United States v. Stewart, 597 F.3d 514

Lead Opinion

DENNIS JACOBS, Chief Judge, joined by RICHARD C. WESLEY and PETER W. HALL, Circuit Judges, concurring in the denial of rehearing in banc.

I concur in the decision of the Court to deny in banc rehearing in this case. But because I do so notwithstanding my agreement with the panel dissent, I owe an explanation.

I

With respect to Stewart’s sentence, the amended panel majority opinion identifies a single procedural error and remands so that the district court can have an opportunity to consider that error, and much else besides. The panel majority acknowledges the unaccountable gap between the offense committed and the sentence imposed — the sentence is said to be “strikingly low” — but believes that review for substantive reasonableness should follow after the district court has had an opportunity to address procedural error (the one identified by the panel majority, the ones detailed by Judge Walker, and the ones hypothesized by Judge Calabresi). This two-step is not announced as an inflexible sequence for all cases, which is to the good, because procedural error and substantive error are permeable concepts. But in this instance, I think postponing the consideration of substantive reasonableness was a mistake and a missed opportunity.

The single procedural error identified by the panel majority is the failure of the district court to decide whether Stewart committed perjury or otherwise obstructed justice. At the same time, nothing in the panel majority opinion — as amended — does or should preclude the district court from rethinking Stewart’s sentence and its component considerations from scratch. The panel majority seems to encourage that. So do I.

II

Judge Walker’s dissent identifies several salient additional procedural errors, and I agree that these additional errors should be addressed by the district court on remand. I cannot improve on Judge Walker’s anatomy of the case, and no purpose would be served by repetition here or by point-by-point endorsement. I will limit myself to three observations.

[A] The terrorism enhancement is the dominant sentencing consideration in this case. The district court erred in discounting it to zero.¹ That is an error both procedural and substantive in nature, highlighting one reason that the two-step sequencing of review for procedural and substantive error makes so little sense in this case.

Any discount based on the fortuitous lack of harm resulting from Stewart’s offense is error (whether procedural, substantive, or both). For the reasons set forth in Judge Walker’s dissent, I agree that injury and death can serve as aggravating factors in sentencing for the crime of material support to terrorism, but that the absence of injury and death cannot serve as mitigating factors.

[B] The district court did not decide whether Stewart abused her position of trust, or her special skills as a lawyer. The panel majority recognizes this omission and, without classifying it as procedural error, directs that on remand “[t]he district court should also consider whether Stewart’s conduct as a lawyer triggers the special-skill/abuse-of-trust enhancement under the Guidelines, see U.S.S.G. § 3B1.3, and reconsider the extent to which Stewart’s status as a lawyer affects the appropriate sentence.” Like Judge Walker, I do not believe that this direction goes far enough.

Judge Walker observed that Stewart’s violation of the Special Administrative Measures jeopardizes an accused’s right to counsel, among other rights. I offer a related concern, which underscores both the applicability of the special-skill/abuse-of-trust enhancement in this case and the seriousness of Stewart’s crime. The trust that Stewart betrayed was conferred upon her as a lawyer for the purpose of assuring that her client would have post-conviction access to counsel. That trust was reposed in her as an officer of the Court, notwithstanding the horrible security dangers that would result from betrayal. Her offense tends to erode judicial confidence that lawyers can be entrusted with national secrets — or (as in this case) with the means to trigger or promote a mass slaughter of innocents. Stewart’s misuse of her special skills and her abuse of trust thus transcend the effect in a single case. The defense of certain sensitive criminal cases and the prosecution of certain sensitive cases of constitutional tort are impaired unless counsel can draw upon a fund of confidence and trust, and Stewart’s offense has debased that currency. See Arar v. Ashcroft, 585 F.3d 559, 578 & n. 11 (2d Cir.2009) (in banc) (citing Stewart’s offense to demonstrate that “the undertakings of counsel” cannot “necessarily abate[ ]” the “risk” of “inadvertent or deliberate disclosure of information harmful to our own and other states”).

[C] The panel majority observes that the district court, in its consideration of the 18 U.S.C. § 3553(a) factors, “found that Stewart’s opportunity to repeat ‘the crimes [for] which she had been convicted will be nil’ because she ‘will lose her license to practice law’ and ‘will be forever separated from any contact with Sheikh Omar Abdel Rahman.’ ” The panel majority does not weigh in on this finding. But Judge Walker does: “This is wrong. One does not need a law license in order to materially support terrorism or to defraud the U.S. government.” I agree with Judge Walker.² One can assist terrorism in many ways, few of which require a license to practice law. While the abuse of her law license is a basis for enhancing Stewart’s sentence, its loss is not a basis for a reduction, at least with respect to the likelihood of recidivism. And as set forth below, Stewart is reported as having expressed the view that she would do what she did again, but “might handle it a little differently” to evade detection. So for her, supporting and promoting terrorism remains acceptable; everything else (law license or not) is technique.

Ill

Notwithstanding the foregoing discussion, I have voted against in banc review at this juncture for the following reasons.

[A] The panel majority opinion makes no law with which I disagree. It identifies one procedural error, which I agree is an error; it encourages the district court to consider the errors identified by Judge Walker, as I do; it declines to reach substantive error without, however, purporting to bind other panels to do the same. In my view, the panel majority opinion is a missed opportunity, and fails to give the district court sufficient guidance. But it does not make law for other cases; it scarcely makes law of the case.

[B] This appeal was under consideration by the panel for two full years; additional lengthy delay would be an institutional disservice. The district court docket reflects that the process of resentencing has been put in motion so that it can take place with conscientious speed following tile issuance of our mandate. Stewart and her family are entitled to know what her sentence ultimately will be within the half-decade following her conviction.

[C]When our remand “effectively undoes the entire ‘knot of calculation’ ” that fixed the original sentence, “ ‘the spirit of the mandate’ requires de novo sentencing.” United States v. Rigas, 583 F.3d 108, 118 (2d Cir.2009). Following such a remand, the “district court [i]s required to resentence [a defendant] in light of the circumstances as they st[and] at the time of [ ] resentencing.” Werber v. United States, 149 F.3d 172, 178 (2d Cir.1998). Since circumstances have changed in the prolonged interval following imposition of the original sentence, the district court should have the opportunity to consider any developments in the first instance.

Media reports (which require skeptical vetting) reflect that Stewart has promoted her criminal conduct as a matter of principle and as an aspirational norm of ethical law practice. After her conviction and sentencing, Stewart participated in a law school conference, “Legal Ethics: Lawyering at the Edge, Unpopular Clients, Difficult Cases, Zealous Advocates.” It attracted “a standing-room crowd of 150 people, most of them law students”:

[Stewart] admitted to having been “cavalier” in the way she followed certain regulations governing communications with her client, but argued that the human bond between a lawyer and client is critical to the lawyer’s role as legal adviser.

“I was representing a client, and I would do it again, but I would do it in a way that would better insulate me,” she said. Her main regret was having been unaware that the government was secretly taping her conversations with Mr. Rahman, she said.

Paul Vitello, Hofstra Polite as Lawyer Guilty in Terror Case Talks on Ethics, N.Y. Times, October 17, 2007, at B3 (correction appended) (“N.Y. Times article”). If accurately reported, these comments call into doubt the district court’s finding that Stewart’s offense is an aberration in an otherwise admirable career. In addition, these comments seek to corrupt the young by enlisting law students in the project of degrading legal practice.³

Moreover, on the day before she was remanded to prison, Stewart gave a radio interview in which she was asked: “[Wjould you do anything differently today, would you do anything differently back then, if you knew what you kn[o]w today?” Stewart responded, in part: “I would do it again. I might handle it a little differently, but I would do it again.” Interview by Amy Goodman, Host, Democracy Now!, with Lynne Stewart, in New York, N.Y. (Nov. 18, 2009), available at http://www.democracynow.org/2009/ll/18/ exclusive_civil_rights_attorney_lyime_ stewart.

I am not in a position to make findings on these points, but in my view the district court should consider these additional circumstances at Stewart’s resentencing.⁴ Indeed, section 3553(a) requires the district court to “impose a sentence, sufficient but not greater than necessary, to comply with [enumerated] purposes” including “the need for the sentence imposed ... to promote respect for the law.” 18 U.S.C. § 3553(a)(2)(A). Stewart’s comments— that she readily would repeat her offense, taking care only to evade detection — also illuminate “the nature and circumstances of the offense” and “the history and characteristics of the defendant.” Id. § 3553(a)(1). On remand, the district court may wish to explore and consider Stewart’s post-conviction and post-sentencing comments as part of the section 3553(a) analysis.

[D] Finally, the issue of substantive reasonableness may be obviated or mitigated following the district court’s correction of the procedural error identified by the panel majority, the district court’s consideration of the procedural and substantive errors identified by Judge Walker (and underscored in part in this opinion and in the accompanying opinion dissenting from the denial of rehearing in banc), and the district court’s findings as to subsequent developments. In any event, these matters will affect the analysis. And this Court may have an opportunity after remand to reach issues that are not decided by the panel majority.

Notes

1 . The panel majority opinion states: “Whether or not the district court applied the terrorism enhancement to Stewart in its Guidelines calculation may be subject to disagreement.” However, it is clear that the district court applied the terrorism enhancement in its initial Guidelines calculation; it determined a total offense level of 41, a criminal history category of VI, and a Guidelines sentence of 360 months. It is also clear that the district court later dissipated the terrorism enhancement based primarily on (i) the lack of harm resulting from Stewart's offense and (ii) the notion of atypicality. The district court thereby erred (whether procedurally, substantively, or both), when it effectively eliminated the terrorism enhancement based on considerations that seem highly dubious for the reasons forcefully stated in Judge Walker's dissent. Cf. United States v. Ressam, 593 F.3d 1095 (9th Cir.2010) (in a terrorism case, r-emanding as procedurally defective a 22-year sentence on the ground, inter alia, that the district court was insufficiently cognizant of the Guidelines range of 65 years to life).

2 . Based in part on the district court's finding regarding recidivism, Judge Walker identifies procedural error in the district court's overvaluation of Stewart’s personal characteristics in the context of the section 3553(a) analysis: "In sum, though we will rarely identify procedural error in the weight a sentencing judge assigns to relevant factors, this is one of those rare cases where the record of a defendant's personal characteristics simply cannot bear the weight necessary to support the challenged sentence.” In footnote two of the panel concurrence, Judge Calabresi observes that "the amount of weight a particular factor can bear” is "normally considered [a] substantive judgment!].” I express no view as to whether the error identified by Judge Walker regarding the loss of Stewart's law license is procedural or substantive. But error it is; and this factor illustrates my earlier point that the two kinds of error are permeable.

3 . The law school put out that Stewart was invited to the conference as a “unique case study” rather than as “an expert in ethics.” Hofstra Law, http://law.hofstra.edu/NewsAnd Events/Conferences/EthicsConference/2007/ index.html (last visited Jan. 11, 2010). That distinction was evidently lost on her. In fairness to the students, her reception was said to be cool, and Nathan Samuel, then a second-year law student, walked out midway through Stewart’s address, returning only to pose an aggressive question. See N.Y. Times article.

4 . Stewart has thus repeatedly affirmed that she would do it again. So much for Judge Calabresi's inventive idea that the district court might consider on remand that the terrorism enhancement should be abated for offenses committed before September 11, 2001, when the enormity of terrorism may not have been fully appreciated. Since the panel majority invites the district court to consider this possibility, it is worth a footnote to show that Stewart herself has not had an epiphany. Indeed, it is hard to identify anyone in this country who would have discounted terrorism after the Beirut barracks, Khobar Towers, the embassy bombings, the 1993 World Trade Center attack, Oklahoma City, the U.S.S. Cole, and the conspiracy to blow up the New York Federal Building and the tunnels into Manhattan.

1 . The maximum sentence available is 180 months per MS count. See 18 U.S.C. §§ 2339A(a), 2339B(a)(l). We assume for purposes of this opinion that the terms were ordered to be served concurrently.

---

Concurrence Opinion

POOLER, Circuit Judge,

concurring in the denial of rehearing en banc:¹

Respectfully, I cannot join in Chief Judge Jacobs’ concurrence in this case as I believe it mistakenly asks the district court to apply the panel’s dissenting opinion, rather than the panel’s majority opinion. The nostra sponte en banc poll, predicated on the rationale set forth in the dissent, did not succeed. The majority opinion therefore stands. As pointed out in the majority opinion, the district court should, of course, take account of the panel dissent. But the decision of the panel is the law of the Circuit for this case on remand and for future cases, unless and until it is overruled by the Supreme Court or by this Court en banc.

Opinions dissenting from denial of rehearing en banc² are not uncommon in this Circuit. They are nonetheless oddities. When such an opinion is filed, there is an extant panel decision resolving the appeal. The active judges declined to revisit that decision en banc. The panel decision is therefore the Court’s decision. Other judges may have views on the matter, but the case is not before them, and what they may say about it has as much force of law as if those views were published in a letter to the editor of their favorite local newspaper.

Yet the unsuccessful request for an en banc rehearing becomes an occasion for any active judge who disagrees with the panel to express a view on the case even though not called upon to decide it. By employing the simple tactic of calling for an en banc poll, active judges provide themselves with an opportunity to opine on a case that was never before them. This amounts to an exercise in free speech rather than an exercise of any judicial function.

Here, the ability of active judges to seek en banc review and then opine on a case on which they did not sit works particular mischief. Once the mandate issues, the panel will remand for resentencing, with an opinion that actively encourages the district court to pay due regard to the panel dissent. It strikes me as inappropriate for other members of the Court to add their views as to what the district court should do on remand. This case may return to this Court on a subsequent appeal. Judges who may rule on a case in the future should not express their views on the outcome of a future appeal before the district court even considers the issues on remand.

The panel’s majority opinion, as commented on by the panel dissent, provides the district court with ample guidance from this Court, rendering further advisory opinions from others unnecessary. Both Gall v. United States, 552 U.S. 38, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007), and Kimbrough v. United States, 552 U.S. 85, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007), stand squarely for the proposition that the district court is in the best position to make decisions about sentencing. We follow the teachings of those cases best when we defer to the district court’s ample discretion. Having remanded to the district court, we should afford it an opportunity to undertake its task without further muddying of the waters.

1 . Judges Guido Calabresi and Robert D. Sack, both members of the panel majority, are now senior judges. As such, neither voted in the en banc poll nor can be expected to respond to the views expressed by non-panel judges.

2 . Chief Judge Jacobs "concurring” opinion is, in substance, a dissent from a portion of the views on the merits of the panel majority.

---

Dissenting Opinion

JOSÉ A. CABRANES, Circuit Judge,

dissenting, with whom Judge REENA RAGGI concurs:

I respectfully dissent from the order denying rehearing en banc on the matter of the sentence imposed on defendant Lynne Stewart.¹ In my view, en banc rehearing was necessary to finish the job the panel majority started — to decide, not simply to identify and pass over, the legal issues so vividly presented in the sentencing of Stewart.²

I do not necessarily disagree with the opinion of Chief Judge Jacobs concurring in the denial of rehearing en banc, at least insofar as he usefully identifies some of the issues that the panel majority has avoided or not decided. But I respectfully dissent from the denial of en banc review because I think that the en banc Court should have decided the overlooked issues at this time. Judge Jacobs believes that we should delay a rehearing on the ground that “this Court may have an opportunity after remand to reach issues that are not decided by the panel majority.” Jacobs, J. Op. at 519. We may not, however, have such an opportunity, as we do not know whether either party will deem it appropriate to appeal the sentence imposed by the District Court after remand. Judge Jacobs also votes against rehearing en banc because the “panel majority opinion makes no law with which [he] disagree^].” Id. at 517. But the problem is not the issues that the panel majority does decide; it is the issues that the panel majority does not decide that require rehearing en banc. In putting off a decision for a speculative “second appeal,” the panel and the en banc Court failed to perform their duty to the Bench and Bar to decide the important issues presented to them and thereby clarify the law of our Circuit.

As it stands, the en banc poll was defeated, and the case will return to the District Court for re-sentencing. In the event that the case does return to this Court for a second appeal, I wish to point out some of the critically important issues that the panel majority failed to decide. I do not provide any detailed discussion of what outcome the panel should have reached on these issues; rather, I merely highlight the matters that the panel neglected, that the en banc Court should have decided, and, therefore, that remain open for possible decision in the future.

I. The Panel Majority’s Means of Avoidance of Sentencing Issues Squarely Presented

Stewart, a member of the legal profession, was convicted of numerous charges, including providing material support to terrorists,³ specifically by facilitating eommunieation by the notorious incarcerated terrorist Sheik Omar Abdel Rahman to his legion of worldwide followers. To understand the seriousness of Stewart’s crime, it is important to understand that Rahman was convicted in the Southern District of New York of leading an extraordinary terrorist conspiracy intent on murdering Egyptian President Hosni Mubarak and bombing tunnels and buildings in New York City, including the 1993 bombing of the World Trade Center. See United States v. Rahman, 189 F.3d 88 (2d Cir.1999). His conviction was secured only after a lengthy investigation and nine-month trial that posed a sufficiently serious threat to the life of the trial judge that, for years, the judge and his family had to endure an around-the-clock protective detail. Rahman was sentenced to life imprisonment and designated to a maximum security federal prison, undoubtedly to provide the strongest possible deterrent, short of death, to his further pursuit of terrorism. As Rahman’s attorney, Stewart was one of the few people permitted to have contact with him in prison and, even then, only subject to certain conditions. See United States v. Stewart, 590 F.3d 93, 163-64 (2d Cir.2009) (“Stewart II”) (Walker, J., concurring in part and dissenting in part) (detailing conditions under which Stewart was given access to Rahman). Stewart abused this position of trust to defeat the very purpose of Rah-man’s prosecution and incarceration, providing him with the material support most important to his ability to continue to pursue terrorist objectives — namely, a means to communicate those objectives to his followers. The Sentencing Guidelines, which strive to identify the sentences generally imposed by judges around the country for crimes falling within the heartland of a particular offense, see generally Kimbrough v. United States, 552 U.S. 85, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007), yielded a recommended sentence for Stewart of a term of imprisonment of 360 months. Instead, the District Court sentenced Stewart to just 28 months, less than one-tenth of the advisory sentence. See Stewart II, 590 F.3d at 93, 163 (Walker, J., concurring in part and dissenting in part).

The unreasonableness of this sentence for a crime whose ultimate object — -terrorism — threatens countless innocent lives, would appear obvious.⁴ In reviewing this sentence, the panel majority was confronted with several questions regarding reasonableness that are of vital importance not only to terrorism cases in particular but to sentencing proceedings in general. Reading the panel majority’s opinion, however, one finds no answers to those questions. One finds, instead, a three-part technique in which the panel avoids making law and, as a result, remands without identifying the full scope of the District Court’s sentencing errors.

In Step One, the panel majority declares uncertainty about a particular legal question but nevertheless offers speculative and general observations that appear to state the relevant law (this is entirely dicta, of course, in view of the panel’s stated reluctance to decide the issue presented). In Step Two, the panel majority decides to remand to the district court for “clarification” of its view on the matters at issue in light of the panel’s tour of the legal horizon. In Step Three, the panel directs that any later appeals in the case be referred to the original panel.

By remanding for “clarification,”⁵ the panel majority not only delays its decision on the sentencing questions clearly presented but also casts those questions as unripe for appellate review, thereby effectively insulating the panel majority’s opinion from review by the en banc Court or by the Supreme Court. And in retaining control over any subsequent appeal, the panel majority ensures that, if the questions are ever answered in this Court, the panel will be the one answering them (at least in the first instance).

II. The Sentencing Issues Left Undecided

I hasten to emphasize that this is not a situation in which a court has skirted a few minor issues or avoided reaching a single, difficult question that is not squarely presented for decision. Despite a record of thousands of pages and more than twenty-one months of deliberations on appeal, the original November 17, 2009 opinion of the panel majority identified only one error (a failure to consider Stewart’s perjury) and either ignored altogether or skimmed over at least five sentencing issues of paramount importance. See United States v. Stewart, No. 06-5015-cr(L), 2009 WL 3818860, Slip Op. 7525, 7623 (2d Cir. Nov. 17, 2009) (“Stewart I ”). On December 23, 2009, the panel majority filed an amended opinion that briefly identified some of these issues but still left them undecided.⁶ See Appendix A. Each of these issues, which I summarize below, should have been taken up by the en banc Court in light of the panel majority’s reluctance or inability to decide them.

A. The Reasonableness of Stewart’s Sentence

Most notably, the panel majority declined to assess the substantive reasonableness of Stewart’s 28-month sentence in light of the seriousness of her crime of conviction. The panel majority did finally say, in its amended opinion, that “Stewart’s sentence is strikingly low in light of what the district court correctly described as the ‘irreducible core of [her] extraordinarily severe criminal conduct.’ ” Stewart II, 590 F.3d at 143 (Maj. Op.) (quoting Sent’g Tr. 118). But despite its belated glancing observation that it had “serious doubts” about the sentence’s reasonableness, the panel majority even in its second (amended) opinion could manage to say no more than that it “th[ought] it appropriate to hear from the district court further before deciding the issue.” Id. at 151. In fact, there is no reason to hear further from the District Court. The record makes clear that Stewart’s sentence was carefully considered and fully explained by the District Court. Indeed, both in its original November 17, 2009 opinion and in its amended December 23, 2009 opinion, the panel went well out of its way to celebrate the presiding judge in the District Court for his meticulousness and attention to detail — “We would be remiss if we did not, at the outset, commend the district court for its thoroughness, thoughtfulness, and effectiveness in the conduct of these unusually lengthy, difficult, and sensitive proceedings. Much of what follows simply reports what it did and tracks what it said.” Stewart I, Slip Op. 7527-28; Stewart II, 590 F.3d at 98-99. The District Court clearly decided that the lack of harm attending Stewart’s support and various circumstances relating to her career and health supported a minimal sentence. See Sent’g Tr at 113-14, passim. In these circumstances, it is the responsibility of an appellate court to decide whether, on the “thorough[ ]” record presented, Stewart I, Slip Op. 7527, the identified factors could bear the extraordinary mitigating weight assigned to them by the District Court. See United States v. Cavera, 550 F.3d 180, 191 (2d Cir.2008) (en banc).

Observing all of this, one might be drawn to the conclusion that what the majority really hopes to “hear” from the District Court after remand is not a further explanation for the sentence imposed, but rather, the pronouncement of a sentence sufficiently higher than the original so that the important issues relating to the mitigation of terrorism crimes can be avoided. But an appellate court does not identify significant sentencing error by winks and nods that it hopes the district court will understand and act on when correcting less significant errors identified as the basis for remand. In any event, in declining to decide the issues before it, the panel majority missed a rare opportunity to clarify the law of sentencing. Indeed, if there ever was a case that afforded an opportunity to further develop the “abuse of discretion” and “shocks the conscience” standard, see United States v. Rigas, 583 F.3d 108, 123 (2d Cir.2009), it was this case, where the District Court sentenced to only 28 months in prison a member of the bar who aided a particularly nefarious and notorious terrorist to continue pursuing his deadly objectives.⁷

The panel majority declined to review the “substantive” reasonableness of Stewart’s sentence apparently based on a notion that this issue could not be reached on a first appeal. But there is no reason for sentencing review to require two or more appeals. The panel majority — and the en banc Court — did have the ability to review for both “procedural” and “substantive” reasonableness, on this appeal. See, e.g., United States v. Ressam, 593 F.3d 1095 (9th Cir.2010) (finding procedural and substantive unreasonableness in the sentence imposed on a convicted terrorist, and noting that the district court’s errors and sparse record made appellate review of substantive reasonableness difficult, but nonetheless identifying the sentence as substantively unreasonable and providing the district court with numerous citations to instances of substantive unreasonableness for guidance on remand). To suggest, as the panel majority did, that courts of appeals are effectively required to undertake a two-step appellate review whenever any sentencing issue is identified as “procedural” is to muddle the law of our Circuit and to promote a rigid and unworkable distinction between “procedural” and “substantive” issues.⁸

Under the approach taken by the panel majority, the presence of any “procedural” errors in a sentence would necessitate a remand to the district court and then, perhaps, a second- — maybe even a third, or fourth? — appeal to excise each “procedural” error before the appellate court can (finally, at long last) review for “substantive” reasonableness. Indeed, the panel majority would appear to suggest a second appeal for substantive unreasonableness even in the circumstance where an appellate panel finds procedural error and where correction of that error would not result in a significant sentence modification, but the panel has “serious doubt” about the sentence’s substantive reasonableness.

As it happens, there is no definitive ruling by the Supreme Court or our Circuit that requires any such elaborate, wasteful, and time-consuming process. In Gall v. United States, the Supreme Court explained that an appellate court “must first ensure that the district court committed no significant procedural error ... [and,] [assuming that the district court’s sentencing decision is procedurally sound, the appellate court should then consider the substantive reasonableness of the sentence imposed.” 552 U.S. 38, 51, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007). The Supreme Court did not, however, require a paso doble of sentencing. Gall states only that if a sentence is free from “procedural” errors, then the appellate court should proceed to consider the “substantive” reasonableness of the sentence imposed. There is no basis for suggesting that Gall states the converse — that if a sentence is procedurally unreasonable a court should not proceed to consider the substantive unreasonableness of that sentence. See, e.g., Ressam, 593 F.3d at 1130-31 (“The procedural errors identified in the district court’s decision rendered the sentence imposed on Ressam both procedurally and substantively unreasonable.”).

In any event, in United States v. Cavera we stated: “Where we find significant procedural error, one proper course would be to remand to the district court so that it can either explain what it was trying to do, or correct its mistake and exercise its discretion anew, rather than for the appellate court to proceed to review the sentence for substantive reasonableness.” 550 F.3d at 190 (emphasis added; citation omitted). The en banc Court in Cavera did not hold that a two-step appellate review is the only proper course. Thus, nothing in our existing sentencing law prevented the panel majority — or the en banc Court — from ruling on both the “procedural” and the “substantive” reasonableness of Stewart’s sentence in the course of this (first) appeal.

While a two-step appellate review may be “one proper course,” id., in some cases, it was not the proper course here. First, as I have noted, where an appellate court is presented with a lenient sentence that is such an extreme variance from the Guidelines in a case involving “ ‘extraordinarily severe criminal conduct,’ ” Stewart II, 590 F.3d at 148 (Maj. Op.) (quoting Sent’g Tr. 118), the appellate court can and should decide whether the sentence “shocks the conscience.”⁹ Stewart II, 590 F.3d at 163 (Walker, J., concurring in part and dissenting in part) (“The ‘informed intuition of the appellate panel’ has a place in appellate review.” (quoting Rigas, 583 F.3d at 123)); see also Gerard E. Lynch, Letting Guidelines Be Guidelines (and Judges Be Judges), Ohio St. J.Crim. L. Amici: Views from the Field, Jan. 2008, at 5 (“[A]ppellate review of reasonableness of sentences can play a valuable part in [the sentencing] process.... ”). The “shocks the conscience” standard is difficult to define, but of course courts apply it routinely, just as they also routinely apply, where necessary, the concepts of “abuse of discretion,” “arbitrary and capricious,” and “manifest injustice.” ¹⁰ An opportunity to clarify these standards was missed here.

Finally, even if the line between “procedural” and “substantive” errors can be drawn with precision in some cases, in this case any potential “procedural” errors so infected the sentencing as a whole that it is virtually impossible to separate “procedural” from “substantive” unreasonableness. As discussed in more detail below, the panel declined to decide whether the District Court committed several critical “procedural” errors in sentencing the defendant to 28-months’ imprisonment. To have addressed whether these were, in fact, “procedural” errors would have been part and parcel of determining whether the defendant’s 28-month sentence was “substantively” unreasonable.

B. The “Nature” and “Seriousness” of Stewart’s Offense

Notably, the panel majority also failed to decide whether the District Court erred proeedurally in assessing the “nature” and “seriousness” of Stewart’s material support offense under 18 U.S.C. § 3553(a)(1)-(2) when the District Court effectively disregarded the element of terrorism after recognizing the nature of her offense, sentencing Stewart as if her offense involved more benign criminal activity. In its original opinion, the panel majority said nothing whatsoever about this issue. See Appendix A. In its amended opinion, rather than decide the straightforward question of whether the District Court committed error in this respect, the panel majority describes hypothetical instances in which the District Court “might have” erred, conjecturing that if the District Court conducted its analysis in certain enumerated ways, “it would ... be error.” Stewart II, 590 F.3d at 151 n. 37 (Maj. Op.). See Appendix A. Incredibly, this additional text offers no conclusions of law and provides no answers, and it is particularly ironic in light of Judge Calabresi’s admonition against issuing advisory opinions. Id. at 152 (Calabresi, J., concurring). This affords no guidance to the District Court as to whether it can reasonably accord no weight whatsoever to the terrorism element of Stewart’s material support crime, much less whether a 28-month sentence falls within the range of reasonable sentences for criminal conduct as serious as Stewart’s. This issue could have been, and should have been, taken up by the en banc Court.

C. Lack of Actual Harm

The panel majority declined to decide whether it was error for the District Court to rely on the lack of actual harm resulting from Stewart’s actions as a basis for a downward variance. In its original opinion, the panel majority stated in a footnote: “As a procedural matter, we conclude that a district court may rely on the fact that no harm resulted from the criminal act at issue.” Stewart I, Slip Op. at 7604 n. 33 (Maj. Op.). It provided no further explanation and did not apply its conclusion to Stewart’s case or the District Court’s handling of the issue. The dissent, however, forcefully explained why lack of harm — ■ particularly when a consequence of vigilant law enforcement efforts — cannot bear mitigating weight in assessing the seriousness of a material support of terrorism crime.¹¹ See Stewart I, Slip Op. 7644 (Walker, J., concurring in part and dissenting in part); see also Stewart II, 590 F.3d at 163 (Walker, J., concurring in part and dissenting in part).

In its amended opinion, the panel majority takes even less of a stance on the issue, pushing the matter instead to the District Court — again without further instruction or legal analysis. See Appendix A. The panel majority claims only that it “make[s] no ruling on th[at] issue now[,] ... not[ing] simply that it is a serious issue to be given consideration by the district court upon reevaluating Stewart’s sentence.” Stewart II, 590 F.3d at 150 (Maj. Op.). Inasmuch as no additional fact-finding is needed or even requested, this approach to appellate decisionmaking — deciding to avoid deciding — was an inadequate response to the “serious issue” in question. The en banc Court should have stepped in and stepped up.

D. Abuse of Trust

As Judge Walker notes in his separate opinion, the panel majority “properly faults the district court for failing to ‘explain how and to what extent the sentence reflected the seriousness of the crimes of conviction in light of the fact that Stewart was ... a member of the bar when she committed them.’ ” Id. at 179 (Walker, J., concurring in part and dissenting in part) (quoting id. at 148 (Maj. Op.)). But as Judge Walker observes, this “does not go far enough.” Id. Rather, once again the panel majority identifies an important issue — “issue spotting,” in the parlance of law school exam settings — but fails to decide it, stating only (and tentatively), in a footnote, that the District Court “may address this issue on remand.” Id. at 151 n. 37 [sic] (Maj. Op.) (emphasis added). As the panel majority recognizes, “[t]he question ... remains whether, because [Stewart] was an experienced and dedicated lawyer acting as such when she broke the law in the manner that she did, her punishment should have been greater than it was.” Id. at 148. Yes, indeed, the question does “remain.” And it too could have been, and should have been, answered, if not by the panel then, by the en banc Court.

E. “Other Issues”

Finally, in addition to the numerous issues already flagged (but left undecided), the panel majority in its amended opinion inscrutably alludes to “other issues” raised in the separate opinions of Judges Walker and Calabresi (adding, quixotically, that the majority opinion’s “silence” on those issues does not “mean that the majority has adopted Judge Calabresi’s views or rejected Judge Walker’s”). Id. at 151. Once again, the message here to the District Court and to the public is that the silence of the panel majority is golden— and without significance. The existence of these “other issues”- — -whatever they may be — clearly suggest the need for an en banc review that unflinchingly would decide all of the issues presented by this ease as to the reasonableness of Stewart’s 28-month sentence.

❖ ‡ *

The only sentencing error squarely found by the panel majority was the District Court’s failure to consider Stewart’s alleged perjury on the stand. In identifying this single procedural error, the panel majority manages to remand this case for resentencing while leaving unresolved each of the serious issues, central to Stewart’s material support of terrorism, summarized above.

Maybe the District Court will understand the winks and nods conveyed by the panel majority in the direction of some of these serious errors. If not, maybe the government will appeal the District Court’s sentencing decision; and if so, maybe the panel will confront these issues squarely after Stewart has been resentenced. And maybe the en banc Court will have a chance to revisit them thereafter. Maybe.

Appendix A

The changes from the original opinion to the amended opinion demonstrate how the panel majority failed to address several important issues:

[[Image here]]

[[Image here]]

[[Image here]]

Appendix B

“In material support convictions after United States v. Booker, 543 U.S. 220, 245 (2005), district courts have generally imposed sentences of at least ten years per material suppoH count, with considerably higher total sentences.” Stewart II, 590 F.3d at 166 n. 4 (Walker, J., concurring in part and dissenting in part).

The following are the sentences that have been given for material support of terrorism (“MS”) following Booker:_

[[Image here]]

1 . Senior Circuit Judge John M. Walker Jr., the author of the panel's minority opinion concurring and dissenting in part, while not authorized to participate in the en banc poll, has endorsed the views expressed in this opinion.

2 . To be clear, active judges of this Court requested a poll on whether to rehear en banc the judgment of the panel only insofar as it addressed the sentence imposed on Stewart by the District Court. There was no request for a poll, much less a vote by the en banc Court, on the judgment insofar as it affirmed the conviction of Stewart or her co-defendants.

3 .Stewart was convicted of one count of conspiring to defraud the United States in violation of 18 U.S.C. § 371; one count of providing and concealing material support to terrorists for a conspiracy to murder persons in a foreign country in violation of 18 U.S.C. § 2339A and 18 U.S.C. § 2; one count of conspiracy to provide and conceal such support in violation of 18 U.S.C. § 371; and two counts of making false statements in violation of 18 U.S.C. § 1001.

4 . For comparison, the Sentencing Guidelines recommend a 28-month sentence for a relatively modest fraud, see U.S.S.G. § 2B1.1 (providing for base offense level of 18 and 27-33 month sentence for first-time offender in fraud involving less than $200,000), or drug transaction, see id. § 2D 1.1 (providing for same base offense level and sentencing range for first-time offender in drug crime involving less than 40 grams of heroin or less than 200 grams of cocaine). But the Guidelines signal that any crime promoting terrorism is to be viewed as extremely serious by providing for a minimum base offense level of 32. See id. § 3A1.4(a). At the same time, the strong need to deter terrorism is evident from the Guidelines recommendation that a terrorism defendant be accorded a criminal history of VI, the highest level possible, without regard to his actual criminal record. See id. § 3A1.4(b).

5 . Of course it is common for an appellate court to remand issues to the district court for further consideration before the appellate court speaks to the issue. This case, however, did not call for such action. The issues in this case were serious, and in most instances had already been addressed by the District Court — it was for our Court to decide either that the expressed views of the District Court were aligned with the law of our Circuit or that they were error.

Punting in a case like this is not "judicial restraint.” Judicial restraint is the refusal to reach out to decide issues that the case does not present. To refuse to decide issues that are squarely presented is an abdication of judicial responsibility.

6 . In addition to joining the panel decision to avoid answering numerous important questions of sentencing law, and remanding to the District Court for "further consideration” of those questions — many already clearly addressed in the District Court's conclusions at sentencing — one member of the panel majority has written a concurring opinion that offers suggestions to the District Court as to how it might approach these and other open questions on remand. The concurring opinion notes that, despite the panel’s rejection of Stewart’s claim of selective prosecution, Stewart II, 590 F.3d at 161-62 (Maj. Op), the District Court may still consider the possible relevance to Stewart’s sentence of the "disparities between [Stewart] and other individuals who were not charged at all.” Stewart II, 590 F.3d at 161-62 (Calabresi, J., concurring). Specifically, it is suggested that the District Court consider the fact that the U.S. Attorney did not seek an indictment of Ramsey Clark for the allegedly similar conduct of "issuing a statement to the media on behalf of [Sheik] Rahman.” Id. at 159.

I will not expand on Judge Walker’s sensible observations as to how and why there is no comparison between Stewart’s conduct and that of Clark. See Stewart II, 590 F.3d at 163-68 (Walker, J., concurring in part and dissenting in part). I note simply that the invitation for further inquiry on this point comes without regard for the separation of powers that confers prosecutorial authority exclusively on the executive branch and narrowly limits judicial inquiry into the exercise of that authority. See, e.g., Harlan Assocs. v. Inc. Vill. of Mineóla, 273 F.3d 494, 499 (2d Cir.2001) (limiting selective prosecution challenge to prosecution's reliance on invidious factors or attempts to hamper exercise of constitutional right).

More troubling still, the suggestion fails to explain how the public would be well served by remedying the executive’s failure to prosecute a second person who may have committed a crime by having the judiciary mitigate the sentence of a person whose commission of the crime was proved beyond a reasonable doubt. If any comparison should be done in this case by the District Court (and, for that matter, the general public), I respectfully submit that it should be to compare Stewart’s sentence with the array of criminal defendants actually charged and convicted of material support to terrorism. See Appendix B.

7 . Judge Raggi and I do not need to be reminded that Gall v. United States, 552 U.S. 38, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007), and Kimbrough v. United States, 552 U.S. 85, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007), teach deference to district courts' broad discretion in sentencing, see, e.g., United States v. Jones, 531 F.3d 163, 170-74 (2d Cir.2008), although we note that this Court had to convene en banc to make that lesson the law of the Circuit, see Cavera, 550 F.3d 180. As Jones recognized and Cavera reaffirmed, however, Gall and Kimbrough do not "grant district courts 'a blank check to impose whatever sentences suit their fancy.’" Cavern, 550 F.3d at 191 (quoting Jones, 531 F.3d at 174). "At the substantive stage of reasonableness review, an appellate court may consider whether a factor relied on by a sentencing court can bear the weight assigned to it ... under the totality of circumstances in the case.” Cavera, 550 F.3d at 191. On such review, "we do not consider what weight we would ourselves have given a particular factor.” Id. We will reverse as substantively unreasonable only those "outlier” sentences demonstrating such "actual abuse of a district court's considerable sentencing discretion,” see Jones, 531 F.3d at 174, as to "shock the conscience,” see Rigas, 583 F.3d at 123. Not surprisingly, we rarely see such cases. This, however, is one of them.

8 . The Supreme Court, in Rita v. United States, 551 U.S. 338, 362 n. 2, 127 S.Ct. 2456, 168 L.Ed.2d 203 (2007), reviewed for "reasonableness” of the sentence as a whole, without distinguishing "substantive” from "procedural” errors. Other Circuit courts have noted the distinction is a matter of labels, rather than one of the merits of the error. See, e.g., United States v. Engle, 592 F.3d 495, 500-01 n. 1 (4th Cir.2010) (noting that even though the government described its arguments as going to the substantive reasonableness of the sentence and, in the Court’s view, those arguments challenged procedural reasonableness, the Court would, "of course, consider the government’s arguments on their merits, without regard to whether the government attached the correct descriptive label to those arguments”).

The attempt to rigidly separate "substantive” and "procedural” questions is as old as the common law, and one long ago recognized as well-nigh impossible. See Guaranty Trust Co. ofN.Y. v. York, 326 U.S. 99, 109, 65 S.Ct. 1464, 89 L.Ed. 2079 (1945) ("Matters of 'substance' and matters of 'procedure' are much talked about in the books as though they defined a great divide cutting across the whole domain of law. But, of course, ‘substance’ and procedure’ are the same keywords to very different problems.... Each implies different variables depending upon the particular problem for which it is used.”); id. (noting that, in that case, it was "immaterial whether statutes of limitation are characterized as ‘substantive’ or ‘procedural’ ”); Hanna v. Plumer, 380 U.S. 460, 472, 85 S.Ct. 1136, 14 L.Ed.2d 8 (1965) (acknowledging that some matters are capable of "falling within the uncertain area between substance and procedure” and "are rationally capable of classification as either”). Compare Rita v. United States, 551 U.S. 338, 127 S.Ct. 2456, 2473, 168 L.Ed.2d 203 (2007) (Stevens, J., concurring) (noting that a sentencing court’s explanation for issuing a sentence based on its dislike for Yankees fans would go to "substantive” reasonableness), with id. at 2483 n. 6 (Scalia, J., concurring in part and concurring in the judgment) (contending that precisely the same explanation for a sentence would be a matter of "procedural” reasonableness).

9 . I note, on this point, that in sentencing defendants for material support of terrorism after United States v. Booker, 543 U.S. 220, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005), district courts have generally imposed sentences of at least 120 months per material support count, with considerably higher total sentences than Stewart’s 28 months. See Appendix B.

10 . In Rigas, 583 F.3d at 123, we reviewed the sentences imposed by the district court using the standard of whether they would constitute a "manifest injustice,” "shock the conscience,” or "otherwise compel a conclusion that they are substantively unreasonable.” We also evaluated whether the district court’s decision "[could] be located within the range of permissible decisions.” Id.; see also Travellers Int'l, A.G. v. Trans World Airlines, Inc., 41 F.3d 1570, 1574 (2d Cir.1994) (quoting United States v. U.S. Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 92 L.Ed. 746 (1948), for the less pithy standard of whether an appellate court is "left with the definite and firm conviction that a mistake has been committed” in holding a finding by the district court to be "clearly erroneous”).

11 . While we do not "categorically proscribe” a sentencing court from considering the full range of facts relevant to a defendant and the crime of conviction in deciding on an appropriate sentence, see Cavera, 550 F.3d at 191 (excepting only invidious facts), we will consider whether a fact "can bear the weight assigned to it under the totality of the circumstances in the case,” id. In this case, the District Court and the general public are entitled to a clear decision from our Court as to whether or not the District Court erred— either procedurally or substantively — in assigning heavy mitigating weight to the lack of harm resulting from Stewart’s material support of terrorism.

Case Details

Case Name: United States v. Stewart

Court Name: Court of Appeals for the Second Circuit

Date Published: Feb 23, 2010

Citations: 597 F.3d 514; 2010 WL 610232; 2010 U.S. App. LEXIS 3772; 06-5015-cr (L), 06-5031-cr (con), 06-5093-cr (con), 06-5131-cr (con), 06-5135-cr (con), 06-5143-cr (con)

Docket Number: 06-5015-cr (L), 06-5031-cr (con), 06-5093-cr (con), 06-5131-cr (con), 06-5135-cr (con), 06-5143-cr (con)

Court Abbreviation: 2d Cir.