greiner, Docket No. 03-2242(L) (2d Cir. June 3, 2005) (Walker, Leval, and Katzmann) (Op. by Leval): Too much has been made of this exceedingly narrow decision in the press (see, e.g., yesterday’s front page article in the New York Law Journal) and in the blogsphere (see, e.g., Professor Berman’s description of it as a “major habeas ruling”). Contrary to some accounts, this decision did not rule that New York’s persistent felony offender law comports with the Sixth Amendment as construed in the line of cases beginning with Apprendi (2000), running through Ring (2002) and Blakely (2004), and concluding (at least for now) with Booker (2005). Rather, this case holds merely that the New York courts’ determinations in 2000 and 2001 – when only Apprendi had been decided – that this law did not violate the Sixth Amendment was not an “unreasonable” reading of Apprendi, and thus that the writ should be denied under the AEDPA. This case says nothing about whether the state court decisions are correct in light of the post-Apprendi cases listed above. Indeed, it seems clear that the New York statute violates the rule of both Ring (2002) and Blakely (2004), even if it is not “unreasonable” to conclude otherwise vis-à-vis solely Apprendi.

The essentials are as follows. The Circuit denied the habeas writ to three § 2254 petitioners sentenced to extended prison terms under the New York persistent felony offender statute, N.Y. Penal Law § 70.10. The defendants each argued in state court that their extended sentences violated the Sixth Amendment as construed in Apprendi. The New York courts rejected this argument in 2000 and 2001 – after Apprendi but well before either Ring, Blakely, or Booker. In an opinion by Judge Leval, the Circuit concluded that the state courts’ determination that the persistent felon statute did not violate the Sixth Amendment was not an “unreasonable” interpretation of Apprendi, the only “clearly established” federal law “as determined by the Supreme Court” at the time of the state court decisions.

Section 70.10 requires two findings by the judge before she may impose the enhanced sentence: [1] the defendant must have been “previously convicted of two or more felonies,” and [2] the judge must be “of the opinion that the history and character of the defendant and the nature and circumstances of his criminal conduct indicate that extended incarceration and life-time supervision will best serve the public interest.” New York Penal Law § 70.10(2). The defendants’ core challenge was that finding [2] violates the Apprendi rule that “[o]ther than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt.” Apprendi v. New Jersey, 530 U.S. 466, 490 (2000). (Finding [1], of course, fits squarely in the Almendarez-Torres exception to the rule.). The state courts rejected this argument, and the defendants’ cases became final in 2000 and 2001.

The Circuit’s review of these decisions via the § 2254 petitions was necessarily deferential, restricted by AEPDA to upholding the decisions unless they were “contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court.” 28 U.S.C. § 2254(d). Crucially, the reasonableness vel non of the state court decisions was to be considered only in light of the Supreme Court cases in existence at the time the state courts issued their decisions. And as noted, only Apprendi had been decided at this time.

The Circuit accepted the State’s argument that “the New York courts did not apply Apprendi unreasonably in concluding that the sentencing judge may make the determinations upon which persistent felony defender sentencing turns.” (p. 14). This was so because the “amorphous” finding that a persistent felon sentence is required “to serve the public interest” differed from the kind of fact-finding at issue in Apprendi.

As Judge Gleeson pointed out in granting habeas to one of the petitioners (only to be overturned by the Circuit), however, § 70.10 is functionally “identical” to the capital sentencing scheme struck down in Ring v. Arizona. Brown v. Greiner, 258 F. Supp.2d 68, 92 (E.D.N.Y. 2003). In the Arizona scheme ruled unconstitutional by Ring, a sentence of death – though theoretically authorized by a jury verdict convicting a defendant of first-degree murder -- was unavailable without a judge finding at least one aggregating circumstance and the absence of sufficiently substantial mitigating circumstances. See Ring v. Arizona, 536 U.S. 584, 592 (2002). Similarly, a Class A-1 felony-length sentence is unavailable for defendants under § 70.10 without finding [2], also made by the judge. For Judge Gleeson, this identity was diagnostic of the unreasonableness of the state court decision upholding the extended sentence under § 70.10: “[B]oth statutes require judicial findings, after a hearing, of at least one aggravating factor before the enhanced sentence is available, [and] both violate Apprendi.” 258 F. Supp.2d at 92.

This seems true enough – and the Circuit did not necessarily disagree with Judge Gleeson’s point. Rather, the Circuit side-stepped it by emphasizing (in a critical footnote) that under AEDPA and in light of the timing of the state court decisions challenged in the § 2254 petitions, “[t]he universe of ‘clearly established Federal law, as determined by the Supreme Court of the United States’” contains only Apprendi and the cases preceding it. Accordingly, Ring (as well as Blakely and Booker) was irrelevant to the determination at hand. The Circuit decided that it was reasonable for the New York court to conclude that the “vague, amorphous assessment” involved in finding [2] was “something quite different from the precise finding of a specific fact” in Apprendi. (p. 19) The idea is that finding [1] (the determination of prior convictions) sets the Class A-1 felony ceiling on the sentence, and finding [2] involves a court’s “traditional role” in deciding what sentence to apply below that maximum. (p. 20)

The Court’s reasoning is curious. Why is the vagueness of a factual finding sufficient to exclude it from the Apprendi rule? After all, a judge is still required to make factual findings, after a hearing, before she can impose an enhanced sentence under § 70.10. One is left wanting a more persuasive argument, particularly in light of Judge Gleeson’s point that Ring ultimately foreclosed the state courts’ reasoning. A better argument would have been that the Apprendi majority had flatly rejected a Ring-type argument in Apprendi itself. Recall that prior to Apprendi, Walton v. Arizona had upheld the capital sentencing scheme later overturned in Ring, and that Apprendi explicitly left Walton untouched. Over Justice O’Connor’s protests, Apprendi, 530 U.S. at 538 (O’Connor, J., dissenting), the Apprendi majority distinguished “state capital sentencing schemes requiring judges . . . to find specific aggravating factors before imposing a sentence of death.” Id. at 496 (opinion of the Court). Given the similarity between the Arizona scheme and § 70.10, Apprendi’s refusal to overrule Walton is compelling evidence that the New York courts’ determination that § 70.10 did not violate the Sixth Amendment was not an “unreasonable” reading of Apprendi.

Once the universe of review expands to encompass Ring, of course, the New York courts’ application of Apprendi becomes much less reasonable. Indeed, § 70.10 seems clearly to violate the rule of Ring. Moreover, Judge Leval’s attempt to distinguish the New York law on the basis of the “amorphous[ness]” of the finding required for an enhanced sentence founders in light of Blakely, which ruled that a state judge’s equally “amorphous” finding that the defendant had acted “with deliberate cruelty” in committing the offense cannot increase the statutory maximum. Accordingly, this decision is limited to cases that became final before any of the post-Apprendi cases had been decided.

A final note: The New York Court of Appeals has a case pending on direct appeal that raises the identical issue of whether § 70.10 violates the Sixth Amendment. In deciding this question on direct appeal, that Court must of course apply all the post-Apprendi cases and need not defer to any prior decisions (as AEDPA required in Brown v. Greiner). And according to the Court’s public summary of the issue, the defendants wisely base their arguments on both Ring and Blakely. One hopes the Court of Appeals will abide by these decisions in striking down § 70.10 as a violation of the Sixth Amendment.

(By James Darrow, a rising 3-L at Stanford Law School, and Yuanchung Lee).

Circuit Grants Habeas Based on Trial Court's Refusal to Admit Exculpatory Testimony of a Defense Witness from a Previous Trial

Christie v. Hollins, Docket No. 03-2878 (2d Cir. May 27, 2005) (Newman, Pooler, Katzmann) (Op. by Newman): A fabulous week for habeas petitioners: This is the second time in one week that the Circuit has reversed a district court's denial of habeas and granted the writ. See Henry v. Poole, below. Here, Judge Newman concludes that the trial court's refusal to allow the defense to introduce a witness's exculpatory testimony from the defendant's first trial (at which the jury acquitted the defendant of the two more serious counts, and hung on the third count), despite the defendant's herculean efforts to ensure the presence of the witness at the second trial, violated the defendant's right to present a defense at trial. This was so even applying AEDPA deference: The state courts' determination that (1) the defense had not made diligent efforts to securethe witness's presence at trial, and (2) the absence of the testimony made no difference to the outcome, was an unreasonable application of clearly established federal law.

Circuit Excises Section 3553(b)(2) in Light of Booker, and Will Review Departure Decisions under Abuse of Discretion Standard

United States v. Selioutsky, Docket No. 04-2740 , (2d Cir. May 27, 2005) (Newman, Wesley, Hall) (Op. by Newman): This is yet another opinion by Judge Newman fleshing out some important details of the post-Booker sentencing (and appellate) world left unresolved by Booker itself. In this case, the Circuit (1) excises Section 3553(b)(2), which required mandatory application of the Guidelines, except for a few very narrow departure grounds, in child pornography and other sex offense cases; and (2) holds that it will review departure decisions by district courts under the pre-PROTECT Act standard of review -- abuse of discretion. Additionally, the Court clarifies that although Booker implements a general standard of reasonableness in evaluating the propriety of a sentence on appeal, the Court will review legal questions leading to the imposition of that sentence de novo, questions of fact under the clearly erroneous standard, and mixed questions of law and fact under either the de novo standard or the clearly erroneous standard depending on whether the question is more legal or more factual (citing Vasquez, 389 F.3d 65 (2d Cir. 2004)). Finally, the Court also continues its trend -- beginning with Rubenstein -- of evaluating whether the district court properly determined the advisory Guidelines range (including departures), and of remanding for resentencing when it determines that that range was improperly calculated (again including departures) -- even while paying lip service to the rule that a sentence may be upheld as reasonable even where the Guidelines range was improperly calculated. Nonetheless, the Court admitted that even where an error led to the imposition of the final sentence, it may affirm that sentence if the record shows that (1) the district court would have imposed the same sentence absent the error, and (2) the sentence itself is reasonable.

The essential facts are very simple. Defendant pled guilty to possessing child pornography and faced a Guidelines sentence of 60 months (because the range of 70 to 87 months exceeded the 5-year statutory maximum). He sought a downward departure at the pre-Booker sentencing on the basis of exceptional family circumstances -- defendant lived with and provided some unspecified financial support to his elderly parents, one of whom was scheduled to have surgery soon. Although defendant had a brother in Pennsylvania, the record does not indicate whether that brother would have assisted their elderly parents if defendant were incarcerated. The record also showed some uncertainty about whether defendant intended to move to Florida to rejoin his own family even if he were not incarcerated.

Despite these uncertainties and a relatively weak record, the district court departed downward and imposed a 30-month sentence. The Government appealed.

The first question faced by the Court was whether Section 3553(b)(2) survives Booker. This Section requires sentencing courts to impose a sentence within the Guidelines range in all child sex offense and other similar cases, absent facts warranting a departure specified by the Guidelines. Booker of course excised Section 3553(b)(1), which also generally required a district court to impose a sentence within the Guidelines range absent a departure, but did not address the continuing viability of its sister provision for the simple reason that neither Booker nor Fanfan had been convicted of a child sex offense.

The Circuit disposed of this question quickly, concluding that the same rationale that led to Section 3553(b)(1)'s excision in Booker -- i.e., mandatory Guidelines violate the Sixth Amendment -- yields the same result for Section 3553(b)(2). Importantly, the Circuit ruled that Section 3553(b)(2) was invalid in its entirety, and not simply in its requirement of a Guidelines sentence. Op. at footnote 6. Therefore, in determining whether to depart downward in a child sex offense case, a district court is no longer constrained by the narrow grounds specified in Section 3553(b)(2), and may depart on any ground justifying a departure in any other case.

The Court went on to discuss the applicable standard of review in evaluating the appropriateness of the downward departure. Although Booker excised Section 3742(e) and replaced it with a standard of "reasonableness," the Circuit explained (citing Crosby) that a sentence may be unreasonable not only as to its length, but also because of procedural errors leading to its imposition. And while the Circuit may overlook any such errors and proceed to the ultimate question of whether the sentence imposed is reasonable or unreasonable (regardless of any errors), it may also notice any such errors and remand for resentencing without such errors instead. As the Court explained, such errors "could render a sentence unreasonable under Booker."

The Court clarified that in performing its appellate function of reviewing such "procedural errors" in sentencing, it will be governed by the same standards that it used before Booker. Thus, legal questions will be reviewed de novo, factual determinations will be reviewed under the clearly erroneous standard, and mixed questions of law and fact will be evaluated under either the de novo standard or the clearly erroneous standard depending on the nature of the question presented (following the procedure outlined by Vasquez, cited above). And as for departure determinations, the Court ruled -- citing Koon for support -- that it will use the pre-PROTECT Act standard of review: Abuse of discretion.

Applying that standard, the Court concluded that the departure was not warranted and remanded the case for further fact finding. Specifically, the Court required the district court to determine on remand whether (1) defendant's brother would be available to care for the parents were defendant incarcerated, and (2) defendant actually intended to remain with the parents in Brooklyn (rather than move to Florida to join his own family) were he not incarcerated.

Finally, the Court addressed the possibility that the sentencing court may have imposed the same 30-month sentence even absent the erroneous departure as a "non-Guidelines" (or variance) sentence after Booker. See foonote 7. If the record supported such a finding on appeal, and if the Circuit determined that a 30-month sentence was not unreasonable, any error in departing would be deemed harmless. However, because the record did not clearly demonstrate that the court would have imposed the same sentence absent the departure (it was a pre-Booker sentencing, after all), remand was required.

Circuit Grants Habeas on IAC Claim, and Suggests Need for En Banc Review of Whether New York's IAC Standard Is "Contrary to" the Strickland Standard

Henry v. Poole, Docket No. 03-2884 (2d Cir. May 24, 2005) (Oakes, Kearse, Sack) (Op. by Kearse): This probably happens once in a blue moon: Judge Weinstein denies habeas, but the Circuit reverses and grants habeas. But, alas, that is what occurred in this case. The Circuit, by Judge Kearse, ruled that state trial counsel was ineffective for advancing a "fallacious" alibi defense at trial. The decision also has an interesting discussion on whether New York State's IAC standard, see People v. Benevento, 91 N.Y.2d 708, 697 N.E.2d 584 (1998), is "contrary to" the federal Strickland standard for purposes of AEDPA deference. Although the Circuit held in an earlier case that the two standards are not contrary to each other, see Lindstadt v. Keane, 239 F.3d 191, 198 (2d Cir. 2001), both Judge Kearse and Judge Sack, in a separate concurrence, apparently disagree with that earlier case and suggest that en banc review of the question may be required in an appropriate case. This case is not that case, however, because the Court ruled that habeas should be granted even applying AEDPA deference to the state court's decision on the IAC question.

The opinion is classic Judge Kearse -- very long & detailed. But the essential facts are as follows. Victim is a livery car driver who was robbed at gunpoint by several men shortly after midnight on August 10th. Three weeks later, petitioner Henry is arrested on an unrelated charge, and for unknown reasons put in a lineup for the earlier robbery. Victim IDs Henry the robber with the gun, even though Henry doesn't come close to fitting the discription given earlier by the victim. At trial, the sole evidence against Henry was the driver's shaky ID.

Defense counsel thoroughly challenged the victim's ID of Henry. Additionally, counsel promised to the jury that his client had an alibi, and then called the defendant's girlfriend to testify on this point. The girlfriend testified that the defendant was with her during the day and into the late evening of August 10th -- which is of course about 24 hours after the robbery. Although the prosecutor noticed this on cross, and demonstrated that the girlfriend's evidence did not actually consitute an alibi, defense counsel somehow did not. Counsel continued to characterize the girlfriend's testimony as an alibi during his summation.

The prosecutor exploited this mistake during his summation. After pointing out the obvious fact that the girlfriend's testimony was irrelevant to whether defendant robbed the victim shortly after midnight on August 10th, the prosecutor turned the defense's blunder against the defendant. He argued that this "false" testimony was fabricated by the defendant, and suggested that this was evidence of consciousnes of guilt. (This argument really does not make much sense: The girlfriend's testimony was not "false" in any way, but simply irrelevant. Her testimony would constitute "fallacious" alibi evidence only if she claimed to be with the defendant at the time of the robbery, and this claim proved to be inaccurate. Here, the girlfriend testified truthfully (or at least was not contradicted on this point) that she was with the defendant about 24 hours after the robbery. Again, this is not false, but simply irrelevant.). Defense counsel did not object to this line of argument.

The Circuit concluded that defense counsel's presentation of "fallacious" alibi evidence constituted ineffective assistance. The difficult point was the prejudice component of the Strickland test -- after all, if counsel did not even have to raise an alibi defense, how does his failed attempt to present alibi evidence harm the defendant? After all, counsel (concededly) attacked the victim's ID testimony aggressively and thoroughly, and this line of defense stood independent of the failed alibi defense. Judge Weinstein relied on this point in denying habeas. Nonetheless, Judge Kearse concluded that counsel's error created a reasonable probability of a different result, finding that "the State's weakened case was bolstered . . . by the false alibi evidence." The Court noted that the prosecutor "capitalized on counsel's reliance on the fallacious alibi evidence by arguing that the alibi was fabricated and suggesting that" this was evidence of consciousness of guilt by the defendant. The Court thus concluded that counsel's blunder in presenting "false" alibi evidence prejudiced his client in a case with weak evidence of his guilt. The Court also concluded that the New York Court of Appeals's contrary determination on this question was an "unreasonable application" of Strickland.

One suspects that the outcome of this case depended heavily on the serious doubt concerning Henry's guilt, which was established solely by very shaky ID testimony. Additionally, although the opinion purports to apply AEDPA deference to the New York court's determination that Henry had not demonstrated a Sixth Amendment violation, that is a bit hard to swallow. More likely, the Panel's clear belief that the New York IAC standard is in fact "contrary to" the Strickland IAC standard informed its purported application of AEDPA deference.

Second Circuit Criticizes Supreme Court Rationale in Hodari D.

U.S. v. Swindle, Docket No. 03-1773 (2d Cir. May 11, 2005) (Feinberg, Cardamone, and Parker)(Op. by Feinberg)

In a refreshingly frank decision, the Second Circuit upheld the denial of a suppression motion, finding that it was "constrained by relevant Supreme Court precedent" in affirming the legality of a seizure. In doing so, the Court was highly critical of that precedent -- California v. Hodari D., 499 U.S. 621 (1991) -- finding no "principled basis" for the decision.

The facts of Swindle are straightforward. While on patrol in an unmarked car in Buffalo, police were searching for a suspect whose physical description did not match that of the defendant, Swindle. While patrolling, they saw Swindle exit a "known drug house" that the suspect had supplied in the past, get in a car that resembled a car that the suspect had "been seen near" and drive away. The officers followed the car for a short time and then flashed their lights and attempted to pull him over. Swindle disobeyed the order, continued to drive, committed several traffic infractions, tossed a bag of crack cocaine outside the car, and fled on foot after eventually pulling over. The police caught him and charged him with possessing the crack.

The Court held that the police officer's initiation of the flashing lights constituted an order to stop. The Court further found that the police had no basis under Terry to initiate such a stop. Mr. Swindle did not match the description of the suspect they were looking for, he was not driving the same car as the suspect's (only a similar make to a car the suspect had "been seen near") and he had not committed any traffic infractions. In sum, "Swindle was simply a black man in a high-crime area driving a car that the wanted fugitive had previously been seen 'near." The Court was strongly critical of the officers conduct, calling it "a clear abuse of police authority." The Court thus had "no difficulty concluding that the officers acted unreasonably in ordering Swindle to pull over."

Unfortunately for Mr. Swindle, the Court felt constrained by Hodari D. in finding that the conduct to be evaluated for judging probable cause for a seizure includes the defendant's conduct after the unreasonable order to stop but before he was actually seized. Thus, even though the initial order to stop (by flashing police lights) was unreasonable and without cause, Mr. Swindle's subsequent conduct of violating traffic laws, tossing the drugs and fleeing on foot, did provide the officers with probable cause for the seizure. The Court did suggest, however, that Mr. Swindle might have a civil remedy for the abusive police conduct: "Although we are precluded from from holding that the officers' unreasonable order violated the Fourth Amendment, we believe that it was an abuse of authority for which Swindle and others like him might seek redress under a source of authority such as the Fourteenth Amendment or some provision of state law."

Mr. Swindle also argued that the initial order to stop was itself a seizure. While the Court was sympathetic to this argument, it once again felt constrained by Supreme Court precedent. In addition to Hodari D., the Court cited County of Sacramento v. Lewis, 523 U.S. 833 (1998) holding that "a police pursuit in attempting to seize a person does not amount to a 'seizure' withing the meaning of the Fourth Amendment." The Court was clearly troubled by this rule, noting that "a substantial argument could be made that a broader definition of 'seizure' -- or some other remedy -- is required to adequately protect Fourth Amendment values from the harms flowing from police initiation of Terry stops without reasonable suspicion."

The Court concluded with a final lament about the Hodari D. line of cases:

"Even if the kind of order given in Swindle's case is rare -- and we do not suggest that it is -- we see no persuasive reason for the law to tolerate it. In view of what we believe to be the controlling cases, however, we must affirm a conviction that was achieved with evidence obtained by an abuse of police power. A remedy for Swindle's Fourth Amendment complaint can come only from higher authority."

Whether that "higher authority" refers to SCOTUS or some other more-ethereal being, it would appear that Mr. Swindle's chances of ever receiving satisfaction from his Fourth Amendment claim are equally poor.

Posted by David Patton

A Second 2255 Petition Need Not Satisfy 2244's Stringent "Gatekeeping" Requirements If Filed Before the First Petition Has Become Final

Usama Whab v. United States, Docket No. 05-1214 (2d Cir. May 19, 2005) (Walker, Leval, Duplantier) (Op. by Leval): This is a surprisingly good, pro-defendant opinion -- in a case litigated by a pro se petitioner, no less. (Two cheers for the Court!) Essentially, the Circuit held that if a defendant files a second 2255 motion before his first 2255 petition has become "final" (i.e., before the time for seeking cert. on the first petition has run), the defendant need not satisfy 2244(b)(3)(A)'s rigorous requirements for "second or successive petitions" (including the requirement of prior approval from the Circuit before filing) . While the traditional "abuse of the writ" doctrine would still apply to such petitions, a petitioner need not satisfy 2244's gatekeeping requirements in such a situation.

The essential facts are as follows. After the Circuit affirmed his conviction on direct appeal, Whab filed his first 2255 petition in April 2004. The district court dismissed the petition, and denied as well as a COA. Whab then sought a COA from the Circuit.

While that request was pending, Whab filed a new application in the Circuit in March 2005, seeking permission under 2244 to file a second 2255 petition. While this new application was pending, the Circuit denied the COA request on the first petition, in April 2005.

This opinion deals with Whab's application to file a second 2255 petition. The Circuit ruled that Whab did not need to satisfy 2244's gatekeeping provision because that provision is triggered only when the subsequent petition is filed after the first petition has become final. And because Whab's first 2255 petition was still pending in the Circuit (in the form of the request for a COA) when he filed his second 2255 petition (in the form of the application for permission to file a second petition he filed in the Circuit), 2244's gatekeeping provision does not come into play. As the Court summarized, "so long as appellate proceedings following the district court's dismissal of the initial petition remain pending when a subsequent petition is filed, the subsequent petition does not come within AEDPA's gatekeeping provisions for 'second or successive' petitions." Op. at 4. The Court thus transferred Whab's second 2255 petition to the district court, and instructed the district court to consider the petition without subjecting it to 2244's rigorous standard. Op. at 6.

Finally, the Court explained in a lengthy footnote that its opinion should "not be misconstrued as providing a free pass to prisoners to file numerous petitions before an initially filed petition is finally adjudicated on the merits." Op. 6 n.2. This is so because even though 2244's gatekeeping requirements do not apply in such cases, the traditional "abuse of writ" doctrine does. Id. (citing McClesky v. Zant, 499 U.S. 467 (1991)).

This is true enough -- but only to a point. Section 2244's requirements are far more difficult to satisfy than the court-created abuse of writ standards. Thus, it may well behoove a defendant in such a situation to file a second 2255 before his first has become final.

Does a District Court Have the Authority to Use the Beyond-a-Reasonable-Doubt Standard at Sentencing after Booker?

United States v. Evelyn Gonzalez, Docket No. 04-1956-cr (2d Cir. May 3, 2005) (Meskill, Calabresi, Wesley) (Op. by Meskill): This odd case has already received a good deal of attention from Professor Berman and others. And rightly so, for it seems to suggest -- if only in dicta and only by silence -- that a district court has the authority, after Booker, to use the beyond-a-reasonable-doubt ("BRD") standard of proof to determine facts relevant to Guidelines enhancements.

The facts are simple. The defendant was charged with conspiracy to distribute 5 kilograms of cocaine. She was convicted at trial by a jury. However, in response to two questions, the jury specfically found that Ms. Gonzalez's offense did not involve either 5 kilograms or more, or 500 grams or more, of cocaine.

At the pre-Booker sentencing, the Government relied on Watts to argue that in determining the appropriate offense level, the district court should overlook the jury's verdict and find by a preponderance of the evidence that the defendant had conspired to distribute at least 5 kilograms of cocaine. The district court, "refusing to vitiate the jury's drug weight finding," Op. at 5, rejected this argument and determined -- consistent with the jury's verdict -- that the defendant had conspired to distribute at least 400 grams but less than 500 grams of cocaine. It did so despite "stat[ing] that the two kilograms of cocaine that Gonzalez's husband possessed when he was arrested was foreseeable to Gonzalez and could theoretically form the basis of its sentencing calculation." Id.

The defendant appealed (challenging her conviction & sentence on other grounds), but the Government did not. Regarding the sentence, the Panel of course ruled that a Crosby remand was appropriate. Op. at 10-11.

In addition, the Court rejected the defendant's very odd "rule of lenity" argument; she claimed that the "factual ambiguity" regarding the amount of drugs for which she should be held responsible should be resolved in her favor. The Court quickly rejected this argument, pointing out that the rule applies only to the resolution of ambiguous statutes (and Guidelines), not ambiguous facts.

Moreover, the Court pointed out that using such a rule to resolve factual ambiguities would "straightjacket a district court in exercising its authority -- that endures post-Booker -- to resolve disputed facts by a preponderance of the evidence when arriving at a Guidelines sentence." Op. at 13. The opinion essentially ends there.

Quite surprisingly, the opinion says nothing about the district court's apparent refusal to use the preponderance standard to determine the appropriate Guidelines range. As noted, the district court -- contrary to Watts -- clearly deferred to the jury's conclusion that the Government had not established BRD that the defendant was responsible for more than 500 grams of cocaine when it sentenced the defendant. And if the district court did not have the authority to disregard the preponderance standard in favor of the BRD standard, one would have expected the Circuit to say something about this. It did not, however.

The most logical reading of this case, then, is that while the district court has the "authority" to use the preponderance standard to resolve sentencing disputes, Op. at 13, it also has the authority to use the higher, BRD standard. This is only dicta, however, and only by implication. Nonetheless, the opinion should be of value to attorneys trying to convince a district court to use the BRD standard at sentencing.

A Blanket Order concerning Appeals with Unpreserved Booker Errors

Yesterday, April 29th, the Court issued the following blanket order, which applies to all appeals raising a Booker claim in which the error was not preserved (i.e., primarily cases in which the sentencing occurred before Blakely was decided in June 2004). It states, in is cryptic entirety,

"John M. Walker, Jr., Chief Judge of the United States Court of Appeals for the Second Circuit, today announced the following: In cases pending on direct review involving appeals of sentences imposed prior to United States v. Booker, 125 S. Ct. 738 (2005), in which the sentencing judge’s error in applying the Sentencing Guidelines mandatorily was not preserved for appellate review by an appropriate objection, the Court will be receptive to motions agreed to by all parties to the appeal to remand pursuant to United States v. Crosby, 397 F.3d 103 (2d Cir. 2005), without the need for briefing of the merits of the appeal. However, any panel to which such motions are presented has the prerogative to dispose of such motions as the panel deems appropriate."

I confess that I am not sure what this 2-sentence order means. On the one hand, the first sentence merely states the obvious: Haven't we all been seeking Crosby remands on consent rather than filing full merits briefs? On the other hand, the second sentence seems to take away somewhat what the first sentence gives: Any individual Panel hearing such consent motions may grant, deny, or do anything else it deems approrpiate to such motions. Huh?

Suggested interpretations are welcome.

The Guidelines Grind Continues, even after Booker

United States v. Maloney, Docket No. 03-1753 (2d Cir. April 28, 2005) (Jacobs, Pooler, Sotomayor) (Op. by Sotomayor): Some of us had hoped that Booker, rendering the Guidelines-derived range merely advisory, would put an end to the mind-numbing analyses of poorly drafted Guidelines provisions churned out by the Circuit each week. Such hopes had been increased by the Court's decision in Rubenstein (see Blog below), in which the Court explained that because reasonableness is now the end-all-and-be-all for determining whether a sentence will be upheld on appellate review (rather than the correct application of the Guidelines), and because whether a sentence is reasonable or not is not necessarily dependent on whether it flowed from a correctly calculated Guidelines range, the Court has the authority to overlook Guidelines disputes and simply affirm or vacate a sentence based on its reasonableness (or lack thereof).

In Rubenstein, the Court decided that it would in fact address & resolve the Guidelines dispute in that case because given the significance of that particular dispute (involving a 4-level enhancement), "the influence of [the erroneous imposition of the enhancement] is likely to be so pronounced that it could cause resentencing after remand to be unreasonable." Op. 19. The Court used the same justification in Capanelli to reach a Guidelines issue where a 5-level enhancement was in dispute. The lesson seemed to be that if the Guidelines dispute involved only a 1 or 2-level difference, the Court would skip the dispute and simply evaluate the sentence itself under Booker's reasonableness standard. Such an outcome would please many advocates tired of arguing over one Guidelines technicality after another.

More recently, however, the Court in Fagans (see Blog below) used a different "test" to determine whether Guidelines disputes would be resolved on appeal. That case explained that if the Guidelines dispute is easily resolved, then the Court would address it; but if the dispute were difficult, the Court may simply undertake a reasonableness analysis.

In Maloney, however, the Court undertakes neither the Rubenstein evaluation ("Is the Guidelines dispute significant?") nor the Fagans assessment ("Is the Guidelines dispute difficult to resolve?"). Rather, the Court says simply: "We reach the double counting issue notwithstanding the fact that the Guidelines are now only advisory, (citing Booker), because the district court on remand remains under an obligation to consider 'the sentence that would have been imposed under the Guidelines,' (citing Crosby)." Op. at 5. Oy. Does this mean that every Guidelines dispute must be resolved? One shudders to think.

In any event, the actual Guidelines dispute is simple: Does the 2-level enhancement under 2B1.1(b)(7)(C) (applicable where the conduct involved the violation of a prior judicial order) apply when the offense of conviction is 18 U.S.C. § 228, which punishes anyone who fails to "pay a support obligation with respect to a child who resides in another State ..." The answer is yes. Although this of course constitutes "double counting" in common parlance, it is not impermissible double counting because, as the Court ruled, the Guidelines intended for such double counting.

Circuit Enforces Appeal Waiver, Entered into before Blakely or Booker, to Dismiss Booker Challenge on Appeal

United States v. Morgan, Docket No. 03-1316 (2d Cir. April 27, 2005) (Straub, Pooler, Parker) (Op. by Parker): You win some, you lose some. A mixed bag for defendants from the Circuit today: a good decision in Fagans, and an awful one in Morgan. In a short opinion -- issued "on reconsideration" -- that is equally short on reasoning or analysis, the Circuit joined the 2 or 3 other circuits that have decided this issue and held that an appeal waiver, even if it was entered into before either Blakely or Booker was decided, will be enforced and thus will bar an appeal raising a Booker claim. This is the central passage:

"Morgan entered into his plea agreement [which included a waiver of his right to appeal any sentence within a stipulated range] after Apprendi v. New Jersey, 530 U.S. 466 (2000), but before Booker/Fanfan. Thus, there is every reason to assume that Morgan had knowledge of his Apprendi rights at the time he entered into the plea agreement. That Morgan did not, by contrast, have knowledge of his rights under Booker/Fanfan makes no material difference. His inability to foresee that subsequently decided cases would create new appeal issues does not supply a basis for failing to enforce an appeal waiver. On the contrary, the possibility of a favorable change in the law after a plea is simply one of the risks that accompanies pleas and plea agreements."

Op. at 3-4.

A Preserved Blakely / Booker Error Warrants Resentencing (and Not Merely a Crosby Remand)

United States v. Fagans, Docket No. 04-4845-cr (2d Cir. April 27, 2005) (Newman, Cabranes, Pooler) (Op. by Newman): This is the most recent of a long line of opinions authored by Judge Newman that discuss the application of Booker to cases pending on review. It answers an open question about whether a Crosby remand (i.e., a proceeding in which the district court decides whether to resentence) or whether a full resentencing is the appropriate remedy when the defendant raised a Blakely-based objection at sentencing. And the lesson is simple: If your client raised a Blakely objection to the Guidelines at sentencing (regardless of whether the objection was made in the form of a "Blakely-ized Guidelines" objections or in the form of a "Guidelines as a whole are unconstitutional" objection) , then you get a full resentencing and not merely a Crosby remand. Fagans also discusses a couple of other issues, including one related to criminal history.

(1) Primarily, Fagans holds that where the defendant at sentencing contended that certain Guidelines enhancements -- including those based on facts concerning his criminal history -- could not be applied in light of Blakely, a subsequent Booker claim is considered preserved on appeal. Because the standard of review is, therefore, not "plain error," Crosby's remedy -- the remand proceeding wherein the district court decides whether s/he would have imposed a materially different sentence had s/he been aware of the Booker regime of advisory Guidelines -- is inapplicable. Rather, a good ol' remand for resentencing -- with the defendant present and perhaps an updated PSR -- is required.

Two steps must be taken before reaching this conclusion, however, but Fagans somewhat mysteriously only specifically addresses one (and that merely in a footnote). The first is whether a Blakely argument, made at or before sentencing, that certain enhancements could not be applied because the facts underlying them were not proved beyond a reasonable doubt (or admitted by the defendant), but that the Guidelines are otherwise applicable, is sufficient to preserve a Booker claim, raised on appeal, that the district court erred in treating the Guidelines as mandatory rather than advisory. Here, it appears that the defendant never argued below for what we all know now as the Booker remedy. Nonetheless, the Court concluded that the error was preserved: "Although the Defendant's objection, based on Blakely, to the compulsory use of the Guidelines could be viewed as limited to preserving a Sixth Amendment objection, we think he sufficinetly alerted the District Court to his claim that it was unlawful to use the Guidelines in a compulsory manner." Op. at 4-5 n.1.

The other step that must be taken before reaching the conclusion of remand-for-resentencing is the harmless error standard. The mere fact that the defendant has preserved an error for appellate review doesn't automatically warrant vacatur of the judgment and remand for resentencing. Rather, such a result is required only if the Government cannot satisfy its burden of proving to the "reviewing court [that] , on the record as a whole, . . . that the error did not affect the district court’s selection of the sentence imposed." Williams v. United States, 503 U.S. 193, 203 (1992).

Fagans contains no discussion of harmless error, however. Rather, it simply concludes that because the error is preserved, remand for resentencing is required.

My 2 cents: Most likely, Fagans did not undertake a harmless error analysis because the Second Circuit had already concluded in earlier cases that it is impossible to determine whether a court that sentenced the defendant under the pre-Booker scheme would have imposed the same sentence had it been aware of Booker, e.g., Crosby, 397 F.3d at 117-118, and, therefore, that the Government could not possibly satisfy its burden under the harmless error standard. As Crosby explained, for instance:

"Without knowing whether a sentencing judge would have imposed a materially different sentence, under the circumstances existing at the time of the original sentence, under the circumstances existing at the time of the original sentence, if the judge had discharged his or her obligations under the post-Booker/Fanfan regime and counsel had availed themselves of their new opportunities to present relevant considerations, an appellate court will normally be unable to assess the significance of any error that might have been made. Perhaps in some cases an appellate court could make an educated guess as to the likely outcome of a remand, but that guess might be wrong, absent a clear indication at the original sentencing supporting the inference that the same sentence would have been imposed under the post-Booker/Fanfan regime."

Crosby, 397 F.3d at 117-118. Accord United States v. Williams, 399 F.3d 450, (2d Cir. 2005) ("[A]s a reviewing court[,] we do not know what the sentence would have been absent the error.").

(2) Fagans also discusses an issue touched upon by recent cases such as Rubenstein and Capanelli (see Blog entries below): Whether the Court should resolve Guidelines disputes even though it is remanding for resentencing anyway, or whether it should allow the district court to resolve these disputes in the first instance. Of course, either solution is plausible because while "[i]n many circumstances, an incorrect calculation of the applicable Guidelines range will taint not only a Guidelines sentence, ... but also a non-Guidelines sentence, which may have been explicitly selected with what was thought to be the applicable Guidelines range as a frame of reference," Fagans, Op. at 5, whether a sentence will ultimately be upheld as reasonable is not necessarily dependent on whether it was imposed pursuant to a correctly or incorrectly calculated Guidelines range, see Rubenstein.

In Rubenstein, the Court seemed inclined to resolve this question by evaluating the significance of the Guidelines dispute to the ultimate sentence. Thus, for instance, it decided to resolve a dispute concerning a 4-level enhancement in Rubenstein (and a 5-level enhancement in Capanelli) because "the influence of [the erroneous imposition of the 4-level enhancement] is likely to be so pronounced that it could cause resentencing after remand to be unreasonable." Op. 19. The implication is that more minor Guidelines disputes (for instance, one that involves whether the defendant falls in one or the other of two adjacent ranges) would not be resolved on appeal. In contrast, Fagans resolves this issue by examining how much work the appellate court might have to do, rather than the potential influence of the Guidelines dispute on the ultimate sentence. As the opinion explains,

"If the issue concerning the correctness of a Guidelines calculation is difficult, an appellate court might understandably prefer not to resolve the issue upon an initial appeal of a pre-Booker sentence and instead to remand ... for resentencing [or for a Crosby remand, depending on whether the error is preserved]. On the other hand, if the Guideline calculation issue is not difficult, it might often be preferable to adjudicate the calculation issue prompty so that subsequent sentencing proceedings will occur in light of a correct calculation."

Op. at 5-6. And because the Gudelines issues in Fagans's appeal were easily answered, the Court resolved them before remanding for resentencing.

(3) Finally, Fagans discusses an issue that has already drawn attention from Professor Berman's remarkably thorough and always up-to-date (really, how DOES he do it?) blog: the scope of the prior conviction exception to the Apprendi / Blakely rule, as announced in Almendarez-Torres. Professor Berman focuses on a passage in Fagans, in which the Court approved of the PSR's inclusion of 2 criminal history points based on the fact that the defendant was on probation when he committed the underlying offense: "While the exact scope of the phrase 'fact of a prior' conviction has yet to be determined, see Shepard v. United States, the conviction itself, and the type and length of a sentence imposed seems logically to fall within this exception." Op. at 7.

Yes, this language is quite broad, perhaps too broad in including not just "the conviction itself," but also "the type and length of a sentence" under the Almendarez-Torres umbrella (leaky though it is). But it is not really much to be excited about, given that the discussion simply concerns the calculation of the now-advisory Guidelines range and thus do not implicate any Sixth Amendment issues. Moreover, it is nothing new -- the Second Circuit has long included such peripheral facts concerning criminal history within the Almendarez-Torres exception. E.g., United States v. Santiago, 268 F.3d 151 (2d Cir. 2001). It will take a clear ruling from the Supreme Court before the Circuit changes its mind in this area.

Government's Breach of Plea Agreement Leads to Resentencing

United States v. Vaval, No. 04-121-cr (April 12, 2005)(Winter, Sotomayor, and Parker)(op. by Winter). In this opinion, the Court addressed two important issues regarding guilty plea practice – the sufficiency of the allocution and the government’s obligations under a plea agreement.

Facts
Troy Vaval and his confederates made arrangements to sell firearms to a confidential informant. When the CI showed up to close the deal, Vaval and the others robbed him, stealing the buy money, his wallet and jewelry, and his car.

Vaval pled guilty to one count of robbery of federal property with a dangerous weapon, pursuant to a plea agreement. The agreement listed the relevant statutory maxima, but had "N/A" next to the line for restitution. In addition, although the government retained the right to describe to the court Vaval’s criminal conduct, it agreed to "take no position concerning" where within the applicable guideline range Vaval should be sentenced, and promised not to seek an upward departure. The agreement excused the government from these provisions only if new information relevant to sentencing came to light or Vaval breached the agreement.

During the plea allocution, the court never advised Vaval that he faced a mandatory restitution order. Later, in response to a question from defense counsel, the government ratified its promises that it would not engage in sentencing advocacy.

At sentencing, however, the prosecutor became an advocate. First, he argued for a higher role adjustment than that contained in the plea agreement, claiming that the mistake was due to a misunderstanding of the law. The court rejected this, and adopted the range contained in the plea agreement. Then, when given a second chance to speak, the prosecutor (1) argued that criminal history category to which Vaval was assigned was too low, characterizing his history as "appalling,"(2) called the defendant’s own statement in mitigation "disingenuous," and(3) gave a lengthy, and most damning, description of the offense and Vaval’s conduct, asking the judge to "consider all of that" when deciding on the appropriate sentence.

The judge sentenced the defendant at the top of the range.

The Court’s Decision
Vaval’s first argument was that he was entitled to get his plea back because he was not advised by the court that he faced mandatory restitution. The Court of Appeals made short work of this, noting that the presentence report correctly indicated that full restitution was mandatory. Although Vaval could have moved to withdraw his plea when he saw that, he did not, which the Court viewed as a sure sign that he would still have gone ahead with the plea if he had been advised of the restitution requirement by the district judge. The Court thus concluded that the Rule 11 violation did not amount to plain error.

The Court was a good deal more sympathetic to his argument that the government had breached the plea agreement at sentencing. Although the agreement permitted the government to "advise" the court of information relevant to sentencing, including Vaval’s criminal conduct, here the prosecutor went further, for example in his highly negative characterizations of Vaval’s criminal history and contrition. Also of significance was the fact that the prosecutor’s comments came after the district court had already calculated the guideline range, leaving little doubt that the prosecutor’s true intent was either to obtain a departure or to influence the court’s decision about where to sentence Vaval within that range, both of which were clearly barred by the agreement.

The Court concluded by going into an detailed examination of the available remedies. There are some cases where no remedy is appropriate – for example, where the breach has previously been cured by specific performance, or where the breach iwas so minor that it did not cause the defendant to suffer any meaningful detriment.

Where the government has breached a pledge to refrain from sentencing advocacy, some remedy is required, and the defendant need not affirmatively show that his sentence was increased as a result of the breach. In this situation, the remand can be either for resentencing before a different judge or withdrawal of the plea depending on the circumstances.

In Vaval’s case, the Court concluded that a resentencing before a different judge was the appropriate remedy because it would fully cure the breach.

On Mail Fraud, Property and Cheap Gas

Oscar Porcelli v. United States, Docket No. 04-2000-pr (2d Cir. April 12, 2005): Those driving automobiles in the New York area in the early 1980's will remember Oscar Porcelli fondly as the owner of the Gaseteria chain of discount gas stations, an operation that was able to charge prices for gas significantly lower than all other outlets. The reason for that ability, it turned out, was that Porcelli simply failed to collect state sales taxes on the gas he sold while filing false sales tax returns. Although this conduct was at worst a misdemeanor under New York law, it led to Porcelli's conviction in Federal court of one count of Rackeetering (RICO) and numerous counts of mail fraud. It also led (alas!) to the demise of the Gaseteria chain and to a whole lot of post-conviction litigation.

On direct appeal in 1989, the Circuit recognized that it "pushed the law" to its outer limits to subject Porcelli to a RICO conviction for conduct that was at most a misdemeanor under state law, but affirmed the conviction nonetheless. Judge Newman dissented on the ground that there was no "deprivation" of property (or attempted deprivation) under McNally v. United States, 483 U.S. 350, since Porcelli never even tried to collect the sales taxes from his customers and never obtained or attempted to obtain property belonging to the state.

The Circuit upheld the conviction twice against subsequent post-conviction attacks based on developments in New York law, the first, a holding that the state had no property interest in checks for taxes due where it had not previously obtained possession of those checks, and the second a holding that taxes due the state were not property of the state prior to being remitted.

In this, his third post conviction proceeding, seeking a writ of coram nobis, Porcelli claimed that the U.S. Supreme Court decision in Scheidler v. NOW, 537 U.S. 393, holding that interference with the use of property was not the "obtaining" of property punishable by the Hobbs Act, had undercut his conviction. The Circuit rejected this contention rather quickly, simply holding that the mail fraud statute, unlike the Hobbs Act, punished "any scheme or artifice to defraud, or for obtaining money or property ...." Whether this is an adequate answer to Porcelli's complaint is hardly clear, however, since the crux of Porcelli's arguments has always been that not only did he not obtain, but that he never tried to obtain, the money due in sales taxes.

Indeed, the Court appeared to recognize the problem by going on to address Porcelli's complaint that his conviction was simply wrong because what the Circuit had previously found sufficient to uphold it -- that by deception he concealed the state's cause of action for unpaid taxes from it -- was not an obtaining or an attempt to obtain property from the state. (An analogous deprivation was, of course, not sufficient to be "obtaining" in Scheidler.) The Court, recognizing that this was an "attractive" argument, essentially rejected it as having come too late.

Following this last (or perhaps only most recent) Porcelli installment, it is a little difficult to tell what exactly the Court holds the gist of the property deprivation in the case to be. The Court appears to abandon its former reliance on Porcelli's concealment of a cause of action as the crux of the crime and to return to the notion that "taxes owed to governments .. [are] "property" within the meaning of the mail and wire fraud statutes." But, as Judge Newman explained originally, in Porcelli's case that does not seem to be applicable, since he did not obtain or try to obtain the state's tax revenues; he just passed on the savings to his customers. So, the Court's only real answer may be that it is "too late in history" to think clearly about Porcelli's case at last; it certainly fails to give any other clear justification of its jurisprudence as applied to Porcelli's case.

Some Choice Dicta about the Reasonableness Standard

United States v. Susan Godding, Docket No. 04-3643 (2d Cir. April 19, 2005) (Oakes, Kearse, Sack) (per curiam): Many of us are waiting for a definitive say from the Circuit about the meaning of Booker's reasonableness standard of review, beyond the generic statements in Crosby and Fleming that reasonableness is a "flexible" concept and that the Circuit will "exhibit restraint, not micromanagement" in performing this appellate function. This odd little case is, unfortunately, not that definitive statement. Nonetheless, the Circuit -- while ultimately simply remanding for a Crosby determination by the district court -- suggests that the sentence imposed, as well as some remarks made by the district court at the original sentencing, were unreasonable.

The opinion tells us very little about the offense or the defendant. Ms. Godding worked for a bank and, over a 5-year period, managed to embezzle over $366,000 from her employer. Although her Guidelines range was 24 to 30 months, the district court downwardly departed (pre-Booker) to a sentence of one day's imprisonment, followed by 5 years' supervised release (with the special condition that the first six months be spent in home detention). The opinion does not discuss what grounds the district court relied upon in departing.

The Government appealed, initially arguing that nothing in the record justified a departure. After Booker was decided, however, the Government moved for a Crosby remand.

The Circuit granted that motion. In so doing, however, the Court expressed strong disapproval of some remarks made by the district court at the sentencing. Specifically, while stating that it was not relying on this as a basis for departure, the district court criticized the victim bank for having so lax a system of security and internal control as to allow the defendant to commit her crime over a 5 year period. As the Circuit put it, "[w]hile the court noted that it was not considering the bank's failure in this regard as a factor relevant to a departure, and that it did not consider it a factor in sentencing, it expressed its view that the significance of the pilferage could be attributed to the bank's failure to act sooner." Op. at 5.

The Circuit was "most troubled" by this statement. Id. It warned "that consideration of such a factor and the conclusion that it, and not the defendant's volitional acts, rendered the sum embezzled in this case more significant than it otherwise would have been, would lead us to question the reasonableness of a non-guideline sentence." Id.

This critique was odd enough, given that the district court explicitly said that its criticism of the bank was not in any way a factor in its sentencing decision. (And doesn't the Circuit always take such statements by the district court at face value, no matter how ludicrous?) But the Circuit did not stop there, and went on to criticize the sentence simply for its brevity. As it stated, "Furthermore, we are more broadly concerned that the brevity of the term of imprisonment imposed by this sentence does not reflect the magnitude of the theft of nearly $366,000 over a five-year period." Op. at 5.

It is hard to draw a proper lesson from this decision -- other than that district courts should watch what they say. The interesting parts of the opinion are all dicta, and the opinion does not give us enough facts about the defendant or her crime to evaluate whether the sentence was appropriate.

A potentially troubling aspect of the opinion, though, is that the district court actually had a good point: Too often, the Guidelines' tunnel-vision focus on quantity (be it loss amount or drug weight) as the determining factor in sentencing is irrational. The district judge rightly criticized such a sentencing philosophy as too "simplistic," and correctly noted that it was "fallacious" to equate quantity with culpability. Op. at 4. This is the same criticism made by other judges and academics, including Judge Lynch in his very fine opinion in Emmenegger, 329 F. Supp.2d 416 (S.D.N.Y. 2004). As Judge Lynch put it in a case involving perhaps very similar facts as this one, the Guidelines often place

"undue weight on the amount of loss involved in the fraud. . . . To a considerable extent, the amount of loss caused by [a] crime is a kind of accident, depending as much on the diligence of the victim's security procedures as on Emmenegger's cupidity. Had Emmeenegger been caught sooner, he would have stolen less money; had he not been caught until later, he would surely have stolen more. Nothing about the offense indicates that Emmenegger set out to steal $300,000, no more and no less. Rather, he took advantage of his position to steal various amounts from time to time. The rough magnitude of the theft is relevant to sentencing, but the particular amount stolen is not as significant."

329 F. Supp.2d at 427 (emphasis added). And reading between the lines of Godding, it seems that this is exactly what the district judge was trying to say.

Of course, the Circuit did not directly attack such a criticism of the Guidelines; it only, technically, criticized the district court's "blame-the-victim" approach. But it is worrisome nonetheless.