In
Descamps, Justice Kagan wrote for the Supreme Court on an issue that permeates our work as federal defenders: what impact do prior convictions have on our clients’ freedom? Although written in the context of the Armed Career Criminal Act (ACCA), the Court’s explication of the “modified categorical approach” is potentially relevant to every firearm, drug, illegal reentry, and sex crime case we handle. Trying to understand the ins and outs of the “modified categorical approach” reminds many of us why we are criminal defense and not tax lawyers. But for our clients, advocacy in this area of the law may be the most important thing we can do for them. So we’ve resorted to our resident advice columnist to try to cut through the denseness of this area of the law. We hope this will help defenders spot the issues that must be researched and litigated to benefit our clients.
Dear Defender:
I’ve heard this phrase plenty of times before but was afraid to ask. What is the “modified categorical approach” and what is it modifying?
I’m Perplexed
Dear Perplexed:
Many statutes and Guidelines create legal consequences depending on whether prior convictions fall into certain classes, such as “violent felony”, “crime of violence”, “serious drug offense,” and “drug trafficking offense.” The statutes and Guidelines have definitional sections that list crimes that fall within those classes, such as “burglary,” “aggravated assault,” and “an offense under the Controlled Substances Act” with a maximum punishment of ten years or more. In the seminal case of
Taylor, the Supreme Court held that the ACCA required a federal definition for generic predicate offenses, such as – in that case – “burglary.” Under
Taylor’s “formal categorical approach,” the sentencing judge simply compares “the elements of the statute forming the basis of the defendant’s conviction with the elements of the ‘generic’ crime—i.e., the offense as commonly understood.”
In
Taylor, the Court recognized that some statutes had alternative crimes within the same statute. In this “narrow range of cases,” the sentencing court could examine court documents – the charging instrument and the jury instructions – to determine if the state conviction was for the branch of the relevant crime that matched the generic federal definition of the predicate offense for a “violent felony” under the ACCA.
So the “modified” of the “formal categorical approach” involves analysis of documents beyond simple comparison of the statute of conviction and the generic elements of the predicate offense. You can have statutes that simply list elements (A + B + C = crime) or that have alternate ways of fulfilling the elements (A + B + C or D or E = crime). The categorical approach is modified only when the elements of the generic description of the predicate offense do not match up with the elements of the statute of the prior conviction. If there is an alternative offense within the statute of the prior conviction that matches the generic definition for the predicate crime, then court documents can be examined to identify which offense within the statute formed the basis for the conviction.
Dear Defender:
Now you’ve given me both a basic understanding of the “modified categorical approach” and a splitting headache. Do I really need to know this stuff for all of my cases. Won’t the probation officer figure out what is a predicate for enhancements? And what does it matter if the defendant’s conduct matched the predicate crime regardless of the technical details of the conviction?
Looking For A Shortcut
Dear Shortcut:
The short answers are “Yes”, “No”, and “Are you kidding me?”
We need to be fully on top of this area of the law because the courts have adopted the categorical approach to analyze statutes and Guidelines that potentially cover most federal cases. Think of all the statutes that involve classifications of prior convictions in addition to the Armed Career Criminal Act: drug trafficking priors that increase exposure under 21 U.S.C. § 841(b)(1); sex prior convictions under statutes such as 18 U.S.C. §§ 2251(e), 2252(b), and 2252A(b); the predicate offense for firearms mandatory minimums under 18 U.S.C. § 924(c); the triggering increase in statutory maximums for illegal reentry under 8 U.S.C. § 1326(b).
The modified categorical approach can even apply in figuring out whether a felony occurred at all. Under 18 U.S.C. § 922(g)(9), the felony gun possession statute requires proof that the defendant was previously convicted “of a misdemeanor crime of domestic violence.” Under
Descamps, the district court should be barred from looking beyond the elements of a simple assault statute to find that the conviction factually constituted a domestic violence misdemeanor (although the relationship can be established at trial beyond a reasonable doubt as described in
Hayes). And don’t forget that, in illegal reentry cases, if the immigration judge improperly treated a prior conviction as an “aggravated felony”, you may be able to move to dismiss based on a collateral attack on the removal order under 8 U.S.C. § 1326(d).
Under the Guidelines, the classification of prior convictions within the statutory maximum frequently increases the total offense level (in addition to counting as criminal history points regardless of the classification). For example, the six- and ten-level increases under the Guideline for being a felon in possession of a firearm depend on determinations whether a prior conviction is classified as a “crime of violence” or a “controlled substance offense” under U.S.S.G. § 2K2.1. The same classifications apply to career offender enhancements under U.S.S.G. § 4B1.2, which automatically increase the sentencing range to close to the statutory maximum by increasing the Criminal History Category to VI and adding levels to the total offense level. Every illegal reentry case involves potential drastic increases for prior convictions for an “aggravated felony” and a “drug trafficking offense” under U.S.S.G. § 2L2.1.
And you can’t depend on the presentence report writer to spot these issues. While there are some probation officers who will properly analyze these questions, the application of the categorical approach requires legal advocacy that it is unreasonable to expect of probation officers. The classification of prior offenses may be the single biggest determinant of the punishment your client faces. This is emphatically a job for defense counsel.
On the last point: the categorical approach sometimes can feel counter-intuitive. The facts may seem to place your client in the category even if the conviction is not within the legal definition. This is part of the reason aggressive advocacy in this area is so necessary. At sentencing, our clients already face a system rigged for over-incarceration based on prior convictions that are double-counted and that trigger irrationally long sentences. Justice Kagan pointed out that, by limiting enhancements under the categorical approach, we are protecting our clients’ Sixth Amendment rights, defending against unfair use of facts not formally litigated, and assuring that our clients receive the benefit of any plea bargain in the prior case. We are betraying our duty of zealous representation if we fail to advocate against enhancements using all the authority at our disposal.
Dear Defender:
So if the Court basically is just sticking by Taylor on an issue that covers many sentencing situations, why did the Court bother taking the case?
Supreme Flashback
Dear Flashback:
The Ninth Circuit in a split en banc opinion concluded that sentencing judges could use the modified categorical approach to find a “missing element” where an indivisible statute – a statute not divisible into different elements to state different offenses – described a non-generic predicate offense. In
Aguila-Montes de Oca, the court said the California burglary statute, which was too broad to constitute generic burglary within the federal definition, could be treated as a “crime of violence” under § 1326 by examining statements made during court proceedings to narrow the statute to the generic definition of burglary; that is, although the statute did not require unlawful entry, the sentencing court could examine other material to discover whether the defendant broke in, thereby providing the otherwise missing element. The Ninth Circuit found that the modified categorical approach did not only apply to divisible statutes, expanding greatly the circumstances in which the sentencing judge could supply a missing element to create a predicate offense. The debate between the majority and the concurring judges in
Aguila-Montes de Oca (all agreed that the generic offense was not established) centered on whether the Ninth Circuit was straying from
Taylor and its progeny by allowing a statute that is indivisible and “categorically broader than generic burglary” to be treated as generic burglary based on application of the modified categorical approach. The Supreme Court granted certiorari to resolve the conflict among the Circuits (Sixth and Ninth versus First and Second) on applying the modified categorical approach to indivisible statutes.
Dear Defender:
Justice Kagan sounded pretty exasperated with the Ninth Circuit’s position. What gives? And she spends a lot of time talking about divisible and indivisible statutes. What’s the difference? And isn’t there a body of case law and resources that have given us nice lists of what is and is not a divisible statute?
Post-Reversal Stress Disordered
Dear Disordered:
In
Descamps, the Supreme Court abrogated the Ninth Circuit’s en banc opinion in
Aguila-Montes de Oca, holding that “sentencing courts may not apply the modified categorical approach when the crime of which the defendant was convicted has a single, indivisible set of elements.” In applying the modified categorical approach to indivisible and overbroad statutes, Judge Bybee’s opinion for the Ninth Circuit suggested several times that the Supreme Court lacked clarity in
Taylor and its progeny on how the modified categorical approach operates. In response, Judge Kagan walked us through
Taylor to
Shepard to
Nijhawan to
Johnson showing that the Supreme Court has always limited application of the modified categorical approach to divisible statutes. Justice Kagan adopted Judge Berzon’s critique of the
Aguila-Montes de Oca majority as using a “modified factual” approach, which Justice Kagan said “turns an elements-based inquiry into an evidence-based one.”
And you’re right that the tone was a bit sharp. Justice Kagan’s opinions have generally been notable for her pleasant, straightforward, and conversational voice. But there is an edge to that voice in
Descamps: Justice Kagan is pretty harsh in her treatment of the Ninth Circuit, which dismissed "everything we have said on the subject as ‘lack[ing] conclusive weight.'" “Yet again, the Ninth Circuit’s ruling flouts our reasoning—here, by extending judicial factfinding beyond the recognition of a prior conviction.” This tone should embolden us in challenging Ninth Circuit precedent in this area. Judge Bybee’s opinion in
Aguila-Montes de Oca included footnote 3, which listed a number of prior Ninth Circuit cases that supported the now-reversed majority approach. That means we have to examine all Ninth Circuit precedent that is adverse to our clients to determine whether
Descamps undermines the reasoning in those cases.
So what exactly is a divisible statute? Justice Kagan says the categorical analysis can be applied to a divisible statute, “listing potential offense elements in the alternative,” which can “render opaque which element played a part in the defendant’s conviction.” Judge Berzon uses “divisible statute” as shorthand “to refer to a statute that lists alternative ways that one or more elements can be established.” Both Justice Kagan and Judge Berzon emphasize that the modified categorical approach can only be used to determine under which express statutory alternative the defendant was convicted. As Justice Kagan spells out in footnote 2, the statute must cover different crimes, “not several different methods of committing one offense.”
But identifying a divisible statute and distinguishing it from one that is simply overbroad is a tricky business. As Justice Kagan stated, “A sentencing court, to be sure, can hypothetically reconceive [an overbroad statute that can be violated in multiple ways] in divisible terms.” Fortunately, the Supreme Court discouraged metaphysical manipulation by “a court blessed with sufficient time and imagination” to supply the missing element. For example, a statute forbidding assault with a “weapon” could not be morphed into a statute prohibiting assault with a gun by supplying the means of violating the statute as by a “firearm.” “Whatever the underlying facts or the evidence presented, the defendant still would not have been convicted, in the deliberate and considered way the Constitution guarantees, of an offense with the same (or narrower) elements as the supposed generic crime (assault with a gun).”
In light of
Descamps, we will need to look at many state statutes anew to figure out if they are indivisible. Or as we would hear from Steve Hubachek, former star AFPD and current CJA counsel, every Ninth Circuit case in this area should be considered presumptively invalid. The cases are often tainted with reasoning that deviates from the methodology required by
Descamps. For example, are the California drug statutes indivisible and, therefore, do they fail to match federal generic trafficking crimes, both because substances are proscribed that are not listed in the federal drug statute and because the state drug statute includes means of committing the offense not included in the Controlled Substances Act? Or how about assaults that can be both reckless and intentional? We’ll need to take a fresh look at many statutes that may already have been litigated pre-
Descamps and advocate against their applicability as enhancers. And given the inevitable risks of litigation, how do we assess the value of plea agreements – especially fast-track offers – in light of adverse Ninth Circuit authority that applies the modified categorical approach in a manner that appears to run contrary to
Descamps?
Which brings us to the useful training material available to help us with applying the modified categorical approach. We have a number of great resources: the indefatigable lawyers at the National Immigration Project, who use the same modified categorical approach in applying immigration laws to their alien clients, have posted this
Practice Advisory of how
Descamps is likely to apply, the Sentencing Resource Counsel has this
pre-Descamps summary, and the training branch has a number of resources under “predicate convictions” on
this website. But at least in the Ninth Circuit, we need to be taking a steely-eyed look at all precedent to determine whether the opinions are part of the Ninth Circuit’s precursors to the
Aguila-Montes de Oca deviation. In other words, we can’t assume that the Ninth Circuit’s previous treatment of a particular statute, without our reconsideration in light of
Descamps, is still valid.
Dear Defender:
Does any of Descamps matter in the Ninth Circuit? Justice Kagan said that, because the government waived the argument, the Supreme Court was expressing no opinion on United States v. Mayer, the case in which the Ninth Circuit held that Oregon’s overbroad and indivisible first degree burglary statute fit into the ACCA’s residual clause as a conviction that “otherwise involves conduct that presents a serious risk of physical injury to another.” What should we do about California and Oregon overbroad burglary statutes?
This Can’t Be Right
Dear Mr. Right:
In light of the
Descamps reasoning,
Mayer should be a dead letter. In footnote 6, Justice Kagan, while expressing no opinion on the issue, quoted Chief Judge Kozinski’s dissent from denial of rehearing en banc in
Mayer: "
Compare United States v. Mayer, 560 F. 3d 948, 960–963 (CA9 2009) (holding that Oregon’s burglary statute falls within the residual clause, even though it does not include all of generic burglary’s elements),
with id., at 951 (Kozinski, C. J., dissenting from denial of rehearing en banc) (arguing that the panel opinion ‘is a train wreck in the making’).” Given the mode of analysis required by
Descamps, the
Mayer panel opinion should no longer be viewed as binding – it is in fact the “train wreck” foreseen by the Chief Judge. Under the Ninth Circuit’s rules on precedent (
Miller v. Gammie), a panel decision need no longer be followed where intervening authority from the Supreme Court undermines the panel’s rationale. Unlike the recent
Green case on restitution and
Apprendi, the
Mayer opinion on the ACCA’s residual clause is irreconcilable with the reasoning of
Descamps.
In
Mayer, the Ninth Circuit injected factual questions regarding prosecutorial practices and potential dangerous scenarios that, in the words of Justice Kagan, “subvert” the formal categorical approach and its rationale, which apply to each part of the ACCA including the residual clause. The
Mayer court’s reliance on
James is a mismatch.
James noted "the proper inquiry is whether the conduct encompassed by the elements of the offense, in the ordinary case, presents a serious potential risk of injury to another." The Court stated that, for an offense not to qualify as a predicate, there must be a "realistic probability, not a theoretical possibility," that the statute covers conduct that does not present a serious potential risk of physical injury. In
Mayer, the defendant showed the breadth of the statute by "pointing to . . . other cases in which the state courts did in fact apply the statute" in such a manner, as stated by the Supreme Court in
Gonzales v. Duenas-Alvarez. Even worse, when reviewing the list of burglary tools, the
Mayer court claimed they were “by definition a very dangerous object”, even though any tool “adapted or designed” to facilitate a forcible entry – a skeleton key? a lock-pick? a jimmy? – is included in the means of committing the offense.
We should be advocating in the district courts and on appeal against the precedential value of
Mayer based on the intervening controlling authority of
Descamps. The type of factual inquiry and mini-trials on implementation of statutes under
Mayer are contrary to the
Descamps rationale, based as it is on 1) the text and history of the ACCA’s focus on uniformity and predictability; 2) avoidance of Sixth Amendment concerns regarding sentencing courts making findings of fact that properly belong to juries; and 3) averting "the practical difficulties and potential unfairness of a factual approach." The listing of the particular crime should impliedly exclude the non-generic offense from the residual clause.
If we get no satisfaction in the district court and direct appeal, we should do all we can to obtain review en banc or by a writ of certiorari. If you get such a case, we should be able to organize amicus support for review of this important question upon which the Ninth Circuit should be in full harmony with the Supreme Court’s opinion in
Descamps. And don’t forget that, in any case involving the residual clause of the ACCA, we should also be following Justice Scalia’s lead in his
Sykes and
James dissents, asserting that the residual clause is void for vagueness, as the Sentencing Resource Counsel has outlined for us
at 8-11.
Dear Defender:
If I’m dealing with a divisible statute, and applying the modified categorical approach, does Descamps give us anything to work with on the types of documents the government can use? And where does the Sixth Amendment come into play? Can we at least use Descamps to argue the doctrine of constitutional avoidance?
Believing Apprendi Rules
Dear BAR:
Under the doctrine of constitutional avoidance, the courts construe statutes to avoid serious constitutional problems. In
Descamps, Justice Kagan super-charged the
Shepard plurality’s constitutional concerns regarding consideration of any facts to prove a statutory predicate conviction beyond those established in compliance with the Sixth Amendment.
Shepard expanded the
Taylor trial documents that can be used to identify which part of a divisible statute underlies a conviction to include judicially cognizable facts during guilty pleas. Justice Kagan cited to
Shepard as barring use of extrinsic material for any purpose other than “merely identifying a prior conviction,” because any other use “would (at least) raise serious Sixth Amendment concerns.” The indictment and plea colloquy can be used to determine “which statutory phrase was the basis for the conviction,” but not to establish what the defendant actually did. Among the practical and constitutional reasons for this limitation, Justice Kagan pointed out: “And during plea hearings, the defendant may not wish to irk the prosecutor or court by squabbling about superfluous factual allegations.”
Justice Kagan’s elaboration of the Sixth Amendment limitations on the use of prior convictions should be very useful. To avoid constitutional problems, the sentencing judge should consider only "a restricted set of materials" and only for the very limited purpose of discerning the statute of conviction. The Court quoted
Shepard in stating that Sixth Amendment concerns “counsel against allowing a sentencing court to ‘make a disputed' determination ‘about what the defendant and state judge must have understood as the factual basis of the prior plea,' or what the jury in a prior trial must have accepted as the theory of the crime." So we should object at sentencing to use of police reports and other extraneous material beyond the very narrow category of documents that do not implicate constitutional concerns.
Especially after reaffirming the
Apprendi line of authority regarding fines in
Southern Union and reversing
Harris on mandatory minimums in
Alleyne, the Supreme Court has demonstrated that Sixth Amendment rights at sentencing must be taken seriously. Here’s Justice Kagan spelling out the clear lines on Sixth Amendment protections:
"And there’s the constitutional rub. The Sixth Amendment contemplates that a jury—not a sentencing court—will find such facts, unanimously and beyond a reasonable doubt. And the only facts the court can be sure the jury so found are those constituting elements of the offense—as distinct from amplifying but legally extraneous circumstances. .. . Similarly, as
Shepard indicated, when a defendant pleads guilty to a crime, he waives his right to a jury determination of only that offense’s elements; whatever he says, or fails to say, about superfluous facts cannot license a later sentencing court to impose extra punishment. . . . .So when the District Court here enhanced Descamps’ sentence, based on his supposed acquiescence to a prosecutorial statement (that he 'broke and entered') irrelevant to the crime charged, the court did just what we have said it cannot: rely on its own finding about a non-elemental fact to increase a defendant’s maximum sentence."
This full-throated endorsement of the Sixth Amendment concerns should be useful in layering constitutional arguments with construction of rules or statutes to avoid the serious constitutional problems in this area. These arguments – including pleading and proof regarding the classification of prior convictions under 8 U.S.C. § 1326 and the ACCA – should be re-energized by the Court’s solicitude toward constitutional rights at sentencing. As Judge Thomas’s concurrence reminds us, “Under the logic of
Apprendi, a court may not find facts about a prior conviction when such findings increase the statutory maximum. This is so whether a court is determining whether a prior conviction was entered,.....or attempting to discern what facts were necessary to a prior conviction.” Justice Thomas continues to call for the Court to reconsider
Almendarez-Torres, the outlier case in Sixth Amendment protections creating an exception for the fact of a prior conviction (although it was really a Fifth Amendment right to grand jury indictment case).
Dear Defender:
As I read about Descamps, I’m getting the queasy feeling that some of my prior clients got screwed under the defective Ninth Circuit standard. If I have someone with a prior overbroad conviction that was treated as a statutory or Guidelines predicate for enhancement, what can I do?
Knot 2 L8
Dear Knot:
We need to scour our closed files for cases in which our clients were hurt by the Ninth Circuit’s approval of overbroad statutes to over-incarcerate our clients. The mechanisms for getting back in court will depend on the specifics of the client’s procedural posture. The most important thing is that we have to bring the case before the court in order to provide the judge with the opportunity to do the right thing.
The key point to remember comes from that queasy feeling: when our clients suffer the massive injustice of a sentence enhanced by a previously-tolerated legal error, the equities require redress. As we review the mechanisms for getting back in front of a judge, remember how the equities of our clients’ over-incarceration should drive the construction of the statutes. The most direct route would appear to be a section 2255 motion perhaps layered with alternative invocation of the court’s ancillary jurisdiction, a writ of error coram nobis, and the general habeas corpus statute, 28 U.S.C. § 2241.
Under 28 U.S.C. § 2255(f), the newly recognized rights and the retroactivity issues should allow for filing for relief within one year of
Descamps. The Supreme Court in
Bousley and
Fiore noted that retroactivity doctrine does not apply where the crime – in light of new controlling authority – did not happen. For example, a person with a simple assault who was treated under § 922(g) as having a domestic violence misdemeanor without trial proof or an admission regarding the relationship, or a person treated as an armed career criminal even though the necessary predicates were lacking, should be able to get back in front of the judge based on the claim that
Descamps establishes innocence of the statutory offense. The same logic should apply to career offenders whose radically increased sentences have their roots in 28 U.S.C. § 994(h), as well as other substantial Guidelines enhancements.
For all cases, the first step is to negotiate with the prosecutor. There should be no reason not to take a second look at a person whose sentence was unlawfully increased based on a legal error now identified and corrected by the Supreme Court. Ethical prosecutors may be willing to waive procedural obstacles and simplify the trip back to the sentencing judge. If not, litigation should ensue. Even after the one-year statute of limitations from the
Descamps decision, in
Dretke v. Haley, the Court left open the application of the
Schlup actual innocence gateway past procedural defaults where the defendant received a sentence erroneously enhanced under a recidivist statute.
The Supreme Court has recently given us reason to believe sentencing judges will recognize the manifest injustice of sentences greatly increased by legal error in the Guidelines computation:
• In
Peugh, the Court recognized the seriousness of the Guidelines calculations as being sufficiently influential that the ex post facto clause applies to Guidelines amendments;
• In
Lafler, the Court noted that, for prejudice purposes under the Sixth Amendment right to counsel, “any amount of [additional] jail time has Sixth Amendment significance.”
• In
Alleyne, the Court emphasized the importance of Sixth Amendment jury trial protections that have a practical effect on the judge’s sentencing options in applying
Apprendi to mandatory minimum sentences.
This is not to say that there are not procedural obstacles that we may face in any case. But the important take-away is that your client's unjust sentence cannot be corrected unless you bring the challenge to the court’s attention.
Dear Defender:
Really? A month after the case comes down, you think you know how the courts will apply it? Isn’t this all just blind speculation?
Skeptical
Dear Skeptical:
You’re right....we don’t know how new Supreme Court cases will be applied. Some cases are announced as game-changers then fizzle, others creep into the limelight over the years. We need to study the
Descamps opinion and mine it for all it is worth because of the very strong likelihood that familiarity with these issues will be necessary at some point in the decision tree of many of our cases: Should we file a motion to dismiss? Is the plea offer meaningful? What punishment is realistically likely if you go to trial? What punishment is realistically likely upon a plea of guilty? Should the government be offering a concession to avoid the risk of losing a pretrial or sentencing argument?
In making these decisions, we should be tapping all resources, especially our fellow defenders who have been wrestling with these issues and similar ones for years. Don’t be shy about calling for help, as I did in consulting with such litigators as Brother Hubachek, AFPD Craig Weinerman from Eugene, and AFPD Davina Chen and the Los Angeles brain trust. These answers are only intended as starting points and as encouragement to pursue all the potential paths
Descamps provides to help our clients avoid over-incarceration.
Steve Sady, Chief Deputy Federal Public Defender, Portland, Oregon
