Enhancement provisions in various federal criminal statutes and the sentencing guidelines have become ubiquitous over the last 20 years. Among the more important examples are the Armed Career Criminal Act in 18 U.S.C. § 924(e); the career offender guideline in § 4B1.1 of the sentencing guidelines, which is also incorporated by reference into the base offense level provisions of the firearms guideline; the prior drug conviction enhancement provisions in 21 U.S.C. § 841; enhancements for prior convictions in the child pornography and sex offender statutes; and the “aggravated felony” enhancements in the illegal reentry statute and guideline.

The question arose early on about how to determine whether a prior state conviction qualified under the definition in federal enhancement statutes of, to give just some examples, “violent felony” or “crime of violence,” “aggravated felony,” or “controlled substance offense.” Does a sentencing court look only at the required elements of the state offense of which the defendant was convicted in the prior case? Or can the court go beyond the statute of conviction and delve into what the defendant “really did” by looking at police reports, receiving affidavits, hearing witnesses, and such? (I use quotation marks around “really did” because there of course might be a dispute about what the defendant “really did.”)

As those who practice regularly in federal court know, the Supreme Court decided in 1990 – in Taylor v. United States, 495 U.S. 575 (1990) – that it is the first of these which is the proper inquiry, adopting what the Court called a “categorical approach.” The specific enhancement statute at issue in Taylor was the Armed Career Criminal Act, but this “categorical approach” has been extended to most (though not all) other enhancement statutes.

Under this categorical approach, what matters is what the defendant was convicted of, not what he “really did.” In applying this test, a sentencing court must compare the elements of the statute under which the defendant was convicted with the definition in the federal enhancement statute, and the court must find the prior conviction does not qualify as an enhancement conviction if there is some possible conduct within the prior conviction statute that does not come within the definition in the federal enhancement statute. Put another way, one compares all possible conduct covered by the prior conviction statute to all possible conduct covered by the federal enhancement statute.

So far, so good. (Or at least better than otherwise.) Unfortunately, Taylor recognized an exception for what it termed “a narrow range of cases where a jury was actually required to find all the elements” of the federal enhancement offense. The Court indicated that this could be established by looking to court records such as jury instructions and the charging document. The courts of appeals, and ultimately the Supreme Court inShepard v. United States, 544 U.S. 13 (2005), then went on to extend this exception – which we have come to know as the “modified categorical approach” – to convictions based on guilty pleas. In that context, courts may look at court records such as a transcript of the plea colloquy, a written plea agreement, and/or other “record[s] of comparable findings of fact adopted by the defendant upon entering the plea,” Shepard, 544 U.S. at 21, but not look at documents such as police reports and presentence reports.

So now to the subject of this post, with an apology for the overly extensive introductory discussion. When this “modified categorical approach” first got recognized, courts – and, unfortunately, most attorneys – just assumed it applied to any prior conviction statute and to any facts the permissible court records might show were admitted at the time of the prior plea. But there’s a very strong argument that it shouldn’t be this way. As articulated by Chief Judge Kozinski of the Ninth Circuit in a concurring opinion in Li v. Ashcroft, 389 F.3d 892 (9th Cir. 2004):

[T]his [modified categorical] approach only works when the difference between the [prior] crime of conviction and the generic crime [in the federal enhancement statute] lies in the fact that particular elements in the former are broader than their counterparts in the latter. By contrast, when the crime of conviction is broader because it is missing an element of the generic crime altogether, we can never find that “a jury was actually required to find all the elements of” the generic crime.

Id. at 899 (Kozinski, J., concurring). As put in a later Ninth Circuit en banc opinion adopting Judge Kozinski’s concurring opinion as the law of the circuit – Navarro-Lopez v. Gonzales, 503 F.3d 1063 (9th Cir. 2007) (en banc) – the modified categorical approach is proper only “when a statute is divisible into several crimes.” Id. at 1073.

But here comes the problem: the law of the circuit got changed just four years later. (You say you thought there was this thing called stare decisis?) In last year’s opinion in United States v. Aguila Montes de Oca, 655 F.3d 915 (9th Cir. 2011) (en banc), the Ninth Circuit overruled Navarro-Lopez. The court held, over a strong dissent by Judge Berzon (technically a concurrence in the judgment, but a dissent on this point), that the modified categorical approach can be applied not just to what the court labeled “divisible statutes” but also to what the court labeled “missing element” statutes. In other words, the modified categorical approach can be applied to a prior conviction underany type of statute. So we’ve lost that issue.

For now, that is. Because where there’s problems, there’s solutions. And there’s a potential solution to this problem down the road, which is the reason for this post. It’s called the Supreme Court. Not our friends, you say? Well, on this stuff, they’ve actually given us as many, or more, good decisions as bad ones. Even the majority opinion in Aguila Montes de Ocaacknowledges there’s a split in the circuits on this issue (“a bit of a jumble,” it says), and Judge Berzon argues in her wonderful opinion that every single circuit to address the issue after two more recent, clarifying Supreme Court cases goes the other way. One fine cert petition written by Assistant Federal Public Defender Paresh Patel in Maryland, which is linked here nicely characterizes the circuits’ cases as “mass confusion.”

Given this “mass confusion,” we shouldn’t give up on the issue, and we should keep preserving it in the lower courts. It seems there’s a better chance than on most issues for Supreme Court review down the road, and, as noted in the paragraph above, the Court has been with us in this prior conviction enhancement area as much, if not more, than it’s been against us. Taylor andShepard are two examples of favorable opinions and others can be found in Leocal v. Ashcroft, 543 U.S. 1 (2004); Begay v. United States, 553 U.S. 137 (2008); Chambers v. United States, 555 U.S. 122 (2009); and Johnson v. United States, 130 S. Ct. 1265 (2010). You’ll also note from the recency of those case cites that the Court has been willing to look at these issues pretty frequently.

So don’t give up on this. We’ve got a much better shot than the frog in the cartoon linked here.